LBNL 46990

ERNEST ORLANDO LAWRENCE BERKELEY NATIONAL LABORATORY

EMERGING ENERGY-EFFICIENT INDUSTRIAL TECHNOLOGIES

N. Martin, E. Worrell, M. Ruth, L. Price LBNL R.N. Elliott, A.M. Shipley, J. Thorne ACEEE

Environmental Energy Technologies Division

October 2000

This work was supported by the Climate Protection Division, Office of Air and Radiation, U.S. Environmental Protection Agency through the U.S. Department of Energy under Contract No. DE-AC03-76SF00098.

Disclaimer This document was prepared as an account of work sponsored by the United States Government. While this document is believed to contain correct information, neither the United States Government nor any agency thereof, nor The Regents of the University of California, nor any of their employees, makes any warranty, express or implied, or assumes any legal responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by its trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof, or The Regents of the University of California. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof, or The Regents of the University of California.

Ernest Orlando Lawrence Berkeley National Laboratory is an equal opportunity employer.

EMERGING ENERGY-EFFICIENT INDUSTRIAL TECHNOLOGIES

N. Martin, E. Worrell, M. Ruth, L. Price LBNL R. N. Elliott, A. M. Shipley, J. Thorne ACEEE

October 2000

Report sponsored by Pacific Gas and Electric Company U.S. Department of Energy U.S. Environmental Protection Agency New York State Energy Research and Development Authority Iowa Energy Center Northwest Energy Efficiency Alliance

The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this report.

CONTENTS
CONTENTS....................................................................................................................................................i LIST OF TABLES....................................................................................................................................... iii ACKNOWLEDGEMENTS .........................................................................................................................iv ABBREVIATIONS .......................................................................................................................................1 EXECUTIVE SUMMARY...........................................................................................................................2 Methodology ..................................................................................................................................................2 Summary of Results ......................................................................................................................................2 Conclusions and Recommendations for Future Work...............................................................................6 I. INTRODUCTION .................................................................................................................................8

II. OVERVIEW OF U.S. INDUSTRIAL ENERGY USE ........................................................................9 Introduction ...................................................................................................................................................9 Industry in Context .......................................................................................................................................9 Economic Output.....................................................................................................................................9 Energy Consumption .............................................................................................................................11 Technology Change in Industry .................................................................................................................13 The Future Of Energy Use In Industry .....................................................................................................14 III. METHODOLOGY AND APPROACH ............................................................................................16 Preliminary Screening of Technologies .....................................................................................................16 Rating Criteria.............................................................................................................................................16 Potential for Energy Savings.................................................................................................................16 Investment Cost for New Technology/Replacement Cost for Existing Processes..................................17 Other Benefits........................................................................................................................................17 Rating Preliminary Technologies...............................................................................................................17 Detailed Assessment of Selected Emerging Energy-Efficient Industrial Technologies .........................18 IV. SUMMARY OF FINDINGS...........................................................................................................23 Introduction .................................................................................................................................................23 Summary of Technology Characterizations..............................................................................................23 Energy Savings ............................................................................................................................................25 The Economics of Energy Savings .............................................................................................................27 Environmental Benefits ..............................................................................................................................29 Non-Energy Benefits ...................................................................................................................................29 Likelihood of Success ..................................................................................................................................31 Technologies from a National Energy Policy Perspective........................................................................31 Suggested Actions to Support Technology Development .........................................................................32 V. CONCLUSIONS AND RECOMMENDATIONS FOR FURTHER WORK...................................34 VI. TECHNOLOGY PROFILES..............................................................................................................36 Condensed Methodology and Summary of Assumptions ........................................................................37 Electron Beam Pasteurization (Food-1) ....................................................................................................38 Low Temperature Heat Recovery in the Food Processing Industries (Food-2).....................................40 Membrane TechnologyFood (Food-3) ...................................................................................................42 Cooling and Storage (Food-4) ....................................................................................................................45 Ultrasound Enhanced Dying (Textile-1)....................................................................................................48

Black Liquor Gasification (Paper-1)......................................................................................................... 50 Condensing Belt Drying (Paper-2) ............................................................................................................ 53 Direct Electrolytic Causticizing (Paper-3)................................................................................................ 55 Dry Sheet Forming (Paper-4) .................................................................................................................... 57 Heat Recover PaperEnclosing Hoods (Paper-5) .................................................................................. 60 High Consistency Forming (Paper-6) ....................................................................................................... 62 Impulse Drying (Paper-7) .......................................................................................................................... 64 Clean Fractionation (Chemicals-1) ........................................................................................................... 67 Gas Membrane Technologies (Chemicals-2) ............................................................................................ 69 Heat Recovery Technologies for Harsh Environments in Chemical Manufacturing (Chemicals-3) .. 71 Levulinic Acid for the Manufacture of Chemicals (Chemicals-4) .......................................................... 74 Liquid Membrane Technologies - Chemicals (Chemicals-5) .................................................................. 76 New Catalysts (Chemicals-6) ..................................................................................................................... 78 Autothermal Reforming (or Combined Reforming) (Chemicals-7) ....................................................... 80 Biodesulfurization of Gasoline (Refining-1) ............................................................................................. 86 Fouling Minimization (Refining-2)............................................................................................................ 88 Roller Kiln (Ceramics-1)............................................................................................................................ 90 100 Percent Cullet Use & Cullet Preheating in Container Glass Manufacture (Glass-1) .................... 93 Gas and Heat Recovery at Basic Oxygen Furnace (Steel-1) ................................................................... 96 Near Net Shape Casting/Strip Casting (Steel-2) ...................................................................................... 98 New EAF Processes (Steel-3) ................................................................................................................... 101 Low NOx Oxy-Fuel Burners in Steel Reheating Furnaces (Steel-4) ..................................................... 104 Smelting Reduction Processes (Steel-5) .................................................................................................. 107 Advanced Forming/Near Net Shape Casting (Alum-1) ......................................................................... 110 Efficient Cell Retrofit Designs (Alum-2)................................................................................................. 113 Improved Recycling Technologies (Alum-3) .......................................................................................... 114 Inert Anodes/Wetted Cathodes (Alum-4) ............................................................................................... 118 Continuous Melt Silicon Crystal Growth (Electron-1).......................................................................... 121 Advanced ASD Designs (Motorsystems-1) ............................................................................................. 123 Advanced Compressor Controls (Motorsystems-2)............................................................................... 125 Motor Diagnostics (Motorsystems-4) ...................................................................................................... 129 Motor System Optimization (Motorsystems-5)...................................................................................... 131 Pump System Efficiency Improvements (Motorsystems-6) .................................................................. 133 Switched Reluctance Drives (Motorsystems-7) ...................................................................................... 135 Premium Lubricants (Motorsys-8) ......................................................................................................... 137 Advanced CHP Turbines (Utilities-1) ..................................................................................................... 139 Advanced Reciprocating Engines (Utilities-2)........................................................................................ 141 Fuel Cells (Utilities-3) ............................................................................................................................... 143 Microturbines (Utilities-4) ....................................................................................................................... 146 Advanced Lighting Technologies (Lighting-1)....................................................................................... 149 Advanced Lighting Design (Lighting-2) ................................................................................................. 152 Anaerobic Wastewater Treatment (Other-1)......................................................................................... 157 High-Efficiency/Low NOx-Burners (Other-2) ........................................................................................ 160 Membrane Technology Wastewater (Other-3) ...................................................................................... 163 Process Integration/Pinch Analysis (Other-4)........................................................................................ 166 Process Control and Sensors (Other-5) .................................................................................................. 170 Variable Mining Machine (Mining-1)..................................................................................................... 173 REFERENCES ......................................................................................................................................... 176

ii

LIST OF TABLES
Table ES-1. Summary of Profiled Emerging Energy-Efficient Industrial Technologies ................................4 Table ES-1. Summary of Profiled Emerging Energy-Efficient Industrial Technologies (continued) .............5 Table ES-2. Technologies with High Energy Savings and a High Likelihood of Success..............................6 Table 1. Historical Share of Industrial Primary Energy Use in the United States ...........................................9 Table 2. Manufacturing Value Added by Sector, 1997 .................................................................................10 Table 3. U.S. Manufacturing Energy Consumption and Fuel Share by Fuel Type, 1985, 1994 ...................11 Table 4. 1994 Manufacturing Energy Consumption by Process and Fuel (TBtu).........................................12 Table 5. Manufacturing Primary Energy Consumption by Region: 1985, 1994 ...........................................13 Table 6. U.S. Industrial Energy Use Forecasting Studies..............................................................................15 Table 7. Preliminary Screening Rating System.............................................................................................18 Table 8. Technologies and Measures Considered in Preliminary Screening Analysis..................................19 Table 8. Technologies And Measures Considered In Preliminary Screening Analysis (Continued) ............20 Table 9. Summary of the Profiled Energy-Efficient Emerging Industrial Technologies...............................24 Table 10. Projected 2015 Implemented Primary Energy Savings Potential ..................................................25 Table 11. Projected 2015 Implemented Electricity Savings Potential...........................................................26 Table 12. Projected 2015 Implemented Fuel Savings Potential ....................................................................26 Table 13. Implemented Savings Share of Sector Projected 2015 Energy .....................................................27 Table 14. Technologies with the Lowest Cost of Saved Electricity ..............................................................28 Table 15. Technologies with the Lowest Cost of Saved Fuel .......................................................................28 Table 16. Environmental Benefits .................................................................................................................29 Table 17. Non-Energy Benefits.....................................................................................................................30 Table 18. Factors Contributing to a High Likelihood of Success..................................................................31 Table 19. Technologies with High Energy Savings and a High Likelihood of Success................................32 Table 20. Technologies Requiring Additional R&D .....................................................................................33 Table 21. Candidate Technologies for Field Trials and Demonstration ........................................................33 Table 22. Profiled Emerging Energy-Efficient Industrial Technologies and their Technology Code...........36

iii

ACKNOWLEDGEMENTS
The authors of this report would like to extend heartfelt thanks to the many individuals who provided their valuable time and assistance on this project. Numerous experts, too many to be individually named, generously contributed to the project team by collecting and evaluating information on the technologies profiled. We would like to thank our sponsors, the Pacific Gas and Electric Company (PG&E), the Office of Air and Radiation of the U.S. Environmental Protection Agency (EPA), the Office of Industrial Technologies of the U.S. Department of Energy (DOE) under Contract No. DE-AC03-76SF00098, the New York State Energy Research and Development Authority (NYSERDA), the Northwest Energy Efficiency Alliance (NEEA), and the Iowa Energy Center (IEC). Specifically, we would like to extend our gratitude to Steven Fok (PG&E), Skip Laitner (EPA), Ken Friedman (DOE), Miriam Pye (NYSERDA), Phil Degens (NEEA), and Bill Haman (IEC) who contributed to the development of the project methodology and offered their input at every step along the way. Finally we wish to acknowledge the expertise of our colleagues who assisted in the research, direction, and production of this report. We thank Steven Nadel at ACEEE for his advice and guidance, and Norma Anglani, who researched some of the technologies this summer. We also thank Susan Ziff, Liz Brown, and Renee Nida for their assistance in developing the final report.

iv

Emerging Energy-Efficient Industrial Technologies

ABBREVIATIONS
ACEEE AEO ASD BOF Btu CADDET CHP CO2 DOE EAF EIA EPA EJ GDP GJ/t HVAC IEC kW kWh LBNL LPG MW MBtu MBtu/ton NEMS NOx NYSERDA NEEA OIT PG&E Co. PJ Quads R&D RD&D TBtu ton t TWh American Council for an Energy-Efficient Economy Annual Energy Outlook adjustable speed drive (motors) basic oxygen furnace (steel making) British thermal unit Center for the Analysis and Dissemination of Demonstrated Energy Technologies combined heat and power carbon dioxide U.S. Department of Energy electric arc furnace (steel making) U.S. Energy Information Administration U.S. Environmental Protection Agency exajoule gross domestic product gigajoule per metric ton heating ventilation and airconditioning Iowa Energy Center kilowatt (electric) kilowatt-hour (electric) Lawrence Berkeley National Laboratory liquid petroleum gas megawatt (electric) million Btu million Btu per short ton National Energy Modeling System oxides of nitrogen New York State Energy Research and Development Authority Northwest Energy Efficiency Alliance U.S. DOE Office of Industrial Technologies Pacific Gas and Electric Company petajoule quadrillion Btu research and design research, development, and demonstration trillion Btu short ton (2000 pounds mass) metric ton terawatt-hour (electric)

LBNL and ACEEE

EXECUTIVE SUMMARY
U.S. industry consumes approximately 37 percent of the nations energy to produce 24 percent of the nations GDP. Increasingly, industry is confronted with the challenge of moving toward a cleaner, more sustainable path of production and consumption, while increasing global competitiveness. Technology will be essential for meeting these challenges. At some point, businesses are faced with investment in new capital stock. At this decision point, new and emerging technologies compete for capital investment alongside more established or mature technologies. Understanding the dynamics of the decision-making process is important to perceive what drives technology change and the overall effect on industrial energy use. The assessment of emerging energy-efficient industrial technologies can be useful for: identifying R&D projects; identifying potential technologies for market transformation activities; providing common information on technologies to a broad audience of policy-makers; and offering new insights into technology development and energy efficiency potentials. With the support of PG&E Co., NYSERDA, DOE, EPA, NEEA, and the Iowa Energy Center, staff from LBNL and ACEEE produced this assessment of emerging energy-efficient industrial technologies. The goal was to collect information on a broad array of potentially significant emerging energy-efficient industrial technologies and carefully characterize a sub-group of approximately 50 key technologies. Our use of the term emerging denotes technologies that are both pre-commercial but near commercialization, and technologies that have already entered the market but have less than 5 percent of current market share. We also have chosen technologies that are energy-efficient (i.e., use less energy than existing technologies and practices to produce the same product), and may have additional non-energy benefits. These benefits are as important (if not more important in many cases) in influencing the decision on whether to adopt an emerging technology. The technologies were characterized with respect to energy efficiency, economics, and environmental performance. The results demonstrate that the United States is not running out of technologies to improve energy efficiency and economic and environmental performance, and will not run out in the future. We show that many of the technologies have important non-energy benefits, ranging from reduced environmental impact to improved productivity and worker safety, and reduced capital costs.

Methodology
The assessment began with the identification of approximately 175 emerging energy-efficient industrial technologies through a review of the literature, international R&D programs, databases, and studies. The review was not limited to U.S. experiences, but rather we aimed to produce an inventory of international technology developments. We devised an initial screening process to select the most attractive technologies that had: (1) high potential energy savings; (2) lower comparative first costs relative to existing technologies; and (3) other significant benefits. While some technologies scored high on all of these characteristics, most had a mixed score. We formalized this approach in a very simple rating system. Based on the literature review and the application of initial screening criteria, we identified and developed profiles for 54 technologies. The technologies ranged from highly specific ones that can be applied in a single industry to more broadly crosscutting ones that can be used in many industrial sectors. Each of the selected technologies has been assessed with respect to energy efficiency characteristics, likely energy savings by 2015, economics, and environmental performance, as well as whats needed to further the development or implementation of the technology. The technology characterization includes a one to two-page description and a one-page table summarizing the results for the technology.

Summary of Results
Table ES-1 provides an overview of the 54 emerging energy-efficient industrial technologies. We evaluated energy savings in two ways. The third column of Table ES-1 (Total Energy Savings) shows the amount of total manufacturing energy that the technology is likely to save in 2015 in a business-as-usual scenario. The 2

Emerging Energy-Efficient Industrial Technologies fourth column (Sector Savings) reflects the savings relative to expected energy use in the particular sector. We believe that both metrics are useful in evaluating the relative savings potential of various technologies. Economic evaluation of the technology is identified in the summary table by simple payback period, defined as the initial investment costs divided by the value of energy savings less any changes in operations and maintenance costs. We chose this measure since it is frequently used as a shorthand evaluation metric among industrial energy managers. Payback times for the technologies range from the immediate to 20 years or more. Of the 54 technologies profiled, 31 have estimated paybacks of 3 years or less, with six paying back immediately Energy savings are most often not the determining factor in the decision to develop or invest in an emerging technology. Over two-thirds of technologies not only save energy but yield non-energy benefits. We separated these non-energy benefits into environmental and other categories. We assessed how important the environmental benefits are to the technology adoption decision and listed the nature of the other benefit(s). We include an assessment of the importance of these non-energy benefits. Technologies do not seamlessly enter existing markets immediately after development. The acceptance of emerging technologies is often a slow process that entails active research and development, prototype development, market demonstration, and other activities. In Table ES-1 we summarize the recommendations for the primary activities that could be undertaken to increase the technologies rate of uptake. Over half of these technologies have already been developed to prototype stage or are already commercial but require further demonstration and dissemination. Each technology is at a different point in the development or commercialization process. Some technologies still need further R&D to address cost or performance issues, some are ready for demonstration, and others have already proven themselves in the field and the market needs to be informed of the benefits and market channels needed to develop skills to deliver the technology. Our outlining of recommended actions in Table ES-1 is not an endorsement of any particular technology. Technology purchasers and users will ultimately decide regarding future development. However, the actions specified are intended to help identify whether a technology is both technically and economically viable and whether it is robust enough to accommodate the stringent product quality demands in various manufacturing establishments. Seventeen emerging technologies could benefit from additional R&D. We suggest further R&D for several primary metal technologies, and several cross-cutting motor and utility technologies. In addition to private research funds, several of the identified technologies have received some R&D support from DOE or other public entities, including federal and state agencies. There are also a large number of technologies that already have made some headway into the marketplace or are at the prototype testing stage, and therefore are candidates for demonstration for potential customers to gain comfort with the technology. While we recommend further demonstration and dissemination of these technologies, it was often difficult to understand what is limiting their uptake without more comprehensive investigation of market issues. Some of the technologies in this category are common in European countries or Japan but have not yet penetrated the U.S. market. Others are being newly developed in the United States and face challenges in reducing the risks perceived by potential purchasers. Two technologies, motor system optimization and pump efficiency improvement, are opportunities for training programs similar to those developed by DOE for the compressed air system management. For advanced industrial CHP turbine systems, the major recommended activity is removal of policy barriers. For other technologies, their unique markets will dictate the form of the educational and promotional activities. We urge the reader to follow up on any details in the specific technology profiles provided in Section VI of this report .

LBNL and ACEEE

Table ES-1. Summary of Profiled Emerging Energy-Efficient Industrial Technologies

Technology Advanced forming Efficient cell retrofit designs Improved recycling technologies Inert anodes/wetted cathodes Roller kiln Clean fractionationcellulose pulp Gas membrane technologies chem. Heat recovery technologieschem. Sector Aluminum Aluminum Aluminum Aluminum Ceramics Chemicals Chemicals Chemicals Total1 Energy Savings Medium High Medium High Medium Low Low Sector2 Simple Savings Payback Medium Immediate High 2.7 Medium 4.5 High 4.0 High 1.9 Low 1.9 Low 10.2 2.4 1.5 11.2 Environ. Other3 Benefits Benefits None P Somewhat P Significant P Significant P, Q Significant P Significant P, O Significant Q, O None Significant Significant P, O P, O O Suggested Next Steps R&D Demo Demo R&D Demo Demo Dissem. Dissem., Demo Demo Dissem. R&D Dissem Demo R&D R&D Dissem. Dissem., R&D Dissem., Demo Demo Dissem. Dissem., Demo Dissem., Demo Demo R&D Dissem. Dissem. Dissem., Demo Dissem., Train Dissem., Train R&D Dissem. Dissem., Demo Dissem., Demo Dissem., R&D Dissem. Dissem., R&D, demo Likelihood of Success High High Medium Medium Medium Medium High Medium High Medium Medium Medium High High Low Low Medium Medium High Medium High Medium Low High Medium Medium High Medium Medium Medium Medium High Medium High Medium High

Medium Medium Low Low Low Low

Levulinic acid from biomass Chemicals Liquid membrane technologies Chemicals chem. New catalysts Chemicals Chemicals Autothermal reformingammonia Plastics recovery Plastics Continuous melt silicon crystal Electronics growth Electron beam sterilization Food Food Heat recoverylow temperature Food Membrane technologyfood Cooling and storage 100% recycled glass cullet Hi-tech facilities HVAC Advanced lighting technologies Advanced lighting design Variable wall mining machine Advance ASD designs Advanced compressor controls Compressed air system management Motor diagnostics Motor system optimization Pump efficiency improvement Switched reluctance motor Advanced lubricants Anaerobic waste water treatment High-efficiency/low NOx burners Membrane technology wastewater Process integration (pinch) Sensors and controls Food Glass Crosscutting Crosscutting Crosscutting Mining Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting

Medium Medium 7.9 Somewhat High High 3.7 Significant Medium Medium 2.8 Compelling Medium High Immediate Somewhat High High Medium Medium High High Medium Medium Medium Medium High High High High High High 19.2 4.8 2.2 2.6 2.0 4.0 1.3 3.0 None None Somewhat Somewhat Significant None None None None None None None None Somewhat None None Significant Significant Significant Somewhat Somewhat Somewhat

P P P, Q P, Q P, Q P, Q O P P, Q, O P, Q, O P, S P, Q P, Q P, Q P, Q P, Q P, Q P, Q P, Q O P P P P, Q

Low Low 10.6 High Medium 1.1 Medium Low 0.0 High High 0.4 Low Low Immediate High High High High 1.5 3.0 7.4 0.1 0.8 3.1 4.7 2.3 2.0

Medium Low Medium Medium Medium Low High High High High Low Medium Low Medium

Emerging Energy-Efficient Industrial Technologies Table ES-1. Summary of Profiled Emerging Energy-Efficient Industrial Technologies (continued)
Simple Environ. Other3 Suggested Likelihood Sector2 Technology Sector Savings Payback Benefits Benefits Next Steps of Success Black liquor gasification Pulp & paper High 1.5 Somewhat P, S Demo High Condebelt drying Pulp & paper Medium 65.2 None P, Q Demo Low Direct electrolytic Pulp & paper Low N/A Somewhat P, Q R&D Medium causticizing Dry sheet forming Pulp & paper Medium Medium 48.3 Somewhat Q R&D, demo High Pulp & paper High Medium 3.9 Somewhat P, S Demo Medium Heat recoverypaper High consistency forming Pulp & paper Medium Medium Immediate Somewhat P, Q Demo Medium Impulse drying Pulp & paper High Medium 20.3 None P, Q Demo Medium Biodesulfurization Pet. Refining Medium Medium 1.8 None Q R&D, demo High Fouling minimization Pet. Refining High High N/A None P R&D Low Medium Medium 14.7 Significant P Dissem. Low BOF gas and sensible heat Iron & steel recovery Near net shape casting/strip Iron & steel High High Immediate Somewhat P, Q R&D High casting New EAF furnace processes Iron & steel High High 0.3 Somewhat P Field test High Oxy-fuel combustion in Iron & steel High Medium 1.2 Significant P Field test High reheat furnace Smelting reduction processes Iron & steel High High Immediate Significant P Demo Medium Ultrasonic dying Textile Medium Medium 0.3 Compelling P, Q Demo Medium Advanced CHP turbine Crosscutting High High 6.9 Significant P, Q Policies High systems Advanced reciprocating Crosscutting High High 8.3 Limited P, Q, O R&D, demo Medium engines Fuel cells Crosscutting High High 58.6 Significant P, Q Demo Medium Microturbines Crosscutting High Medium Never Somewhat P, Q, O R&D, demo Medium Notes: 1. High could save more than 0.1% of manufacturing energy use by 2015, medium saves 0.01 to 0.1%, and low saves less than 0.01%. 2. High could save more than 1% of sector energy use by 2015, medium saves 0.1 to 1%, and low saves less than 0.1%. 3. P=productivity, Q=quality, S=safety, and O=other. Total1 Energy Savings High High Low

We assess the technologys likelihood of success in the marketplace. While our study evaluates each technology in relation to a given reference technology, the reality of the market is that technologies compete for market share. We made a judgement (based on the energy savings, cost-effectiveness, importance of non-energy benefits, market conditions, data reliability, and potential competing technologies) as to the likelihood that the technology would succeed in the marketplace. From a national energy policy perspective, it is important to understand which technologies have both a high likelihood of success and a high energy-savings. While various audiences may be interested in sectorspecific or regional-specific technologies, the technologies listed in Table ES-2 are intended to provide guidance to those interested in the impact of energy-saving technologies on a more national level. This table also identifies the recommended next steps appropriate for each technology.

LBNL and ACEEE

Table ES-2. Technologies with High Energy Savings and a High Likelihood of Success
Technology Efficient cell retrofit designs Advanced lighting technologies Advance ASD designs Membrane technology wastewater Sensors and controls Black liquor gasification Near net shape casting/strip casting New EAF furnace processes Oxy-fuel combustion in reheat furnace Advanced CHP turbine systems Autothermal reforming-ammonia Membrane technology - food Advanced lighting design Compressed air system management Motor system optimization Pump efficiency improvement High efficiency/low NOX burners Process integration (pinch analysis) Heat recovery - paper Impulse drying Smelting reduction processes Advanced reciprocating engines Fuel cells Microturbines Inert anodes/wetted cathodes Code Alum-2 Lighting-1 Motorsys-1 Other-3 Other-5 Paper-1 Steel-2 Steel-3 Steel-4 Utilities-1 Chem-7 Food-3 Lighting-2 Motorsys-3 Motorsys-5 Motorsys-6 Other-2 Other-4 Paper-5 Paper-7 Steel-5 Utilities-2 Utilities-3 Utilities-4 Alum-4 Total Energy Savings High High High High High High High High High High High High High High High High High High High High High High High High High Likelihood of Success High High High High High High High High High High Medium Medium Medium Medium Medium Medium Medium Medium Medium Medium Medium Medium Medium Medium Medium Recommended Next Steps Demonstraton Dissemination, demonstration R&D Dissemination, R&D R&D, demonstration, dissemination Demonstration R&D Field test Field test Policies Dissemination Dissemination, R&D Dissemination, demonstration Dissemination Dissemination, training Dissemination, training Dissemination, demonstration Dissemination Demonstration Demonstration Demonstratoin R&D, demonstration Demonstration R&D, demonstration R&D

Advanced forming Alum-1 Medium High R&D Plastics recovery Chem-8 Medium High Demonstration Continuous melt silicon crystal growth Electron-1 Medium High R&D 100% recycled glass cullet Glass-1 Medium High Demo Anaerobic waste water treatment Other-1 Medium High Dissemination., demonstration Dry sheet forming Paper-4 Medium High R&D, demonstration Biodesulfurization Refin-1 Medium High R&D, demonstration *note technologies in this table are listed in alphabetical order based on industry sector

Conclusions and Recommendations for Future Work

For this study, we identified about 175 emerging energy-efficient technologies in industry, of which we characterized 54 in detail. While many profiles of individual emerging technologies are available, few reports have attempted to impose a standardized approach to the evaluation of the technologies. This study provides a way to review technologies in an independent manner, based on information on energy savings, economic, non-energy benefits, major market barriers, likelihood of success, and suggested next steps to accelerate deployment of each of the analyzed technologies. There are many interesting lessons to be learned from further investigation of technologies identified in our preliminary screening analysis. The detailed assessments of the 54 technologies are useful to evaluate claims made by developers, as well as to evaluate market potentials for the United States or specific regions. In this report we show that many new technologies are ready to enter the market place, or are currently under development, demonstrating that the United States is not running out of technologies to improve energy efficiency and economic and environmental performance, and will not run out in the future. The study shows that many of the technologies have important non-energy benefits, ranging from reduced environmental impact to improved productivity. Several technologies have reduced capital costs compared to the current technology used by those industries. Non-energy benefits such as these are frequently a motivating factor in bringing technologies such as these to market. Further evaluation of the profiled technologies is still needed. In particular, further quantifying the nonenergy benefits based on the experience from technology users in the field is important. Interactive effects 6

Emerging Energy-Efficient Industrial Technologies and intertechnology competition have not been accounted for and ideally should be included in any type of integrated technology scenario, for it may help to better evaluate market opportunities. While this report focuses on the United States, state- or region-specific analysis of technologies may provide further insights into opportunities specific for the region served. Regional specificity is determined by the type of users (i.e., industrial activities) in the region, as well as the available technology developers. Combining the region-specific circumstances with the technology evaluations offered in this report may lead to varying policy choices for regional entities such as state governments, state or regional agencies, or utilities. Our selection of a limited set of 54 technologies was an arbitrary constraint based on the funding available. A number of the initial technologies screened appeared very interesting and warrant further study, but were eliminated due to resource constraints. In addition, the initial list of candidate technologies should not be viewed as all-encompassing. The authors are aware that other promising existing technologies exist, and that by their nature new technologies will be continually emerging. Ideally, the effort reflected in this report should be the start of a continuing process that identifies and profiles the most promising emerging energyefficient industrial technologies and tracks the market success for these technologies. An interactive database may be a better choice for it would allow the continuous updating of information, rather than providing a static snapshot of the industrial technology universe. This report identifies and profiles many promising emerging energy-efficient industrial technologies, which can achieve high energy-savings, and have a good likelihood of success due to their economic, environmental, product quality, and other benefits. We recommend next steps that product developers and policy-makers could undertake for each of the most promising technologies. Follow-up assessments are needed to identify additional emerging technologies, and to track the emergence of the technologies profiled in this report.

LBNL and ACEEE

I. INTRODUCTION
Whether one is the general manager of a manufacturing plant looking to increase productivity or an environmentalist seeking to reduce greenhouse gas emissions, it is evident that technology plays a critical role in the nations economy. From a policy-making perspective, the better we understand technology developments the more effective we will be in utilizing our future research dollars and in undertaking sound strategy development. As just one example, few economic models today provide a reasonable characterization of both existing and emerging technologies. But even models with only a limited characterization of technology tend to forecast significantly different energy consumption patterns than those that reflect actual technology choices confronted by consumers and businesses (Laitner 2000a). Inappropriate characterization of technologies can lead to poor analysis and eventually less than optimal policy choices. Getting a clearer picture about emerging technologies will help to: Identify new R&D projects; Identify potential technologies for market transformation activities; Provide common information on technologies to a broad audience of policy-makers; and Offer new insights into technology development and energy efficiency potentials

The development of this report emerged from a desire across several research, development, and policymaking agencies to improve our common understanding of the status of emerging energy-efficient technologies in the industrial sector. Although many technologies save energy, often the driver for their adoption is reductions in capital costs and other non-energy benefits. It is important to better understand the developmental stories and drivers behind the emerging technologies. With the support of PG&E Co.,1 NYSERDA, DOE, EPA, NEEA, and IEC, staff from LBNL and ACEEE produced the current report. The sponsors are not responsible for the reports content, or any errors or omissions. This report focuses on key emerging energy-efficient technologies in the industrial sector. Our goal is to identify and assess these technologies from the viewpoint of both energy and non-energy benefits. While we focus on technologies that show a strong potential for energy savings, we also account for the nonenergy benefits associated with such technologies, since often these non-energy benefits can be the key driver in overall technology adoption. We hope that this assessment further identifies the significant potential available in the United States and other countries for further advancement toward greener production. This work complements the 1998 study, Emerging Energy-Saving Technologies and Practices for the Buildings Sector (Nadel et al. 1998), which provided data on technologies with the largest potential savings, including likely costs, savings, and date of commercialization. Similar to the 1998 report, the goal of this current effort has been to collect information on a broad array of potentially significant emerging energy-efficient industrial technologies and carefully characterize a sub-group of 54 emerging technologies. The characterization of an emerging, energy-efficient technology is somewhat difficult. What was emerging a decade ago may now be standard practice. In this report our use of the term emerging denotes technologies that are both pre-commercial but near commercialization and technologies that have already entered the market but have less than 5 percent of current market share. We also have chosen technologies that are energy-efficient (i.e., use less energy than existing technologies and practices to produce the same product) as well as technologies that often have other non-energy benefits associated with their use.

The PG&E Co. program is funded by California utility customers and is administered by PG&E Co. under the auspices of the California Public Utilities Commission.

Emerging Energy-Efficient Industrial Technologies

II. OVERVIEW OF U.S. INDUSTRIAL ENERGY USE

Introduction
Energy is a key input for our modern U.S. economy. In 1998, the United States consumed 94 Quadrillion Btu (99 EJ)2 of primary energy or 25 percent of world primary energy use (EIA 2000).3 Within the various sectors of the United States, the industrial sector remains a significant energy user, consuming nearly 40 percent of primary energy resources (Table 1). The industrial sector is extremely diverse and includes agriculture, mining, construction, and manufacturing. Table 1 identifies historical industrial energy consumption in relation to U.S. total primary energy consumption. Table 1. Historical Share of Industrial Primary Energy Use in the United States Units Total U.S. Quads (EJ) Total Industry Quads (EJ) Percent share % Source: EIA 2000 1950 34.6 (36.5) 16.2 (17.1) 47% 1970 67.9 (71.6) 29.6 (31.3) 44% 1990 84.1 (88.7) 32.1 (33.9) 38% 1998 94.2 (99.4) 35.4 (37.4) 38%

Energy is necessary to help our industries create useful products; however, we are increasingly confronted with the challenge of moving our economy and society toward a cleaner, more sustainable path of production and consumption. The development and use of cleaner, more energy-efficient technologies can play a significant role in limiting the environmental impacts associated with many industries while enhancing productivity and reducing manufacturing costs.

Industry in Context
Economic Output Industrial activities are still a key component of U.S. economic output. In 1997, industrial activities accounted for 24 percent of gross domestic product (GDP), which that year was $8,300 billion in 1997 dollars, and employed 27 million full and part-time employees (BEA 2000). Within the industrial sector, manufacturing activity (consisting of all industrial activity outside of agriculture, mining, and construction) accounts for 70 percent of industrial value added (BEA 2000). Table 2 identifies the distribution of value added by various manufacturing activities throughout the United States. The table also provides aggregated totals for the four U.S. regions that correspond to the regions that are used in the reporting of manufacturing energy statistics.

In this report we present energy consumption and energy intensity information in both english units (Btus) and standard international units (joules), as the latter is the unit of international communication on energy issues. When appropriate we do note conversion factors. One quadrillion Btu (10^15) equals 0.95 EJ and one metric tonne equals 0.907 short tons. 3 Primary energy reflects the losses associated with the conversion, transmission and distribution of electricity. For the U.S. economy as a whole we use an electricity conversion efficiency of 33 percent. For calculation of primary energy savings in 2015 for the technology evaluation we assume a conversion efficiency of 40 percent accounting for the future efficiency improvement in power generation by 2015 (due to increased use of combined cycles and combined heat and power by 2015).

LBNL and ACEEE

Table 2. Manufacturing Value Added by Sector, 1997

Stone, clay, & glass Primary metals Food & kindred products Textile & apparel Fabric. metals & machiner Lumber and wood Transprt. equip. Chemical Electron. Equip. Paper products Petro. products

West Alaska Arizona California Colorado Hawaii Idaho Montana Nevada New Mexico Oregon Utah Washington Wyoming South Alabama Arkansas Delaware District of Col. Florida Georgia Kentucky Louisiana Maryland Mississippi North Carolina Oklahoma South Carolina Tennessee Texas Virginia West Virginia Midwest Illinois Indiana Iowa Kansas Michigan Minnesota Missouri Nebraska North Dakota Ohio South Dakota Wisconsin Northeast Connecticut Maine Massachusetts New Hampshire New Jersey New York Pennsylvania Rhode Island Vermont U.S. Total U.S. Total ($billion current)

18% 0% 1% 11% 1% 0% 0% 0% 0% 1% 2% 1% 2% 0% 32% 2% 1% 0% 0% 2% 3% 2% 1% 1% 1% 4% 1% 2% 2% 7% 2% 0% 30% 5% 4% 1% 1% 5% 2% 2% 0% 0% 6% 0% 3% 19% 2% 0% 2% 1% 3% 5% 5% 0% 0% 100% 1,378.9

20% 0% 1% 11% 2% 0% 1% 0% 0% 0% 1% 1% 2% 0% 32% 1% 2% 0% 0% 3% 4% 2% 1% 2% 1% 2% 1% 1% 2% 6% 3% 0% 34% 7% 2% 3% 1% 3% 3% 4% 1% 0% 5% 0% 4% 15% 1% 0% 1% 0% 3% 5% 5% 0% 0% 100% 118.5

13% 0% 0% 11% 1% 0% 0% 0% 0% 0% 0% 0% 1% 0% 59% 5% 1% 0% 0% 2% 11% 3% 1% 1% 2% 15% 0% 8% 3% 4% 3% 0% 8% 1% 1% 1% 0% 2% 0% 1% 0% 0% 1% 0% 1% 19% 1% 0% 2% 1% 3% 8% 4% 1% 0% 100% 53.9

26% 0% 1% 7% 1% 0% 2% 1% 0% 0% 8% 0% 5% 0% 43% 5% 3% 0% 0% 3% 6% 1% 2% 0% 4% 5% 1% 2% 3% 6% 3% 1% 21% 2% 4% 1% 1% 2% 3% 1% 0% 0% 4% 0% 3% 10% 0% 1% 1% 1% 0% 2% 4% 0% 0% 100% 42.8

11% 0% 0% 5% 0% 0% 0% 0% 0% 0% 1% 1% 2% 0% 40% 5% 2% 1% 0% 2% 6% 2% 3% 1% 1% 3% 1% 4% 3% 4% 3% 0% 29% 4% 1% 1% 1% 3% 5% 1% 0% 0% 5% 0% 9% 20% 2% 3% 2% 1% 3% 4% 5% 0% 0% 100% 55.0

8% 0% 1% 5% 0% 0% 0% 0% 0% 0% 0% 1% 0% 0% 42% 1% 1% 2% 0% 2% 2% 2% 4% 1% 1% 6% 0% 3% 2% 10% 3% 2% 26% 5% 4% 1% 1% 3% 1% 3% 0% 0% 6% 0% 1% 24% 2% 0% 1% 0% 9% 4% 7% 0% 0% 100% 158.8

25% 1% 0% 18% 0% 1% 0% 1% 0% 0% 0% 1% 2% 0% 43% 1% 1% 0% 0% 1% 0% 2% 9% 0% 2% 0% 4% 0% 0% 20% 0% 0% 15% 5% 2% 0% 1% 1% 1% 1% 0% 0% 4% 0% 0% 17% 1% 0% 1% 0% 10% 3% 3% 0% 0% 100% 35.2

16% 0% 1% 8% 1% 0% 0% 0% 1% 1% 1% 1% 2% 0% 37% 2% 1% 0% 0% 4% 4% 2% 1% 1% 1% 5% 1% 2% 3% 8% 2% 1% 29% 4% 3% 2% 1% 5% 2% 2% 0% 0% 8% 0% 2% 18% 1% 0% 1% 0% 2% 5% 7% 0% 0% 100% 33.7

11% 0% 1% 4% 0% 0% 0% 0% 0% 0% 2% 1% 2% 0% 28% 4% 2% 0% 0% 1% 2% 3% 0% 2% 1% 2% 1% 1% 2% 4% 1% 2% 45% 6% 11% 1% 0% 7% 1% 1% 0% 0% 14% 0% 3% 17% 1% 0% 1% 1% 1% 3% 9% 1% 0% 100% 53.2

18% 0% 1% 12% 1% 0% 0% 0% 0% 0% 1% 1% 1% 0% 26% 1% 1% 0% 0% 1% 1% 2% 1% 1% 1% 2% 1% 2% 2% 9% 1% 0% 39% 8% 4% 2% 1% 7% 3% 2% 0% 0% 8% 0% 5% 17% 2% 0% 3% 1% 2% 4% 5% 0% 0% 100% 258.2

36% 0% 5% 17% 1% 0% 1% 0% 0% 4% 8% 0% 1% 0% 26% 1% 1% 0% 0% 3% 2% 1% 0% 1% 1% 4% 1% 1% 1% 10% 1% 0% 20% 5% 2% 1% 0% 1% 2% 1% 1% 0% 4% 0% 2% 18% 2% 0% 4% 1% 1% 4% 6% 0% 0% 100% 157.3

16% 0% 2% 7% 1% 0% 0% 0% 0% 0% 1% 1% 5% 0% 26% 1% 1% 1% 0% 2% 3% 4% 1% 0% 1% 1% 1% 1% 3% 3% 2% 0% 48% 3% 7% 1% 2% 16% 1% 5% 0% 0% 10% 0% 2% 10% 2% 0% 1% 0% 0% 3% 3% 0% 0% 100% 136.1

18% 0% 1% 12% 2% 0% 0% 0% 0% 0% 1% 1% 1% 0% 30% 1% 1% 0% 0% 3% 3% 2% 0% 1% 1% 6% 1% 1% 2% 4% 4% 0% 26% 5% 3% 1% 1% 4% 3% 2% 1% 0% 4% 0% 3% 26% 2% 0% 4% 1% 3% 11% 4% 1% 0% 100% 276.4

Note: Other manufacturing includes: tobacco products (SIC 21), furniture and fixtures (SIC 25), printing and publishing (SIC 27), rubber and plastics (SIC 30), leather products (SIC 31), instruments and related products (SIC 38), and miscellaneous manufacturing (SIC 39). Source: BEA 2000

10

Other

Total

Emerging Energy-Efficient Industrial Technologies As the table indicates, the South and Midwest accounted for nearly two-thirds of total manufacturing output in 1997. On a state level, the largest contributors to manufacturing GDP were California, Texas, Ohio, New York, Pennsylvania, Illinois, and Michigan, which together accounted for 40 percent of manufacturing output in 1997. Energy Consumption Energy expenditures between 1994 and 1996, the most recent years for available statistics, fluctuated between $60 and $70 million, which accounted for about 2 percent of total costs, although in some industries the share could have been as high as 6 percent (Census 1998). EIA produces periodic detailed statistics on energy consumption in the U.S. manufacturing sector.4 The most recent detailed data available are from 1994. Table 3 summarizes historical energy consumption by fuel between 1985 and 1994. Table 3. U.S. Manufacturing Energy Consumption and Fuel Share by Fuel Type, 1985, 1994 1985 Fuel Net electricity Residual fuel oil Distillate fuel oil Natural gas LPG Coal Coke and breeze Other Total final energy Total primary energy* Source: EIA 1988, 1997 TBtu (PJ) 2,173 (2,293) 505 (532) 185 (195) 4,647 (4,903) 96 (101) 1,304 (1,376) 590 (623) 4,102 (4,328) 13,615 (14,365) 18,027 (19,020) Fuel share 16% 4% 1% 34% 1% 10% 4% 30% 100% 1994 TBtu (PJ) 2,656 (2,802) 441 (465) 152 (160) 6,141 (6,479) 99 (104) 1,198 (1,264) 703 (742) 5,126 (5,408) 16,515 (17,424) 23,113 (24,386) Fuel share 16% 3% 1% 37% 1% 7% 4% 31% 100% % change 22% -13% -18% 32% 3% -8% 19% 25% 21% 21%

We used an electricity efficiency factor of 33 percent to convert from final to primary energy.

As the table indicates, energy use increased by 21 percent between 1985 and 1994, which translates into a growth of about 2 percent per year, slower than the rate of manufacturing economic growth over the same period. There has been a slow transition to more flexible fuels (natural gas, electricity), and the use of other fuels (primarily waste gas and biomass-derived fuels), which combined account for over 80 percent of the total fuel use in manufacturing. The use of oil and coal has declined even though overall energy consumption grew.

We discuss manufacturing energy use in detail in this section due to availability of data. Manufacturing accounts for roughly 70 percent of total industrial energy use.

11

LBNL and ACEEE

Table 4. 1994 Manufacturing Energy Consumption by Process and Fuel (TBtu)

Electricity 28 2,075 284 138 1,367 271 15 457 217 185 46 4 Residual fuel oil 313 106 103 3 14 5 3 5 4 96 2,656 1 9 441 Distillate fuel oil 42 51 29 18 4 49 7 1 35 4 2 9 152 Natural gas 2,396 2872 2702 21 95 53 726 351 30 1 335 9 148 6,141 LPG 15 54 49 2 3 1 25 5 1 19 1 Coal 875 302 299 3 8 3 Coke & other 2,381 687 Total 6,050 5,460 4,153 161 1,489 271 73 1,279 588 185 81 59 351 16 2,760 5,828

Boilers Total Process Use Process heating Process cooling Machine drive Electro-chemical Other process use Non Process Use HVAC Lighting Facility support On-site transport Conventional electricity generation Other Not allocated Total Source: EIA 1997

6 0 13 1,198

4 99

3,039 16,515

These fuels are used to operate a variety of manufacturing operations including process heating, cooling, motor drive, and providing general utilities (e.g., power or steam). Table 4 provides information on the breakdown of fuel use by process for manufacturing in 1994. Based on recent LBNL analysis (Einstein et al. 2000), we reallocated some of the coke/other fuels that were previously not allocated to boiler inputs and coke for process use in the metals industry. As the table indicates, the production of steam in boilers for electricity generation5 and process use accounts for the largest end-use within manufacturing, followed by process heating and machine drive. Process energy efficiency measures that reduce process steam consumption (e.g., drying measures in the pulp and paper industry) can also reduce boiler fuels use, as do direct boiler efficiency measures. The majority of natural gas is used in boilers and for process heating (e.g., furnaces), while half the electricity use is used for machine drive. When total motor systems are accounted for, a recent study found that process motor use accounted for 63 percent of all electricity use in industry in 1994 (Xenergy 1998). The consumption of energy for various processes is not equally divided among all industries. Rather, within manufacturing, there exist a set of activities in which the energy requirements to produce a unit of output are significantly higher than average energy requirements for manufacturing overall. These energyintensive sectors account for 80 percent of primary energy use in manufacturing but only a third of manufacturing value added. Energy-intensive sectors include paper; chemicals; petroleum and stone, clay, and glass products; primary metals; and food and kindred products.6 These industries are often a prime target for emerging energy-saving technologies since they tend to better leverage energy savings. At the same time, some of the energy intensive sectors are growing more slowly and are less likely to make new capital investments as compared to some of the faster growing industries such as electronics and metal fabrication. In both cases, non-energy benefits associated with the investment in energy-saving technologies can be a key factor in justifying the expenditures on new equipment. Table 5 identifies the regional distribution of manufacturing energy use in 1985 and 1994 with a breakdown of manufacturing sectors into energy-intensive and other manufacturing. Similar to the trends in manufacturing value-added, the South and Midwest accounted for the majority (three-fourths) of the
Cogeneration or CHP is an important aspect of industrial process use that is not fully captured in Table 4. In 1994, 128 TWh of electricity was produced by cogeneration in manufacturing, as compared to a net purchased amount of 778 TWh. Over a third of cogenerated electricity was produced by steam turbines connected to boilers or high-temperature processes. 6 Food and kindred products is normally included in energy-intensive since it too consumes a large amount of energy due to the high volume of product throughput even though the manufacturing processes themselves are less energyintensive than the other energy-intensive sectors.
5

12

Emerging Energy-Efficient Industrial Technologies countrys manufacturing energy use. The energy intensive sectors share of energy consumption was roughly three or more times greater than less-intensive sectors in all U.S. regions. Table 5. Manufacturing Primary Energy Consumption by Region: 1985, 1994
1985 NORTHEAST Energy-intensive Other MIDWEST Energy-intensive Other SOUTH Energy-intensive Other WEST Energy-intensive Other US TOTAL Energy-intensive Other Total TBtu (PJ) 1,664 (1,756) 657 (694) 3,793 (4,002) 1,122 (1,184) 6, 859 (7,236) 1,348 (1,423) 2,075 (2,189) 508 (536) 14,381 (15,172) 3,647 (3,848) 18,027 (19,020) Percent 9% 4% 21% 6% 38% 7% 12% 3% 80% 20% 100% 1994 TBtu (PJ) 1,662 (1,754) 740 (781) 4,732 (4,993) 1,545 (1,630) 9,294 (9,806) 1,954 (2,062) 2,536 (2,676) 650 (686) 18,224 (19,228) 4,888 (5,158) 23,113 (24,386) Percent 7% 3% 20% 7% 40% 8% 11% 3% 79% 21% 100%

Source: EIA 1988, 1997

Note: assuming an electricity conversion efficiency of 33 percent for both 1985 and 1994.

Technology Change in Industry

The demand for energy to produce manufactured products is related to the volume and mix of production as well as the efficiency of the equipment used in the manufacturing processes. A broad proxy for efficiency is its inverse, energy intensity, or the amount of energy required to produce a unit of output. Research about the United States has shown that since the first oil price shock in 1973, manufacturing energy consumption would have been significantly higher were it not for decreases in energy intensity.7 As long as they can remain competitive, businesses will often choose to operate existing equipment and technology throughout its useful lifetime, which can run for 20 years or more for large pieces of equipment such as cement kilns or blast furnaces. At some point, however, businesses are faced with investment in new capital stock. At this decision point, new and emerging technologies compete for capital investment alongside more established or mature technologies. Even if a standard technology is chosen, it is likely to be more efficient than the equipment it is replacing. Understanding the dynamics of the decision-making process is important to better perceive what drives technology change and its overall effect on industrial energy use. Barriers for technology transfer in the industrial sector include corporate decision-making rules (e.g., high hurdle rates, split incentives between various parts of a company), lack of information, limited capital or technology availability, shortage of trained personnel (especially in small and medium-sized enterprises), low energy prices, perceived risk, and the invisibility of energy savings. Many new technologies follow a traditional S curve adoption path whereby a small segment of the industry, known as early adopters, embraces a new and unproven technology despite high costs and potential risks. As the technology becomes more common, the perceived risks decrease and the cost of the technology declines. The period needed to achieve a significant market share may vary and depends on the technology characteristics, as well as characteristics of the market and the particular sector. De Canio and Laitner (1997) point out that the current approaches to model technology diffusion tend to underestimate
7

Golove and Schipper (1996) performed a long-term analysis of the U.S. manufacturing sector from 1958 to 1991, which found that declines in energy intensity played the dominant role in limiting actual energy consumption. Belzer et al. (1995) also found that energy intensity declines accounted for over half of the energy savings in the industrial sector.

13

LBNL and ACEEE the rate since they do incorporate cost information (i.e., an investment approach) but lack the representation of the influence of time and the impact of an increasingly critical mass of technology adopters (De Canio and Laitner 1997). Figure 1 shows a typical S curve of the adoption of continuous casting technology in the U.S. iron and steel industry. Although the technology eventually reached saturation during the 30-year period, it took much longer in the United States than in other steel producing countries.8 Many innovation and energy polices focus on accelerating the rate of adoption of specific technologies by reducing the costs or perceived risks of the technology. Figure 1. Continuous Casting Use in the United States (1970-1998)

100 % 80% 60% 40% 20% 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 0%

Source: IISI 2000a Various programs try to reduce several barriers to adoption. A wide array of policies to increase the implementation rate of new technologies has been used and tested in the industrial sector in industrialized countries with varying success rates. We do not discuss general programs and policies in this report but refer to the literature (see Alliance et al. 1997; Bernow et al. 1999; Martin et al. 1999; Interlaboratory Working Group 2000; and Worrell, Bode, and de Beer 1997). With respect to technology-diffusion policies, there is no single instrument to reduce barriers; instead, an integrated policy accounting for the characteristics of technologies, stakeholders, and countries addressed is needed. RD&D projects often reduce risk and lower initial investment costs. Technology procurement programs such as the golden carrot lower the initial risk to technology developers by subsidizing the research and product development for more efficient technologies. Demand-pull programs seek to organize buyer groups to create a more ready market for emerging technologies. Financial incentive programs such as tax credits or other financial instruments seek to underwrite the first cost of the investment by the purchaser. All of these policies aim to more rapidly increase the share of the technology than would have been the case in the absence of a policy instrument (Worrell, Bode, and de Beer 1997).

The Future Of Energy Use In Industry

In recent years, several studies have been undertaken related to modeling and forecasting the future of industrial or manufacturing energy use in the United States. EIA uses the National Energy Modeling System (NEMS) model to develop base case and alternative scenarios for energy consumption in various economic sectors, including industry. Other forecasting studies have incorporated various levels of policies usually working off of the NEMS baseline as the business-as-usual case. Table 6 identifies some of the main characteristics of industrial energy use forecasting studies developed in recent years.

In Italy, South Korea, and Japan, for example, 96 percent or more of steel was continuously cast by 1993, whereas only 85 percent was continuously cast in the United States at that time (IISI 1996).

14

Emerging Energy-Efficient Industrial Technologies Table 6. U.S. Industrial Energy Use Forecasting Studies
Study Annual Energy Outlook 2000 (AEO 1999) Scenarios for a Clean Energy Future (Interlaboratory Working Group 2000) Energy Innovations (Alliance to Save Energy, et al. 1999) Americas Global Warming Solutions (Bernow et al. 1999) Model NEMS NEMS (with modified baseline and policy inputs) NEMS (with the use of LIEF model to estimate policy impacts) LIEF (benchmarked in 1998 to NEMS) Key Policies Reference case incorporates existing policy trends. Includes sensitivity cases to low and high economic growth and low and high oil prices. Voluntary agreements, information programs, investment-enabling programs, regulations, R&D programs, cap and trade system (advanced scenario) Incentives, increased R&D, increased use of recycled feedstock, overcoming barriers to combined heat and power production Technical assistance, information programs, tax credits, R&D

In its Annual Energy Outlook, the EIA main reference case forecasts that primary industrial energy consumption will grow by 0.9 percent per year between 1998 and 2020 from 35.0 Quads (36.9 EJ) to 42.2 Quads (44.5 EJ) (EIA 1999). Similar conclusions are reached with the baseline or business-as-usual forecasts from the Scenarios for a Clean Energy Futures, Energy Innovations, and Americas Global Warming Solutions, which all forecast baseline growth rates from 0.7 to 0.9 percent per year, even though the initial consumption levels vary by study. Others argue that the rise of the Internet economy is more rapidly supplanting our demand for traditional manufactured goods than we currently acknowledge and we may begin to see much slower growth in the business-as-usual case.9 In the three policy change forecasts, there is a consensus that through various policy instruments and further reducing industrial energy consumption, various policies can make a difference in accelerating the rate of technology adoption. All three suggest that it is possible to achieve a future in which industry consumes the same energy as today but has managed to continue to grow economically. The models used to forecast industrial energy use do not include methodologies for technological choice; instead, they include estimated parameters that simulate technological improvements. Therefore, the approaches taken by these models have limited use in exploring how policies can accelerate technology adoption, but can provide useful information on the impacts of accelerated adoption. Enhancing the models to incorporate technology choice is a fertile area for future research. Not all efficiency technologies are potential future winners. Increasing the rate of adoption of efficient technologies often requires additional investment of time and resources to identify, assess, and integrate these new technologies into the marketplace. Our report contributes to this process by identifying what we believe to be some of the key emerging energy-efficient technologies that have the potential to help accelerate U.S. industry towards a more rapid improvement of energy efficiency than would be the case in business-as-usual circumstances. These technologies also can help transition our industrial base to the clean production approaches needed in the near future. The selected emerging energy-efficient industrial technologies would be in the start of the S-curve, as depicted in Figure 1.

Laitner (2000b) argues in particular argue that mainstream forecasts may be overestimating U.S. energy and carbon dioxide emissions in the year 2010 by up to 5 percentwhile significantly underestimating overall U.S. economic growth. In the industrial sector, the production of materials needed for construction (stone, clay, and glass materials) and paper production are particularly likely to face growing competition from the Internet.

15

LBNL and ACEEE

III. METHODOLOGY AND APPROACH

As we noted in the introduction, emerging technologies are defined as those technologies that are either currently under development, but close to commercialization (i.e., could be reasonably expected to enter the marketplace by 2005), or have a low market penetration (i.e., it is commercialized but has achieved a less than 5 percent market share) or are pre-commercial. The set of emerging technologies evolves over time as industry continues to learn about newer and more improved manufacturing methods, and new technologies emerge from the laboratory and enter the marketplace.

Preliminary Screening of Technologies

The first step in our technology assessment was to collect limited information on a broad universe of potential technologies. Our key sources of information included the U.S. Department of Energy, Office of Industrial Technologies; the Centre for the Analysis and Dissemination of Demonstrated Energy Technologies (CADDET); LBNL and ACEEE reports; and information from the sponsors of the project. Due to time constraints we did not attempt to collect any primary data on technology performance. It is important to note that it is likely that promising technologies were over looked. Oftentimes, no one source provided all of the information we sought for our assessment (energy use, energy savings compared to average current technology, investment cost, operating cost savings, lifetime, etc.). We therefore made our best effort to combine readily available information along with expert judgement where necessary. Our three key preliminary screening criteria, described in detail below, were: Potential for energy savings Ratio of first costs of new technology to replacement cost for existing processes Other benefits From these screening criteria we developed an initial scoring ratingwith a maximum rating of 100 pointsto help select technologies for final screening. We also noted whether a technology has a low market penetration or whether the technology is pre-commercial. Below we discuss the rating criteria and scoring criteria, and we then present our initial screening results in Table 8.

Rating Criteria
Potential for Energy Savings We sought to identify technologies that could have a large potential impact in terms of saving energy. These may be technologies that are specific to one process or one industry sector, or so-called crosscutting technologies that are applicable to a variety of sectors. High energy savings technologies were rated as those whose primary energy savings in 2010 could be 0.1 percent or greater of primary energy consumption in 1994. We chose 1994 as our energy base-year since that is the year in which the most recent detailed manufacturing energy consumption statistics were published (EIA 1997)10. Medium energy savings are those in which the industry-wide potential energy savings in 2010 could be between 0.01 percent and 0.009 percent of primary industrial energy consumption. Low energy savings are those in which potential savings could be below 0.009 percent. In estimating primary energy savings, we first identified the specific energy savings of each technology by comparing the energy used by the emerging technology to the energy required by current processes. Our second step was to scale up this savings estimate to see how much energy savingsfor industry overallthis technology would achieve by 2010. For the most part, we derived specific energy savings information from various technology assessment studies noted above. In scaling up the technology-specific energy savings to achieve a rough national estimate of energy savings, we relied on our general knowledge of the various industrial processes to which this technology could be applied. We also took into account structural limitations to the penetration of the technology. For
10

1998 energy data will not be available from EIA until after the conclusion of this project.

16

Emerging Energy-Efficient Industrial Technologies example, a mechanical pulping technology is limited by the total throughput of pulp to which the measure can be applied. Additionally, we recognized that market penetration, in the absence of significant policy support, can take time given the slowness of stock turnover in many industrial facilities. Our estimate of national energy savings is merely meant to be indicative of the relative impact of any particular technology on a national scale. Given time and resource constraints, our goal in calculating potential energy savings in the pre-screening stage was simply to estimate the relative impact (high, medium, low) of any particular technology, not specifically to provide a highly detailed calculation of energy savings. Investment Cost for New Technology/Replacement Cost for Existing Processes Because of the time-consuming nature of collecting detailed cost data, our goal for the preliminary screening was to develop a shorthand indicator of the relative expense of investing in the emerging technology. An emerging technology that's first cost was estimated to be 1 to 5 times or more expensive than replacement using existing practices was given a rating of "3," while a technology that's first cost was estimated as the same or cheaper was rated as "2" or "1," respectively. While such an approach can be useful as a pre-screening device, we realize that it has limitations, primarily that a more expensive technology may still be a good investment particularly if it can provide significant energy savings over a long period of timeThus we have further quantified the economic benefits of each technology, using metrics such as the cost of conserved energy and the internal rate of return. Other Benefits Usually, energy-efficient technologies are not purchased solely for their energy benefits but also because of other, non-energy benefits accrued from their use. We grouped these non-energy benefits into four categories: environmental, productivity, product quality, and safety (see below for details). These additional benefitsand not the energy savingscan often be the determining factor in deciding to purchase the technology. We judged how these other non-energy benefits would affect the technology choice decision. For technologies where these non-energy benefits were thought to be the dominant factor in selecting the technology, a rating of compelling was used. For a technology with non-energy and energy benefits that were judged of equal importance, a rating of significant was applied. For technologies with non-energy benefits, but energy savings drove the technology decision, a rating of somewhat was used. If the technology has no significant non-energy benefits that would influence the technology selection, a rating of none was applied. This terminology was also used in evaluating the non-energy benefits in the final profiles. Environmental benefits refer to reductions in air emissions (e.g., sulfur dioxide, nitrogen oxides, particulate matter, dust) or reductions in waste streams that result from the use of the emerging technology. Productivity improvements can often result if the emerging technology reduces down-time required for operation and maintenance, reduces operation and maintenance costs, or increases yield. Product quality improvements and safety benefits often result from the fact that process energy requirements are more carefully controlled and monitored. In the preliminary screening we noted any other benefits that accompanied a particular technology or measure, but did not attempt to quantify them. We included the presence of other benefits in our rating system.

Rating Preliminary Technologies

Based on the initial screening process, the most attractive technologies are those that: (1) have a high potential energy savings; (2) have lower relative first costs compared to existing technologies; and (3) have other significant benefits. While some technologies score high on all of these characteristics, most have a mixed score. We formalized this approach in a very simple rating system shown in Table 7 below.

17

LBNL and ACEEE

Table 7. Preliminary Screening Rating System

High Medium Low None Energy Savings 40 20 10 N/A Cost 10 20 30 N/A Other Benefits 30 20 10 0

As the rating system above indicates, a technology with high potential energy savings, low cost, and the presence of other high or compelling other benefits would be given a rating of 100. The lowest score a technology can receive is "20," where energy savings is low ("10"), cost is high ("10") and there are no other benefits ("0"). Table 8 below identifies all the technologies considered in our preliminary screening. In some cases we abbreviated the technology description for space considerations.

Detailed Assessment of Selected Emerging Energy-Efficient Industrial Technologies

After establishing a preliminary selection of technologies based on the preliminary screening, our next step in the technology assessment process was to convene a workshop that brought together the research staff and sponsors to review the preliminary technology selection and to refine the criteria for the detailed assessment. Our aim was to develop a broad enough range of criteria that would allow for a thoughtful evaluation and presentation of the technologies, while at the same time not trying to dilute the evaluation with too many data points, which often may rely on too much speculative information. The workshop was held in Washington, D.C. on April 27, 2000, and helped the research and sponsor group develop a consensus for moving forward with the detailed assessment of the short list of technologies. The detailed assessment consists of a 1-2 page write-up of the technology or measure and a final evaluation table detailing 8 main areas:
Market and sectoral information Base-case information New measure information Energy savings information and analysis Cost information and analysis Key non-energy factors Evaluation Sources and Contacts

The write-up describes the measure, including the issues surrounding the analysis of energy savings and cost-effectiveness, key non-energy factors, and background on the evaluation of the technology, and recommends next steps advance the technology. Market information includes a description of the industries to which the technology/measure is applicable (e.g., cement, iron and steel, Crosscutting). We also provide information on the end-uses for the technology (i.e., process, process heating, process cooling, electrochemical processes, utilities, ventilation and space conditioning, lighting, motor and drives), the principal energy types used by the technology (i.e. electricity, gas, oil, coal, biomass, waste fuel, fuels [multiple fossil fuels], other), and the primary market segment (i.e., retrofit, new, replace on failure, original equipment manufacturers). There may be more than one market segment for which the technology is applicable; we used our judgement to identify the most predominant segment. Finally, we also included a key output driver or the energy consumption for our 2015 base-case related to that sector. For example, a steel furnace technology would have as a 2015 basecase reference value the expected steel output for that year. 2015 Base-case includes a description of the current technology or practice, the volume of production or annual operating hours associated used in the baseline and savings analysis, and baseline energy consumption for the existing process (i.e., fuels, electricity, primary). New measure information includes a description of the new technology, energy consumption information (i.e., fuel, electricity, primary energy), information on the current status of the technology (i.e., commercialized, field testing, prototype, research), the expected date of commercialization (if known), and the lifetime of the technology.

18

Emerging Energy-Efficient Industrial Technologies Table 8. Technologies and Measures Considered in Preliminary Screening Analysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 Technology/Measure Ceramic Filters Ramex Tunneller Variable Wall Mining Machine Vibration Fluidized Bed Membrane Technology - Food Electron Beam Sterilization Heat Recovery - Low Temp. Cooling And Storage Heat Recovery Food High Temp. Freeze Concentration Supercritical Extraction Controlled Atmosphere Packaging 4 Or More Effect Evaporator Efficient Cooling Systems Condi-Cyclone Dryers Heat Pump Dryer Ultrasonic Dying Suction Slot Dewatering Direct Contact Water Heating Textile Heat Recovery Dyeing Vacuum System Automated Dyebath Reuse Membrane Technology Textiles Improved Drying Systems Direct Electrolytic Causticizing High Consistency Forming Black Liquor Gasification Impulse Drying Heat Recovery - Paper Dry Sheet Forming Condebelt Drying Flotation Deinking/Stickies Removal Bacterial Reduction Of Sulfur Press Drying Biopulping Fluidized Bed For Biomass Waste Air/Steam Impingement Drying Freeze Concentration Mill Effluent Fiber Loading Equipment/PCC Thermodyne Pulp Dryer Pressurized Groundwood-Super Direct Drying Cylinder Firing Molten Metal Paper Dryer Multi-Port Drying Cylinder Fluidized Bed Heat Exchanger New Refractory Materials Heat Recovery -Printing New Catalysts Clean Fractionation Levulinic Acid From Biomass Liquid Membranes Gas Membranes Heat Recovery Technologies Oxidation Of Benzene To Phenol Corn Fiber Fractionation Sector Mining Mining Mining Mining Food Processing Food Processing Food Processing Food Processing Food Processing Food Processing Food Processing Food Processing Food Processing Food Processing Food Processing Food Processing Textile Textile Textile Textile Textile Textile Textile Lumber And Wood Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Pulp And Paper Printing Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 Technology/Measure Oxy-Burners (Chemicals) Silicones From Sand Chlorate Cathodes for ClO2 Electrodeionization Advanced Chlorine Cells Advanced Cleanroom HVAC Selective Cracking-Ethylene Catalytic Autothermal Oxydehydrogenization Advanced Reactor Design-Methanol Advanced Recovery-Fractionation Melt Crystalization-Benzene Alkane Functionalization Catalysts Dividing Wall Column-Olefins Autothermal Reforming-Ammonia Membrane Reactor /Ammonia Adiabatic Pre-Reformer (Ammonia) Ammonia Process Control Membrane Reactor/Steam Reforming Ammonia Synthesis Using Sorbents Biodesulfurization Fouling Minimization Liquid Membranes In Refining Low Profile FCC Ammonia Absorption Refrigeration Hydrogen Purification Froth Flotation Plastics Recovery Heat Recovery In Plastics Water As Cooling Refrigerant Fluidized Bed/Plastics Recovery Tunnel Kiln Plastics Roller Kiln Innovative Tunnel Kiln Process Control-Glass Tanks Ion-Exchange System - Float Glass New Glass Melting Technologies Efficient Burners For Glass Furnaces Pre-Heat Technologies-Glass Electric Forehearth/Indirect Cooling 100 percent Recycled Glass Cullet Cogen--Exhaust Gas Drying Of Blast Furnace Slag For Blended Cements New Refractory Materials - Cement Fluidized Bed Kiln Mineral Polymers Heat Recovery For Cogeneration Advanced Communition High Efficiency Roller Mills Near Net Shape Casting/Strip Casting New EAF Furnace Processes Smelt Reduction Processes Oxy-Fuel/Reheat Furnace BOF Gas/Sensible Heat Recovery High Levels Of PCI Coke Oven Gas Cogeneration "Pickliq" HCL Regeneration Intelligent Inductive Processing Sector Chemicals Chemicals Chemicals Chemicals Chemicals Cross-Cutting Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Refining Refining Refining Refining Refining Refining Plastics Plastics Plastics Plastics Plastics Ceramics Bricks/Tiles Glass Glass Glass Glass Glass Glass Glass Cement Cement Cement Cement Cement Cement Cement Iron and Steel Iron and Steel Iron and Steel Iron and Steel Iron and Steel Iron and Steel Iron and Steel Iron and Steel Iron and Steel

19

LBNL and ACEEE

Table 8. Technologies And Measures Considered In Preliminary Screening Analysis (Continued)

111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 Technology/Measure Improved EAF Refractories Coke Dry Quenching Non-Recovery Coke Ovens Waste Oxides Recycling In Heat Recovery In Sinter Plants Scrap Pre-Heating Recuperative Burners Steel Strapping/Mini-Mill Improved Recycling Efficient Cell Designs Inert Anodes Advanced Forming Pot Lining Additive Improve Casting Furnace Fy-Gem Grain Refinement Twin Chamber Pulp Lifter Solvent Recovery Using Nitrogen Continuous Melt Silicon Advanced Polysilicon Adv. Electrogalvination Advanced Coating Processes Sensors And Controls Low NOx Burners Advanced Lubricants Motor Diagnostics Compressed Air Management Advanced CHP Turbines Advance ASD Designs Advanced Recip. Engines Advanced Compressor Controls Advanced Lighting Design Membranes- Wastewater Sector Iron and Steel Iron and Steel Iron and Steel Iron and Steel Iron and Steel Iron and Steel Iron and Steel Iron and Steel Aluminum Aluminum Aluminum Aluminum Aluminum Aluminum Aluminum Alumina Chemicals Electronics Electronics Autos Autos Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 Technology/Measure Pump Efficiency Pinch Analysis Switched Reluctance Motor Advanced Lighting Anaerobic Waste Water Treatment Motor System Optimization Fuel Cells Microturbines Metalax Stress Relief Method Energy Management Systems Clean Energy Systems Heat Pumps Written Pole Motor Heat Recovery Turbine Copper Rotor Motor Permanent Magnet Motor Efficient Transformers General Heat Recovery Molten Metal Filtering GFX Drainwater Heat Recovery High-Efficiency Welding Furnace Process Modeling Unconventional Yield Improvement Simulation Programs New Metal Heating Thermal Storage Cooling Low Friction Working Fluids Recuperative Burners Oxy-Fuel Burners Copper Motor Rotors Tube Feeder Sector Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting Cross-Cutting

Savings information identifies electricity, fuel, and primary energy savings for a typical application of the new technology relative to the reference technology. The analyst made an assessment of the rate at which the technology is expected to penetrate the market. We used a simplified, uniform penetration rate to represent a plausible estimate of the market penetration of each measure during the analysis period. We assigned measures to one of three standard penetrations rates (high, medium, and low). These rates are tied to assumptions how readily is the market likely to adopt the measures. In general, the penetration rates assume successful programs, and that the technologies compete against the reference technology but not against each other for the market share. While the market diffusion will be sigmoid as discussed in Section II: Overview of U.S. Industrial Energy Use, we assumed linear penetration. Thus, when the market requires a high level of intervention to successful adopt the measure, annual market-penetration rate was assumed to be 5 percent, with an ultimate penetration of 30 percent in 2015. For measures that require medium market intervention required, the annual market-penetration rate was assumed to be 7.5 percent, for an ultimate penetration of 45 percent in 2015. Where the intervention is low (i.e., the technology is likely to be adopted with little intervention), we assume that market penetration rates will be high: 10 percent per year to an ultimate penetration of 60 percent in 2015. These penetration rates begin in the first year after commercialization, or 2001 for those technologies that are already commercialized. For measures with retrofit as the predominate mode of market deployment, the portion of the market that can be impacted by a technology is assumed to be 100 percent. For replacement (i.e., replace on failure), the portion is assumed to be the period of the study (15 years) divided by the measure life. For new construction, it is the growth in capital investment for the target industry divided by the anticipated total installed capital value in 2015. Finally we estimated the share applications that the technology captures by 2015 (e.g., for which the measure is technically feasible and cost-effective to the end-user on a life-cycle cost basis). Feasible applications refers to the percentage of the total market that the technology is estimated to capture by 20

Emerging Energy-Efficient Industrial Technologies 2015. Any other key assumptions for savings potential are noted in the spreadsheet and profiles narrative, and the total 2015 fuels, electricity, and primary energy savings is calculated. Our 2015 energy savings estimate is relative to the 2015 base-case information identified in the market information section of the table. Cost information and analysis provides an estimate of the technology or measures investment cost ($/unit output), whether that investment is incremental or full cost, and any change in operations and maintenance cost ($/unit output) for adopting the technology. We include three measures of cost-effectiveness: cost of conserved energy11 for electricity, fuels, and primary energy, simple payback12 for the investment relative to the reference technology (years), and internal rate of return (IRR percent).13 Simple payback and internal rate of return are metrics that are often used by industries and financial analysts, while cost of conserved energy has been useful as a cost-effectiveness indicator for the policy community.14 Key non-energy factors are those factors that can significantly affect the decision to purchase a technology. These include the presence of other benefits (productivity, quality, environmental, other [i.e. safety]), and to what extent the technology is currently being promoted. In the evaluation section of this table, researchers identify the major market barriers that could impede the successful implementation of this technology. The technologys likelihood of success (high, medium, and low) is rated based on it its cost-effectiveness, key non-energy factors, and major market barriers. We suggest what next steps are appropriate to accelerate the deployment of the technology. Finally, the analyst provides an assessment of the overall quality of the data used in the analysis using a rating of excellent, good, fair or poor. Finally, we provide information on sources for the key data collected and principal contacts for those interested in follow up analysis. Treatment of Utility Technologies A slightly different approach was used to analyze the power generating technologies in this study. Each of the utility technologies has a unique capacity characteristic, ranging from microturbines with an electric generation capacity of below 300 kW to industrial CHP turbine systems with capacities approaching 50 MW. Therefore a methodology was developed using as a reference the primary energy required to generate 1kWh of grid-supplied electricity at an average delivered efficiency the projected 2015 grid efficiency of 33.4 percent (EIA 1999). This reference case was compared with the fuel and primary energy required to generate 1 kWh of electricity from utility technologies based on the efficiency of the technology. For each of the measures, the electricity savings is 1 kWh (the amount of electricity that would otherwise have been purchased from the grid). The fuel and primary energy savings are dependent on the respective efficiency of each measure. Using this approach, the relative energy savings of each technology was determined on a consistent basis. In determining the cost-effectiveness of the utility technologies, an average industrial electricity price of $0.039/kWh was used. This value is the projected 2015 industrial price for electricity in the AEO 2000 reference case (EIA 1999). Electricity prices vary wildly by region, service provider, and industrial segment. In reality, industrial facilities can pay a price of anywhere between $0.01 to $0.14 on average. This price is determined by any agreements that are established between a manufacturing facility and the local power supplier. The actual rate also depends on several other factors such as time of day (peak demand charges). The electricity price profoundly effects the economics of on-site electricity generation
11 The cost of saved energy is calculated by: (ACAP+ O&M)/E where: ACAP is the capital cost of technology annualized as a loan for the life of the measure, at the default discount rate (i.e., 15 percent), O&M is the change in annual operating cost, and E is the annual non-energy energy savings. 12 The simple payback is calculated by: CAP/(EC-O&M) where: CAP is the capital cost of the technology, EC is cost of the energy saved based on 2015 projected national energy price (EIA 1999), and O&M is the change in annual, nonenergy operating cost. 13 The IRR is calculated from a analysis based on the initial capital cost and the annual cash flow of energy cost savings and change in non-energy O&M for the life of the measure discounted at a rate of 15 percent. 14 While we calculate a general IRR and cost of conserved energy based on average energy savings, we realize that the attractiveness of the investment is very plant specific and that the attractiveness of an individual investment may look different from the technology viewed in national terms.

21

LBNL and ACEEE technologies. For example, a 65 percent efficient 800 kW gas reciprocating engine has a simple payback period of 8 years when compared to purchased grid electricity at $0.039/kWh. At a purchased electricity price of $0.12/kWh, the simple payback drops to less than a year. Ancillary benefits can make electric generating technologies more economically attractive as well. Certain industrial sectors, such as the pharmaceutical, semiconductor and microelectronics sectors, demand highquality power, often with a reliability target of six-nines (i.e., 99.9999 percent). While the average reliability of the U.S. electric grid hovers near 99 percent, this is not reliable enough for many of these applications. Many of these industries must employ stand-by power systems to meet their requirements. No accepted methodology exists for determining this ancillary reliability benefit of an on-site power generating technology. In principle this can be determined by calculating the revenues that would be lost during a grid outage. This value however is highly site specific. In high-value applications, such as semiconductor or pharmaceutical manufacturing and data-centers where loss revenues can easily exceed $1 million/hour, this consideration can make generating technologies, even with high initial capital costs, economically attractive (Elliott and Spurr 1999). All the utilities, except for industrial CHP turbine systems, were evaluated in electricity generation only mode. It is also worth noting that all of the utility technologies in this study become more efficient when operated with heat recovery (cogeneration or CHP mode). The increased efficiency allows for a higher rate of return as well as lower combustion-related emissions per unit of generated energy. This option is particularly attractive in industries with high large process heat or space conditioning demands such as the food, chemicals, paper, and microelectronics industries. In conclusion, it should be noted at times the lack of reliable data (especially if the technologies are precommercial) can impede a thorough assessment. When this is the case, we note this and do not attempt to stretch the analysis beyond its ability to be supported by the underlying data. Our goal is to provide as thorough an assessment of the various emerging technologies as possible, given the available information.

22

Emerging Energy-Efficient Industrial Technologies

IV.

SUMMARY OF FINDINGS

Introduction
The industrial sector is a significant energy user, consuming nearly 40 percent of U.S. primary energy resources and producing about a quarter of GDP. The development and use of cleaner, more energyefficient technologies can help limit the negative environmental impacts associated with many industries while enhancing productivity and reducing manufacturing costs. This study aims to identify and evaluate emerging energy-efficient technologies for use in the industrial sector. In this section we rank these technologies by overall energy savings, electricity savings, fuel savings, share of sector savings, and environmental benefits. We conclude with suggested actions to support the development of these technologies and evaluations of the likelihood of technologies to succeed in the marketplace.

Summary of Technology Characterizations

Based on a comprehensive literature review and the application of some basic initial screening criteria (see Section 3:Methodology and Approach), we identified and developed profiles for 54 technologies. The technologies themselves range from highly specific technologies that can be applied in a single industry to more broadly Crosscutting technologies, which can be used in many industrial sectors. Table 9 summarizes the results of the individual analyses. We evaluated energy savings in two different ways. The third column in Table 9, Total Energy Savings, shows the amount of total manufacturing energy that the technology is likely to save in 2015 in a businessas-usual scenario. The fourth column, Sector savings, shows the savings relative to expected energy use in the particular sector. We believe that both metrics are useful in evaluating the relative savings potential of various technologies. Economic evaluation of the technology is identified in the summary table as Simple Payback, defined as the initial investment costs divided by the value of energy savings less any changes in operations and maintenance costs. We chose this measure since it is frequently used as a shorthand evaluation metric among industrial energy managers. Payback periods for the technologies range from the immediate to 20 years or more. Of the 54 technologies profiled, 31 have estimated paybacks of 3 years or less, with six paying back immediately. The individual profiles also include estimations of internal rate of return and cost of saved energy. Energy savings are often not the determining factor in the decision to develop or invest in an emerging technology. Over two-thirds of these technologies not only save energy but yield environmental or other non-energy benefits. These non-energy benefits include: increases in productivity, worker safety, product quality, and capacity; and reduced capital and operating costs. Technologies are not simply developed and then seamlessly enter existing markets. The acceptance of emerging technologies is often a slow process that entails active research and development, prototype development, market demonstration, and other activities. In Table 9 we summarize the recommendations for the primary activities that should be undertaken to increase the rate of uptake of these technologies. Over half have already been developed to prototype stage or are already commercial but require further demonstration and dissemination. While data on many of the technologies were readily available and appeared reliable and self-consistent, for some technologies the analyst faced significant challenges. Each analyst judged the relative quality of data ranging from poor to excellent. The data quality judgement is also given in Table 9 for each technology. Below we review the key parameters in greater detail. In the following sections we evaluate how the technologies can be grouped relative to each parameter of interest.

23

LBNL and ACEEE

Table 9. Summary of the Profiled Energy-Efficient Emerging Industrial Technologies

Technology Advanced forming Efficient cell retrofit designs Improved recycling technologies Inert anodes/wetted cathodes Roller kiln Clean fractionation - cellulose pulp Gas membrane technologies-chem Heat recovery technologies chemi Levulinic acid from biomass Liquid membrane technologieschem New catalysts Autothermal reforming-Ammonia Plastics recovery Continuous melt silicon crystal growth Electron beam sterilization Heat recovery - low temperature Membrane technology - food Cooling and storage 100% recycled glass cullet Black liquor gasification Condebelt drying Direct electrolytic causticizing Dry sheet forming Heat recovery paper High Consistency forming Impulse drying Biodesulfurization Fouling minimization BOF gas and sensible heat recovery Near net shape casting/strip casting New EAF furnace processes Oxy-fuel combustion in reheat furnace Smelting reduction processes Ultrasonic dying Variable wall mining machine Hi-tech facilities HVAC Advanced lighting technologies Advanced lighting design Advance ASD designs Advanced compressor controls Compressed air system management Motor diagnostics Motor system optimization Pump efficiency improvement Switched reluctance motor Advanced lubricants Anaerobic waste water treatment High efficiency/low NOx burners Membrane technology wastewater Process Integration (pinch analysis) Sector Aluminum Aluminum Aluminum Aluminum Ceramics Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Chemicals Plastics Electronics Food Food Food Food Glass Pulp & Paper Pulp & Paper Pulp & Paper Pulp & Paper Pulp & Paper Pulp & Paper Pulp & Paper Pet. Refining Pet. Refining Iron & Steel Iron & Steel Iron & Steel Iron & Steel Iron & Steel Textile Mining Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Total Energy Savings1 Medium High Medium High Medium Low Low Medium Low Low Medium High Medium Medium High Medium High Medium Medium High High Low Medium High Medium High Medium High Medium High High High High Medium Low Medium High High High Medium High Low High High Medium Medium Medium High High High Sector Est. Simple Savings2 Life Payback3 Medium 15 Immed. High 15 2.7 Medium 15 4.5 High 10 4.0 High 30 1.9 Low 15 1.9 Low 15 10.2 Medium 10 2.4 Low 20 1.5 Low 10 11.2 Medium High Medium High High Medium High Medium High High Medium Low Medium Medium Medium Medium Medium High Medium High High Medium High Medium Low High High High Medium Low High Low High High Low Medium Low Low Medium Low 20 30 20 7 10 25 10 15 25 30 20 10 20 20 20 20 15 15 30 20 40 10 40 10 25 20 4 20 15 15 1.5 15 10 10 15 0.5 20 20 10 15 7.9 3.7 2.8 Immed. 19.2 4.8 2.2 2.6 2.0 1.5 65.2 N/A 48.3 3.9 Immed. 20.3 1.8 N/A 14.7 Immed. 0.3 1.2 Immed. 0.3 10.6 4.0 1.3 3.0 1.1 0.04 0.4 Immed. 1.5 3.0 7.4 0.1 0.8 3.1 4.7 2.3 Environ. Other4 Benefits Benefits Next Steps None P R&D Somewhat P Demo Significant P Demo Significant P, Q R&D Significant P Demo Significant P, O Demo Significant Q, O Dissem None P, O Dissem, demo Significant P, O Demo Significant O Dissem Somewhat Significant Compelling Somewhat None None Somewhat Somewhat Significant Somewhat None Somewhat Somewhat Somewhat Somewhat None None None Significant Somewhat Somewhat Significant P P Q, P P, Q P, Q P, Q O P, S P, Q P, Q Q P, S P, Q P, Q Q P P P,Q P P R&D Dissem Demo R&D Data Quality Good Fair Good Good Fair Good Good Fair Good Good Fair Fair Fair Excellent

R&D Fair Dissem Fair Dissem, R&D Fair, Poor Dissem, demo Fair Demo Good Excellent Demo Demo Good R&D Good Good R&D, demo Demo Good Demo Fair Good Demo R&D, demo Excellent R&D Fair Good Dissem R&D Good Field test Fair Fair Field test

Significant P Demo Good Fair Compelling P, Q Demo None P,S Demo Fair None P Dissem. Fair None Q, P, O Dissem, demo Excellent None P, Q, O Dissem, demo Good None P, Q R&D Good Good None P, Q Dissem None P, Q Dissem Good None Somewhat None None Significant Significant Significant Somewhat Somewhat P, Q P, Q P, Q P, Q P, Q O P P P

Dissem, demo Good Dissem, train Good Dissem, train Good R&D Good Dissem. Good Dissem, demo Good+ Dissem, demo Poor Dissem, R&D Fair, Poor Fair Dissem. R&D, demo, Fair Sensors and controls Crosscutting High Medium 10 2.0 Somewhat P,Q dissem Advanced CHP turbine systems Crosscutting High High 10 6.9 Significant P, Q Policies Excellent Advanced reciprocating engines Crosscutting High High 7 8.3 Limited P, Q, O R&D, demo Excellent Fuel cells Crosscutting High High 7 58.6 Significant P, Q Demo Good Good Microturbines Crosscutting High Medium 7 Never Somewhat P, Q, O R&D, demo Notes: 1. "High" could save more than 0.1% of manufacturing energy use by 2015, medium is 0.01 to 0.1%, and low is less than 0.01%. 2. "High" could save more than 1% of sector energy use by 2015, medium is 0.1 to 1%, and low is less than 0.1%. 3. Immed is immediate. 4. P is productivity, Q is quality, S is safety, O is other.

24

Emerging Energy-Efficient Industrial Technologies

Energy Savings
Depending on the particular technology and application, the technologies will reduce electricity consumption, fuel consumption, or both. Table 10 presents the 28 technologies having high total energy savings, rated according to their primary energy savings (i.e. accounting for losses in the production and delivery of electricity). These savings represent the estimated 2015 implemented savings under a businessas-usual scenario (i.e. excluding policy efforts to stimulate adoption of a specific technology). As would be expected, the Crosscutting technologies (motor systems, lighting, utilities) save the largest amount of primary energy, followed by selected specific technologies in energy-intensive sectors (steel, petroleum, paper, aluminum, and chemicals). However, this does not mean that sector-specific technologies should be overlooked, as many of these may save substantial amounts of energy in a particular sector, or may have important additional benefits (see below). Table 10. Projected 2015 Implemented Primary Energy Savings Potential
Technology Motor system optimization Advanced reciprocating engines Compressed air system management Pump efficiency improvement Advanced CHP turbine systems Advanced lighting design Advanced lighting technologies Fuel cells Near net shape casting/strip casting Sensors and controls Fouling minimization Membrane technologywastewater Microturbines Black liquor gasification Efficient cell retrofit designs Process Integration (pinch analysis) Autothermal reformingAmmonia Condebelt drying Electron beam sterilization Inert anodes/wetted cathodes Smelting reduction processes Impulse drying Membrane technologyfood Advance ASD designs New EAF furnace processes Heat recoverypaper High efficiency/low NOx burners Oxy-fuel combustion in reheat furnace Code Motorsys-5 Utilities-2 Motorsys-3 Motorsys-6 Utilities-1 Lighting-2 Lighting-1 Utilities-3 Steel-2 Other-5 Refin-2 Other-3 Utilities-4 Paper-1 Alum-2 Other-4 Chem-7 Paper-2 Food-1 Alum-4 Steel-5 Paper-7 Food-3 Motorsys-1 Steel-3 Paper-5 Other-2 Steel-4 Sector Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Iron and steel Crosscutting Petroleum refining Crosscutting Crosscutting Pulp and paper Aluminum Crosscutting Chemicals Pulp and paper Food processing Aluminum Iron and steel Pulp and paper Food processing Crosscutting Iron and steel Pulp and paper Crosscutting Iron and steel Primary Energy TBtu (EJ) 1502 (1585) 777 (820) 563 (594) 502 (530) 484 (510) 408 (430) 231 (244) 185 (195) 138 (146) 136 (143) 123 (130) 118 (125) 67 (71) 64 (68) 46 (49) 38 (40) 38 (40) 34 (36) 34 (36) 34 (36) 32 (34) 30 (32) 27 (28) 25 (26) 24 (25) 22 (23) 21 (22) 21 (22)

Electricity is a unique energy source, with significant emissions and a large infrastructure supporting its generation and delivery. Many industries, including electric utilities, will find it important to focus on technologies that save electricity. Table 11 identifies the top 20 technologies in terms of electricity savings. Our estimate of savings is based on an economically feasible market penetration in 2015 under business-asusual conditions. As Table 11 indicates, the Crosscutting technologies concerning motor systems, lighting, and utilities are expected to have the most significant impact in terms of savings along with selected sectorspecific technologies. The most important sector-specific technologies are black liquor gasification (a potentially large self-generation technology in the pulp and paper sector) and technologies that reduce electricity use in the aluminum and electric arc furnace/secondary steel sectors. According to EIA, the total forecast of electricity use for the U.S. industrial sector in 2015 is 13,000 TWh (EIA 1997). While the top technology only represents 1 percent of total forecast electricity use, this is still a significant amount, representing $7 billion in electricity expenditures alone. Since electricity is one of the most high-quality and expensive energy inputs, small reductions in electricity expenditures can have a large impact on reductions in operating costs for various manufacturing establishments.

25

LBNL and ACEEE

Table 11. Projected 2015 Implemented Electricity Savings Potential

Technology Motor system optimization Advanced reciprocating engines Advanced CHP turbine systems Advanced ASD designs Compressed air system management Fuel cells Pump efficiency improvement Advanced lighting design Advanced lubricants Microturbines Advanced lighting technologies Black liquor gasification Advanced compressor controls Switched reluctance motor Near net shape casting/strip casting Efficient cell retrofit designs Inert anodes/wetted cathodes New EAF furnace processes Electron beam sterilization Biodesulfurization Code Motorsys-5 Utilities-2 Utilities-1 Motorsys-1 Motorsys-3 Utilities-3 Motorsys-6 Lighting-2 Motorsys-8 Utilities-4 Lighting-1 Paper-1 Motorsys-2 Motorsys-7 Steel-2 Alum-2 Alum-4 Steel-3 Food-1 Refin-1 Sector Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Crosscutting Pulp and paper Crosscutting Crosscutting Iron and steel Aluminum Aluminum Iron and steel Food processing Pet. Refining Electricity Twh 176 156 79 72 66 65 59 48 46 40 27 10 9 7 6 5 4 3 3 2

Table 12 identifies the top 14 technologies in terms of fuel savings. Unlike the electricity savings, the technologies highlighted in this table are primarily sector-specific; although Crosscutting technologies (membranes, sensors, process integration) show strong potential for fuel savings. The fuel savings below tend to reflect better utilization of low-quality or by-product fuels, improved heat recovery, or better direct application of process heating. Similar to electricity savings, no one technology represents an overwhelming proportion of industrial fuel consumption in 2015 (estimated at 31,960 TBtu), but each of the technologies in Table 12 represent a savings in energy expenditures between $30 and $900 million per year. Table 12. Projected 2015 Implemented Fuel Savings Potential
Technology Membrane technology wastewater Fouling minimization Sensors and controls Near net shape casting/strip casting Impulse drying Autothermal reforming-Ammonia Process Integration (pinch analysis) Membrane technologyfood Condebelt drying Smelting reduction processes Dry sheet forming Oxy-fuel combustion in reheat furnace High efficiency/low NOx burners Heat recoverypaper Code Other-3 Refin-2 Other-5 Steel-2 Paper-7 Chem-7 Other-4 Food-3 Paper-2 Steel-5 Paper-4 Steel-4 Other-2 Paper-5 Sector Crosscutting Pet. Refining Crosscutting Iron and steel Pulp and paper Chemicals Crosscutting Food processing Pulp and paper Iron and steel Pulp and paper Iron and steel Crosscutting Pulp and paper Fuel Savings TBtu (PJ) 276 (292) 123 (130) 111 (117) 86 (91) 64 (67) 38 (40) 37 (39) 36 (37) 34 (36) 32 (34) 28 (30) 23 (24) 21 (23) 20 (21)

The presentation in the last three tables focuses on aggregate energy savings. These technologies are dominated by measures that are applicable in a broad range of industries (e.g., Crosscutting) or in the dominant energy-intensive industries. In Table 13 we identify those technologies that offer important energy savings to their industry sector. These technologies have a high share of energy savings relative to energy use within the specific sector, where we define high as having a greater than 1 percent share of sectoral primary energy use. While the savings for a given technology may be modest in absolute terms, it may be important to the limited sector in which it is applicable. As noted in Section 2: Overview, there is significant regional variation in the distribution of industry. Many of the energy-intensive industries are concentrated in a few states. The industrial sector in some other states may be dominated by less energy26

Emerging Energy-Efficient Industrial Technologies intensive industries. Thus, states with concentrations of these industries may find these technologies of significant interest. Table 13. Implemented Savings Share of Sector Projected 2015 Energy
Technology Continuous melt silicon crystal growth Motor system optimization Roller kiln Hi-tech facilities HVAC Efficient cell retrofit designs Near net shape casting/strip casting Advanced reciprocating engines Inert anodes/wetted cathodes Compressed air system management Pump efficiency improvement Advanced CHP turbine systems Fouling minimization Advanced lighting design Electron beam sterilization Black liquor gasification Advanced lighting technologies Membrane technologyfood 100% recycled glass cullet for container glass Smelting reduction processes Fuel cells New EAF furnace processes Autothermal reformingAmmonia Code Electron-1 Motorsys-5 Ceramics-1 HVAC-1 Alum-2 Steel-2 Utilities-2 Alum-4 Motorsys-3 Motorsys-6 Utilities-1 Refin-2 Lighting-2 Food-1 Paper-1 Lighting-1 Food-3 Glass-1 Steel-5 Utilities-3 Steel-3 Chem-7 Sector Electronics Crosscutting Ceramics Crosscutting Aluminum Iron and steel Crosscutting Aluminum Crosscutting Crosscutting Crosscutting Pet. Refining Crosscutting Food processing Pulp and paper Crosscutting Food processing Glass Iron and steel Crosscutting Iron and steel Chemicals Share of sectoral savings 20.0% 11.5% 8.2% 7.4% 6.6% 6.4% 5.9% 4.9% 4.3% 3.8% 3.7% 3.4% 3.1% 2.0% 1.8% 1.8% 1.6% 1.5% 1.5% 1.4% 1.1% 1.0%

The Economics of Energy Savings

As we noted earlier in the section, payback is frequently used as a shorthand evaluation metric among industrial energy managers. For evaluating technologies from the perspective of cost-effective energy savings we use cost of saved energy as our metric. Thirty technologies have a cost of saved electricity of less than 4.6/kWh (the average industrial electricity price in 1996), with thirteen technologies having a net negative cost (i.e. costs of saved electricity is less than zero). Most of these were measures that achieved significant immediate energy savings while costing less than the reference technologies or not requiring significant capital outlays. Twenty-nine technologies had costs of saved fuel of less than $2.8/MBtu (the estimated average fuel price for industry in 1996), with 19 with values of zero or less than zero. These technologies share much in common with the top electricity savings technologies, and in fact 16 of the technologies appear on both lists.

27

LBNL and ACEEE

Table 14. Technologies with the Lowest Cost of Saved Electricity

Technology Advanced forming/near net shape technology Levulinic acid from biomass (biofine) Cooling and storage Advance ASD designs Advanced lubricants Anaerobic waste water treatment Dry sheet forming High Consistency forming Impulse drying BOF gas and sensible heat recovery Near net shape casting/strip casting New EAF furnace processes Microturbines Advanced compressor controls Sensors and controls Heat recovery technologies chemicals Advanced reciprocating engines Direct electrolytic causticizing Black liquor gasification Efficient cell retrofit designs Advanced CHP turbine systems Pump efficiency improvement Motor system optimization Compressed air system management Hi-tech facilities HVAC Inert anodes/wetted cathodes Advanced lighting technologies Oxy-fuel combustion in reheat furnace Variable wall mining machine Advanced lighting design Code Alum-1 Chem-4 Food-4 Motorsys-1 Motorsys-8 Other-1 Paper-4 Paper-6 Paper-7 Steel-1 Steel-2 Steel-3 Utilities-4 Motorsys-2 Other-5 Chem-3 Utilities-2 Paper-3 Paper-1 Alum-2 Utilities-1 Motorsys-6 Motorsys-5 Motorsys-3 HVAC-1 Alum-4 Lighting-1 Steel-4 Mining-1 Lighting-2 Cost of Saved Electricity $/kwh <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 0.000 0.001 0.006 0.007 0.008 0.008 0.008 0.010 0.010 0.012 0.015 0.022 0.029 0.034 0.035 0.041 0.046

Table 15. Technologies with the Lowest Cost of Saved Fuel

Technology Advanced forming/near net shape technology Clean fractionation-cellulose pulp Levulinic acid from biomass (biofine) Cooling and storage 100% recycled glass cullet for container glass Anaerobic waste water treatment Membrane technology wastewater Black liquor gasification Direct electrolytic causticizing Near net shape casting/strip casting Oxy-fuel combustion in reheat furnace Smelting reduction processes Ultrasonic dying Advanced CHP turbine systems Advanced reciprocating engines Fuel cells Continuous melt silicon crystal growth High Consistency forming Biodesulfurization Sensors and controls Roller kiln Membrane technologyfood Process Integration (pinch analysis) Plastics recovery High efficiency/low NOx burners Autothermal reforming-Ammonia Heat recovery technologieschemicals Heat recovery paper Heat recovery food industrylow temperature Code Alum-1 Chem-1 Chem-4 Food-4 Glass-1 Other-1 Other-3 Paper-1 Paper-3 Steel-2 Steel-4 Steel-5 Textile-1 Utilities-1 Utilities-2 Utilities-3 Electron-1 Paper-6 Refin-1 Other-5 Ceramics-1 Food-3 Other-4 Chem-8 Other-2 Chem-7 Chem-3 Paper-5 Food-2 Cost of Saved Fuel $/mbtu <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 <0 0.0 0.0 0.0 0.20 0.57 0.59 0.86 0.86 0.94 1.13 1.63 2.09 2.48

28

Emerging Energy-Efficient Industrial Technologies

Environmental Benefits
For some industries, the cost of complying with environmental regulation can be an important driver in the decision to invest in particular energy-efficient technologies, especially in the non-attainment areas. Of the 54 technologies profiled, 20 had environmental benefits that were either compelling or significant. These technologies are presented in Table 16. The benefits mainly fall in the areas of reduction of wastes and emissions of criteria air pollutants. The use of environmentally friendly emerging technologies is often most compelling when it enables the expansion of incremental production capacity without requiring additional environmental permitting. In selected cases, the decision to invest in these technologies based on tgeir environmental criteria is part of a larger, long-term business strategy towards sustainable development and staying ahead of the regulatory curve.

Table 16. Environmental Benefits

Technology Improved recycling technologies Inert anodes/wetted cathodes Roller kiln Clean fractionation - cellulose pulp Gas membrane technologies-chemicals Levulinic acid from biomass (biofine) Liquid membrane technologies chemicals Autothermal reforming-Ammonia Plastics recovery Code Alum-3 Alum-4 Ceramics-1 Chem-1 Chem-2 Chem-4 Chem-5 Chem-7 Chem-8 Significant Significant Significant Significant Significant Significant Significant Significant Compelling Significant Significant Significant Significant Significant Significant Significant Compelling Significant Significant Environmental Benefits Reduced emissions and reduced scrap metal waste - eases compliance for environmental regulation No CO2 emissions and reduction of perfluorocarbon emissions Reduced NOx emissions, a major concern of ceramics and glass manufacturers Uses a renewable feedstock, reduces 1.8 million tons of waste by 2010 Decreases CO2 emissions by 0.1325 tons/ton product per year Reduces landfill waste and uses a renewable feedstock Decreases CO2 emissions and other combustion related emissions 50% reduction in NOx emissions Reduced land filling of plastics from automobile shredder residue Reduces NOx and SOx emissions the primary sources of air pollutants from the glass industry Reduced volume of spent lubricant for disposal Reduced sludge production; in other applications, the biochemical oxygen demand (BOD) level can be significantly reduced (CADDET, 1996) Reduction of NOx emissions by 30-70% Reduced CO and PM emissions NOx emission reduction of up to 70-90% Lower air and water emissions of sulfur and poly-aromatic hydrocarbons Reduces volume of waste water, while reducing salt and urea Reduces combustion related emissions per unit of fuel input Little to no NOx emissions

100% recycled glass cullet for container Glass-1 glass Advanced lubricants Motorsys-8 Anaerobic waste water treatment High efficiency/low Nox burners BOF gas and sensible heat recovery Oxy-fuel combustion in reheat furnace Smelting reduction processes Ultrasonic dying Advanced CHP turbine systems Fuel cells Other-1 Other-2 Steel-1 Steel-4 Steel-5 Textile-1 Utilities-1 Utilities-3

Non-Energy Benefits
While energy and environmental concerns factor into technology investment decisions at many industrial facilities, it is frequently the productivity and product quality benefits that most frequently ensure the adoption of a technology. Improvements in productivity and quality contribute significantly to the economic attractiveness of a given technology and may indeed be the largest deciding factor in technology investments. Thirty-five technologies in this study had significant or compelling productivity, quality, or other non-energy benefits.

29

LBNL and ACEEE

Table 17. Non-Energy Benefits

Technology Ultrasonic dying Advanced forming Direct electrolytic causticizing Motor diagnostics Code Textile-1 Alum-1 Paper-3 Motorsys-4 Product Productivity Quality Benefits Benefits Compelling Compelling Compelling None Compelling Somewhat Compelling Somewhat None None None None Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant Significant None Significant Significant Somewhat None None None None None Significant Significant Significant Significant Significant Significant Significant Significant Somewhat Somewhat Somewhat Somewhat Somewhat Somewhat Somewhat Other Non-energy Benefits None None None Somewhat May be able to avoid plant capital expansions due to increased production Significant Investment 10% less than conventional installation None None Significant Investment 10% less below conventional installation None None None None Significant Makes the production of levulinic acid economical None None None None Significant Reduced fan speed can reduce worker noise exposure Somewhat Can allow expansions without needing to upgrade utility service, and can allow for peak load shaving Somewhat Can allow expansions without needing to upgrade utility service, and can allow for peak load shaving Somewhat Ability to downsize equipment and free up space None None None None None None Significant May avoid need for addition compressor purchase or allow retirement of existing compressor with resulting reduced O&M and salvage value Significant May avoid need for addition compressor purchase or allow retirement of existing compressor with resulting reduced O&M and salvage value Somewhat Safety Significant Lower production costs Significant Improved working conditions and safety None Significant Added energy savings with use of controls and sensors; faster start-up Significant Added energy savings w/ task lighting; reduced HVAC load; faster start-up

Liquid membrane technologiesChem-5 chemicals Biodesulfurization Refin-1 Dry sheet forming Paper-4 Chem-2 Gas membrane technologies chemicals Oxy-fuel combustion in reheat Steel-4 furnace New EAF furnace processes Steel-3 Efficient cell retrofit designs Alum-2 Fouling minimization Refin-2 Levulinic acid from biomass Chem-4 (biofine) Advanced CHP turbine systems Utilities-1 High Consistency forming Paper-6 Sensors and controls Other-5 Electron beam sterilization Food-1 Motor system optimization Motorsys-5 Advanced reciprocating engines Microturbines Pump efficiency improvement Utilities-2 Utilities-4 Motorsys-6

Near net shape casting/strip Steel-2 casting Continuous melt silicon crystal Electron-1 growth Impulse drying Paper-7 Condebelt drying Paper-2 Advance ASD designs Motorsys-1 Advanced lubricants Motorsys-8 Advanced compressor controls Motorsys-2

Compressed management

air

system Motorsys-3

Significant

Somewhat

Inert anodes/wetted cathodes Alum-4 Clean fractionationcellulose Chem-1 pulp Variable wall mining machine Mining-1 Switched reluctance motor Motorsys-7 Advanced lighting technologies Lighting-1 Advanced lighting design Lighting-2

Significant Somewhat Somewhat Somewhat Somewhat Somewhat

Somewhat None None Significant Somewhat Somewhat

30

Emerging Energy-Efficient Industrial Technologies

Likelihood of Success
It is difficult to predict how likely a technology is to be successful in the marketplace. Many factors will contribute to the outcome, including: changes in market conditions; the value of the energy and non-energy benefits; strategic considerations; and competition from other technologies. Based on all these factors, we made a qualitative assessment of the likelihood that a technology would succeed in the marketplace. Table 18 presents the twenty-one technologies rated with a high likelihood of success. They span the range of industrial sectors, but in general tend to have among the shortest paybacks. In addition, all have energy and/or non-energy benefits, which help account for the high likelihood of success.

Table 18. Factors Contributing to a High Likelihood of Success

Technology Advanced forming Efficient cell retrofit designs Gas membrane technologies chemicals Levulinic acid from biomass (biofine) Plastics recovery Continuous melt silicon crystal growth 100% recycled glass cullet Advanced lighting technologies Advance ASD designs Motor diagnostics Anaerobic waste water treatment Membrane technology wastewater Sensors and controls Black liquor gasification Dry sheet forming Heat recoverypaper Biodesulfurization Near net shape casting/strip casting New EAF furnace processes Oxy-fuel combustion in reheat furnace Advanced CHP turbine systems Code Alum-1 Alum-2 Chem-2 Chem-4 Chem-8 Electron-1 Glass-1 Lighting-1 Motorsys-1 Motorsys-4 Other-1 Other-3 Other-5 Paper-1 Paper-4 Paper-5 Refin-1 Steel-2 Steel-3 Steel-4 Utilities-1 Est. Life Simple Environ. (yr) Payback Benefits 15 Immediate None 15 2.7 Somewhat 15 10.2 Significant 20 20 7 25 4.0 15 15 20 10 10 30 20 20 15 20 40 10 10 1.5 2.8 Immediate 2.0 1.3 1.1 Immediate 0.8 4.7 2.0 1.5 48.3 3.9 1.8 Immediate 0.3 1.2 6.9 Significant Compelling Somewhat Significant None None None Significant Somewhat Somewhat Somewhat Somewhat Somewhat None Somewhat Somewhat Significant Significant Other Benefits P P Q, O P, O P Q, P Q, P, O P, Q P, Q O P P,Q P, S Q P, S Q P,Q P P P, Q

Technologies from a National Energy Policy Perspective

From a national energy policy perspective, it is important to understand which technologies have both a high likelyhood of success and a high energy-savings. While various audiences may be interested in sector-specific or regional-specific technologies, the technologies listed in Table 19 are intended to provide guidance to those interested in the impact of energy-saving technologies on a more national level. This table also identifies the recommended next steps appropriate for each technology.

31

LBNL and ACEEE

Table 19. Technologies with High Energy Savings and a High Likelihood of Success
Technology Efficient cell retrofit designs Advanced lighting technologies Advance ASD designs Membrane technology wastewater Sensors and controls Black liquor gasification Near net shape casting/strip casting New EAF furnace processes Oxy-fuel combustion in reheat furnace Advanced CHP turbine systems Code Alum-2 Lighting-1 Motorsys-1 Other-3 Other-5 Paper-1 Steel-2 Steel-3 Steel-4 Utilities-1 Total Energy Savings High High High High High High High High High High Likelihood of Success High High High High High High High High High High Recommended Next Steps Demo Dissem., demo R&D Dissem., R&D R&D, demo, dissem. Demo R&D Field test Field test Policies Dissemination Dissem., R&D Dissem., demo Dissem. Dissem., training Dissem., training Dissem., demo Dissemination Demo Demo Demo R&D, demo Demo R&D, demo R&D R&D Demo R&D Demo Dissem., demo R&D, demo R&D, demo

Autothermal reforming-ammonia Chem-7 High Medium Membrane technology - food Food-3 High Medium Advanced lighting design Lighting-2 High Medium Compressed air system management Motorsys-3 High Medium Motor system optimization Motorsys-5 High Medium Pump efficiency improvement Motorsys-6 High Medium High efficiency/low NOX burners Other-2 High Medium Process integration (pinch analysis) Other-4 High Medium Heat recovery - paper Paper-5 High Medium Impulse drying Paper-7 High Medium Smelting reduction processes Steel-5 High Medium Advanced reciprocating engines Utilities-2 High Medium Fuel cells Utilities-3 High Medium Microturbines Utilities-4 High Medium Inert anodes/wetted cathodes Alum-4 High Medium Advanced forming Alum-1 Medium High Plastics recovery Chem-8 Medium High Continuous melt silicon crystal growth Electron-1 Medium High 100% recycled glass cullet Glass-1 Medium High Anaerobic waste water treatment Other-1 Medium High Dry sheet forming Paper-4 Medium High Biodesulfurization Refin-1 Medium High *note technologies in this table are listed in alphabetical order based on industry sector

Suggested Actions to Support Technology Development

Each technology is at a different point in its development or commercialization process. Some technologies still need further R&D to address cost or performance issues. Other technologies are ready for demonstration. Some technologies have already proven themselves in the field, and only need the market to become informed about the technologys benefits and market channels to develop skills to deliver the technology. Table 9 outlined the recommendations to support future development of the technologies. Note that these recommendations are not an endorsement of any particular technology. Future development will ultimately be decided by the technology purchasers and users. However, the recommended actions are intended to help clarify whether a technology is both technically and economically viable, and to help eliminate market barriers that would otherwise slow or inhibit the technologies deployment. Seventeen emerging energy-efficient industrial technologies can benefit from additional R&D. As Table 20 indicates, we suggest further R&D for several primary metal technologies (e.g., advanced forming, inert anodes/wetted cathodes in aluminum and near net shape casting in steel) and several Crosscutting motor and utility technologies (e.g., advanced ASD designs, switched reluctance motor, advanced reciprocating engines, micro-turbines, sensors, and controls). In addition to private research funds, several of the identified technologies have received some R&D support from the DOE or other public entities, including federal and state agencies.

32

Emerging Energy-Efficient Industrial Technologies Table 20. Technologies Requiring Additional R&D
Technology Advanced forming Inert anodes/wetted cathodes Continuous melt silicon crystal growth Electron beam sterilization Membrane technologyfood Advance ASD designs Switched reluctance motor Direct electrolytic causticizing Dry sheet forming Fouling minimization Near net shape casting/strip casting New catalysts Biodesulfurization Advanced reciprocating engines Microturbines Membrane technology wastewater Sensors and controls Code Alum-1 Alum-4 Electron-1 Food-1 Food-3 Motorsys-1 Motorsys-7 Paper-3 Paper-4 Refin-2 Steel-2 Chem-6 Refin-1 Utilities-2 Utilities-4 Other-3 Other-5

There are, however, a large number of technologies that already have made some headway into the marketplace or are at the prototype testing stage. The technologies presented in Table 21 represent excellent candidates for demonstrations. For some, field trials are needed to gain operating experience, but with others, demonstration is required for potential customers to gain comfort with the technology.

Table 21. Candidate Technologies for Field Trials and Demonstration

Technology Advanced forming Efficient cell retrofit designs Improved recycling technologies Roller kiln Clean fractionationcelluose pulp Heat recovery technologieschemicals Levulinic acid from biomass (biofine) Plastics recovery Membrane technologyfood Cooling and storage 100% recycled glass cullet for container glass Variable wall mining machine Black liquor gasification Condebelt drying Dry sheet forming Heat recoverypaper High Consistency forming Code Alum-1 Alum-2 Alum-3 Ceramics-1 Chem-1 Chem-3 Chem-4 Chem-8 Food-3 Food-4 Glass-1 Mining-1 Paper-1 Paper-2 Paper-4 Paper-5 Paper-6 Technology Impulse drying Biodesulfurization New EAF furnace processes Oxy-fuel combustion in reheat furnace Smelting reduction processes Ultrasonic dying Advanced lighting technologies Advanced lighting design Motor diagnostics Anaerobic waste water treatment High efficiency/low NOx burners Membrane technology wastewater Sensors and controls Advanced CHP turbine systems Advanced reciprocating engines Fuel cells Microturbines Code Paper-7 Refin-1 Steel-3 Steel-4 Steel-5 Textile-1 Lighting-1 Lighting-2 Motorsys-4 Other-1 Other-2 Other-3 Other-5 Utilities-1 Utilities-2 Utilities-3 Utilities-4

While we recommend further demonstration and dissemination of a given technology, it is often difficult to understand what is limiting a technologys uptake without a more comprehensive investigation of market issues. Some of the technologies in this category are common in European countries or Japan but have not yet penetrated the U.S. market. Others are being newly developed in the U.S. and face challenges in reducing the perceived risks by investors. Two technologies, motor system optimization (motorsys-5) and pump efficiency improvement (motorsys-6), are opportunities for training programs similar to those developed by the DOE for the compressed air system management (motorsys-3). For advanced industrial CHP turbine systems (utilities-1), the major recommended activity is removal of policy barriers. For others, their unique markets will dictate the form of the educational and promotional activities. We urge the reader to follow up on details in the specific technology profiles.

33

LBNL and ACEEE

V. CONCLUSIONS AND RECOMMENDATIONS FOR FURTHER WORK

While many profiles of individual emerging technologies are available, few reports have attempted to impose a standardized approach to the evaluation of these technologies. This report provides a way to review technologies in an independent manner, based on information on energy savings, economics, nonenergy benefits, major market barriers, likelihood of success, and suggested next steps to accelerate deployment of each of the analyzed technologies. This report serves much the same purpose for the industrial sector that Emerging Energy-Savings Technologies and Practices for the Building Sector (Nadel et al. 1998) did for the buildings sector. In fact, several of the technologies covered in the buildings study were chosen again here. These technologies are Crosscutting technologies, which not only span across industries but also across economic sectors. However, important differences exist between these two reports due to the unique characteristics of the buildings and industrial sectors. In buildings, much of the energy use is related to the building itself (e.g., heating, cooling and lighting), while in industry most of the energy is associated with the process of turning raw materials into useable products. With so much energy focused on specific processes, many processrelated technologies have narrow markets of applicability. By contrast, the buildings sector can be divided into fewer, larger groups (e.g., office buildings, warehouses, apartments, hospitality, and single-family homes) which share many common applications. Most of the industrial sector technologies are less broadly applicable. There is also significant variation in energy use within industry group, and non-uniform geographic distribution of industry groups. Thus a specific process technology may have limited national impact, while being of critical importance to a region in which the affected industry is concentrated. State- or regionspecific analysis of technologies could provide further insights into unique regional opportunities. Combining the region-specific circumstances with the technology evaluations offered in this report may lead to varying policy choices for regional entities such as state governments, state or regional agencies, or utilities. A regional focus could also identify different technologies that need to be assessed, specific to particular regions. Our selection of a limited set of 54 technologies was an arbitrary constraint based on the funding available for this study. A number of the initial technologies screened appeared very interesting and would deserve further study, but were eliminated due to our resource limitations. In addition, the initial list of candidate technologies should not be viewed as all encompassing. The authors are aware that many other promising technologies exist, and by their nature new technologies will be continually emerging. Ideally, the effort reflected in this report would become the beginning of a continuing process that identifies and profiles the most promising emerging energy-efficient industrial technologies, and tracks the market success for those technologies previously profiled. An interactive database would allow the continual updating of information, rather than providing a static snapshot of the industrial technology universe. The quality of data on emerging technologies is varied, especially for less fully developed technologies. For technologies yet to enter the commercial market, costs remain an issue of speculation, frequently tinted by optimism. In addition, further quantification of the other benefits based on the experience from technology users in the field is an important area for future analysis. It would be useful to revisit many of these technologies in a few years to update the information, as well as to see what lessons can be learned from experience in the marketplace. Non-energy benefits are as important, if not more important, than energy savings in determining the market potential of energy-efficient technologies. In our analyses, we have indicated the importance of these benefits by rating them as driving the adoption decision (compelling), of equal importance to other factors (significant), or supporting a decision to adopt (somewhat). Almost all technologies chosen have some non-energy benefits. A more detailed assessment of these benefits may help to better evaluate future market potential for these technologies.

34

Emerging Energy-Efficient Industrial Technologies While our study evaluates each technology in relation to a given reference technology, the reality of the market is that technologies compete for market share. Interactive effects and inter-technology competition (e.g., paper drying technologies) have not been accounted for, but ideally should be in any type of integrated technology scenario. We expect that the data collected in this study will prove valuable to modelers who evaluate technology choices in the market. The authors will explore this issue further in a forthcoming companion report. These observations lead us to suggest that an appropriate follow-on activity would be to establish an ongoing emerging industrial technology characterization effort. This effort would involve setting up a database to catalog emerging energy-efficient industrial technologies as they are identified. Our preliminary screening list of 174 technologies could form the initial basis for this database. Each year, a number of technologies would be selected for a more detailed assessment, as has been done for this study. In addition, every few years some of the detailed technology profiles would be revisited to update the information and track the technologies success in the marketplace. This database, along with the associated detailed technology assessments, would be a valuable resource to researchers, modelers, product developers, and policy-makers, all of whom need standardized information regarding these important emerging energy-efficient industrial technologies.

35

LBNL and ACEEE

VI. TECHNOLOGY PROFILES

The technology profiles for our study are presented in the following section. Table 22 identifies the final technologies profiled. Details regarding the methodology and assumptions used in the analysis are discussed in Section 3: Methodology and Approach. Table 22. Profiled Emerging Energy-Efficient Industrial Technologies and their Technology Code Technology
Electron Beam Sterilization Heat Recovery Food Industry - Low Temperature Membrane Technology - Food Cooling and Storage Ultrasonic Dying Black Liquor Gasification Condebelt Drying Direct Electrolytic Causticizing Dry Sheet Forming Heat Recovery - Paper High Consistency Forming Impulse Drying Clean Fractionation - Cellulose Pulp Gas Membrane Technologies-Chemicals Heat Recovery Technologies - Chemicals Levulinic Acid From Biomass (Biofine) Liquid Membrane Technologies-Chemicals New Catalysts Autothermal Reforming-Ammonia Plastics Recovery Biodesulfurization Fouling Minimization Roller Kiln 100% Recycled Glass Cullet For Container Glass BOF Gas and Sensible Heat Recovery Near Net Shape Casting/Strip Casting New EAF Furnace Processes Oxy-Fuel Combustion In Reheat Furnace Smelting Reduction Processes Advanced Forming/Near Net Shape Technology Efficient Cell Retrofit Designs Improved Recycling Technologies Inert Anodes/Wetted Cathodes Continuous Melt Silicon Crystal Growth Advance ASD Designs Advanced Compressor Controls Compressed Air System Management Motor Diagnostics Motor System Optimization Pump Efficiency Improvement Switched Reluctance Motor Advanced Lubricants Advanced CHP Turbine Systems Advanced Reciprocating Engines Fuel Cells Microturbines Advanced Lighting Design Advanced Lighting Technologies Hi-Tech Facilities HVAC Anaerobic Waste Water Treatment High Efficiency/Low NOx Burners Membrane Technology Wastewater Process Integration (Pinch Analysis) Sensors and Controls Variable Wall Mining Machine Code Food-1 Food-2 Food-3 Food-4 Textile-1 Paper-1 Paper-2 Paper-3 Paper-4 Paper-5 Paper-6 Paper-7 Chem-1 Chem-2 Chem-3 Chem-4 Chem-5 Chem-6 Chem-7 Chem-8 Refin-1 Refin-2 Ceramics-1 Glass-1 Steel-1 Steel-2 Steel-3 Steel-4 Steel-5 Alum-1 Alum-2 Alum-3 Alum-4 Electron-1 Motorsys-1 Motorsys-2 Motorsys-3 Motorsys-4 Motorsys-5 Motorsys-6 Motorsys-7 Motorsys-8 Utilities-1 Utilities-2 Utilities-3 Utilities-4 Lighting-1 Lighting-2 HVAC-1 Other-1 Other-2 Other-3 Other-4 Other-5 Mining-1

36

Emerging Energy-Efficient Industrial Technologies

Condensed Methodology and Summary of Assumptions

This page briefly describes the information contained in each major segment of the technology profile tables and any major assumptions that entered the analysis. For a more detailed discussion of the project approach and methodology, please turn to Section 3: Methodology and Approach. Market Information includes a description of the industries to which the technology/measure is applicable. We also provide information on the end-uses for the technology, the principal energy types used by the technology, and the primary market segment. There may be more than one market segment for which the technology is applicable; we used our judgement to identify the most predominant segment. Finally, we also included a key output driver or the energy consumption for our 2015 base-case related to that sector. Reference Technology includes a description of the current technology or practice, the volume of production or annual operating hours associated used in the baseline and savings analysis, and baseline energy consumption for the existing process. New Measure Information includes a description of the new technology, energy consumption information, information on the current status of the technology, the expected date of commercialization (if known), and the lifetime of the technology. Savings information identifies electricity, fuel, and primary energy savings for a typical application of the new technology relative to the reference technology. The analyst made an assessment of the rate at which the technology is expected to penetrate the market. The penetration rates assume that the technologies compete against the reference technology but not against each other for the market share. We assumed a linear penetration curve. The penetration rates begin in the first year after commercialization, or 2001 for those technologies that are already commercialized. For measures with retrofit as the predominate mode of market deployment, the portion of the market that can be impacted by a technology is assumed to be 100 percent. For replacement (i.e., replace on failure), the portion is assumed to be the period of the study (15 years) divided by the measure life. For new construction, it is the growth in capital investment for the target industry divided by the anticipated total installed capital value in 2015. Feasible applications refers to the percentage of the total market that the technology is estimated to capture by 2015. Cost-Effectiveness provides an estimate of the technology or measures investment cost ($/unit output), whether that investment is incremental or full cost, and any change in operations and maintenance cost ($/unit output) for adopting the technology. We propose to include three measures of cost-effectiveness: cost of conserved energy for electricity, fuels, and primary energy, simple payback for the investment relative to the reference technology (years), and internal rate of return (IRR percent). Simple payback and internal rate of return are metrics that are often used by industries and financial analysts, while cost of conserved energy has been useful as a cost-effectiveness indicator for the policy community. Key Non-Energy Factors are those factors that can significantly affect the decision to purchase a technology. These include the presence of other benefits (productivity, quality, environmental, other [i.e. safety]), and to what extent the technology is currently being promoted. In the Evaluation section of this table, researchers identify the major market barriers that could impede the successful implementation of this technology. The technologys likelihood of success (high, medium, and low) is rated based on it its cost-effectiveness, key non-energy factors, and major market barriers. We suggest what next steps are appropriate to accelerate the deployment of the technology. Finally, the analyst provides an assessment of the overall quality of the data used in the analysis using a rating of excellent, good, fair or poor.

37

LBNL and ACEEE

Electron Beam Pasteurization (Food-1)

Radiation pasteurization entails subjecting food to controlled amounts of ionizing radiation that has sufficient energy to knock electrons from the outer rings of atoms of the foods to create free radicals and ions, resulting in the destruction of bacteria and pathogens. The radiation used does not have sufficient energy to split atoms that would cause the exposed objects to become radioactive. The U.S. Food and Drug Administration has approved the following sources of ionizing radiation for the treatment of foods: Gamma rays produced by the natural decay of radioactive cobalt-60 or cesium-137 isotopes X-rays with a maximum energy of five million electron volts (MeV) Electrons with a maximum energy of 10 MeV

Electron beam technology has perhaps the greatest potential for the safe, effective, and cost-efficient radiation pasteurization of meat, dairy, and canned goods. In electron beam systems, a multi-stage electron accelerator generates a dense beam of high-energy electrons. This beam is magnetically focused and scanned across the target, providing saturation of the food product with electrons that deposit their energy and break the chemical bonds of its atoms. Electron beam sterilization has been used in medical devices for more than 40 years, but only in recent years have the problems of relatively low penetration ability and device complexity been solved. Electron beam pasteurization competes with the other radiation treatments as an alternative to thermal pasteurization. Thermal pasteurization is the primary technology employed in the dairy and canning industries. In the traditional pasteurization process for milk, the liquid is raised to a temperature of 162 degrees Fahrenheit (72 C) for 15 seconds followed by rapid cooling to 44 degrees (7 C). Liquid foods such as milk, fruit juices, beer, and wine are pasteurized using plate-type heat exchangers consisting of a large number of thin, vertical steel plates that are clamped together in a frame. The plates are separated by small gaskets that allow the liquid to flow between each successive plate. After the process is completed, the product is packaged under aseptic conditions to prevent recontamination of the product. The technology uses over 90 percent less energy than conventional pasteurization techniques. During the 1970s, several companies, including Varian Associates, Proctor and Gamble, and Siemens began renewed research in the application of x-ray technology for medical equipment and their involvement in the improvement of accelerated electron technology raised performance parameters to a new level. The major disadvantage of electron beams has been that the electrons dont penetrate more than an inch and a half into an organic object. Improvements in the equipment design have overcome this problem to a certain degree. The greatest advantage of electron beam pasteurization is that it is quite versatile. The technology can be utilized to treat products that would normally undergo thermal treatment as well as products that cannot withstand the high temperatures of traditional pasteurization. Meat products and fresh fruits and vegetables can be irradiated to kill bacteria and molds. One of the largest market barriers that face this technology is the stigma that is associated with irradiated foods. None of the countrys major food companies will publicly acknowledge interest in food irradiation (Skerret, 1997), but developments such as the Clinton Administrations food-safety initiative may renew interest in this area. Economics will play a large role in determining which of the alternative approaches to thermal pasteurization will ever become widely used in food processing. Food is a relatively inexpensive commodity, therefore even slight decreases in processing costs can have a big impact on consumer prices. Electron beam processing currently adds an additional ten cents or so per pound of product (www.techreview.com/articles/nd97/skerrett.html), but demonstration facilities such as SureBeam Corporations electronic pasteurization system in Sioux City, Iowa (www.surebeamcorp.com/food/systems.php), could soon bring about lower costs. Electron beam pasteurization techniques can be a viable option for foods that cannot withstand high temperatures, such as meats, cheeses, fruits, and vegetables. In order for this technology to truly enter the marketplace, the initial capital and installation costs will need to come down, in all but the more expensive specialty food markets.

38

Emerging Energy-Efficient Industrial Technologies Electron Beam Pasteurization Data Table

Units Electron Beam Pasteurization Food-1 Replace thermal pasteurization Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

tons

Food Process heating, other Natural gas, electricity, coal New, replace on failure 93150000

SIC 20 Okos, et al. 1998

Assume 15% increase over USDA 1999 figures http://www.usda.gov/nass/aggraphs/milkprod.htm

heat pasteurization of milk (raise from ambient to 162 F for 15 sec, then cool to 44 F) 1.00 tons 4.40E+02 kWh Okos, et al. 1998 1.25 MBtu Okos, et al. 1998 4.99 MBtu Okos, et al. 1998, assume 42.5% heating efficiency, 19.5% cooling efficiency electron beam pasteurization of milk 50.00 kWh 0.00 MBtu 0.43 MBtu Commercialized 1995 10 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 389.62 1.25 4.57 low 8% 2903.45 9.28 34.04 100 Incremental 10 0.08 24.04 6.55 19.2 0.37% Significant Significant None H,M,L Medium Public perception Low Testing on safety Fair Implementing electron beam is capital intensive 89% 100% 91% Industry must overcome negative stigma of irradiated foods

http://www.surebeamcorp.com/food/ebeamtech.php, 50kwh/ton http://www.biosterile.com/foodpast.htm

Electron beam pasteurizes in a few seconds and does not require heating Does not alter the taste or quality of the food

Several companies are involved in promoting the technology Negative publicity involving irradiated food. Banned in EU Testing for safety must be done to allay fears of public

H,M,L E,G,F,P

http://www.usda.gov/nass/aggraphs/milkprod.htm Okos, et al. 1998 http://www.oit.doe.gov/factsheets/petroleum/pdf/gasbiopet.pdf Thayer, et al. 1996

39

LBNL and ACEEE

Low Temperature Heat Recovery in the Food Processing Industries (Food-2)

Food processing industries play a vital role in the U.S. economy and in foreign trade. Classified under Standard Industry Code (SIC) 20, together these industries account for a large portion of U.S. industrial energy use, ranking as the fifth largest energy-using industry after petroleum refining, chemicals, primary metals, and paper manufacturing. Unlike other energy-intensive industries, the food industry does not produce a homogenous output and operates at significantly lower temperatures. Therefore, energy consumption in food industries comes from a wide range of production activities. Some of the energyconsuming activities of the sector include roasting, baking, cooking, frying, drying, freezing, refrigeration, pasteurization, evaporation and distillation. There are also energy demands for supplying buildings with heat, light, and air conditioning. In 1994, roughly two-thirds of final energy demand for the food industries was fuel for boilers to provide steam and heat to the various processes (EIA 1997). There are many opportunities to take advantage of heat recovery in food processing. Heat recovery describes a situation where excess heat from some production process is utilized in another process step. Heat recovery can be accomplished by using all or part of the exhaust gas from one process as the inlet gas to another process. Alternatively, a piece of equipment called a heat exchanger can capture the heat in the exhaust and transfer it to another flow of gas or liquid. Heat exchangers are commonly used throughout industrial processes, and there are numerous manufacturers producing many varieties of heat exchangers, including heat pumps, plate recuperators, tube recuperators, heat tubes, run-around coils, and economizers. Energy savings from heat recovery in the food industry depend upon finding applications where heat recovery is economical and improves the process. Case studies illustrate that there are many potential applications for cost-effective heat recovery applications in the food industry. In some of these projects, excess heat from one energy-intensive process step was used in another process step. At bakeries, heat exchangers were installed in the exhaust stacks of the ovens where bread was baked. The heat recovered was used during the dough-rising stage (CADDET 1994a, CADDET 1997c) or provided hot water for other processes. At vegetable processing plants, excess heat from frying (CADDET 1995a) or steam peeling (CADDET 1999a) was used to provide hot water to the facility for use at other process steps. Two other projects at beverage facilities used heat exchangers to improve process integration; one was at a brewery (CADDET 1999b) and one was at a whisky distillery (CADDET 1994b). Heat recovery also has important applications for drying processes. Important drying processes in the food industry include the drying of grains and beans for storage or fodder (CADDET 1994c), drying malt for breweries (CADDET 1997b), and pulp drying in the sugar processing industry. For drying, the material is typically treated with heated dry air. The gases leaving a dryer will have high moisture content and still contain residual heat. Heat exchanger systems capture both the heat remaining in the gases as well as the latent heat in the water vapor of these exhaust gases, and transfers this heat to the inlet gases for the dryer. Most heat exchangers used in food processing are constructed of stainless steel, and this meets the requirements of most applications. In applications where significant amounts of dissolved chloride exists in the material being passed through the heat exchanger, which is common for preserved or prepared foods, the potential for corroding stainless steel is high. In these cases, the common choice is to use heat exchangers made of nickel, nickel-steel alloy, or titanium. Plastic heat exchangers may one day be used in these corrosive applications, but for now they are too costly and do not meet the design specifications for the food industry. The eight projects cited above occurred around the world Netherlands, UK, Australia, and Canada and ranged in total project costs from $13,000 to over $1 million. These projects totaled annual energy savings of 290,000 MMBtu (306,000 GJ), with an average capital cost of roughly $16 per MMBtu ($15/GJ) saved annually. With the average price of primary energy around $4 per MMBtu ($3.8/GJ), the payback period for these projects averages around 4 years. We estimate that food industry energy consumption in 2015 will be approximately 1700 TBtu (1790 PJ) of primary energy (AEO 1999). Roughly 50 percent of this will be fossil fuel for boilers meeting steam demand for food processing. Another 15 percent will be fossil fuel used directly in processes, and 35 40

Emerging Energy-Efficient Industrial Technologies Heat Recovery Applications in the Food Industry Data Table
Units
Heat Recovery Applications in the Food Industry food-2 Use of heat exchangers at various applications in the food industry.

Notes

Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources

TBtu

Food Industry Process heat Fuels, Electricity Retrofit 1710.5

EIA99

Minimal use of heat recovery in drying systems and to preheat boiler feed water TWh TBtu TBtu 58 985 1478.3 EIA, 1997 EIA, 1997

Use of heat recovery technologies to lower energy consumption in drying and to reduce losses from boilers TWh 58 TBtu 958 1451 TBtu Commercialized, Research Depends on specific application 1995 Distributor claims heat exchanger in non-corrosive environment has indefinitely lifetime if properly maintained Years 25 TWh/% TBtu/% TBtu/% % TWh TBtu TBtu $/Mbtu-s $ $/kWh $/Mbtu $/Mbtu Years % 0 27.0 27.0 Low/Medium 30% 0 9.4 9.4 16 Full cost 0 2.48 2.48 2.48 4.8 20% Somewhat Somewhat None H,M,L Low Awareness Low Promote pinch analysis Fair Fears of fouling and corrosion Estimate of capital investment based on sample projects Rough estimate value of average productivity benefits 0% 3% 2%

Discount rate for all CCE calculations is 15%

Increased throughput, less heat and vapor discharged to facility and atmosphere

H,M,L E,G,F,P

Own estimates based on literature survey EIA, 1999 EIA, 1999 Numerous CADDET reports Conversation with distributor (West Chem Equipment), author judgement Author judgement Numerous CADDET reports George Fisher Co.(Sam Wharry), West Chem Equipment

percent will be fuels consumed to meet the food industrys electricity demand (AEO 1999). Of the total energy consumed for steam demand, 20 to 25 percent is lost due to boiler inefficiencies (AEO 1999, Drescher et al. 1997). The use of heat recovery systems can lower the boiler losses to 12 to 16 percent in industries where there are opportunities for heat recovery (Drescher et al. 1997). Assuming that 20 percent of energy use falls into this category and is retrofit for heat recovery by 2015, 14 TBtu (15 PJ) of energy savings can be attained. For drying systems, we assume that 16 percent of food industry energy use is used for drying (CADDET Newsletter 1997). Estimates of the potential savings from heat recovery systems for drying range from 10 to 50 percent for various projects (Drescher et al. 1997, Mercer 1994). We assume that 20 percent of the drying energy demand is appropriate for heat recovery and retrofit by 2015, and that these projects on average reduce dryer energy demand by 25 percent. Under these conditions, energy consumption is lowered by 9.5 TBtu (10 PJ).

41

LBNL and ACEEE

Membrane TechnologyFood (Food-3)

The food and kindred products industry (SIC 20) comprises a wide variety of activities. The sector is large and growing. The value of shipments exceeds $400 Billion (Drescher et al. 1997), and is also a large energy consumer. Primary energy use in 1994 was 1480 TBtu (1560 PJ), equivalent to 5 percent of total industrial energy use in the U.S. Primary energy consumption in 2015 is estimated to be 1700 TBtu (1790 PJ) (AEO 1999). The main energy consuming sub-sectors are corn milling, sugar, meat packing, soybean oils, beverages, and dairy. In the food industries, membranes are used to concentrate, fractionate and purify liquid products. In the food and beverage industry, four types of membrane processes are important: microfiltration (MF), ultrafiltration (UF), nanofiltration (NF) and reverse osmosis (RO). Gas separation is only used in the fruit and vegetable sector for packaging in a nitrogen atmosphere. The differences between the membrane methods are the separation capabilities due to size and molecular weight, although the separation capabilities do overlap. Table 11 provides an overview of potential membrane applications in different food industries. Due to the diverse activities in the food industry, quantitative estimates of the potential capacities for membrane applications are difficult to make. Below we discuss some of the major potential applications in more detail. Overview Of Potential Membrane Applications In The Food Industry. Sector Applications Dairy Whey concentration Milk concentration Desalting of salt whey Fractionation of proteins Concentration of egg and egg white Recovery of by-products from waste water Cold stabilization of beer Clarification of wine Removal of alcohol from beer and wine Pretreatment of water Upgrading of citrus juices Preconcentration of dilute sugar solutions Syrup concentration Recovery of sugar from rinse water Concentration of tomato juice Concentrating juices Juice flavor and aroma concentration Membrane type RO RO ED* UF UF UF MF UF RO RO RO UF UF UF/RO RO/UF RO/UF UF/RO

Grain milling Beverage

Sugar

Fruits and vegetables

*Electrodialysis Source: Kseoglu et al. 1993, Maaskant et al. 1995, KMS 2000. We focus on the dairy, beverages, fruit and vegetables industry. Almost 40 percent of the total membrane market of over $1 Billion in the U.S. is found in the food industries (Wiesner and Chellam 1999). The dairy industry is the most important sector using membranes in the U.S. (Dziezak 1990). The fruit and vegetable industry has a large potential for improved energy efficiency using membranes. The beverage sector is also an important sector for applying membranes. For example, membranes can be used for the removal of alcohol from beer and the treatment of water, but this application may decrease rapidly because of the possibility of producing beer without alcohol. In the sugar sector, membranes are used in almost 20 percent of the potential applications in countries like The Netherlands. In the dairy and fruits and vegetable industries, membrane technology is considered proven in many applications. Barriers to implementation include the lack of information, as well as the need for specific membranes in specific applications. Major suppliers are APV (Denmark), Koch Membrane Systems (U.S.), Osmonics (U.S.), U.S. Filter (U.S.). Research is directed at new applications, more efficient and longer lasting

42

Emerging Energy-Efficient Industrial Technologies Membrane Technology in the Food Industry Data Table
Units Membranes Food-3 Process Applications of Membranes in the Food Industry Market Information: Industries Food End-use(s) Separation Energy types Fuel Market segment New 2015 basecase use Tbtu 1712.0 Reference technology Description Evaporation, Clarification Throughput or annual op. hrs. N/A. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources GWh TBtu TBtu 930 147.8 155.7 Notes

SIC 20

primary primary energy energy use use AEO, EIA 1999 2000 forecast forecast ofof Food Food

Estimated energy use in 'membrane-eligible' process uses, 15% of 1994 pump use Estimated energy use in 'membrane-eligible' process uses, 15% of 1994 fuel use

Membrane technology replaces existing separation processes GWh 2662 TBtu 88.7 TBtu 111.3 Commercial, Research 1990 10 -1732 59.1 44.4 Medium 60% -1039 35 26.6 450 Full cost -55 N/A 0.59 0.78 2.2 45% Somewhat Somewhat Somewhat H,M,L Medium Specificity, Unknown Medium Dissemination, R&D Fair, Poor AEO 2000 EIA, 1996 (MECS 1994) Estimate based on case-studies and Eichhammer, 1995 Wiesner and Chellam, 1999 Author estimate Estimate based on payback period of case-studies (Caddet) (CADDET) Case-studies (Caddet) (CADDET) Actual costs vary heavily depending on application Actual costs vary heavily depending on application -186% 40% 29% Many applications commercial; new membranes under development

Years GWh/% TBtu/% TBtu/% % GWh Tbtu Tbtu $/Mbtu-s $/Mbtu-s $/kWh $/Mbtu $/Mbtu Years %

Actual savings depend on application Actual savings depend on application Actual savings depend on application Rough estimate, based on current uses

Discount rate for all CCE calculations is 15%

Reduced resource use Improved quality (drinks) Reduced water use

H,M,L E,G,F,P

membranes. Federal research programs (e.g. ATP) support development of membrane technology, as well as development of specific applications (e.g. DOE, EPA). Dairy industry. Worldwide, many thousands of m2 membranes have been installed in the dairy industry. It also is the sector with the longest history of the use of membranes, which are used for the desalting of whey

43

LBNL and ACEEE (NF)15, the concentration of whey (RO), the conversion of milk into cheese and soft cheese and the preparation of egg white and egg yolk. For example, a nano-filtration unit was installed in 1996 for whey concentration at a dairy plant in The Netherlands, replacing a two-stage evaporation process (CADDET 1998a). The system reduced steam use by almost 70 percent (from 13 MBtu/ton (15 GJ/t) dry solids to 3.6 MBtu/ton (4.2 GJ/t) dry solids), while power consumption increased from 89 kWh/ton (98 kWh/t) to 153 kWh/ton (168 kWh/t) dry solids. Net energy savings were 8.8 MBtu/ton (10.2 GJ/t) water removed. Additional savings were achieved in the use of sodium hydroxide and nitric acid, as well as reduced transport costs and emission charges, reducing the payback period to 1.3 years (CADDET 1998a). The investment costs were $9.3 ft2 ($100/m2) (CADDET 1998a). Current developments in dairy industry are the reduction of bacteria in milk and the clearing of dairy fluids. The application of membranes in the dairy industry is considered to be in an important phase for implementation on a large scale. Beverages. Water treatment is an important application of membranes in the beverages industry (Comb 1995). For example, membranes are used by Coca-Cola (in Salina, KS) and membranes are also used for juice concentration and for alcohol recovery in the production of non-alcoholic beers (Gach et al. 2000). A number of breweries (e.g. Miller Brewing Co.) already apply membranes for alcohol removal from beer. Nevertheless, potential exists for further application and development. Replacement of plate membranes by new spiral membranes at the Heineken brewery in Den Bosch, The Netherlands, reduced pumping energy and water demand, and resulted in savings of 0.17 kWh/gallon beer (4.6 kWh/100 liter beer). At investments of $0.06/gallon (1.7$/100 liter) production capacity, the simple payback period was just over 4 years (CADDET 2000a). Fruits and Vegetables. There have been several demonstration projects using membranes in the fruits and vegetables industry. At Golden Town Apple Products in Canada, a combination of ultra-filtration and reverse osmosis was used for apple juice concentration (CADDET 1996a). In this process, the juice is heated to about 140F (60C) and afterwards passed through the reverse osmosis membrane and the ultrafiltration membrane. The system has a maximum capacity of 3,000 l/hr for feedstock, 1,500 l/hr for final concentrate and 1,500 l/hr for water removed by reverse osmosis. It is most economical for small systems that need to remove no more than 4,500 to 9,000 pounds (2040 to 4080 kg) of water an hour. The energy savings are estimated to be 66 percent compared to an evaporation process, while the volume of the equipment is reduced by 50 percent as are the transportation costs. The payback period of the combined system is about 2.5 years (CADDET 1996a). It is extremely difficult to estimate the potential energy savings from implementation of membranes in the food industry without a detailed study. For specific applications, energy savings may be up to 40-55 percent of the energy needs for distillation and evaporation. Research is aimed at increasing the number of applications, increasing product quality, lifetime, and increasing energy savings. A European study estimated that membranes could be used to replace 15 percent of fuel using applications in the food industries (Eichhammer 1995). Based on this estimate, we assume that fuel savings are on average 40 percent, while electricity use increases by 10 percent of the fuel savings (expressed as final or site energy). Additional production savings are achieved through product quality, reduced water use, and lower operation costs. The investment and operating costs depend heavily on specific application, and may even be site-specific. However, for the purposes of this study we make a general estimate, noting that the costs may vary widely in practice. Generally, capital costs are expressed per unit of surface area, while about half of the capital costs are for the system components (e.g. pumps, piping) (Wiesner and Chellam 1999). System costs may vary between $6/ft2 and $37/ft2 (200$/m2 and 1300$/m2). Based on the different case studies we estimate an average payback period of 3 years, including non-energy benefits. Membrane life of a properly operated facility may easily exceed 10 years (Wiesner and Chellam 1999). We assume a lifetime of 10 years. The energy savings and cost estimates are rough. Given the large potential application area and potential energy savings, an in-depth study into membrane applications, energy savings, and capital and operational cost benefits is warranted.
15

By the mid-1990s more than 10,000 m2 for the desalting of whey had already been installed in the U.S. dairy industry (Maaskant et al. 1995).

44

Emerging Energy-Efficient Industrial Technologies

Cooling and Storage (Food-4)

Refrigeration in the food sector is a large energy consumer and is mainly used for freezing or cooling of meat, fruit, vegetables, as well as for frozen products (e.g. ice-cream, juice). Refrigeration in industry is mostly done by means of compression cooling and in some cases by absorption cooling (Mottal 1995). Electricity use for refrigeration in the food and beverages industry is estimated at 11.1 TWh (Xenergy 1998), mainly used by compressors. Many options exist to improve the performance of industrial refrigeration systems. System optimization and control strategies combined show a large potential for energy efficiency improvement of up to 30 percent (Brownell 1998). Opportunities include system design, component design (e.g. adjustable speed drives), as well as improved operation and maintenance practices. We focus on new system designs. Adjustable speed drives and process control systems have been discussed elsewhere. New system designs include the use of adsorption heat pumps, gas engine driven adsorption cooling, new working fluids (e.g. ammonia, CO2) and alternative approaches (e.g. thermal storage). Due to the wide variety, we focus on selected technology developments in the areas of gas engines, thermal storage and new working fluids. Gas engines can be applied to drive the compressor instead of an electric motor. A gas engine is used as the direct drive, and the system can follow refrigeration loads by using variable engine speed. The waste heat of the engine can be used to preheat water or for space heating at the plant. GRI has developed a system, marketed by Thermopower Corporation, which has been tested in ice production, food processing, and chemical industries (GRI 1997). Other suppliers market similar products. NYSERDA supported an innovative demonstration at a dairy plant with a gas engine with an absorption chiller. Without the absorption sub-cooling, the project would have saved 52 percent on a primary energy basis. With the absorption cooling the project decreased primary energy use by 77 percent (CADDET 1996b). The gas engine compressor system (without absorption cooling) has capital costs twice as high as a chiller system, and a payback period of about 2 years. A similar system installed at Pittsburgh (PA) cooling warehouse had a payback period of 1.9 years (CADDET 2000b). The gas engine-absorption cooling system has substantially higher capital costs compared to an electric chiller system (almost a factor 3 higher), but the large energy savings and reduced peak energy use result in a payback period of 4 years. The use of a gas engine may result in higher onsite NOx emissions, although offsetting high peaking power plant emissions. Hence, in non-attainment areas extra NOx-reduction measures need to be installed. Thermal storage is an old technology in the sense that it has been used for several centuries for seasonal cooling. Thermal storage has been re-discovered for applications in the food industry to shave peak loads by using off-peak power to generate ice, which is stored in a so-called ice pond and used for cooling. Several plants operate thermal storage systems in the U.S., combined with innovative cooling concepts, e.g. a fermentation plant in Rochester (NY), a cheese factory in Corfu (NY), a food services company in Clark County (NV) and a vegetable and food processing plant in Placentia (CA). Energy savings vary depending on the plant. The fermentation plant in Rochester (NY) reduced cooling energy needs by 80 percent compared to the existing mechanical chiller system. This system had a payback period of up to 4 years (CADDET 1993a). In other applications the savings were not always fully documented or are much smaller. The load shift accounts for the productivity increase, as it allows the use of low-priced electricity at the off-peak hours. Given the current peaking power-supply problems in California, the Midwest and Texas, peak power is a highly valuable commodity, making this technology economically attractive. Other major trends are a reduction of refrigerant charges and the development of new working fluids. Traditionally, the most common working fluids for compression heat pumps are ammonia and CFCs or HCFCs. R&D is directed toward alternative working fluids, especially for the CFCs and HCFCs due to the Montreal Protocol. These alternative working fluids can save energy. Savings of 2 to 20 percent have been reported (Trepp et al. 1992, Lorentzen 1993a, Lorentzen1993b). Recent developments include the use of natural refrigerants such as CO2 (Stene 1999). CO2 is suitable for cooling of storage facilities. In Japan research has also looked at metal hydride systems for commercial cooling, as well as for small-scale systems. A first working prototype was demonstrated in 1995 at a very small scale (for a vending machine), and the technology has been demonstrated for a warehouse of 1100 ft2 (100 m2) at storage temperatures of 40oF (-40oC). The system can be designed in a wide variety of scales (10 10,000 kW), and reduces power use by approximately 20 percent compared to traditional CFC-containing systems (JNT 1996).

45

LBNL and ACEEE

Cooling and Storage in the Food Industry Data Table

Units
Cooling and Storage Systems Food-4 Innovative designs of cooling/refrigeration equipment in food preservation

Notes

Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources

Food Motor and drives Electricity New N/A.

SIC 20 Excluding motor systems, lighting, HVAC

Cooling demand in selected subsectors is unknown in 2015

Estimated energy consumption for cooling in the food industry TWh TBtu TBtu 11 0 94.4 Xenergy, 1998

Innovative designs of cooling/refrigeration equipment in food preservation TWh 9 TBtu 0 75 TBtu Thermal storage, other technologies being developed Commercial 1990 Years 15 TWh/% TBtu/% TBtu/% % TWh TBtu TBtu $/Mbtu-s $/Mbtu $/kWh $/Mbtu $/Mbtu Years % 2 0.0 18.9 Low 40% 1 0.0 7.5 32 Full cost -6 -0.53 -0.53 -0.53 2.6 38% None None Somewhat Somewhat Low Caddet, 1990 CADDET, 1990 (on (on primary primary energy energy basis) basis) Credit for shift of peak electricity use (on primary energy basis) 20% N/A. 20% Estimates based on case studies

Off-peak electricity use

H,M,L

H,M,L E,G,F,P

Unknown, New Medium Demonstration,Dissemination Estimates based on case studies Fair

Xenergy, 1998 CADDET, Caddet, 1993; 1993a; Caddet, CADDET, 1997 1997j Author's estimate Author's estimate Caddet, 1990 CADDET, 1990 CADDET, Caddet, 1990 1990

46

Emerging Energy-Efficient Industrial Technologies For the technology characterization, we assume a potential for energy efficiency improvement of 20 percent on average, which can be achieved using different technologies, e.g. thermal storage, natural gas engine (not for non-attainment areas) and the use of new refrigerants in small-scale industrial applications. Higher energy savings are possible in specific cases, as outlined above. Given the incentives for reduction of peak power use and expected peaking-power shortages in important food producing regions, we assume that there is a substantial interest in implementing new refrigeration equipment in the food industry. Hence, we estimate that between 2000 and 2015 40 percent of the potential may be realized. Capital costs will depend heavily on the specific site and cooling conditions, as well as technology implemented. Hence, the costs and profitability of the investment will vary widely. We base the cost estimate on the thermal storage system installed at Kirk Produce, Placentia (CA) (CADDET 1990). The cost savings because of switching to off-peak hours electricity use have been accounted as a productivity benefit. Other benefits may occur, such as increased product quality (CADDET 1990), but have not been taken into account in the cost estimates. Most technologies, except for the use of selected new refrigerants, have been demonstrated commercially. Hence, dissemination of the results among other potential users is needed, as is demonstration of new concepts or innovative combinations of efficient cooling systems.

47

LBNL and ACEEE

Ultrasound Enhanced Dying (Textile-1)

The textile industry in the United States is a mature industry that used 476.5 TBtus of Primary energy, or 2.3 percent of manufacturing energy in 1994 (EIA 1997). While some of the more labor intensive portions of the industry have moved over seas in the part two decades, the less labor intensive, more energy intensive portion of the industry have stayed in this country, largely through the application of technology to increase productivity. One of the most energy intensive sectors in textiles is dyeing and finishing. The dying of fabric in the textile industry involves two physical processes: the transport of the dyestuffs into the fibers, and the dyestuffs uptake by or reaction with the fibers, resulting in a fast color. Traditionally these actions are accomplished by the application of time, temperature, and pressure. The addition of chemicals, such as salt and urea, to speed the process is usually needed. Conventional dying processes are capital and energy intensive, and the presence of salt and urea in the waste stream creates pollution abatement challenges (Mock 2000). In addition, ultrasound has been shown to enhance the washing phase in which unreacted dyestuffs are removed from the fabric. Both time required for washing and volume of water required are reduced. The application of ultrasound in the dying process offers a number of advantages. Research has indicated that in the presence of ultrasound, the transport and uptake of dye by the fabric can be significantly accelerated. These results occur because the ultrasound energy causes the fiber to swell while reducing surface tension. In addition, the ultrasound allows for a more rapid reaction of the dye with the fabric, because the ultrasound energy preferentially heats the dyestuffs in the fabric. All these benefits of accelerated dyeing can be achieved at lower temperature and atmospheric pressure without the need to add chemicals to the dyestuffs. In addition, the use of ultrasound allows for precise control of the color shade, thus significantly reducing variations in shade commonly experienced with conventional dyeing processes (McCall, Cato and Grady 1992). Because of the need to maintain a uniform ultrasound field, the technology is only applicable to web dying. In web dyeing, a single thickness of fabric is dyed continuously on machine referred to as a dye range. The fabric web is transported under tension through the various stages of the process on drums (dyeing, fixing, washing and drying) similar to a papermaking machine. Research has shown particular efficacy for the application of ultrasound in continuous dying of cotton fabric (McCall, Cato and Grady 1992, Grady 2000, Mock 2000). Web-dyed fabric represents about 0.13 percent of domestic value of textile shipments, with dyeing representing about 0.12 percent of the energy used in the textile industry (Census 1996, Census 2000). The application of ultrasound reduces thermal energy used directly by the process by 10 percent. In addition, savings are realized in the reduced treatment of wastewater. The volume of spent dyestuffs and wash-water are reduced, and treatment is made easier because the concentration of salt and urea in the waster stream is reduced (Mock 2000). In addition, the lack of salt and urea in the dyestuff may also allow for recycling of the dyestuff (McCall, Cato and Grady 1992). Because of the lack of data and the variations in how wastewater is treated between different plants, we have been unable to estimate these additional energy savings. Ultrasound can be retrofitted to existing, dye ranges or be engineering into new systems. Because ultrasound decreases dyeing, fixing and washing times, the through-put for the equipment can be increased significantly at the same operating cost, thus reducing the fixed cost associated with the operation of a range. In addition, because less dye stuffs and chemicals are required to dye a lot of fabric, variable costs are also reduced. ACEEE has assumed that production for a range could be increased by 50 percent, resulting in a corresponding annual unit fixed and variable O&M costs reduction of $330,000 per million yards of fabric. A number of barriers exist to the deployment of this technology. Because this sector of the industry is mature, operating profits are low, and significant international looms, companies are hesitant to make new capital outlays. In addition, most of the dyeing equipment manufacturers are not domestic and most of the ultrasound development has occurred in this country. The technology has been demonstrated at the bench scale, and while research into the science has continued, commercialization activities have been suspended due to a discontinuation of federal textile research funding. Foreign equipment manufacturers have not 48

Emerging Energy-Efficient Industrial Technologies stepped into the breach. One of the principal researchers in the area indicates that the primary activity need to more the technology into market is funding to demonstrate a commercial prototype (Grady 2000). The United States-Asian Environmental Partnership has also identified this technology as one of six key emerging textile technologies (USAEP 1999). Ultrasound Enhanced Dying Data Table
Units Notes Ultrasonic Dying Textile-1 Replace existing continuous web drying with ultrasonic enhanced dying Market Information: Industries Textile SIC 226 End-use(s) Other Energy types Electricity, gas Market segment Both new dye ranges and as a retrofit to existing ranges New, retrofit 2015 basecase million 4291 Based on 1999 Cotton Broadwovens production (Census 2000) , sq.yds. scaled using EIA 2000 growth projection (Honeycutt 2000). Reference technology Description Web dying of cotton broadwovens using a continuous dye range 1,000,000 Throughput or annual operating hours sq.yds 92.2 Electricity use kWh 1992 energy intensity is based on prorating MECS Energy use for textiles to cotton broadwovens as reported in Current Industrial Reports (Census 1998). Fuel use MBtu 53.0 53.8 Primary Energy use MBtu New Measure Information: Description Apply ultrasound to web dying, reducing temperature, contact time, & eliminating salts & urea Electricity use kWh 92.2 Fuel use MBtu 10% by reduced contact time and increased dye transfer rate at lower 47.7 temperatures. 48.5 Primary Energy use MBtu Current status Bench-scale prototype Date of commercialization 2005 Estimated average measure lifetime Years 10 Savings Information: Electricity savings kWh/% 0 0% Fuel savings MBtu/% 5.3 10.0% Primary energy savings MBtu/% 5.3 9.9% Penetration rate 45% penetration in 2010 Medium Feasible applications % 23% Feasible 50% of cotton broadwovens Other key assumptions for savings Does not include savings in wash water treatment Electricity savings potential in 2015 GWh 0 5.1 Fuel savings potential in 2015 TBtu Primary energy savings potential in 2015 TBtu 5.1 Cost Effectiveness Investment cost $ 100,000 Cost to add ultrasound generators (McCall et al, 1992) Full cost Type of cost Cost of generators incremental to range cost -330,000 Change in annual costs (O&M/other benefits) $ Increases throughput by 50% for same O&M Cost of conserved energy (electricity) $/kWh NA Cost of conserved energy (fuel) $/MBtu -58,526 Cost of conserved energy (primary energy) $/MBtu -58,526 Discount rate for all CCE calculations is 15% Simple payback period Years 0.3 Internal rate of return % 330% Key non energy factors Productivity benefits Compelling Increases throughput, thus expanding capacity and reducing unit O&M cost for existing capacity Product quality benefits Compelling More first quality fabric due to Reduced improved control. Environmental benefits Reduces volume of waste water, while reducing salt and urea Compelling Other benefits None Current promotional activity H,M,L Low Commercialization suspended due to funding cuts. Evaluation Major market barriers Lack of domestic manufacturers, commercial equipment, and increased first cost Likelihood of success H,M,L Medium Recommended next steps Funding of production prototype demonstration Data quality assessment E,G,F,P Fair Sources: 2015 basecase Census 2000 and Honeycutt 2000 Basecase energy use Derived from EIA 1997 New Measure energy savings McCall et al 1992, Grady 2000 Lifetime Grady 2000, Mock 2000 Feasible applications Mock 2000 Costs McCall et al 1992, Grady 2000 and Mock 2000 Key non energy factors McCall et al 1992, Grady 2000 and Mock 2000 Principal contacts Perry Grady, NCSU 919/515-3255 Additional notes and sources

49

LBNL and ACEEE

Black Liquor Gasification (Paper-1)

The pulp and paper industry is a large industrial energy user, with an estimated primary energy consumption of 2,970 TBtu (3133 PJ) in 1994 (EIA 1997). Of this amount, we estimate that 1,348 TBtu (1422 PJ) of biomass is used in boiler plant facilities to produce steam for various paper manufacturing processes (EIA 1997). Boiler plants that burn biomass materials are primarily located at Kraft pulp mills, which currently account for nearly 80 percent of the pulp produced in the U.S. (Kincaid 1998). In standard integrated Kraft mills, the spent liquor produced from de-lignifying wood chips (called black liquor) is normally burned in a large recovery boiler, named that because the black liquor combustion is used to recover the chemicals used in the delignification process. Because of the relatively high water content of the black liquor fuel (the fuel is usually combusted at a solids content of 65-75 percent), the efficiency of existing recovery boilers is limited. Electricity production capacity is also reduced since recovery boilers produce steam at lower pressures for safety reasons. One of the new technologies being developed as an alternative to direct combustion of the black liquor is the gasification of the liquor and its subsequent combustion in gas turbines designed to accommodate the lower energy content black liquor gas. The gasification of the liquor converts it into a more useable energy source (Worrell, Bode, and de Beer 1997). While there are also technologies directed solely on gasification, in this assessment we focus on the combined cycle turbine technology combined with gasification. This combination technology has potential to produce significantly more electricity than the current boiler/steam turbine systems, and even make the mill an electricity exporter. The two main types of gasification are low temperature/solid phase and high temperature/smelt phase. The gasification produces a fuel gas that needs to be cleaned to remove undesired impurities for the power system and to recover pulping chemicals. Low temperature gasification is based on a fluidized bed at atmospheric pressure and a temperature 1290F (700C) or lower, below the melting point of inorganic salts that comprise most of the char from black liquor. Sodium carbonate is used as the bed material and is precipitated out and reused (Worrell, Bode, and de Beer 1997, Berglin et al. 1996). The key manufacturer of this process is Manufacturing and Technology Conversion International (MTCI), a U.S. firm (Larson and Raymond 1997, Larson et al. 2000). High temperature gasification occurs at 360 lbs/in2 (2.5 Mpa) and above the melting point of the inorganic salts 1740F (950C) or higher, and chemicals are recovered in a smelt. Higher temperatures lead to higher carbon conversion rates but also may lead to more corrosion in the reactor vessel (Worrell, Bode, and de Beer 1997). The synthesis gas is water quenched (producing low-pressure steam) and cleaned before being fired in the turbine. Kvaerner has done significant research and development of high-temperature systems and the first commercial demonstration of a pressurized, oxygen-blown gasifier will take place in Pieta, Sweden (Larson and Raymond 1997, Larson et al. 2000). Energy savings estimates for this technology vary but are potentially significant. Existing recovery boilers consume roughly 27 Mbtu (28.5 PJ) of black liquor and other biomass per air dried ton of chemical pulp with power production efficiencies using steam turbine systems of 10 percent (Consonni et al. 1998, Larson et al. 1997). While increased fuel inputs are required for gasification systems, and increased electricity inputs are required (especially for gas compression in the combined cycle system), power efficiencies are much higher, thereby allowing for significant primary energy savings. Based on an electricity production capacity of 2,000 kWh/ton, which represents an average of the range of outputs from the various systems (output ranges from 1200-3000 kWh/ton), we assume a primary energy savings potential of 6 Mbtu/ton (7 GJ/t) pulp. Currently, there are no full-scale gasifier/combined cycle plants operating. However, the first fully commercial high temperature air-blown black liquor gasifier plant was installed in 1997 at Weyerhauser in Bern, North Carolina (Erikson and Brown 1999). (A low-temperature demonstration gasifier was demonstrated at the same site in the early 1990s.) The current gasifier is a high-temperature unit developed by Kvaerner (Erikson and Brown 1999). This 734 klb/day unit provides incremental recovery capacity with the product gas being burned in a boiler.

50

Emerging Energy-Efficient Industrial Technologies Black Liquor Gasification Data Table

Units Black liquor gasification Paper-1 Black liquor gasification Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Mton

Pulp and Paper Utilities Biomass New 70.0

SIC 26

EIA 1999. Kraft pulp throughput estimate in 2015

Tomlinson recovery boilers tons 1 kWh 28 MBtu 26.6 MBtu 26.9

Per air dry ton pulp. Recovery boiler sizes range from XX - YY Power output of 800-850 kWh/ton pulp Range of 24-27. HP steam output of 11 to 14 Mbtu/ton pulp

Black liquor gasification with combined cycle gas turbine, BLGCC Power consumption ranges fm. 68-400 kWh/ton. Power production ranges fm. 1200-3000 kWh/ton pulp kWh 267 Fuel consumption ranges 22-38 Mbtu/ton. Steam output of ranges from 7-14 Mbtu/ton pulp MBtu 28.7 MBtu 31.0 Gasifier demo facilities operating, but not with CC turbines Research Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 30 961 -2.1 6.1 Medium 15% 10097 -22 63.7 20 Incremental 5 0.01 -3.82 1.33 1.5 69% Somewhat None Somewhat Somewhat Medium Technical, marketing H,M,L E,G,F,P High Demonstration Excellent 34% Assume a power production of 2000 kWh/ton. -8% Increased black liquor/biomass consumption 23% Assume feasible for 15% of 2015 chemical pulp production Savings applied to feasible applications for 2015 output Savings applied to feasible applications for 2015 output 2% savings. Primary energy consumption of 3549 Tbtu in 2015 Full investment cost range - $300/t pulp. replacing at end of life. Operation and maintenance. Range from $2-7 shown in literature.

Discount rate for all CCE calculations is 15%

Increased throughput, fuel flexibility, reduced capital costs Reduced emissions Reduced explosion risk comapred w/ conventional systems. Demonstration facilities already constructed Refractories, gas cleanup, chemical recovery systems. High cost of initial units Encourage demonstration project of combined cycle facility

H,M,L

EIA, 1999 Larson et al., 2000; Consoni et al.,1998 Larson et al., 2000; Consoni et al.,1998; Lorson et al., 1997 Worrell et al.,1997a Larson et al., 2000, Robinson, 2000. Larson et al., 2000 Lason et al. 2000; Sadowski et al.1999; OIT 1999a Eric Larson, http://www.princeton.edu/~cees

The U.S. Department of Energy is interested in promoting both black liquor and biomass gasification through its support of research and demonstration projects under the Office of Industrial Technology. The DOE recently issued two rounds of solicitations on biomass and black liquor gasification. Georgia Pacific will be demonstrating an MTCI low pressure atmospheric system in a soda chemical pulping mill, a process that produces a feed similar to black liquor (Robinson 2000). Plans to demonstrate a high-temperature

51

LBNL and ACEEE system, a joint Kvarener/Air systems project, were unsuccessful but other demonstration projects may emerge in the second round of DOE solicitations. Recent research on capital cost estimates for gasification systems by a research team that consisted of staff from Princeton University and Weyerhaueser found a range of costs for high-temperature and lowtemperature combined cycle facilities. These costs ranged $270-340 $/ton ($300-375/t) pulp for a commercially developed technology as a new installation (Larson et al. 2000), and depend in part on the additional complexity introduced into the mill related to gas cleanup and chemical recovery. In general, the gasification combined cycle system is expected to have equal or slightly higher costs when compared to conventional systems while at the same time allowing for increased throughput and most importantly, increased power generation (OIT 1999, Larson et al. 2000). Therefore, while annual operations costs were estimated to increase slightly to between $2-7/ton ($2-8/t) compared to existing systems, the credit a mill receives from reduced electricity purchases make the cost-effectiveness of such a system relatively attractive. The electricity buyback price therefore becomes an important driver in project economics (Larson et al. 1998, Sadowski, et al. 1999). The gasification systems also are expected to improve environmental performance, with fewer particulates and nitrogen oxides than in conventional systems (OIT 1999). Finally, gasifiers are less likely to explode; this provides additional safety benefits. The opportunities for this technology are large. A majority of the recovery furnaces and conventional power boilers in existing pulp and paper plants are 20 to 30 years old and more than half of them will need to be replaced or upgraded in the near future (OIT 1999, Larson and Raymond 1997). Analysis from the industry/DOE teams gives the technology a high rating (Erikson and Brown 1999). However, additional research and demonstration are needed before gaining market acceptance. Some key areas include: developing adequate clean up systems for the medium Btu gas, improving refractory reliability, demonstrating cost-effective chemical recovery (especially sulfur separation), and demonstrating overall system integration (Larson and Raymond 1997, Oscarsson 1999). It seems clear that a cost-shared, public-private partnership that involves several companies would be needed to help overcome the technological barriers and to reduce risks given the high capital cost of initial units. This is beginning, but more active involvement will be required. For this technology to be successful in the marketplace, further development, testing, and demonstration will be necessary in the U.S. For the near term, given additional developmental barriers, we believe that there is a medium likelihood of achieving significant market penetration in the near term (by 2015) with increasing successes likely in the slightly longer term.

52

Emerging Energy-Efficient Industrial Technologies

Condensing Belt Drying (Paper-2)

The pulp and paper industry is a large industrial energy user, with an estimated primary energy consumption of 2,970 TBtu (3133 PJ) in 1994 (EIA 1997). Papermaking (as opposed to pulp production) is usually divided into four basic steps: 1) stock formation and forming, 2) pressing (mechanical dewatering), 3) evaporative drying, and 4) finishing. Of these steps, the drying is the most energy-intensive since it requires evaporation of the water on the web. We estimate that of this amount, 27 percent (787 TBtu) (830 PJ) of primary energy was used for paper drying in 1994 (Martin et al. 2000). In current drying practices, after the paper sheet is formed and pressed to remove excess water and promote bonding of fibers, and no more water can be removed mechanically, the sheet moves through a series of 4050 steam heated cylinders, with the final consistency being about 90-95 percent solids content. With the Condensing Belt (or Condebelt) drying technology being developed by Valmet (a Finnish company), the paper is dried in a drying chamber by contact with a continuous hot steel band, heated either by steam or hot gas, rather than being run through the steam-heated cylinders. On the other side of the sheet are three layers: a fine wire gauze, a coarse wire gauze, and an externally cooled steel band. The evaporated water passes through the wire gauze and condenses on the steel band. The condensate is removed by pressure and suction (de Beer 1998). The benefit of the Condebelt technology is that it has the potential to completely replace the drying section of a conventional paper machine, and has a drying rate 5-15 times higher than conventional methods (Lehtinen 1995). Based on results from pilot plant tests performed by Valmet, de Beer (1998) estimates that for larger drying machines where losses through the seals of the drying chamber can be better controlled, steam savings are 10 to 20 percent of existing processes, while electricity consumption is expected to remain equal. The first commercial installations of the Condebelt technology were in Finland (1996) and South Korea (1999). These two plants produce industrial and packaging paper grades, and this technology may be applicable initially to continuous paperboard production (Huovila and Ojala 1999, Dimond 2000). Because the Condebelt has a higher drying rate than standard drying machines, the size of the Condebelt dryer can be reduced. The two plants have been constructed as add-on technologies to existing facilities, with minimal energy savings. However, larger savings are possible if the Condebelt were constructed as a full replacement. This technology is still in the early commercialization stage. Total costs for the installation of a paper machine including the forming and pressing can range from $850-1,300/ton (Hekkert and Worrell 1997). Initial cost for the demonstration facility were $260/ton paper (de Beer 1998). One estimate suggests that the cost of installing Condebelt for a greenfield (or newly constructed) plant would run up to double the cost of an existing cylinder machine (Ojala 2000, Ronkainen 2000). Other estimates suggest that the cost would be roughly the same (Worrell, Bode, and de Beer 1997, Hekkert and Worrell 1997). We assume a 25 percent increase from existing costs. Operations and maintenance costs are not expected to change significantly from current practice. Installation of Condebelt technology is expected to result in increased productivity (increased throughput, less capital expenditure) while also allowing for some improvement in product quality (Retulainen and Hmlinen 1999, de Beer 1998). There do not appear to be any significant technical barriers although no full scale large commercial operations have been installed in the U.S. There are, however, other competing commercial and emerging drying technologies that may limit rapid uptake by the U.S. market. While there appear to be limited technical barriers for this technology, it remains to be proven for a variety of paper grades (aside from linerboard) and has yet to make headway in the U.S. market. We believe that the market penetration for this technology by 2015 could be medium to low, and that several demonstration projects would probably be necessary to see how the technology fares under U.S. conditions.

53

LBNL and ACEEE

Condensing Belt Drying Data Table

Units Condebelt drying paper-2 Condensing belt drying Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Notes

Mton

Pulp and Paper Process heating Fuels, electricity New, retrofit 132.5

SIC 26

EIA, 1999.paper throughput estimate in 2015

Drying section, paper production tons 1 kWh 19 MBtu 8.6 8.7 MBtu Condensing belt drying system kWh 19 MBtu 7.3 7.5 MBtu Commercial 1996 Years 20 kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu 0 1.3 1.3 Medium 20% 0 34 34.1

Motor drive for machine rollers Steam use in drying cylinders

De Beer, 1998b De Beer, 1998b

Worrell et al., 1997a 0% De Beer, 1998b 15% De Beer, 1998b 15% Applicable to most paper grades. Demo currently w/ linerboard

Primary energy consumption of 3549 Tbtu in 2015

Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources

$ $ $/kWh $/Mbtu $/Mbtu Years %

260 Incremental 0 N/A 32.24 32.24 65.2 -9% Significant Somewhat None

Assume full paper machine costs of $1000/ton. Assume 50% drying end. $600/ton full investment cost. (De Beer, 1998; Worrell et al. 1997)

Discount rate for all CCE calculations is 15%

Reduced capital expenditure (small machines), higher production rate Improvement in strength properties

H,M,L

Low Marketing Low

One major supplier, non-US

H,M,L E,G,F,P

US demonstration at commercial scale Good EIA, 1999 Elaahi & Lowitt, 1988; Nilsson et al, 1995; Giraldo & Hyman, 1994; Jaccard & Willis, 1996 De Beer, 1998b Worrell et al., 1997a Retulainen, E., Hmlinen, A. 1999 De Beer, 1998b Retulainen, E., Hmlinen, A. 1999 Timo Ojala (Timo.Ojala@valmet.com)

54

Emerging Energy-Efficient Industrial Technologies

Direct Electrolytic Causticizing (Paper-3)

The pulp and paper industry is a large industrial energy user, with an estimated primary energy consumption of 2,970 TBtu (3133 PJ) in 1994, or 13 percent of manufacturing energy use (EIA 1997). Kraft pulp production is the predominant process, accounting for nearly 80 percent of the pulp produced in the U.S. and more than 700 TBtu (738 PJ) of primary energy (Kincaid 1998, Martin et al. 2000). In a typical Kraft mill, the extraction and reuse of the pulping chemicals following chemical or Kraft pulping consists of three stages: black liquor concentration, energy recovery, and recaustization of the remaining liquor. The concentration usually takes place in Multiple Effect Evaporators (MEEs) and Direct Contact Evaporators (DCEs) to drive up the final solids concentration to 70-80 percent. The black liquor is sprayed into the recovery boiler where the remaining water evaporates. The organic components of the solids burn, thereby releasing the heat that dries the liquor transferring heat to boiler tubes for heat generation. The heat of this combustion smelts the remaining inorganic chemicals, which flow from the furnace and are ready for recaustization. The smelt from the recovery boiler is mixed with some weak white liquor to form green liquor. This green liquor consists mostly of sodium carbonate (Na2CO3) and sodium sulfide (Na2S). The green liquor is recausticized by the addition of calcium hydroxide (Ca(OH)2) under controlled temperature and agitation. This recaustization converts the sodium carbonate back to sodium hydroxide (NaOH) and leaves a precipitate of calcium carbonate (CaCO3). The precipitate is removed, leaving white liquor that can be reused to pulp more wood. The calcium carbonate precipitate also feeds back into the process in the lime kiln, where it is heated to produce lime (CaO) which is then dissolved in water to produce the calcium hydroxide used in recaustization. The lime kiln is usually fueled by oil or gas, and requires on average 1.9 Mbtu/ton (2.3 GJ/t) pulp fuel and 14 kWh/ton (15 kWh/t) pulp electricity (Elaahi and Lowitt 1988, Jaccard and Willis 1996, Nilsson et al. 1995). Direct electrolytic causticizing is a process where, rather than using the traditional causticizing process and equipment, an electrolysis cell is used to remove carbonate from a molten smelt solution of sodium carbonate, sodium sulfide and sodium sulfate. Carbon (from carbonate) is removed from the system in the form of carbon monoxide and carbon dioxide. Sodium oxide (Na2O), the desired electrolytic product, is contacted with water to produce sodium hydroxide (NaOH). This product is then used for white liquor production early in the Kraft cycle (Wartena 2000). This technology is pre-commercial and is being developed by the Institute of Paper Science and Technology in Atlanta, Georgia. Initial funding was cost shared by the U.S. Department of Energy (DOE 1998) and the research is now being directly funded by industry. An electrolysis cell was designed and assembled in the laboratory. The next steps include running tests on the laboratory scale cell with mill smelt, improving the understanding of electrolytic fundamentals, and constructing a pilot scale plant (Pfromm 2000). The pilot scale plant will require commercial investment but interest has already been expressed. Initially, the technology will probably be used as an add-on for causticizing a partial stream of Kraft smelt, especially for mills facing capacity bottlenecks. The technology is compatible with smelt produced from combined cycle black liquor gasification and could be a component of more advanced mill designs (Pfromm 2000). On a final energy basis the recausticizing using electrolytic cells is expected to consume up to 50 percent less than the existing lime kiln configurations in plants, with a cell consumption estimated at 272 kWh/ton pulp (247 kWh/t) (Pfromm 2000). However, since the production of electricity is currently associated with losses of nearly two-thirds of the initial heating value of the fuels at the power plant, on a primary energy basis the consumption of the cells is expected to be on par or slightly below existing systems. The compelling driver, therefore, of this technology, is the capital cost savings. All recausticizing equipment of a 1,000 ton-per-day mill would be replaced by one electrolytic cell (5m2) (DOE 1998). Recent estimates for capital expenditures are $22 million for a 1,000 ton pulp/day mill, or roughly $60/ton pulp16. The electrochemical approach also promises to simplify the control of the process and improve product quality.

These costs are lower than the costs for rebuilding existing facilities. These are rough estimates based on similar costs to operate electrolytic reduction cells in aluminum facilities, and include costs for inert anode cells, cell controls, and the potline.

16

55

LBNL and ACEEE

Direct Electrolytic Causticizing Data Table

Units Direct electrolytic causticizing paper-3 One step causticizing process in Kraft mills Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Pulp and Paper Process heating Fuels, electricity New 70.0

SIC 26

Kraft pulp production estimate, Annual Energy Outlook, 2000

Existing recausticizing process (including lime kiln) ton Based on energy use per ton pulp 1 kWh Martin et al., 2000 14 MBtu Jaccard and Willis, 1996; Nillson et al., 1995 2.1 MBtu 2.3 Direct causticizing using electrolytic reduction kWh Pfromm, 2000 272 MBtu 0.0 MBtu 2.3 Pre-commercial 2005-2010 Years 10 kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % -258 2.1 0.0 low 10% -1811 15 -0.3 -10 Incremental 0 0.01 -0.93 40.17 N/A N/A Compelling Somewhat Somewhat H,M,L Somewhat Technical Medium Good EIA, 2000 Martin et al. 2000 Pfromm, 2000 ASME, 1999 Author estimate Pfomm, 2000 Pfromm, 2000 Peter Pfromm, IPST (peter.pfromm@ipst.edu) Assume that cell costs will be lower than existing costs -1912% Pfromm, 2000 100% 0% Initially mills with capacity bottlenecks

Discount rate for all CCE calculations is 15% No payback: net energy savings too low

Reduction of capital investment requirements Better control of causticizing process may lead to better product quality Reduction of dust and other emissions from lime kilns on site Pre-commercial but technology has support Still need technology to be tested Continued R&D

H,M,L E,G,F,P

While the promise of this technology is strong, and there is industry support, it is still in its early stages of development. It is unlikely that a full-scale working cell will likely be deployed until 2005, but given success, we could expect to see the technology begin to make headway initially for cases of incremental capacity expansion at particular mills, and later with replacement of existing technology.

56

Emerging Energy-Efficient Industrial Technologies

Dry Sheet Forming (Paper-4)

The pulp and paper industry is a large industrial energy user, with an estimated primary energy consumption of 2,970 TBtu (3133 PJ) in 1994 (EIA 1997). Papermaking (as opposed to pulp production) is usually divided into four basic steps: 1) stock formation and forming, 2) pressing (mechanical dewatering), 3) evaporative drying, and 4) finishing. Of these steps, the drying is the most energy-intensive since it requires evaporation of the water on the web. We estimate that of this amount, 27 percent (787 TBtu) (830 PJ) of primary energy was used for paper drying in 1994 (Martin et al. 2000). In the forming step, the continuous slurry that has been prepared is formed into a uniform web. The most common forming machines are twin wire formers (for thin and multi-layered sheets). Both machines deposit low consistency wood-pulp (less than 1 percent wood-pulp) onto a moving wire mesh which allows water to drain away. While originally conceived as a paper technology, dry web forming has developed into its own industry niche of non-wovens, which involves the production of light absorbent paper-like materials used in personal hygiene products. In dry web forming, the non-woven is produced without the addition of water. The fibers can be disbursed either through a carding technique or through an air layering technique. In the former, the fibers are disbursed mechanically while in the latter, fibers are suspended in air and paper is formed from this suspension. Fiber-to-fiber bonding is obtained by adding resins to the fibers or by spraying a polymer-latex onto the web formed. The air laying technique allows for a higher production rate and better control and most dry forming is done through this technique (de Beer et al. 1998b, Pivko 1999b). The advantage of dry sheet forming is the significant savings in energy requirements for evaporating water from the sheet in the later drying stage. De Beer (1998b) estimates a savings of 50 percent in drying fuel requirements and an increase in electricity consumption in an air layering plant of 150-250 kWh/ton paper to maintain the air stream and motor drive for the equipment. Airlaid drying/dry sheet forming technology was invented simultaneously by a Danish inventor named Karl Kroyer and by the Japanese at Honshu paper company (Pivko 1999b). Early conception of the technology occurred in the 1940s, but commercialization of todays processes took place in the early 1980s through Moeller and Jochumsen (M&J), a Danish firm (Pivko 1999b). Today, other producers of dry formed technology include Dan Web (Denmark) and Honshu Paper Co. (Japan) (de Beer 1998, Pivko 1999b). United Paper Mills-kymmene had originally licensed the Dan-Web technology (used at Walkisoft, Finland) but is now not a manufacturer of this technology (Pivko 1999b). Worldwide, installed capacity with this technology is about 350 ktons, and is growing rapidly. New capacity additions expected over the next couple of years are 120 ktons, mostly in North America (Pivko 2000). Current installed capacity in North America is estimated to be only 0.1 percent of total paper production in this region (Pivko 1999a, FAO 2000). The largest capacity plant is being constructed in North Carolina (Ward 2000). The primary products currently being produced with this technology are personal hygiene products (diapers, feminine hygiene, adult incontinence, training pants for babies, baby wipes), and some specialty areas (tableware, medical products, hot towels in restaurants). This is a small percentage of the overall paper tissue market as production has already shifted into the non-wovens. We estimate that the market replacement potential is 5 percent of U.S. paper production (Kincaid 1998). If this technology becomes applicable to the paper industry, direct investment costs could be one-third to one-half a conventional non-integrated paper mill (de Beer 1998). Operation and maintenance costs are also expected to be lower (de Beer 1998). However, the technology does not have the same type of machine speed as paper producers (1,500 m/min as compared to up to 6000 m/min on conventional paper machines) (Pivko 2000). Total costs for the installation of a paper machine (including the forming and pressing) can range from $850-1,300/ton (Hekkert and Worrell 1997). Air-laid technologies are slightly more expensive. A 55,115 ton (50,000 tonne) state-of-the-art plant in North Carolina under construction is being built at an estimated cost of $1,500/ton, the first project where costs have dropped below $2,000/ton (Pivko 2000). These lower costs may expand the potential market opportunities for this type of material. Aside from potential cost efficiencies that are associated with this technology, wastewater pollution is eliminated thereby allowing a more flexible location of paper mills closer to demand centers.

57

LBNL and ACEEE

Dry Sheet Forming Data Table

Units Dry sheet forming paper-4 Dry sheet forming Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Mton Paper drying tons kWh MBtu MBtu

Pulp and Paper Process heating Fuels, electricity New, retrofit 132.5

SIC 26

EIA, 1999.paper throughput estimate in 2015

1 480 9.2 13.3

Martin et al., 2000 Fuel use primarily in drying, not forming stage; Martin et al.,2000

Dry sheet forming kWh MBtu MBtu

Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years %

710 4.6 10.6 Commercial 1985 20 -230 4.3 2.3 Medium 5% -1521 28 15.5 350 Incremental 0 -0.24 13.02 23.87 48.3 N/A. None Significant Somewhat -48% 47% 18%

De Beer, 1998b De Beer, 1998b

Worrell et al., 1997

Currently applied only to specialty products; Pivko, 1999

Cost ranges from $1,500 to $2000/tonne (Pivko, 1999)

Discount rate for all CCE calculations is 15%

Improved product quality for personal hygiene products Reduction in water waste Technology already in the marketplace In niche market. Technology not likely applicable for broader application For the niche market Research/demonstration on applicability to other grades

H,M,L

H,M,L E,G,F,P

Technical High Good

EIA, 1999 Elahi & Lowitt, 1998; Nillson et al., 1995; Jaccard & Willis, 1996 De Beer, 1998 Worrell et al., 1997a Pivko, 1999 Pivko, 1999 De Beer, 1998; Pivko, 1999 Ivan Pivko, Notabene Associates Inc. ibpivko@aol.com; 941-3838404

58

Emerging Energy-Efficient Industrial Technologies It appears that dry sheet forming technology will continue to be developed for specialty applications and in the near future will not be adapted for production of standard paper grades. Rather, the higher quality product has caused a restructuring of tissue production for personal hygiene products in standard paper mills to non-wovens dry forming technology.

59

LBNL and ACEEE

Heat Recover PaperEnclosing Hoods (Paper-5)

The pulp and paper industry is a large industrial energy user, with an estimated primary energy consumption of 2,970 TBtu (3133 PJ) in 1994 (EIA 1997). Papermaking (as opposed to pulp production) is usually divided into four basic steps: 1) stock formation and forming, 2) pressing (mechanical dewatering), 3) evaporative drying, and 4) finishing. Of these steps, the drying is the most energy-intensive since it requires evaporation of the water on the web. We estimate that of this amount, 27 percent (787 TBtu) (830 PJ) of primary energy was used for paper drying in 1994 (Martin et al. 2000). In the drying section, steam filled rollers dry paper through the evaporation of water in the web. A typical drying machine may have up to 40-50 steam heated drying cylinders (de Beer 1998, Elaahi and Lowitt 1988). Heat recovery technologies are primarily directed at this initial stage of the drying section. In the middle of this section is the size press which can apply coating to the paper. The size press must be placed so that the paper can continue drying after coating because the coating itself must also dry. The last stage in the papermaking process is the Calendar stack, which is a series of carefully spaced rollers that control the thickness and smoothness of the final paper. There is a strong link between pulp consistency and steam demand on the drying section. Here, pulp enters with a consistency of 40-45 percent and paper exits the machine with a consistency of 90-95 percent (de Beer 1998, Abrahamsson et al. 1997). Typically 2 kg water are evaporated per kg of paper and 6.7 kg of air is required to remove 1 kg of water vapor (de Beer 1998). In the paper making process, the heat, which is mainly required in form of low-pressure steam, is transferred to the web via the steam-injected cylinders. As the water vapor exits the web, carried away by pre-heated air, and the web is dried, saturated lowpressure steam is released. The goal of more advanced waste heat recovery systems is to convert this lower quality steam into more useful heat. Existing equipment based on canopy air-to-air heat recovery systems recover about 15 percent of the energy contained in the hood exhaust air. There are several systems for heat recovery that can improve energy efficiency. One new system involves the installation of enclosed hoods and sensors on the drying section of the paper machine. Paper machines with enclosed hoods can require up to one-half the amount of air per ton of water evaporated than paper machines with canopy hoods. Thermal energy demands are reduced since a smaller volume of air is heated. Electricity requirements in the exhaust fan are also reduced optimizing drying efficiency (Elaahi and Lowitt 1988, CADDET 1994d). Another promising system further upgrades this waste heat by means of heat pumps and mechanical vapor recompression (MVR) (Van Deventer 1997, Abrahamsson et al. 1997). A different technology approach, which involves the heating provided to the cylinders, is to use stationary siphons to better extract the exhausted steam from the cylinders (Morris 1998). The heat can also be recuperated from the ventilation air of the drying section and used for heating of the facilities (de Beer et al. 1994). In 1994, U.S. paper drying consumption averaged about 8.6 Mbtu/ton (10 GJ/t) (Martin et al. 2000). Roughly 20 percent of the heat is required for air heating. By enclosing hoods, air heating requirements are minimized because of higher rates of heat recovery from the captured steam. Optimizing ventilation and using sensors control on the machine allows steam savings of 0.65 Mbtu/ton (0.75 GJ/t) paper and electricity savings of 5.7 kWh/ton (6.3 kWh/t) paper (CADDET 1994d). Conchie (1993) claims further savings of 0.86 Mbtu/ton (1 GJ/t) in a UK tissue mill. The savings was achieved by replacing the worn out Yankee hood and adding two novel features to the machine: the counter current series air flow (mainly of interest to manufacturers of tissue and machine-glazed papers) and humidity sensors (of general relevance to all papermakers). By using MVR to produce superheated steam from the water vapor extracts from the web, Van Deventer (1997) estimated steam savings of 50 percent and an increase in electricity consumption of 145 kWh/ton (159 kWh/t). Improved siphon technology can achieve savings up to 0.76 mbtu/ton (0.88 GJ/t). CADDET (1994d) notes a cost of $9.5/ton paper and additional O&M costs of $0.07/ton paper for the installation of a closed hood system that optimizes ventilation (CADDET 1994d, Conchie 1993). The addition of technologies to upgrade the heat (e.g. MVR and heat pumps) is estimated to be more expensive,

60

Emerging Energy-Efficient Industrial Technologies Heat Recovery Paper Data Table

Units Heat Recovery Paper (Enclosing hood) Paper-5 Heat recovery in paper drying Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Mton

Pulp and Paper Process heating Fuels, electricity New, retrofit 132.5

SIC 26

EIA, 1999.paper throughput estimate in 2015

Drying section paper production tons 1 kWh 18 MBtu 1.8 MBtu 2.0

Martin et al., 2000 (electricity share for the whole drying section) 20% of Fuel use in drying is for air heating; de Beer, 1998

Enclosing hood in the drying section of papermaking allows to recover the heat necessary for air heating kWh CADDET, 1994f; Van Deventer, 1997 12 MBtu CADDET, 1994f; Van Deventer, 1998 1.1 MBtu 1.2 Commercial Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 20 6.30 0.76 0.81 Medium 20% 166.9 20.1 21.6 9.5 Full cost 0.07 0.25 2.09 1.95 3.9 25% Somewhat None Somewhat Somewhat Medium CADDET 1994f; Conchie, 1993 O&M costs CADDET (1994f) 35% 41% 41% Author estimate, based on stock turnover of larger machines Based on lifetime of other drying technologies

Discount rate for all CCE calculations is 15%

Increased throughput Reduced emissions Safety. The steam is not discharged indoor Installations do already exist in EU Information

H,M,L

H,M,L E,G,F,P

Medium Continued demonstration Good EIA, 1999 Martin et al. 2000 (CADDET 1994f) (Conchie 1993). De Beer 1998; Martin et al. 2000 Both paper mills and waste paper mills (CADDET 1994f) (Conchie 1993). (CADDET 1994f) (Conchie 1993). CADDET 1994f; Willem van Zanten (w.van.zanten@novem.nl)

$17.6/ton paper (de Beer et al. 1994). Because the heat exchangers require frequent cleaning, the additional O&M costs will amount to $1.6/ton paper. In addition to energy savings, enclosing hoods and optimizing ventilation can also increase productivity. In one installation, the payback from increased material throughput (an additional 5,500 tons) was less than 1.5 years (CADDET 1994d). Enclosing hoods and optimizing ventilation can be a successful technology in the marketplace for all paper grades, and there might be a likelihood of achieving significant market penetration for the future. It is most likely that this technology would be installed for larger newer machines, so rapid market penetration is limited.

61

LBNL and ACEEE

High Consistency Forming (Paper-6)

The pulp and paper industry is a large industrial energy user, with an estimated primary energy consumption of 2,970 TBtu (3133 PJ) in 1994 (EIA 1997). Papermaking (as opposed to pulp production) is usually divided into four basic steps: 1) stock formation and forming, 2) pressing (mechanical dewatering), 3) evaporative drying, and 4) finishing. Of these steps, the drying is the most energy-intensive since it requires evaporation of the water on the web. We estimate that of this amount, 27 percent (787 TBtu) (830 PJ) of primary energy was used for paper drying in 1994 (Martin et al. 2000). In the forming step, the continuous slurry that has been prepared is formed into a uniform web. The most common forming machines are Fourdrinier machines (for thin sheets) and twin wire formers (for multilayered sheets). Both machines spray low consistency pulp (less than 1 percent pulp) onto a moving wire mesh which allows water to drain away. In high consistency forming, the process pulp enters at the forming stage, and has more than double the consistency (3 percent) than normal furnish. This measure increases forming speed and leads to energy savings in the pressing section, due to reduced de-watering and vacuum power requirements (Dudley, 2000; Martin et al. 2000). High consistency forming (HCF) was patented in the 1960s and developed by Ahlstrom Corporation in the 1970s (Dudley, 2000). Pilot plants using various technology configurations were tested in the mid-1980s, and the SymFlo HC design was commercialized in 1985. However, the technology has been slow to catch on, and currently there are only a few large scale operating installations worldwide (Dudley, 2000). The primary market currently for this technology is liquid packaging, although there is a possibility to expand it to the folding carton sector as well (Dudley, 2000). The driving force for its adoption is the potential savings in material, as the forming can lead to a 5-8 percent savings in basis weight (Eklund 2000; Dudley, 2000). The reason for this savings is that the technology uses a multi-ply (more than one headbox) configuration. This allows for setting the fibers in a bulkier three-dimensional configuration than normal headbox configurations, or in industry terms there are more fibers in the Z-direction This bulkier sheet allows the papermaker to achieve an equal caliper at less fiber weight (Dudley, 2000). The technology can be installed as an add-on technology to existing processes with some modifications. It is mainly geared toward paper grades where stiffness is highly valued. HCF forming is not viable for lightweight papers. It requires a minimum basis weight of 100 grams/square meter (due to fluid dynamics issues) and HCF formed in a single layer at such low basis weights is subject to tearing due to low machine direction tensile strength (Dudley, 2000, Elaahi and Lowitt 1988). Weyerhauser has been using a high consistency forming machine to produce liquid packaging board (e.g. cups, milk cartons) and this technology is also being deployed by International Paper for similar applications (Eklund 2000). Initial expectations for this technology were that it would yield significant energy savings. Early tests of the technology found an energy savings in the drying section of 10-15 percent (Elaahi and Lowitt 1988, Nomura et al. 1989). In practice these drying energy savings have not materialized. Rather, energy requirements for stock preparation and for vacuum and dewatering requirements (i.e. less pumping power) are reduced. We estimate a 20 percent saving in stock preparation estimated at 50 kWh/ton (Dudley 2000, Martin et al. 2000). Since this forming technology reduces the water content of the sheet entering the press section, less material (e.g. wire) is needed for dewatering. This can result in a 10-15 percent savings in capital costs for the wet end of the machine since it allows for reductions of the size of both the forming and drainage area (Dudley 2000, Eklund 2000). We estimate a capital cost savings of $10/ton ($11/t) paper and a slight increase in operation and maintenance costs of $0.6/ton ($0.7/t) paper (Dudley 2000, Jaccard and Willis 1996). Product quality is also expected to improve since the paper would have increased strength. The driving force behind the adoption of this technology is the potential savings in fiber (5-8 percent) and improved product quality for selected markets. As of now, high consistency forming has not penetrated broadly into the U.S. market. However, given the potential savings in material costs, we expect that for

62

Emerging Energy-Efficient Industrial Technologies particular boxboard applications (e.g. liquid packaging), this technology has a large potential to gain a broader acceptance into the U.S. market in the near term. High Consistency Forming Data Table
Units High Consistency forming paper-6 High Consistency forming Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Mton

Pulp and Paper Process heating Electricity New, retrofit 132.5

SIC 26

EIA, 1999.paper throughput estimate in 2015

Paper production tons kWh MBtu MBtu

1 480 9.2 13.3

Martin et al., 2000 Fuel use primarily in drying, not forming stage; Martin et al.,2000

Install high consistency former kWh 430 MBtu 9.2 MBtu 12.8 Commercial 1985 Years 20 kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ 50 0.0 0.4 Medium 9% 612 0 5.2 -10 Incremental 0.6 -0.02 -2.26 Immediate N/A Significant Significant Somewhat H,M,L Low Technical, marketing Medium

No change in electricity consumption

Demonstrated in Japan and Canada SymFlo HC model Worrell et al., 1997a 10% 0% Savings attributed to reduced energy use in drying stage 3% Exclude light grades of paper

$70/tonne for new installation Assume no additional cost for installation of high consistency former Operations and maintenance

$ $/kWh $/Mbtu $/Mbtu Years %

Discount rate for all CCE calculations is 15%

Potential of up to 8% reduction in raw materials requirements Less water use

H,M,L E,G,F,P

Demonstration Fair EIA, 1999 Jaccard and Willis, 1996, Giraldo & Hyman, 1994; Elaahi & Lowitt, 1988 Dudley 2000; de Beer 1998 Worrell et al., 1997a Dudley 2000; Eklund 2000 Dudley 2000; de Beer 1998 Eklund, 2000 Weyerhauser, Longview mill

63

LBNL and ACEEE

Impulse Drying (Paper-7)

The pulp and paper industry is a large industrial energy user, with an estimated primary energy consumption of 2,970 TBtu (3133 PJ) in 1994 (EIA 1997). Papermaking (as opposed to pulp production) is usually divided into four basic steps: 1) stock formation and forming, 2) pressing (mechanical dewatering), 3) evaporative drying, and 4) finishing. Of these steps, the drying is the most energy-intensive since it requires evaporation of the water on the web. We estimate that of this amount, 27 percent (787 TBtu) (830 PJ) of primary energy was used for paper drying in 1994 (Martin et al. 2000). In current drying practices, after the paper sheet is formed and pressed and no more water can be removed mechanically, the sheet moves through a series of 40-50 steam heated cylinders, with the final consistency being about 90-95 percent solids content. In conventional papermaking the web has a moisture content of 45-50 percent before entering the drying section. Impulse drying is a technology that improves the mechanical dewatering of paper and reduces the amount of water that needs to be removed in the drying section. In impulse drying the paper web is subjected to very high temperatures at the press nip in order to drive moisture out of the web so that the moisture content is significantly reduced (to 38 percent or less) before entering the drying phase (OIT 1999). The technology involves pressing the paper between one very hot rotating roll (300-900F) and a static concave a conventional shoe press. The pressure is about ten times higher than that in press and Condebelt drying (de Beer et al. 1998b, Boerner and Orloff 1994). Ultimately, consistencies of the sheet can be increased to 55 percent for board and 78 percent for lightweight paper using impulse drying, but the paper still needs to be fed through a conventional drying system after this stage (de Beer 1998). The impulse dryer can be retrofitted into an existing machine or incorporated into new models. For new machines, the size and costs of the paper machine can be reduced compared to existing processes, thereby making it more cost-effective. Also the drying rate can be significantly increased (50-500 times). This technology first began development in 1980 at the Institute of Paper Science and Technology in the U.S. with the collaboration of Beloit (de Beer 1998). The patents for this technology were originally licensed to Beloit and are now owned by Valmet, a Finnish company. While impulse drying is applicable to many grades of paper, initial U.S. efforts were directed toward the drying of newsprint and linerboard (de Beer 1998, IPST 1998). Successful production of reeled impulse dried linerboard took place in September, 1998 when an impulse dryer was tested on a Beloit paper machine (Orloff et al. 1999). Beloit research facilities tested a variety of pilot scale configurations, including the addition of a short and regular shoe press and hover press, to eliminate delamination problems (Orloff and Crouse 1999). Most recent, test trials have documented an increase in speed and an increase in speed, press dryness, and strength characteristics compared to existing technology (Orloff et al. 2000, Larsson and Orloff 2000). Given the higher consistencies of the paper or board entering the conventional drying section, drying energy consumption is significantly reduced. de Beer (1998) assumes potential steam consumption reductions of 40-50 percent with a small increase in electricity consumption of 5-10 percent (de Beer 1998). However, these estimates assume that the rotating roll is heated by fuel. Both the Canadian and U.S. pilot tests were based on electric induction heating of the rotating roll which reduces primary energy savings to closer to 15 percent (Orloff et al. 1999, CADDET 1995b). Incremental installation costs range from $70-100/ton paper although these cost data are not based on actual full-scale operating facilities (Jaccard and Willis 1996, Worrell, Bode, and de Beer 1997). Operation and maintenance costs are not expected to change since additional costs for the impulse dryer maintenance are reduced by the shorter machine (de Beer 1998). Impulse drying has been shown to produce paper which is thinner, smoother and stronger than that yielded by the conventional drying process (CADDET 1995b, IPST 1998, Orloff and Crouse 1999, Orloff et al. 2000). Trials with a South African furnish demonstrated increased production speeds by 14 percent and

64

Emerging Energy-Efficient Industrial Technologies Impulse Drying Data Table

Units Impulse drying paper-7 Impulse drying Market Information: I ndust r i es Enduse( s) Ener gy t ypes M ar ketsegm ent 2015 basecase use Reference technology D escr i pt i on Thr oughputorannualop.hr s. El ect r i ci t y use Fueluse Pr i m ar y ener gy use New Measure Information: D escr i pt i on El ect r i ci t y use Fueluse Pr i m ar y Ener gy use C ur r entst at us D at e ofcom m er ci al i zat i on Est .avg.m easur el i f e Savings Information: El ect r i ci t y savi ngs Fuelsavi ngs Pr i m ar y ener gy savi ngs Penet r at i on r at e Feasi bl e appl i cat i ons Ot herkey assum pt i ons El ec svgs pot ent i ali n 2015 Fuelsvgs pot ent i ali n 2015 Pr i m ar y ener gy svgs pot ent i ali n 2015 Cost Effectiveness I nvest m entcost Type ofcost C hange i n ot hercost s C ostofsaved ener gy ( el ec) C ostofsaved ener gy ( f uel ) C ostofsaved ener gy ( pr i m ar y) Si m pl e payback per i od I nt er nalr at e ofr et ur n Key non energy factors Pr oduct i vi t y benef i t s Pr oductqual i t y benef i t s Envi r onm ent albenef i t s Ot herbenef i t s C ur r entpr om ot i onalact i vi t y Evaluation M aj orm ar ketbar r i er s Li kel i hood ofsuccess R ecom m ended nextst eps D at a qual i t y assessm ent Sources: 2015 basecase Basecase ener gy use N ew m easur e ener gy savi ngs Li f et i me Feasi bl e appl i cat i ons C ost s Key non ener gy f act or s Pr i nci palcont act s Addi t i onalnot es and sour ces Notes

Mt on

Pul p and Paper Pr ocess heat i ng Fuel s,el ect r i ci t y N ew ,r et r of i t 132. 5

SI C 26

EI A,1999. papert hr oughputest i m at ei n 2015

Dr yi ng sect i on,paperpr oduct i on t ons 1 kW h 19 M Bt u 8. 6 M Bt u 8.7 I m pul se dr yi ng syst em kW h 170 M Bt u 6. 2 M Bt u 7.6 C om m er ci al 1996 Year s 20 kW h/ % M Bt u/ % M Bt u/ % % GW h Tbt u Tbt u $ $ $/ kW h $/ M bt u $/ M bt u Year s % 151 2. 4 1. 1 M edi um 20% 4009 64 29.5 70 I ncr em ent al 0 0. 07 4. 66 10. 04 20. 3 0% 810% 28% 13%

M ot ordr i ve f orm achi ne r ol l er s St eam use i n dr yi ng cyl i nder s

Or l of fetal . ,1999 Or l of fetal . ,1999

W or r el letal . ,1997a,At l as pr oj ect

I ni t i alpenet r at i on i n new spr i ntand l i ner boar d

Pr i m ar y ener gy consum pt i on of3549 Tbt ui n 2015 Ful li nvest m entcost$75100

Discount rate for all CCE calculations is 15%

Si gni f i cant Som ew hat N one H, M, L Low M ar ket i ng M edi um

R educed capi t alexpendi t ur e( sm al lm achi nes) ,hi gherpr oduct i on r at e I m pr ovem enti n st r engt h pr oper t i es

O ne m aj orsuppl i er ,nonUS

H, M, L E, G, F, P

U S dem onst r at i on atcom m er ci alscal e G ood EI A,1999 El aahi& Low i t t ,1988;N i l sson etal ,1995;G i r al do & H ym an, 1994;Jaccar d& W i l l i s,1996 D e Beer ,1998 W or r el letal . ,1997a M ar t i n etal . ,2000 D e Beer ,1998b;W or r el letal . ,1997a D e Beer ,1998b;O r l of fetal .1999 D .O r l of f ,I SPTG eor gi a( davi d. or l of f @ i pst . edu)

65

LBNL and ACEEE reduced basis weight (i.e. increased strength) by 2.5-5 percent, with an overall 20 percent improvement in productivity (Orloff et al. 2000). The drying section can also be reduced, resulting in lower capital costs. It allows an existing paper mill to operate at increased speeds (thus increasing production capacity), and allows for a new paper machine to significantly reduce the number of conventional drying rollers. While the technology is promising, there were problems initially with the paper delaminating or sticking to the roll (Boerner and Orloff 1994, Orloff and Crouse 1999). Recent research has focused on inhibiting sheet delamination through impulse drying at elevated ambient nip-opening pressures or through controlled depressurization (Orloff and Crouse 1999, Orloff et al. 2000). These new methods may actually improve the operational flexibility of the technology. Still, there is concern that technical obstacles for commercialization might be insurmountable (Ronkainen 2000). The creation of a commercial market for impulse drying has not yet become a reality and the development of full-scale commercial demonstration units will still be needed to help transition this technology to market. There does not yet appear to be significant backing for a large scale U.S. demonstration project and researchers at the Swedish pulp and paper research institute have recently stated that there is still a lot of work to be done before commercial application will be reality (Luiten 2000).

66

Emerging Energy-Efficient Industrial Technologies

Clean Fractionation (Chemicals-1)

Polyols are a major feedstock in the chemical industry. Polyols include ethylene and propylene glycol. The goal of the technology is to use a low-value feedstock in an innovative and economical manner. The majority of corn fiber in the United States (over 10 billion pounds) is sold as animal feed for approximately $0.04 per pound. Corn fiber fractionation aims to create a high value product from this inexpensive feedstock. Ethylene glycol and propylene glycol are used to produce antifreeze, polyester, fiberglass reinforced plastics for use in boat hulls, and construction pipes. Efficient separation technology may turn a corn wet mill into an economical producer of ethylene glycol, propylene glycol, and other sugar-derived products for the commodity chemical industry. Currently, most of these chemicals are produced from petrochemical reagents. Corn fiber fractionation promises to turn a corn wet mill into a low-cost producer of industrial chemicals. The innovative fractionation technology is being developed to cleanly and selectively remove hemicellulose from the corn fiber and to subsequently separate and isolate the xylose and arbainose fraction. Hemicellulose makes up 60 to 70 percent of the weight of corn fiber, and xylose and arabinose make up about 60 to 70 percent of the weight of the hemicellulose. Catalytic corn conversion of xylose and arabinose into ethylene and propylene glycol would produce a valuable feedstock with a very wet milling large market and a variety of applications.
corn fiber Currently most ethylene glycol and High value propylene glycol are produced in a similar niche market chemicals reaction by acid or thermal catalysis of ethylene oxide and propylene oxide, Xylose, arabinose food respectively. The yield of glycol in this additives process usually exceeds 95 percent. The Bioactive conditions of the reaction are dependent of compounds whether or not an acid catalyst is used. Less severe reaction conditions are needed in the presence of a catalyst. The nonEthylene glycol Propylene gylcol xylitol catalytic process requires higher temperatures and pressures. Fractionation of corn fiber offers 12 percent overall energy savings in the production of glycols and a payback period of under two years. Other benefits include the use of a plant-based, renewable corn feedstock.

Ethylene glycol and propylene glycol are used to produce a variety of products such as antifreeze, and also as a major feedstock for the plastics industry. In this analysis, we have considered the plastics industry as the main market for these two chemicals for the sake of calculating their future demand. According to the Annual Energy Outlook 2000, the plastics industry is expected to grow by 2.9 percent annually until 2015. We have also assumed that 30 percent of all new capacity for producing ethylene and propylene glycol. The new facilities will be located at current wet-milling sites. This would allow the mills to sell a much more high-valued product, in addition to the feed that they already sell. Some of the barriers to the implementation of this technology include the variable price of corn and the relatively high initial capital equipment costs of equipping a corn mill with fractionation equipment. The success of this technology is perhaps most highly dependent on a low corn purchase price. The barrier of capital cost could perhaps be overcome by the attractive payback of this technology.

67

LBNL and ACEEE

Clean Fractionation Data Table

Units Clean Fractionation Chem-1 Replace petrochemical glycol production Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Notes

tons

Chemicals Process heating, other Electricity, gas, coal, other New 5600000

SIC 28 The process requires a heated and pressurized reaction vessel

Scaled up production (1997) using EIA 1999 growth in plastics industry - this represents 0.01% of the energy use in the U.S. chemical industry and 0.5% of the energy use in industrial organic chemicals

Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources

Petrochemically derived ethylene glycol and propylene glycol (by acid or thermally catalyzed hydration of ethylene or propylene 1.0 tons 4.67% electric - EIA 1997 0.000056 kWh 3.9 MBtu 61% natural gas, 26.9% other, 6.72% coal - EIA 1997 3.9 MBtu Ethylene glycol production uses 4.09 * 10^9 btu/ton - DOE 2000a Separation of corn fiber to produce ethylene glycol, propylene glycol, and xylitol 0.00012 kWh 10.98% electric - EIA 1997 3.4 MBtu 45% coal, 39.31% n.g., 2.89% other, 0.578% resid fuel oil - EIA 1996 3.4 MBtu OIT 1999 Pilot plant 2005 15 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % -6.88E-05 0.464 0.464 Medium 13% -0.000050 0.338 0.338 130 Incremental -66.7 646484.40 -95.84 -95.84 1.9 52% Somewhat None Significant Significant Medium -123% 12% 12% New capacity additions Savings potential applied to 47,000 tons of fractionated product in 2015 Savings potential applied to 47,000 tons of fractionated product in 2016 Savings potential applied to 47,000 tons of fractionated product in 2017 OIT 1999 Assumed 30% of production savings based on OIT 1999

H,M,L

Uses a renewable feedstock, reduces 1.8 million tons of waste by 2010 Lower production costs Eastman Chemical has pilot plant and moving towards scale-up

H,M,L

Equipping corn mills with process and distribution equipment Infrastructure Medium Co-funding of production scale demonstration E,G,F,P Good EIA 1999 DOE-OIT 2000a OIT 1999 OIT 1999 Personal Communication with Joe Bozell, NREL (303) 384-6276 OIT 1999 OIT 1999

68

Emerging Energy-Efficient Industrial Technologies

Gas Membrane Technologies (Chemicals-2)

Gas separations are some of the most challenging processes in the chemical industry. Mixtures are difficult to separate in their gas phase and generally are separated with the use of adsorbents. If useable products are desired after separation, mixtures are frequently brought into the liquid phase (through pressure or temperature control) and extracted. The use of membrane technology to separate or purify liquids is now well established. Since the early 1990s, this technology has been used in a variety of industries including the food and beverage industry, water companies and the chemical industry (CADDET 1999c). One of the most energy-intensive unit operations in the chemical industry is separation (DOE-OIT 2000a). Separation technologies include distillation, fractionation, and extraction. Certain mixtures of chemicals cannot be separated beyond a certain point by standard distillation processes and must undergo extraction. Azeotropic mixtures such as isopropyl alcohol and water fall into this category. Extraction takes advantage of the relative solubilities of solutes in immiscible solvents. If the solutes are in an aqueous solution, an organic solvent that is immiscible with water is added. The solutes will dissolve either in the water or in the organic solvent. If the relative solubilities of the solutes differ in the two solvents, a partial separation occurs. The upper, less dense solvent layer is physically separated from the lower layer. The separation is enhanced if the process is repeated on each of the separated layers. Gas membranes offer an alternative to liquid-liquid extraction that uses much less energy. This technology can be used to separate organic mixtures. The example of separating a mixture of methanol and water, membrane separation uses 17 percent less fuel than liquid-liquid extraction. Separation processes account for one quarter of the process energy to produce isopropyl alcohol. Membrane separators also tend to cost about 10 percent less than traditional separation units. The annual operating costs of membranes tend to run a bit higher than other separators. Membranes must be replaced rather frequently and foul easily. The aim of this gas separation technology is to avoid the energy consumption associated with the condensation of an azeotropic vapor mixture of methanol and water. The process is necessary because the azeotropic nature of the original mixture makes it impossible to separate the two fractions by simple distillation. The membrane used in this example acts like a molecular sieve. It separates a mixture of methanol and water by allowing the water molecule to pass through the filter, while retaining most of the methanol. The entire process takes place in the vapor phase. A large potential market for gas membrane separators is mobile and stationary fuel cells. One of the types of fuel cells that has promise for mobile applications is the proton exchange membrane (PEM) fuel cell. The U.S. Department of Energy along with the U.S. Department of Transportation has been conducting research and demonstration projects in this area. Progress in gas membrane technologies will aid the commercialization of this technology as well. The market for gas membrane separators will encompass every portion of the chemical industry. While industrial organic chemicals will dominate the market for membranes, other industries such as the food and pulp and paper industries can benefit from improvement of membrane processes. The organic chemical industry is forecasted to grow by 15 percent between the years 2000 and 2015. The market for membranes remains even larger because of the relatively few processes for which they are currently used for separation. Membrane science continues to be evolving. Membranes with varying qualities are continuously being developed for the separation of specific gas mixtures. One of the ways in which membranes could be improved is by increasing their lifetime and by decreasing their sensitivities to fouling. Many gas membranes for example are fouled by exposure to sulfur. Sulfur-resistant membranes would be a great improvement for many processes in the petrochemical industries.

69

LBNL and ACEEE

Gas Membrane Technologies Data Table

Units Gas Membrane Technologies Chem-2 Replace liquid/liquid extraction Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Notes

tons

Chemicals Process heating, other Electricity, gas, coal, other New, replace on failure 6915000.0

SIC 28

Projected production of methanol EIA 1999, DOE 2000a. As much as 50 percent (2135 trillion btu/year) of the energy use in the organic chemical industry is used for separations.

Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources

Distill the mixture to its azeotropic point, then do a liquid/liquid extraction 1.0 tons 11.36 kWh CADDET 1999a 2.32 MBtu Assumed that the heat for distillation required 20% more energy than mem sep. 2.42 MBtu Distill the mixture to its azeotropic point, separate with gas membrane kWh 1.93 MBtu 1.93 MBtu CADDET 1999a Commercialized 1997 15 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 11.36 0.386 0.483 Medium 3% 1.96 0.067 0.084 -1.0 Incremental 1.62 0.13 3.75 3.00 10.2 8% None Somewhat Significant Significant Medium Availability High Establish markets Good 100% 17% 20% Estimated market share of separated methanol Savings potential applied to 5% of the methanol market in 2015

1,100,000 for installation of new unit CADDET 1999c (for 20 ton/day facility) Operating costs are lower, but membrane must be replaced frequently ($11,823 annually for 20ton/day facility)

Based on fuel mix in US from EIA 1997

H,M,L

Decreases CO2 emissions by 0.1325 tons/ton product per year Investment 10% less below conventional installation Morton International BV has operating facility Limited production of specific membranes Membrane technology can be used in the food and petrochem industries

H,M,L E,G,F,P

CADDET 1999c, EIA 2000 CADDET 1999c CADDET 1999c CADDET 1999c Assumption of analyst CADDET 1999c CADDET 1999c

70

Emerging Energy-Efficient Industrial Technologies

Heat Recovery Technologies for Harsh Environments in Chemical Manufacturing (Chemicals-3)

The use of heat recovery in the chemicals industry is very common. The various processes used in the industry require extreme temperatures and often need rapid changes in temperature. Heat exchangers are used throughout the industry to provide efficient use of energy and to improve process control. Compact heat exchangers, which have a comparatively large amount of surface area in a given volume, are highly efficient and offer greater flexibility. There are, however, applications where conventional heat exchanger technology cannot be applied. This is true when the conditions include excessive temperature or high pressure, or when the medium passing through the heat exchanger is corrosive or caustic. Recent advances in the construction of heat exchangers have yielded equipment that can be used in environments that were previously too extreme (Reay 1999). These advances have come in two ways. First, new materials that are resistant to corrosion are used in the construction of heat exchangers. Second, novel designs and manufacturing techniques have led to heat exchangers that can tolerate higher temperatures and pressures. Because heat exchangers can now be used in these extreme situations, more heat from the process can be captured and utilized, therefore leading to energy savings. This profile highlights two opportunities for heat exchangers under harsh conditions. The production of sodium hydroxide (caustic soda) begins with the electrolysis of brine to generate chlorine and an aqueous solution of sodium hydroxide. The sodium hydroxide solution needs to be evaporated to various extents to yield the concentrations of the marketed products. This solution is highly corrosive and reaches temperatures up to 195C during evaporation. Conventional heat exchangers are damaged and show leakage in this application. They can be replaced by printed circuit heat exchangers (PCHE), which are made by chemically milling channels into flat metal plates that are then diffusion bonded into blocks. Nickel is used to resist corrosion. An installation of this type of PCHE at a sodium hydroxide plant required an investment of roughly $20,000, and led to annual energy savings of 3,980 Mbtu (4,200 GJ). The greatest benefit of this project was that is permitted large increases in production, generating additional revenue of approximately $200,000 for the plant (CADDET 1992). Another corrosive application for novel heat exchangers is the production of nitric acid. Nitric acid is produced by burning ammonia in air at high temperature, then oxidizing the product and absorbing it into water. During production, the gas must be cooled from roughly 900C to below 60C before absorption can take place. The heat released while the gas is cooled down to 200C is captured for use in other processes at the plant. Below 200C, the gas can condense and corrode the heat exchanger, so heat recovery has not been possible. To avoid the corrosion problem, the heat exchanger can be constructed with a nickel/chromium alloy and can be designed to prevent acid re-evaporation. A plant adopted this novel approach and recorded annual energy savings of 130,000 Mbtu (137,000 GJ). This led to a three-year payback on the $1.2 million investment (CADDET 1993b). The PCHE used in the sodium hydroxide application was produced by Heatric Ltd, located in Dorset, UK. This company reports that the majority of their sales for compact heat exchangers are to offshore oil processing plants because these heat exchangers are much more compact and lighter than their conventional counterparts, making them economical in offshore applications where expensive structural supports are needed (McCormack 2000). Heatric has experience with the chemical industry, including both sodium hydroxide and nitric acid production, and report that the drivers behind the use of compact heat exchangers are corrosion resistance, multi-stream capability, and debottlenecking. Many other companies market compact heat exchangers to the chemical industry, including Alfa Laval and APV. U.S. production of sodium hydroxide and nitric acid in 1997 was 11.8 million tons (10.7 million tonnes) and 9.5 million tons (8.6 million tonnes), respectively. These two products fall under the industrial groupings Alkalis and Chlorine (SIC 2812) and Nitrogenous Fertilizers (SIC 2873). Energy consumption for these two classifications totaled just over 500 TBtu of primary energy use for 1994, which accounts for one-eighth of total energy use in the chemical industry (AEO 1999). Assuming that these sub-sectors follow the same growth path forecast for the chemical industry as a whole, and assuming they experience the same changes in energy intensity expected across the sector, their primary energy use will grow to 613 TBtu by 2015. 71

LBNL and ACEEE

Heat Recovery Chemicals Data Table

Units Notes Heat Exchangers for the Chemical Industry Chem-3 Novel heat exchangers for aggressive environments: high temperature and pressure, corrosive products Market Information: Industries Chemical products Specifically, manufacture of sodium hydroxide and nitric acid, SIC 2812 and SIC 2873 End-use(s) Process heat Energy types Fuels Market segment Retrofit 2015 basecase use Tbtu 613.2 1994 Use by SIC 2812, SIC 2873, assuming same growth and change in intensity as the chemicals sector as a whole, EIA 1997, EIA 1999 Reference technology No heat recovery in highly corrosive applications Description Throughput or annual op. hrs. 17 Electricity use TWh Combined output of sodium hydroxide (10.6 Mt) and nitric acid (8.6 Mt), 1997 356 Fuel use TBtu These energy values for summed for 2 four-digit sectors: 2812 Alkalis and 502.2 Primary energy use TBtu Chlorine, and 2873 Nitrogenous Fertilizers, MECS 1994 New Measure Information: Description New heat exchanger design to tolerate harsh environments and allow for greater heat recovery 17 Electricity use TWh Fuel use TBtu 334 480 Primary Energy use TBtu Commercialized, Research Current status Depends on specific application Date of commercialization 1995 Est. avg. measure life Years Despite high tolerance, will eventually foul in extreme environments 10 Savings Information: Electricity savings TWh/% 0 0% Fuel savings TBtu/% 22.0 6% Primary energy savings TBtu/% 22.0 4% Medium Penetration rate 30% Feasible applications % Other key assumptions Savings observed in the case studies are for typical plants of each type. 0 Elec svgs potential in 2015 TWh 8.1 Fuel svgs potential in 2015 TBtu 8.1 Primary energy svgs potential in 2015 TBtu Cost Effectiveness 8 Investment cost $/Mbtu Estimate of capital investment based on sample projects Full cost Type of cost 0 Change in other costs $ Too variable to quantify 0.01 Cost of saved energy (elec) $/kWh 1.63 Cost of saved energy (fuel) $/Mbtu 1.63 Cost of saved energy (primary) $/Mbtu Discount rate for all CCE calculations is 15% 2.4 Simple payback period Years 42% Internal rate of return % Key non energy factors Moderate (site-specific) Productivity benefits Improve plant operation (debottle-necking), increased yield Product quality benefits None Environmental benefits None Other benefits Small volume and weight lowers installation costs relative to standard heat Somewhat exchangers Current promotional activity H,M,L Medium Evaluation Major market barriers Awareness,Perceptions Fears of fouling and corrosion, conservatism in user industries. Likelihood of success H,M,L Medium Recommended next steps Dissemination,demonstration projects in US link with training about process integration & pinch analysis Data quality assessment E,G,F,P Own estimates based on literature survey Fair Sources: 2015 basecase EIA, 1999; EIA 1997 Basecase energy use EIA, 1997 New measure energy savings Average of CADDET studies Lifetime Author judgement Feasible applications Author judgement Costs Average of CADDET studies Key non energy factors Principal contacts Heatric, Ltd. (Des McCormack), Additional notes and sources

72

Emerging Energy-Efficient Industrial Technologies While these heat exchangers are also applicable to other harsh chemical environments, we focus in this assessment on nitric acid and sodium hydroxide production. We estimate that these heat exchanger technologies could be adopted in 30 percent of all nitric acid and sodium hydroxide production in the U.S. by 2015. This would lead to energy savings of 8.1 TBtu per year. Based on the reported projects, the cost for these installations would be roughly $8 per MMBtu saved annually, which, given an average fuel price to the chemical industry of $3.42 per MMBtu, indicates a payback period of 2.4 years. The payback could be considerably faster depending on the site-specific factors of the application. For example, if the introduction of the heat exchanger alleviates a bottleneck in production the increase in output provides a large productivity benefit (Reay 1999). In the sodium hydroxide project, the increased productivity lowered the payback period to five weeks (CADDET 1992). Other benefits that accompany compact heat exchangers in these applications are reduced maintenance and replacement costs, and lower installation costs due to the reduced size and weight of the equipment. This technology is likely to be successful. The entire class of compact heat exchangers accounts for only 510 percent of the sales at this time, but their sales are increasing much more rapidly than total heat exchanger sales (Reay 1999). The key to promoting these technologies is to disseminate information and establish demonstration projects in the U.S. that illustrate the use of compact heat exchangers in harsh environments.

73

LBNL and ACEEE

Levulinic Acid for the Manufacture of Chemicals (Chemicals-4)

Biomass can be used as a raw material to produce large numbers of chemicals (or chemical intermediates), yet to date, many of these products have failed in the marketplace because they frequently are not economically viable and face high production costs. Levulinic acid (LA) holds promise as an inexpensive feedstock for producing many industrial chemicals and products. The two chemicals that could significantly increase the market for levulinic acid are methyltetrahydrofuran (MTHF), a fuel additive, and delta-amino levulinic acid (DALA), a biodegradable herbicide/pesticide. This measure is actually a group of process technologies that aims to create a greater market for levulinic acid from biomass by improving the production methods of MTHF and DALA. The Biofine Corporation developed a process using acid hydrolysis of 6-carbon sugars as the key step for LA production. This process minimizes side product formation and the resulting separation problems associated with them by significantly improving the tradition engineering of the LA production process through a novel, two reactor system. The figure below demonstrates the path for DALA and MTHF production. The technology is being demonstrated on a one ton/day scale at a facility in South Glens Falls, New York. The MTHF is produced in a greater than 80 percent yield via a Waste Cellulosic Material single stage catalytic hydrogenation process. The process for forming DALA Biofine affords a product with a purity of greater than 90 percent, giving a Levulinic Acid process that is commercially viable. The DALA process is currently being improved in three New Process New areas: converting by-products of DALA production for use in plastics manufacturing, using new DALA (biodegradable insecticide) MTHF (fuel additive) reagents to simplify the production of DALA, and purifying the final product to remove a salt generated during production. To minimize waste streams, solvent and by-products are being recovered and reused at each step of the process. Currently, levulinic acid has a worldwide market of about one million pounds per year at a price of $4-6 per pound. Large-scale commercialization of the Biofine process could produce levulinic acid for as little as $0.32 per pound, spurring increased demand for LA and its derivatives. The current levulinic acid demonstration plant in South Glens Falls, NY uses paper mill sludge as the raw material. The plant was originally producing 1-2 tons of levulinic acid per day and has increased its output to 4-6 tons per day. Commercialization efforts have been underway with support from Biometics, Inc. Engineers are currently attempting to increase production capacity and reduce operation and maintenance costs. The demonstration plant is still on a pilot-scale. Many cost issues can be resolved once the plant is operating at full scale. Once the economics are in place, other producers of levulinic acid may consider implementation of the Biofine method of production. Levulinic acid holds the most promise as a precursor to the fuel additive MTHF. Especially in the American gasoline market, where per capita fuel consumption continues to increase, MTHF is becoming an important product.

74

Emerging Energy-Efficient Industrial Technologies Levulinic Acid Data Table

Units Levulinic Acid from Biomass (biofine) Chem-4 Replace dehydrative treatment with acid Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

tons

Chemicals Process heating, other Electricity, gas, coal, other New 150000

DOE 2000a, EIA 1999 - currently, levulinic acid is a niche chemical, but new product applications could increase the market.

Dehyrdrative treatment of biomass or carbohydrates with acid 1 tons www.epa.gov/greenchemistry/sba99.htm 85 kWh www.epa.gov/greenchemistry/sba99.htm 42 MBtu www.epa.gov/greenchemistry/sba99.htm 42.7 MBtu Acid hydrolysis of 6-carbon sugars 80 kWh 38.0 MBtu 38.7 MBtu Demonstration 2002 20 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 5.00E+00 4.000 4.043 High 15% 0.113 0.090 0.091 1000 Incremental -640 -96.05 -120.06 -118.80 1.53 65% Significant None Significant Significant High $344,000 for installation of demonstration scale plant www.pnl.gov/news/1998/98mthf $0.32/lb www.epa.gov/greenchemistry/sba99.htm 6% 10% 9%

www.epa.gov/greenchemistry/sba99.htm www.epa.gov/greenchemistry/sba99.htm 4-6 ton per day plant in South Glens Falls, NY

Produces high yield capacity with less waste and fewer byproducts Reduces landfill waste Makes the production of levulinic acid economical Biometics, Honeywell, NYSERDA involved in development

H,M,L

H,M,L E,G,F,P

High Scale-up Good

The demonstration plant is increasing capacity

DOE 2000a, EIA 1999 www.epa.gov/greenchemistry/sba99.htm www.epa.gov/greenchemistry/sba99.htm www.epa.gov/greenchemistry/sba99.htm www.epa.gov/greenchemistry/sba99.htm www.epa.gov/greenchemistry/sba99.htm

75

LBNL and ACEEE

Liquid Membrane Technologies - Chemicals (Chemicals-5)

The U.S. chemicals industry makes up more than 10 percent of the U.S. manufacturing gross domestic product. This is greater than any other sector including food, machinery, motor vehicles, aerospace, and electronics (Census 1996). The industry employs nearly 850,000 workers at 12,000 plants nationwide (Census 1996). The industry produces so many products that it eludes clear definition. Most industrial chemicals, in fact, are consumed by chemical-related businesses. Steel and aluminum mills, paper mills, semiconductor manufacturers, drug companies, carpet mills, and battery producers are all relatively large customers. The chemical industry uses many different fuel sources for its energy needs (i.e. natural gas, electricity, coal, and fuel oils), with nearly 50 percent of the total used as feedstocks (DOE-OIT 2000a). Although they vary widely from product to product, energy expenditures can represent a significant portion of manufacturing costs in the industry. One of the most energy-intensive unit operations in the chemical industry is separation. Separation technologies include distillation, fractionation, and extraction. Certain mixtures of chemicals cannot be separated beyond a certain point by standard distillation processes and must undergo extraction. Azeotropic mixtures such as isopropyl alcohol and water fall into this category. Extraction takes advantage of the relative solubilities of solutes in immiscible solvents. If the solutes are in an aqueous solution, an organic solvent that is immiscible with water is added. The solutes will dissolve either in the water or in the organic solvent. If the relative solubilities of the solutes differ in the two solvents, a partial separation occurs. The upper, less dense solvent layer is physically separated from the lower layer. The separation is enhanced if the process is repeated on each of the separated layers. Liquid membranes offer an alternative to liquid-liquid extraction, and use much less energy. This technology can be used to separate both aqueous and organic mixtures. The example of separating a mixture of isopropyl alcohol and water, membrane separation uses 60 percent less fuel than liquid-liquid extraction. Separation processes account for one quarter of the process energy to produce isopropyl alcohol. Membrane separators also tend to cost about 10 percent less than traditional separation units. The annual operating costs of membranes tend to run a bit higher than other separators. Membranes must be replaced rather frequently and foul easily. The market for liquid membrane separators will encompass every portion of the chemical industry. While industrial organic chemicals will dominate the market for membranes, other industries such as the food and pulp and paper industries can benefit from improvement of membrane processes. The organic chemical industry is forcasted to grow by 15 percent between the years 2000 and 2015. The market for membranes remains even larger because of the relatively few processes for which they are currently used for separation. One of the largest barriers facing liquid membranes is limited production. Liquid membranes are highly specific with regards to the compounds that they can separate, therefore differing processes will require differing membranes. More research and development is needed to improve the performance of these technologies.

76

Emerging Energy-Efficient Industrial Technologies Liquid Membrane Technologies - Chemicals Data Table
Units Liquid Membrane Technologies Chem-5 Replace liquid/liquid extraction Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Notes

tons

Chemicals Process heating, other Electricity, gas, coal, other New, replace on failure 805680

SIC 28

2015 production of isopropyl alcohol AEO 2000,DOE 2000a (15% up from 1997). As much as 50 percent (2135 trillion btu/year) of the energy use in the organic chemical industry is used for separations.

Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources

Distill the isopropyl/water mixture to its azeotropic point, then do a liquid/liquid extraction 1.0 tons 120.90 kWh 11% electricity EIA 1997 8.36 MBtu 89% fuel EIA 1997 9.39 MBtu 25% of energy (4693 btu/lb) is for separation. DeBeer 1994 Distill the mixture to its azeotropic point, separate with liquid membrane 120.90 kWh 3.34 MBtu Technology saves 60% of separation fuel input MBtu 4.38 Commercialized 2000 10 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 0.00 5.016 5.016 Medium 20% 0.0 0.81 0.81 -7 Incremental 17 3.11 3.11 11.2 6% None None Significant Significant High Availability Medium Establish markets Good $62.6/ton for full installation of membrane separator DeBeer 1994 Operating costs are lower, but membrane must be replaced frequently 0% 60% 53%

Fuel mix in US from EIA 1997

H,M,L

Decreases CO2 emissions Investment 10% less than conventional installation Dow Chemical promoting Limited production of specific membranes Membrane technology can be used in the food and petrochem industries

H,M,L E,G,F,P

DOE 2000a, EIA 2000 DeBeer 1994 DeBeer 1994 DeBeer 1994 DeBeer 1994 DeBeer 1994

77

LBNL and ACEEE

New Catalysts (Chemicals-6)

Catalysis is the phenomenon by which certain chemicals (catalysts) can speed up a chemical reaction without undergoing any permanent chemical change themselves. They can be recovered after a reaction and used repeatedly (although most catalysts have finite lifetimes). Catalysts lower the activation energy required for a reaction to complete. Without the right catalyst, many reactions do not progress. Furthermore, the chemical nature of the catalyst can have a radical effect in selecting reaction pathways leading to different chemical products. Over recent decades there has been enormous progress in understanding the underlying molecular mechanisms, which has had an explosive effect on the development of new catalyst systems. About 80 per cent of processes in the chemical industry now depend on catalysts to work efficiently, and the number is rising. New catalysts are being designed and new catalytic processes being devised that aim to produce cleaner and more efficient chemical processes. These use less energy, and environmentally acceptable agents (for example, air or oxygen as an oxidant instead of hydrogen peroxide) and perhaps water as a solvent, resulting in less noxious waste. Since the chemical industry is so diverse and produces such a large quantity of chemicals, the exact impact of new catalysts is difficult to predict. For this analysis, a very common industrial chemical, ethylene, was used as an example. Current production of ethylene consumes 8,197 Btus/lb. A new catalyst could lower the energy consumption by 20 percent. One of the more promising areas in catalytic research is the area of nanoscale catalysts. One typical objective of nanoscale catalyst research is to produce a material with exceedingly high selectivity at high yield in the reaction product or product slate, that is, chemicals by design, with the option of altering the product by changing the surface functionality or composition at the nanoscale. For instance, new catalysts with increasing specificity are now being fabricated in which only one or two spatial dimensions are of nanometer size. A second objective is to discover nanoscale materials or structures with exceedingly high storage capacity per unit volume and weight for gases such as H2 or CH4, which would then be more economic for use either as a combustion fuel or as the means to power fuel cells for ultralow-emission vehicles or for electric power generation. A third objective is to fabricate molecular sieving membranes using inorganic crystalline materials such as zeolites. For molecular sieving membranes, one critical challenge rests on discovering ways to create large-scale, thin, nearly defect-free membranes. Most large chemical companies have a research group that is devoted solely to the development of new catalysts. Improvements are constantly occurring, although many of the technologies are proprietary. One of the barriers that faces catalyst research is the high cost of catalytic ligands. Many of these, especially in the specialty pharmaceutical industries can cost upwards of $50,000 per pound. This presents a problem in the commodity chemical sector where products are sold for minimal or no profit. The largest market for catalysts may be in the industries with high-valued products such as pharmaceuticals and biotechnology.

78

Emerging Energy-Efficient Industrial Technologies New Catalysts Data Table

Units New Catalysts Chem-6 Replace pyrolisis of hydrocarbons to produce ethylene (example) Market Information: Industries Chemicals End-use(s) Process heating, other Energy types Electricity, gas, coal, other Market segment New, retrofit 2015 basecase tons 27743750 Notes

SIC 28

2015 production of ethylene EIA 1999, DOE 2000a (15% up from 1997). This is an example - virtually all chemical processes use catalysts, therefore improvements would greatly lower energy use in the industry.

Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources

Distill the isopropyl/water mixture to its azeotropic point, then do a liquid/liquid extraction tons 1.0 kWh 11% electricity EIA 1997 528.40 14.59 MBtu 89% fuel EIA 1997 16.39 MBtu Process energy of 8197 Btu/lb, DOE 2000a Pyrolysis of hydrocarbons kWh 423 11.67 MBtu 13.11 MBtu R&D 2005 20 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 105.68 2.917 3.278 Low 15% 439.8 12.14 13.64 50 Incremental 4 0.11 4.11 3.66 7.9 11% None None Somewhat H,M,L Medium Catalyst ligands costs $30k-$50k per pound Operating costs are lower, but catalyst must be replaced frequently 20% 20% 20%

DOE 2000a DOE 2000a Catalysts could save 20% of primary energy use (average)

Fuel mix in US from EIA 1997

Decreases need for process heat and pressurization All major chemical companies are involved in catalyst research

H,M,L E,G,F,P

Medium R&D Fair DOE 2000a, EIA 2000 DOE 2000a, EIA 2000 DOE 2000a, and judgement of analyst Judgement of analyst Judgement of analyst DOE 2000a, EIA 2000

79

LBNL and ACEEE

Autothermal Reforming (or Combined Reforming) (Chemicals-7)

The production of nitrogenous fertilizer is very energy-intensive and the nitrogenous fertilizer industry produces a variety of fertilizers and other nitrogen-compounds. Ammonia is the most important intermediate chemical compound, used as the basis for almost all fertilizers. In the U.S. ammonia is one of the major chemicals produced, with an estimated production of 18.0 Million tons (16.3 Mt) (CMA 1996). About 80 percent of the ammonia is used for fertilizer production and the remainder is used for a variety of products, mainly explosives and plastics. Annual fuel use is estimated at 254 TBtu (excluding feedstocks) (268 PJ) while 349 TBtu (368 PJ) natural gas is used as feedstock. Natural gas is the main fuel used for ammonia manufacture in the U.S. Electricity consumption is estimated at 3.9 TWh. We estimate the energy intensity of ammonia manufacture at 33.8 MBtu/ton (39.3 GJ/t) (including feedstocks, HHV) and 127 kWh/short ton (140 kWh/t), resulting in an estimated primary energy consumption of 35.2 MBtu/short ton ammonia (HHV), equivalent to 37.1 GJ/t ammonia (LHV) (Worrell et al. 2000). No recent new ammonia plants have been built in the U.S., limiting uptake of new autothermal reforming technology to expansion projects and retrofit of existing plants. The ammonia synthesis starts with the production of syngas from natural gas. Reforming takes place in two stages, the primary and the secondary reformer. The inputs for the reforming process are NG (mainly CH4), water (steam) and air. Desulfurized CH4 is heated and led into the primary reformer. Because the reforming reaction is endothermic, heat has to be supplied externally in this stage. Product gas from the primary reformer, a mixture of H2, CO and CO2 (still containing CH4), is passed to the secondary reformer. Here it is mixed with pre-heated air and passed over a nickel catalyst at 1100C. CH4 is partly burned with oxygen from the air to generate the energy that is necessary for the steam reforming reaction. Oxygen is also used to oxidize a part of the CH4 to CO and H2. The syngas leaves the reactor at a temperature of approximately 980C. In some processes (ICI-AMV, Braun), excess air is supplied to the secondary reformer, so the primary reformer can be smaller and facilitates milder reforming conditions (Worrell and Blok 1994). In the development of new, more efficient syngas production processes, more attention is paid to syngas production using the partial oxidation method. The chemical reaction is: partial oxidation: CH4 + 1/2 O2 CO + 2H2 (1)

Reforming processes that combine steam reforming and partial oxidation are the most efficient. These processes are called advanced processes. In one of these advanced processes, autothermal reforming (ATR) process, both reaction (1) and (3) play an important role. The processes of partial oxidation and steam reforming are highly integrated, i.e. both reactions take place in one reactor. This reactor has similarities to the secondary reformer of the steam reforming process. The reactions that take place are combinations of combustion and steam reforming (Christensen and Primdahl 1994). In the combustion zone, the reaction is: combustion zone: CH4 + 3/2 O2 CO + 2 H2O (2)

This reaction is without CO2 production because CO is the primary combustion product, which is converted to CO2 by a slow secondary reaction. In the thermal and catalytic zones, the reactions (1) and (2) occur to form H2. The oxygen content of the oxidant in the reforming process depends on the application of the syngas. For the production of NH3, air is needed because this contains the N2 necessary for the synthesis of ammonia. Autothermal reforming has been used since the 1960s on a small scale. Research has been done to upgrade the process and make it suitable for large scale production. Haldor Topse has used autothermal reforming in small scale designs since the 1960s, and has adapted its design for CO-rich synthesis gas, especially for methanol production (Rostrup-Nielsen 1993). Statoil, the Norwegian oil company, is building a new methanol factory on the north-west coast of Norway. The plant will use an ATR designed by Haldor Topsoe with a capacity of 2,400 tpd. Other producers (e.g. Lurgi) also market ATR methanol plants. Uhde in Germany developed the Combined Autothermal Reformer (CAR) process and has built a demonstration plant. According to Uhde the energy requirement would be around 24.5 MBtu/ton NH3 (LHV) at plant scales larger than 500 tpd (Christensen and Primdahl 1994, Marsch and Thiagarajan 1993). 80

Emerging Energy-Efficient Industrial Technologies Autothermal Reforming Data Table

U ni t s
Autothermal Reforming Chemicals-7 Autothermal or combined eforming replaces conventional steam reforming

N ot es

Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality beneifts Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources

Mtons

Ammonia Making Process Heating Natural Gas New, Expansion 18.0

SIC 2873

1994 production; no growth expected in this industry

Average steam reforming ammonia plant tpy Plants size varies between 33,000 and 1,800,000 tons/year 1 kWh 127 MBtu 33.8 34.9 MBtu Autothermal reformer kWh 127 MBtu 26.8 27. 9 MBtu Commercial 1996 Years 30 kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 0 7.0 7.0 Low 30% 0 38 37.8 55 Retrofit -0.5 N/A 1.13 1.13 3.7 26% Somewhat Significant H,M,L Low Slow market development for ammonia limits uptake technology Commercial Medium Marketing, Retrofit assistance Fair Worrell et al., 2000 Worrell et al., 2000 Czuppon et al., 1996 Author's estimate Author's estimate Christensen and Primdahl, 1994; Smit et al., 1994 EFMA, 1995 Mike Grant, Haldor Topsoe, Houston, TX (281) 228 5095 0% 44% 20% Slow market development for ammonia limits uptake technology

Technology can also be used for hydrogen and methanol production

Due to slow market, assume retrofit existing plants Estimate

Authothermal plants have lower maintenance and production costs

H,M,L E,G,F,P

81

LBNL and ACEEE The advanced KRES (Kellogg Reforming Exchanger System) process (Czupon 1994) is also based on autothermal reforming. The KRES system has first been installed in Canada (at Ocelot Ammonia, Co, Kitimat, BC) to provide syngas for an equivalent of 350 tpd ammonia. The KRES process has also been integrated in a new ammonia plant design: Kellogg Advanced Ammonia Process (KAAP), of which the first have been built in Trinidad and other countries. The specific energy consumption of the KRES-process is estimated at 25.9 MBtu/ton (HHV) (equivalent to 27.2 GJ/tonne (LHV), including feedstocks (Czuppon et al. 1996). For 2015 we assume that retrofit of an existing ammonia plant by replacing the reformer with an autothermal reformer and integrating into the plant may reduce natural gas use to 26.8 MBtu/ton ammonia (HHV) (Czuppon et al. 1996), or (28.1 GJ/tonne, LHV). We assume that power consumption does not increase (assuming the KRES-process without additional oxygen consumption). The capital costs for a new greenfield plant for a modern ammonia plant using autothermal reforming are smaller than current technology (Czuppon et al. 1996). Exact investments are not given by the developers of the processes, and will depend on the local situation and capacity. Christensen and Primdahl (1994) estimated the investment costs of an autothermal reformer to be lower than the investments in both the primary and the secondary reformer of the AMV-ICI process. The reformers in the AMV-process count for 21 percent of the total investment (300 US$/tonne). The investments for an ATR will probably be around than 55 US$/ton (Smit et al. 1994), assuming retrofit of an existing plant. O&M costs are lower compared to that of a conventional steam reformer. An autothermal reformer, reducing the fluegasses from a fired primary reformer, may reduce NOx emissions by 50 percent (EFMA 1995). Next steps include the commercial application of an integrated autothermal reformer.

82

Emerging Energy-Efficient Industrial Technologies

Plastics Recovery (Chemicals-8)

In the United States, plastics production has grown significantly over the past two decades at rates of 3-8 percent annually with total plastics production of 29 million tons (26 Mt) in 1996 (Chemical and Engineering News 1997). Some of the main plastic products include polyethylene, (low (PET) and high density(HDPE)), polypropylene, polystyrene, and polyvinyl chloride (PVC); markets have seen particularly strong growth in PVC, polypropylene, and high-density polyethylene. While not as energy-intensive as the production of bulk chemicals, the production of plastic materials in SIC 2821 accounts for an important share of chemical energy use due to the large volume of production. Primary energy consumption for plastics and resins production in 1994 was 400 TBtu (400 PJ) or 2 percent of manufacturing energy consumption. While some progress has been made in recovering plastics from some waste streams, the overall recovery rate for post consumer waste in the U.S. is extremely low, about 9 percent (Denison, 1997). There are still large opportunities to greatly increase recycling in the U.S. In many cases, economics prevent the increase in recycling since the cost of collecting and processing post-consumer plastics is higher than the cost of producing virgin materials (Kobler 2000). Complex or mixed waste streams are particularly challenging to separate and make pure enough to be useful. Aside from the PET and HDPE bottle markets, one of the single largest concentrated supplies can be found in automobile shredder residue (ASR). ASR includes plastics, rubber, glass, fibers, and amounts to 3 to 5 million tons (2.7-4.5 Mt) annually (DeGaspari 1999). It is estimated that 20-31 percent of this is 20 different types of plastic materials; however, the two major types of plastic are polypropylene and ABS (acrylonitrile, butadiene, and styrene) (Kobler 2000). Currently, virtually no post-consumer plastics are used in todays new vehicles (USCAR 1998, Salyp 2000, Kobler 2000). Of these plastics, thermoplastics such as polypropylene, polyethylene, polycarbonate, nylon, and polyurethane can potentially be melted and re-used while thermosets do not re-melt and are more challenging to recycle (Kobler 2000, Betts 1999). Various technologies are being developed to recover and reuse plastics. MBA polymers developed a mechanical separation process that allows plastics of similar densities to be separated for reuse. This has been used to separate and recover different plastics from computer housings. Early development of this process was partially supported by the U.S. Department of Energys NICE3 program, the Vehicle Recycling Partnership (VRP), and the American Plastics Council (APC) (OIT 1999, Biddle 2000, Yester 2000). Argonne National Laboratory has developed a separation technology called froth flotation to separate and recover ABS and HIPS from appliance wastes (USCAR 2000, DeGaspari 1999, Daniels 2000, Kobler, 2000). This technology is receiving developmental support from the U.S. Department of Energys Office of Industrial Technology in collaboration with the VRP and the American Plastics Council (OIT 1999) and could be applied to ASR in the future. A small-scale trial using appliance waste streams was conducted at the Appliance Recycling Centers of America with support from the VRP but no pilot plant has been constructed (Daniels 2000, Yester 2000). Recovery Plastics International (RPI) has developed a skin flotation technology that recovers about 80 percent of the plastic stream from ASR (Kobler 2000). This technology has also received R&D funding support from VRP and a one- ton/hour pilot plant is currently operating. In the froth flotation technology, plastics of similar densities are placed in an aqueous solution, and the wetting characteristics of various plastic types are selectively adjusted. This preparation allows for small gas bubbles to attach to particular plastics thereby allowing for separation in the solution (USCAR 1999, DeGaspari 1999). The skin flotation technique at RPI puts on a skin of plasticizer on the plastic surface selectively which makes it hydrophobic. That targeted plastic type, which preferentially absorbed the plasticizer, is the only one to float (Kobler, 2000). Only with skin flotation technology has raw ASR been used as the primary feedstock material, and also is able to separate out plastics from rubbers improving the quality of the separated product (Kobler 2000). Energy savings from this system can be significant. Including the embodied energy in plastics, savings estimates range from 50-75 Mbtu/ton (58-87 GJ/t) material recycled (Daniels 2000, Richman 2000). Fisher and Mark (1999) note that the plastics content of ASR is about 13 percent by weight. Based on this analysis we estimate a savings of 13 Mbtu/ton (15 GJ/t) (Fisher and Mark 1999, Lipinsky and Wesson 1995).

83

LBNL and ACEEE

Plastics Recovery Data Table

Units Plastics recovery Chem-8 Plastics recovery for ASR Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Mtons Notes

Plastics Process heating Fuels New 0.3

SIC 2821

Based on assumption of growth in automobile plastics content

Plastics manufacture for automobiles ton 1 kWh 887 MBtu 49.4 MBtu 57.0 Advanced recovery technologies kWh MBtu MBtu 1065 7.8 16.9 Near commercial 2002 20 -178 41.6 40.1 Medium 70% -40 9 9.0 225 Full 0 N/A 0.86 0.90 2.8 36% Somewhat None Compelling H,M,L Medium Technical H,M,L E,G,F,P High Fair -20% 84% 70%

Worrell et al., 1994

Electricity for mechanical recycling and for ASR & other thermoplastics recovery

Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years %

Costs of $150-300/ton recovered material Assume that operations competitive in cost with virgin plastics (Kobler, 2000)

Discount rate for all CCE calculations is 15%

May be lower cost than existing processes Reduced landfilling

Need to further develop and demonstrate technology Still significant support for the technology. High activity in Europe as well. U.S. demonstration, regulatory changes

Salyp, 2000; DeGaspari, 1999 Worrell et al., 1994 Kobler, 2000 Author estimate Author estimate Daniels, 2000; Kobler, 2000 DeGaspari, 1999 Recovery Plastics Int'l (801-973-4774); Ed Daniels, ANL (630-2522000)

Aside from energy savings, the environmental implications of recycling technologies are significant. Roughly 25 percent of the weight of the vehicle is currently landfilled, which includes plastics, foam, 84

Emerging Energy-Efficient Industrial Technologies copper, trace metals, rubber, and fluff. ASR recycling is estimated to divert at least 40 percent of this currently landfilled materials (Kobler 2000). Installation cost estimates for plastics separations technologies vary in range from $100-350/ton ($110390/t) recovered material based on annual recovery capacity (Daniels 2000, Kobler 2000). Operations costs are claimed to be competitive or lower than existing virgin plastics and estimates have been given of 15-20 cents/pound for the RPI process and 50-75 cents/pound for the ANL process (Daniels 2000, Kobler 2000). The relative payback will also depend on the market price for the various recovered materials, assuming they meet market specifications. Ranges for the virgin prices for various polymers are shown in the table below (Kobler 2000). Material Polypropylene (PP) PP Filled ABS PUR foam PC Nylon Market price (/lb.) 32-40 40-50 55-70 30-35 60-85 80-98

These processes are both pre-commercial. Paybacks are estimated to be 2 years or less on new plant investment although the technology has not been fully deployed (Kober 2000, USCAR 2000, Daniels 2000) These recovery technologies are pre-commercial. MBA polymers began commercial operations 1999 and processes several million pounds of recovered plastics per month of computer housings (MBA Polymers 2000). The company claims to have an operational commercial separations process applicable for ASR at its Richmond (CA) facility (Biddle 2000). An ANL froth flotation system was demonstrated in the U.S. at the Appliance Recycling Centers of America in Minneapolis, Minnesota, but no permanent demonstration facility has been constructed (USCAR 2000). In 1999, Argonne signed a licensing agreement with N.V. Salyp, a recycler in Belgium, to incorporate the foam cleaning system into demonstration facilities, but it is not clear whether Salyp will also incorporate the plastics recycling component of the system into their manufacturing process (DeGaspar, 1999, Fisher 2000). Also, skepticism has been raised on the efficacy of the froth flotation technology as compared to other technologies in producing a high quality product (Schedler 2000). RPI may have the most market ready system as they have been operating a 1-ton/hour demonstration plant in Utah since 1998. RPI claims that it is within a year or two of commercialization (Kobler 2000, USCAR 1998). Aside from technical and economic feasibility, the full commercialization of this technology is dependent on changes in U.S. environmental regulations. Existing regulations promulgated in 1976 under the Toxic Substances Control Act are unclear but apparently do not allow the reintroduction of any product containing more than 2 parts per million of toxic polychlorinated Biphenlys (PCBs) (Kobler 2000, USCAR 2000, EPA 2000a)17. Shredder residue on average has concentrations of 10-30 ppm PCBs, however this residue is primarily on the surface of the plastics and generally not embedded in the plastic material itself (Kobler 2000). ASR technologies that wash the plastic surfaces are able to remove the PCBs and produce products below the 2 ppm PCBs level. Clarification in the regulations to account for this will help to stimulate the ASR plastics recovery market (Kobler 2000). R&D support by the VRP was curtailed because of this issue but there is optimism that this will be remedied soon and the EPA is looking into modification possibilities (Yester 2000, Fisher 2000). Other key issues in the development of large-scale recovery facilities include ensuring access to a consistent source and volume of ASR streams so that recovered plastics customers such as auto manufacturers can be ensured of continued uninterrupted supply. Were this technology successful, it could significantly affect the plastics supply market for automobiles and other applications requiring higher-end plastics. We believe that there is a high likelihood of the potential for a growing domestic market over the near term assuming the resolution of the regulatory issues.
The original legislation states that no person may manufacture, process, or distribute in commerce or use any polychlorinated biphenyl in any manner other than in a totally enclosed manner (i.e. any manner that would expose human beings to PCBs) (EPA 2000).
17

85

LBNL and ACEEE

Biodesulfurization of Gasoline (Refining-1)

As the overall sulfur content of gasoline has increased over the past few years, gasoline manufacturers have had to find better and more efficient ways of desulfurizing their supplies. The average sulfur content of gasoline in the U.S. gasoline pool is about 300 ppm. In California, standards instituted in 1996 require gasoline that has a sulfur content of 30 ppm. Technologies that can consistently and economically deliver fuel with no more than 50 ppm sulfur will be required when EPA Tier II air regulations take full effect in 2003. Biodesulfurization, the process in which live microorganisms selectively remove sulfur from fuel, promises to deliver low-sulfur gasoline economically and with fewer environmental emissions. Currently, the Merox process is the primary technology employed for the removal of sulfur in gasoline. In this process, gasoline and a small quantity of air are processed over a heterogeneous catalyst at high temperatures and pressures. The gasoline then comes in contact with a caustic solution to remove sulfur. The caustic solution is then contacted with air and a catalyst, thereby converting the extracted compounds to disulfides. The advantage of oxidative biodesulfurization processes is that the reaction takes place at ambient temperatures and pressures and produces non-toxic by-products, eliminating the need for collateral processing of hydrogen sulfide. Biocatalysis is more selective than the Merox process and has the ability to target individual groups of sulfur which contain species such as mercaptans, alkylmercaptans, and polysulfides. The biocatalytic process may be designed as a batch process in which the reactants and biocatalyst are maintained in a reaction vessel for a period of time. Alternatively, the bioprocess can be designed as a continuous flow process in which the reactants are only brought into contact with the biocatalyst for a limited period of time. The initial capital investment for a biodesulfurization unit will be around $18 million for a 25,000 barrel per day facility in 2015 (OIT 1999). This is a significant improvement over the estimated $36 million for a standard desulfurization facility. The yearly operating and maintenance costs run a bit higher for a biodesulfurization unit about an additional $620,000 annually. The unit saves enough energy to deflect this cost and result in a payback of just under two years. Once EPA Tier II regulations come into effect, virtually all gasoline produced for domestic use will require desulfurization. Gasoline production in 2015 is estimated to be 11.73 million barrels per year, according to the Annual Energy Outlook 2000. Biodesulfurization is estimated to use 10-15 percent less energy than the Merox process. The decrease in energy use is attributable to the lower temperature and pressure of the bioprocess as well as the reduced need for separation of subsequent streams. The biotechnology is still in the bench-scale test state. A market-ready product can be expected in 2005. Future developments for this technology include: elucidation of the desulfurization pathway including the isolation, identification, and quantification of the pathway intermediates; enhancement of solvent tolerance of the catalyst; definition of the basis for the required genetic improvements of the organisms; and determination of the rate and extent of gasoline desulfurization. Most refineries and gasoline processing facilities operate continuous reactions. It is relatively easy to maintain and operate a batch bioreactor, but it requires significant startup time to initiate the microbial activity and allow products to accumulate. Future research is needed to develop continuous flow reactions since these processes are more prone to contamination with undesired microorganisms, making quality control difficult to maintain.

86

Emerging Energy-Efficient Industrial Technologies Biodesulfurization Data table

Units Biodesulfurization of Gasoline Refin-1 Replace hydrodesulfurization Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

million barrels

Refining Process heating, other Electricity New, replace on failure 4,281

SIC 2911 http://www.oit.doe.gov/factsheets/petroleum/pdf/gasbiopet.pdf DOE-OIT 1998a 2% annual growth -personal communication w/ J. Decicco 2000

Hydrodesulfurization of the feed to the fluid catalytic cracking unit (Merox Process) 1.0 barrels 11.14 kWh DOE-OIT 1998a 0 MBtu DOE-OIT 1998a 0.09 MBtu DOE-OIT 1998a Biocatalytic removal of sulfur from gasoline 9.99 kWh Assumed 10% savings over conventional technology http://www.oit.doe.gov/factsheets/petroleum/pdf/gasbiopet.pdf MBtu 0 MBtu 0.09 http://www.oit.doe.gov/factsheets/petroleum/pdf/gasbiopet.pdf Bench scale trials 2005 Years 15 kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ 1.15 0 0.01 Medium 45% 2210 0 18.85 -131506 Incremental 72118 43268 5072392.48 1.823 55% ANL 1998 Based on 25,000 bbl/day facility 18M for biodesulf vs. 36M for hydro 10% 0% 10%

$ $/kWh $/Mbtu $/Mbtu Years %

O&M costs are higher for biodesulfurization units

Significant None H,M,L Medium Capital and equipment intensive High Research, scale-up Excellent

Biodesulfurization does not reduce octane the way hydrodesulfurization does

H,M,L E,G,F,P

Improve biocatalyst stability, faster kinetics, broader substrate specificity

EIA 1999 Personal communication with John Decicco 2000 http://www.oit.doe.gov/factsheets/petroleum/pdf/gasbiopet.pdf Argonne National Laboratory 1998 EIA 1999 Argonne National Laboratory 1998

87

LBNL and ACEEE

Fouling Minimization (Refining-2)

The petroleum refining industry is one of the largest energy consumers in the manufacturing sector. Primary energy consumption in 1994 was 3,300 TBtu (3,500 PJ), or 16 percent of total manufacturing energy consumption. Modern refineries are complex integrated systems that transform crude oil into transport fuels, residual fuel oil, and other products. The energy required for processing a unit of crude oil input in a complex refinery is roughly equal to about 10 percent of the energy content of the input crude, although this can vary (WEC 1995). The main processes in refining involve crude distillation (the separation of crude oil into various distillate products through pyroprocessing and fractionation), conversion (the addition of hydrogen into hydrocarbon chains to produce higher quality products), reforming (the reorganization of hydrocarbon molecules to increase the octane) and finishing or treating processes (removal of sulfur and other impurities) (WEC 1995). Crude distillation alone consumes about 45 percent of the energy content of the oil (Worrell 1994). In a complex refinery most processes occur under high temperature and pressure conditions; the management and optimization of heat transfer among processes is therefore key to increasing overall energy efficiency. Fouling, a deposit buildup in units and piping which impede heat transfer, require the combustion of additional fuel. For example, the processing of many heavy crude oils in the U.S. increases the likelihood of localized coking deposits in the heating furnaces, thereby reducing furnace efficiency and creating potential equipment failure. An estimate by the Office of Industrial Technology at the U.S. Department of Energy noted that the cost penalty for fouling could be as much as $2 billion annually in material and energy costs (OIT 1999). Several methods of investigation have been underway to attempt to reduce fouling including the use of sensors to detect early fouling, physical and chemical methods to create high temperature coatings (without equipment modification), the use of ultrasound, as well as the improved long term design and operation of facilities. The U.S. Department of Energy initially funded preliminary research into this area, but funding has been discontinued (Huangfu 2000, Bott 2000). Initial analysis on fouling effects of a 100,000 bbl/day crude distillation unit found an additional heating load of 12.3 kBtu/barrel (13.0 MJ/barrel) processes (Panchal and Huangfu 2000). Reducing this additional heating load could results in significant energy savings. This technology is still in the conceptual and basic research stage and therefore it is difficult to assess capital costs at this time. Argonne National Laboratory (ANL) has been the lead in working with the refining industry in the area. Progress so far has included: a basic understanding of fouling mechanisms developed (for example, the presence of iron sulfide in crude oil and its link to fouling), the development of a threshold fouling model by ANL, the testing of prototype fouling detection units, the development of a Heat Exchanger Design Handbook (1999 Edition) incorporated ANLs petroleum fouling threshold model, and the preparation of a guideline document on Heat Exchanger Fouling in the Crude Oil Distillation Unit (Panchal 2000). It is likely that a well-designed heat exchange network would have fewer cleaning requirements, thereby saving in operations and maintenance costs. Also, were this technology to be more fully developed, it would have a potentially large market, given the size of the U.S. refining sector. While the issue of fouling is now on the radar screen of plant managers (there is a bi-annual Fouling Mitigation conference held by Argonne and the American Institute for Chemical Engineers), a stronger commitment by the refining industry would be needed to advance this technology to the next stage of development. Some sources believe that the future development of in this area is expected to be in the area of Condition-Based Maintenance of Heat-Transfer Equipment that will be based on Knowledge-Based and Monitoring -Based Mitigation of Fouling/Corrosion (Panchal 2000).

88

Emerging Energy-Efficient Industrial Technologies Fouling Minimization Data Table

Units Fouling Minimization refin-2 Improve heat exchanger operations Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

mill. bbl/day Domestic refining bbl kWh MBtu MBtu

Refining Process heating, other Fuels New, replace on failure 8712.6

SIC 2911 DOE-OIT 1998a EIA 1999

1.0 5.17 0.47 0.5

EIA 1995b (Petroleum Supply Annual); EIA 1997

Fouling minimization practices kWh 5.17 MBtu 0.40 0.4 MBtu Bench scale trials 2005 Years 15 kWh/% MBtu/% MBtu/% % GWh TBtu TBtu $ $ $/kWh $/Mbtu $/Mbtu Years % 0.00 0.07 0.07 Low 20% 0 123 123 N/A N/A NA NA NA N/A N/A Significant None None H,M,L Low Technical Low Fair 0% 15% 14%

Potential for reduction in 30% of heating energy use

Not currently available Not currently available

Reduce downtime

Need for further R&D

H,M,L E,G,F,P

EIA 1999 EIA 1997 Panchal and Huangfu, 2000

Ehr-ping Huangfu, U.S. Dept. of Energy (202-586-5000)

89

LBNL and ACEEE

Roller Kiln (Ceramics-1)

Roller kilns can be used in the manufacture of structural clay products and ceramics. Structural clay products are mainly building bricks, roof tiles and sewer pipes. In the U.S. approximately 15.1 million short tons (13.6 Mt) of clay are consumed for the production of these products (Virta 1998). Of this, 98 percent is used for making bricks. In 1998 8.26 billion bricks were produced with a value of $1.5 billion. Bricks are manufactured in most states, but production is concentrated in Alabama, North Carolina, Texas, Georgia, Ohio, South Carolina, Missouri, Arkansas, California, and Pennsylvania. Ceramic production in the U.S. is concentrated mainly in the production of tiles (42 percent) and sanitary ware (18 percent). Tile production in 1997 is estimated at 154 million m2 (Virta 1998). Total clay consumption for ceramic production was 1.84 million short tons (1.67 Mt) in 1997 (Virta 1998). No statistical information is available on energy use for production of ceramics and structural clay products. A recent survey of ceramic industry kilns showed that energy use for tunnel kilns varied between 2.52 and 3.82 MBtu/short ton brick (2.93-4.44 GJ/t) (Whittemore 1999). Intermittent kilns used between 2.91 and 8.46 MBtu/ton (3.38-10.4 GJ/t) (Whittemore 1999). We assume a predominant use of tunnel kilns with an average specific fuel consumption of 3.6 MBtu/ton (4.2 GJ/t). Based on the production volumes and clay consumption we estimate fuel use for baking at 46 TBtu (49 PJ) for bricks and 14 TBtu (15 PJ) for ceramic products, or a total fuel consumption of 60 TBtu (63 PJ). Average specific fuel consumption for both industries is estimated at 4.5 MBtu/ton dry product (5.2 GJ/ton dry product). Natural gas is probably the main fuel used, although there are kilns that use other fuels (e.g. oil, and even wood chips for example at a plant in Mississippi), while sawdust is added to the bricks and partially combusted in the baking process. Bricks, tiles and other ceramics are baked from clay. The clay is formed, dried and then baked. Previously flame and ring-kilns were used with long production cycles (up to 14 days). Today, the most common process is the tunnel kiln. Tunnel kilns have a relatively short production cycle of 75-140 hours. The firing process in the tunnel kiln is automated, and consists of three zones through which the bricks travel: preheating, baking and cooling. A new technology is the rapid firing technology for bricks and tiles, called the roller kiln. In the rapid firing process the clay is prepared dry with appropriate additives to maintain the forming and baking characteristics required. The amount of water is thus reduced to 6-8 percent (compared to 18-20 percent in the current process). The fired products are transported on refractory rollers, rather than on lorries. A roller kiln makes it possible to reduce the heating time (to approximately 8-9 hours (Tomasseti 1995) and use shorter firing curves). The flue gas volumes in the roller kilns are lower, compared to the tunnel kiln, reducing the heat losses (Elmi 1993). This reduces not only the heat demand, but also the power consumption for air circulation. Roller kilns are the state-of-the-art for the production of sanitary ware and wall and floor-tiles. They can be found in modern facilities across the world, and also in the U.S. (e.g. in Ohio and Texas). They are not yet used in the production of bricks in the U.S. In The Netherlands a roller kiln was demonstrated for sanitary ware (CADDET 1993c). The kiln reduced energy consumption by 60 percent relative to the previously used tunnel kiln and reduced the specific energy consumption to 4.2 MBtu/ton product (3.8 GJ/t) (CADDET 1993c), compared to 9.3 MBtu/ton (10.8 GJ/t) (LHV). The performance can be even further improved by heat recovery from the flue gases. The technology is under investigation for more massive products like tiles and bricks. In Italy a new plant produces 50,000 lightened and specially shaped bricks per day using the rapid firing technology. In 1996, two roller kilns for bricks were in operation in Europe (Italy and Germany) while two were under construction in Indonesia and Mexico. The plant in Italy was designed to consume 1.2 MBtu/ton (1.4 GJ/t) (LHV) (Tomasseti 1995). Initially it consumed 1.4 MBtu/ton (1.6 GJ/t) (LHV) (Tomasseti 1995). We estimate average energy consumption for a future roller kiln in the U.S. at 1.85 MBtu/ton dry product (2.15 GJ/t) (HHV, using the 1997 production volume structure). To shorten the firing time in the kiln the heat distribution needs to be optimal, and the temperature needs to be distributed evenly through all material travelling through the kiln. The first roller kilns have a single

90

Emerging Energy-Efficient Industrial Technologies Roller Kiln Data Table

Units
Roller Kiln Ceramics-1 Energy-efficient roller kiln replacing tunnel kiln

Notes

Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources

Mt

Clay Products Process Heating Natural Gas New 14.6

SIC 325, 326

Estimated 1994 energy use 71 Tbtu: 60 Tbtu fuel, 11 Tbtu electric, slow growth expected

Tunnel Kiln kWh MBtu MBtu Roller Kiln kWh MBtu MBtu 100,000 tpy 0 4.5 4.5

Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years %

0 1.9 1.9 Commercial 1993 30 0 2.7 2.7 Medium 15% 0 6 5.8 10 incremental 0 N/A 0.57 0.57 1.9 N/A None Somewhat Significant 0% 59% 59%

Commercial for ceramics, first uses for small-capacity brick kilns

Competition of improved tunnel kilns Half of kiln turnover will use roller kiln technology

Estimated investment costs are 10$/ton annual capacity over a tunnel kiln Incremental costs over that of a tunnel kiln

Reduced NOx emissions

H,M,L

Low Demonstration for bricks Medium Demonstration for bricks Fair Slow stock turnover of kilns

H,M,L E,G,F,P

Virta 1998: production equal to 1997 Author's estimate on basis of Whittemore (1999) Tomassetti 1995; CADDET 1993c Author's estimate Tomassetti 1995; Elmi 1993 Tomassetti 1995; CADDET 1993c Author's estimate

91

LBNL and ACEEE layer of products, while new designs have a double layer. This is well suited for ceramic products. However, it less suited for the larger capacity brick kilns. Developers like the Italian firm Mori (the main developer of the rollerkiln for ceramics) are trying to develop multi-layer kilns. Other suppliers of roller kilns are SACMI (an Italian firm with an U.S. subsidiary), Lex Kiln (CA), and Keller (Germany). Investment costs for a tunnel kiln with a capacity of 110,000 tons/year (100,000 t/year) were estimated at $2.1 million (Tomasseti 1995), equivalent to approximately $19/ton-capacity ($21/t-capacity). Tomassetti (1995) expects roller kilns to be less expensive than a tunnel kiln. Kilns for sanitary ware have a lower capacity and higher typical investment costs. The roller kiln for ceramics as described in the demonstration project in The Netherlands had higher investment costs of $38/ton-capacity ($41/t), with a payback period of 2 to 2.5 years with Dutch gas prices of $3.00/MBtu ($2.80/GJ) (CADDET 1993c). The 1994 natural gas price for the stone, clay and glass industries in the U.S. was $2.83/Mbtu ($2.68/GJ), which would give an average U.S. payback period of 2.7 years. For this study we will assume that the investments of a roller kiln are equal to that of a tunnel kiln, both for bricks and other ceramic products. The maintenance costs are lower or equal compared to that of a conventional tunnel kiln. Roller kilns will likely be implemented when the conventional tunnel kilns need to be replaced, or when expanding capacity at an existing facility. Competing technologies will be more efficient tunnel kilns as developed in Europe and the U.S., or retrofitting existing facilities with improved insulation with low thermal mass materials (LTM), LTM-carts, and improved heat recovery. The DOE NICE3-program sponsors the demonstration of a new kiln with LTM-insulation in Southern California, reducing energy use by half and reducing NOx emissions by 40 percent. R&D is needed to develop materials that can hold the heavy weight of tiles and bricks while withstanding the stresses of rapid heating and cooling. R&D is also directed at the construction of a kiln with a good air circulation and at ensuring good brick quality. The applicability of the technology for different types of bricks should be demonstrated, before implementation is feasible (Elmi 1993).

92

Emerging Energy-Efficient Industrial Technologies

100 Percent Cullet Use & Cullet Preheating in Container Glass Manufacture (Glass-1)
The glass industry is a capital- and energy-intensive industry and serves four distinct markets: glass containers, fiberglass for insulation and structural applications, flat glass for windows, and specialty glass such as tableware, light bulbs, television tubes, and fiber optics. The industry had shipments of about $27 billion, and spent over $1.9 billion on new capital equipment in 1997. The industry spent about $1.4 billion on fuels and energy, representing over 5 percent of its value of shipments in 1997. The glass industry includes major corporations but also small businesses and is spread across the nation, concentrated in Ohio, Pennsylvania, California, North Carolina, Texas, Indiana, Michigan, New Jersey, New York, and Wisconsin. In 1997, the glass industry produced approximately 21 million tons (19 Mt) of glass products and used 2.25 million tons (2.04 Mt) of recycled glass, which are know as cullets. The glass industry used in excess of 250 TBtu (260 PJ) annually (EIA 1996). Nearly 80 percent of this energy is supplied by natural gas for the glass melting, with electricity accounting for the majority of the remaining energy used (EIA 1996). We focus on the production of container glass. In the U.S., the glass industry produces approximately 10.3 million tons (9.4 Mt) of glass containers annually and more than 650,000 tons (590,000 t) are also imported (GPI, 2000). Approximately 35 percent (or 3.8 million tons (3.4 Mt)) of all glass containers available to consumers are recycled, of which 2.25 million tons (2.04 Mt) are recycled into glass containers. The other recovered glass containers may be used for secondary recycling, e.g. abrasives, asphalt. The production of container glass in the U.S. consumed 66 TBtu (70 PJ) natural gas, 2 TBtu (2 PJ) oil and 4.3 TWh in 1994 (EIA 1997). This is approximately 114 TBtu (120 PJ) on a primary energy basis. The average specific fuel consumption based on the MECS data is estimated to be 6.6 MBtu/ton (7.7 GJ/t), of which an estimated 5.8 MBtu/ton (6.7 GJ/t) is used in the smelting furnace. Energy use for container glass furnaces could be somewhat lower, although wide variations in energy intensity exist. Although glass containers already contain on average over 20 percent cullets in the U.S., higher use of cullets is possible. In Europe container glass manufacturers sometimes use 80 percent cullets, while the first furnaces using 100 percent cullets are now being introduced. Increasing cullet use by 10 percent will reduce fuel use by approximately 2.5 percent (Enneking 1994). Increasing cullet use to 100 percent will allow larger energy savings as the temperature can be lowered below the typical melt temperature of 1550oC, since the sand does not need to be melted. We assume energy savings of 19 percent on fuel for glass melting, or equivalent to 1.13 MBtu/ton (1.31 GJ/t) glass for furnaces switching to 100 percent cullet use. Energy is also saved in the production of soda ash, which constitutes approximately 20 percent of the raw material feed. We assume that 0.15 ton of soda ash is used per ton of container glass in the U.S. Increasing cullet use to 100 percent will save energy use for soda ash manufacture with 1.3 MBtu/ton glass (1.5 GJ/t) (assuming 8.8 Mbtu/ton soda ash (10.2 GJ/t)) (Enneking 1994). However, increased use of cullets will lead to increased processing of recovered glass, as the quality of the cullets becomes more important to maintain product quality. We assume 0.17 MBtu/ton (0.20 GJ/t) (Enneking 1994) for glass separation and cleaning, consuming approximately 0.13 MBtu/ton glass (0.15 GJ/t). Net energy savings at the glass plants are estimated at 1.0 MBtu/ton (1.2 GJ/t), and indirect energy savings at 1.3 MBtu/ton glass (1.5 GJ/t). Energy efficiency can be further improved by batch cullet preheating. Especially in oxy-fuel-fired furnaces preheating is an efficient way to recover the heat contained in the flue gases. Currently, the fluegasses can be used to generate steam. Cullet preheaters have been under development since the 1980s. Commercial applications can be found in a few kilns around the world (e.g. in Germany, The Netherlands). Cullet preheating development projects are ongoing in the United States (supported by DOE, NYSERDA) and Europe. In the cullet preheater the cullets are preheated to a temperature of 570 1000oF (300-540oC) in direct contact with the flue gases (OIT 1999, Lubitz 1999). Preheating reduces energy use in the furnace, reduces oxygen use, and improve productivity of the furnace by reducing melting time and increasing furnace longevity (OIT 1999). For a cullet load of 90 percent the fuel savings are estimated at 0.40 MBtu/ton, HHV (0.42 GJ/t, LHV) (Lubitz 1999). Higher preheating temperatures may lead to fuel savings of 0.5 MBtu/ton glass (0.6 GJ/t) (OIT 1999).

93

LBNL and ACEEE

Glass Cullet Pre-Heating Data Table

U ni t s N ot es 100% Recycled Cullet Feed and Cullet Preheating Glass-1 Use of 100% recycled cullet feed into glass melting furnace, combined with cullet preheating in an oxy-fuel furnace Market Information: Industries SIC 3221 Glass Containers End-use(s) Process heating Energy types Natural Gas, Oil Market segment New, Retrofit 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Mtons 12.3 Based on 1997 container glass production and AEO2000 forecast

Average Glass Melting Furnace ton/yr 95500 kWh 39 MBtu 5.8 MBtu 6.1

250 metric tonnes per day

Use of 100% recycled cullet feed into glass melting furnace, combined with cullet preheating in an oxy-fuel furnace kWh 39 MBtu 4.4 MBtu 4.7 Demonstration 2000 Years 25 kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu 0 1.4 1.4 Medium 25% 0 4 4.3 New furnace would have lower capital cost than conventional furnace Assuming retrofit 0% 24% 23%

$ $ $/kWh $/Mbtu $/Mbtu Years %

10 Full cost -2 N/A -0.33 -0.33 2.0 49% Marginal

Di scountr at ef oral lC C E cal cul at i ons i s 15%

Depends on cullet quality Significant H,M,L Medium Cullet quality, lifetime High Full scale demonstration Good

H,M,L E,G,F,P

Feasible applications: Fair EIA, 1999 EIA, 1997 Enneking, 1994; Pieper et al., 1995 Pieper et al., 1995 Pieper et al., 1995 Enneking, 1994; Lubitz, 1999; Portner, 1999

94

Emerging Energy-Efficient Industrial Technologies Due to increased productivity, a furnace with cullet preheating can be smaller than a furnace without preheating. This leads to reduced capital costs. The typical investment costs for a furnace without preheating are estimated at $55 million and with cullet preheating at $51 million for a furnace capacity of 276 tons/day (250 tonnes/day) (Pieper et al. 1995). Hence, the replacement costs would be $40/annual ton ($44/annual tonne) lower than a conventional furnace. It is also possible to add a cullet preheater to an existing furnace, as was done at PLM Glasindustrie, Dongen, The Netherlands. The specific costs for this project were $9.9/ton ($10.9/t) (Dfl2.45 Million for a 320 tonnes/day furnace, 1996) (CADDET 1997d). The change in operating costs depends on the oxygen costs. At oxygen costs of 8.5cts/Nm3, the cullet preheater would reduce operating costs by approximately $4/ton (based on German conditions, 1995) (Pieper et al. 1995). We will assume production cost savings of $2/ton. Campaign life (i.e. the period that the furnace is used continuously, before being rebuilt) of a furnace is about 10-12 years, while the total operating lifetime is approximately 21-25 years (Pieper et al. 1995). The reduced fuel use and lower flame temperature will lead to reduced X emissions, while SOx emissions can be reduced if the sodium sulfate content of the raw material is reduced. In a German oxyfuel-furnace with cullet preheating the NOx emissions were reduced to 0.5 lb./ton glass (0.25 kg/tonne) (Lubitz 1999), or less than 500 mg/Nm3. Oxyfuel furnaces without preheating can achieve about 0.6-0.7 lb./ton glass (0.300.36 kg/tonne) (Portner 1999). Uncontrolled PM emissions may increase without sufficient emissions control equipment, as cullet preheating may increase the emissions (Enneking 1994). Hence, efficient gas cleanup is needed. Also, high preheating temperatures and long preheating times may lead to increased COemissions, due to combustion of organics in the cullet-mix (Enneking 1994), and dioxin emissions. At a furnace with preheating in Germany dioxin emissions of 0.04 ng/Nm3 were measured (ng = 10-9) (Lubitz 1999). The main barrier is quality control of the cullets, e.g. sorting and removal of inert contamination. To reduce CO emissions from the preheater, organic matter should be removed as much as possible from cullets. Based on glass collection schemes in Europe we estimate that approximately 60 percent of waste glass can be recovered, and that 60 percent of that can be used for glass manufacture, estimated at 4.8 Million tons by 2015. Hence, we estimate the maximum penetration of this technology at 25 percent of the 2015 container glass production. At an average lifetime of 25 years for a glass furnace 60 percent of the current furnace capacity will be replaced by 2015. If all retired furnaces would be replaced by new pre-heating, oxy-fuel furnaces the energy savings would be higher. We have not accounted for this in the estimated savings. Savings could be an additional 17 TBtu (18 PJ) due to installing pre-heating furnaces. Future needs consist of two parts. Firstly, demonstration of the technology at commercial scale at a US plant. Secondly, the collection of waste glass has to become more effective (i.e. larger volume of waste glass recovered) and efficient (i.e. increased separation on color, and at low cost), so that more waste glass can be used by the container industry.

95

LBNL and ACEEE

Gas and Heat Recovery at Basic Oxygen Furnace (Steel-1)

The iron and steel industry is one of the largest industrial energy consumers. The U.S. iron and steel industry is made up of integrated steel mills that produce pig iron from raw materials (iron ore, coke) using a blast furnace and steel using a basic oxygen furnace (BOF) and electric arc furnace steel mills that produce steel from scrap steel, pig iron, or direct reduced iron (DRI) using an electric arc furnace (EAF). In 1994, 2,180 TBtu (2,300 PJ) (about 11 percent of manufacturing primary energy use) was consumed in the production of about 100 Mton (91 Mt) of crude steel products (EIA 1997). Within steelmaking, the largest energy use is required to reduce iron ore in the integrated mills and to re-melt steel scrap in electric arc furnace mills (Worrell et al. 1999). Primary steel is produced using the basic oxygen furnace (BOF). In the BOF process liquid pig iron (hot metal), scarp and limestone are mixed. Oxygen is injected to reduce the carbon content of the hot metal from about 5 percent to less than 2 percent. Steel contains less than 2 percent carbon. At the same time some other impurities are reduced in the steel. The BOF process replaced the last open hearth furnace in the U.S. by 1992 due to its greater productivity and lower capital costs. Several configurations exist depending on the way the oxygen is injected. BOF crude steel production in 1994 was 61 million tons (55.3 million metric tons). The 2015 production is assumed to be 66.7 million tons (60.5 million metric tons) (AEO 1999). Fuel and electricity consumption in the BOF is estimated at 18 TBtu (19 PJ) and 1.7 TWh, respectively in 1994. Energy intensity for this process step in 1994 was 0.30 MBtu/short ton fuel (0.3 GJ/t) and 27 kWh/short ton steel (30 kWh/t) (Worrell et al. 1999). In the U.S. no BOF gas seems to be recovered (Margolis 1996). According to the EPA, the BOF-process is an important source of CO emissions, emitting 617,000 tons in 1992, or equivalent to 21 lb./short ton liquid steel (Margolis 1996). Carbon in the hot metal reacts to carbon monoxide (CO), which is emitted as BOF gas. The BOF gas has a heating value between 7.4 and 9.1 MJ/Nm3 (mean value of 8.5 MJ/Nm3, LHV) (IISI 1998). By reducing the amount of air entering over the converter, the CO is not converted to CO2. The BOF gas can be recovered and used as fuel gas in the steel plant or for steam and power production. The hot off-gases must be cooled before gas cleanup, and the heat can be recovered by generating steam and hot water. BOF gas combined with sensible heat recovery (repressed combustion) is the single most energy-saving process improvement in this process step, making the BOF process a net energy producer. Repressed combustion is very common in integrated steel plants in Europe and Japan as an efficient means for energy recovery, emission control and dust recycling (IISI 1998). Repressed combustion reduces CO and dust emissions and, since the metal content of the dust is high, about 50 percent of the dust can be recycled in the sinter plant or in the steel plant (Stelco 1993, IISI 1998). Two systems exist for gas cleanup. The dry system uses a dry cylindrical precipitator while the wet system uses a venturi scrubber and a wet precipitator. The wet system uses about 8 kWh/tonne (7.3 kWh/t) liquid steel, while the dry system uses only 2 kWh/tonne (1.8 kWh/t) liquid steel. The wet system also needs additional water and a water clarification system (IISI 1998). The dry system needs an additional pelletizing plant, but allows recycling of the dust in the steel plant rather than the sinter plant. In this analysis we assume a dry system, as many integrated plants have closed sinter plants for environmental reasons and it has lower investment costs. The amount of gas recovered depends on the hot metal charge in the BOF as the main source of carbon. Assuming a hot metal charge of 1800 lb./ton liquid steel (900 kg/tonne liquid steel), approximately 2860 cu.ft. (or 81 Nm3) of BOF gas can be recovered, accounting for flaring and air leakage into the system (IISI 1998). This is equivalent to 607 kBtu/ton (706 MJ/t). Steam recovery can be up to 120 lb./ton of steel (60 kg/tonne) (IISI 1998). We assume steam recovery of approximately 100 lb./ton, equivalent to 130 kBtu/ton (150 MJ/t). The total fuel savings are equivalent to 737 kBtu/ton (857 MJ/t) with increased power consumption of 2 kWh/ton. The costs will depend on the need for extra gas holders, as well as the size and layout of the BOF plants. We assume that installation is only feasible for large scale BOF plants with annual capacities of around 3 million tons/year (2.7 Mt). This is assumed to be 45 percent of the 2015 BOF capacity. Estimated capital costs are $20/ton crude steel ($22/t), based on plants in Japan (Inoue 1995) and The Netherlands (Worrell et al. 1993). There are additional O&M costs. 96

Emerging Energy-Efficient Industrial Technologies Gas and Heat Recovery at Basic Oxygen Furnace Data Table
Units BOF gas and sensible heat recovery steel-1 Recovery of BOF-Gas and Heat Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Mtons

Iron and Steel Process Heating Fuel, Steam Retrofit 66.7

331 Steelmaking Converter in Primary Steelmills Recovery of BOF-gas and Steam

Basic Oxygen Furnace Plant with 2 or 3 converters ton steel Capacity may vary between 1 and 5 Million tons/year 1 kWh 27 MBtu 0.3 0.5 MBtu Repressed Combustion system with waste heat boiler and dry gas cleaning system kWh 29 MBtu -0.4 -0.2 MBtu Commercial Technology common in Europe and Japan and in all new BOF-steel plants 1980's Years 30 kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % -2 0.7 0.7 Low 23% -30 11 10.8 20 Full cost 0.1 -1.57 4.27 4.37 14.7 3% Small None Significant H,M,L Low Capital Cost H,M,L E,G,F,P Low Good Strict environmental regulation for PM and CO may make technology attractive Cost data are assessed to be fair EIA, 1999 Worrell et al., 1999 IISI, 1998 Based on review steelmaking facilities Worrell et all, 1993; Inoue, 1995 IISI, 1998; VAI Voest Apline Industries, Pittsburgh, PA -7% 246% 136% No plants recover BOF-gas in the U.S. 50% of Large Scale BOF-plants by 2015

Discount rate for all CCE calculations is 15%

Recovery of iron-containing dust and recycling in steel plant Reduced CO and PM emissions

97

LBNL and ACEEE

Near Net Shape Casting/Strip Casting (Steel-2)

The iron and steel industry is one of the largest industrial energy consumers. The U.S. iron and steel industry is made up of integrated steel mills that produce pig iron from raw materials (iron ore, coke) using a blast furnace and steel using a basic oxygen furnace (BOF) and electric arc furnace steel mills that produce steel from scrap steel, pig iron, or direct reduced iron (DRI) using an electric arc furnace (EAF). In 1994, 2,180 TBtu (2,300 PJ) (about 11 percent of manufacturing primary energy use) was consumed in the production of about 100 Mton (91 Mt) of crude steel products (EIA 1997). Currently, the casting and rolling process is a multi-step process. The liquid steel is first cast continuously into blooms, billets, or slabs. Liquid steel flows out of the ladle into the tundish (or holding tank), and then is fed into a watercooled copper mold. Solidification begins in the mold, and continues through the caster. The strand is straightened, torch-cut, then discharged for intermediate storage (Kozak and Dzierzawski 2000). Most steel is reheated in reheating furnaces, and rolled into final shape in hot and cold rolling mills or finishing mills. A recent LBNL study estimated that casting and rolling consumed 332 TBtu (350 PJ) of primary energy in 1994 (Worrell et al. 1999). The reheating furnaces are usually gas and oil operated and consume roughly 2.8 MBtu/ton (3.3 GJ/t) of energy. Near net shape casting is a new technology that integrates the casting and hot rolling of steel into one process step, thereby reducing the need to reheat the steel before rolling it. As applied to flat products, instead of casting slabs in a thickness of 120-300 millimeters, strip is cast directly to a final thickness between 1 and 10 mm. (De Beer et al. 1998a, Opalka 1999, Worrell, Bode, and de Beer 1997). The steel is essentially cast and formed into its final shape without the reheating step. An intermediate technology, thin-slab casting casts slabs 30-60 mm thick and then reheats them (the slabs enter the furnace at higher temperatures than current technology thereby saving energy). This technology is already commercially applied in the U.S. and other countries. The energy consumption of a thin strip caster is significantly less than the current process of continuous casting. For the intermediate thin slab casting process, energy consumption is 0.8 MBtu/ton (0.9 GJ/t) fuel and 39 kWh/ton (43 kWh/tonne) electricity (Fleming 1995). Near net shape casting is expected to consume even less energy. Based on average U.S. practices in 1994, we estimate a primary energy savings of 4.0 Mbtu/ton crude steel (4.7 GJ/t) based on the difference between energy consumed in the current process and energy consumed in near net shape casting (Worrell et al. 1999). In the U.S., near net-shape casting has so for been applied to the production of near net beams. This technology was introduced by Nucor at their joint venture company Nucor-Yamato Steel Company in Blytheville, Arkansas and later applied at Nucor's plant in Berkeley County, South Carolina (Worrell et al. 1999, Wechsler 2000) and is also being used by Chaparral steel., all electric arc furnace producers (Worker 1998). Currently, two German suppliers, SMS and Mannesmann-Demag, supply near net casters for flat products using the thin slab technology(Worrell, Bode, and de Beer 1997). No strip caster for carbon steel products has yet been built and operated in full scale and production capacity. However, a demonstration strip caster for flat rolled carbon steel operated at full scale (though at reduced capacity due to molten steel constraints) from 1995 through 1999 in Australia, and the first commercial strip caster for flat rolled stainless steel products came on line in 1999 in Japan's Nippon Steel corporation casting line (Isenberg-O'Loughlin 1998, Opalka 1999). A flat rolled carbon steel caster has not yet been commercially applied for flat rolled products in the U.S but the successful Australian caster is now in the process of being relocated to Nucor's plant in Crawfordsville, Indiana. It is expected to begin first production in December 2001 (33Metalproducing.com 2000a, Wechsler 2000) Based on a review of the 1999 casting roundup and other literature, we estimate that the current U.S. market share for near net shape products or thin strip products is less than 5 percent (Iron and Steelmaker 1999). However, there is a large effort to develop new potential applications and markets. More than 30 R&D projects have been undertaken on this technology (DeBeer 1999). Large research programs are

98

Emerging Energy-Efficient Industrial Technologies Near Net Shape Casting Data Table
Units Near net shape casting/strip casting steel-2 Replace current continuous casting with direct near net shape casting Market Information: Industries Iron and Steel End-use(s) Process heating Energy types Gas, electricity Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources New 115.6 Notes

SIC 331

Mtons

Greenfields & refit of existing facilities. Some retrofit applications AEO 2000, continuous casting output

Continuous casting/hot rolling tons kWh MBtu MBtu 1 206 2.8 4.6 Unit consumption presented. Casters range from 150 kton/y to 3,000 kton/y Worrell et al., 1999 Worrell et al., 1999 Worrell et al., 1999

Near net shape casting/thin strip casting kWh 30 MBtu 0.3 MBtu 0.6 Commercialized 1995 Years 20 kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu 176 2.5 4.0 High 30% 6093 86 138 90% 90% 90%

Worrell et al., 1997, DeBeer, 1998a Worrell et al., 1997. DeBeer, 1998a estimates 0.0 Near net beams but not yet flat rolled products No flat rolled caster yet commercial Worrell et al., 1999

Apply to non high end steel products, Worrell et al.,1999 Savings applied to feasible applications for 2015 output Savings applied to feasible applications for 2015 output 6% savings. Assume 15% less than conventional casting systems. Full retrofit cost $103 Worrell et al. 1997

$ $ $/kWh $/Mbtu $/Mbtu Years %

-18 Incremental -40 -0.24 -17.35 -10.82 -0.4 #DIV/0! Significant Somewhat Somewhat

Discount rate for all CCE calculations is 15% Based on $2/Mbtu average 1994 primary energy for steel

Reduced production time, reduced capital costs Improved surface properties Reduced emissions Conferences, marketing by suppliers, research consortiums Also, CSP flat rolling plants limited

H,M,L

High Technical challenges High R&D Good

H,M,L E,G,F,P

Significant literature; limited field data EIA, 1999 Worrell et al. 1999 Worrell et al., 1997 Worrell et al. 1999 SMS, 1995; Tomasseti, 1995, Kuster, 1996 DeBeer, 1998a Flemming, 1995; Tomasseti, 1995, Kuster, 1996, Worrell et al. 1999

99

LBNL and ACEEE ongoing, with the cooperation of European and Japanese steel companies (Worrell 1995, Opalka 1999). The U.S. Department of Energy has identified near net casting as one of its focus technologies in the steel Industries of the Future and is currently devoting its effort to evaluating strip cast steel in conventional applications (DOE 1999) Capital costs for near net shape casting plants are expected to be lower than current practice due to the elimination of the reheating furnaces. Estimates on the reduction of capital costs have ranged from 30-60 percent below current practice (SMS 1995, Tomasseti 1995, Kuster 1996). Given that this technology is still new, we currently estimate a capital cost 15 percent below conventional continuous casting. Operations and maintenance costs are also expected to drop by 20-25 percent, although these reductions will depend on local circumstances (Worrell 1995, Tomasseti 1995). Tomasseti 1995 has also noted that integration of casting and rolling has also significantly reduced dust emissions resulting in a near dust-free environment. While this technology looks promising, there are also some important technical challenges that need to be addressed. The US steel industry noted in their technology roadmap the need to develop a better knowledge of the variations of heat transfer, develop new models, sensors, and control systems, develop new techniques of liquid flow control, and finally to develop post-processing steps to improve strip steels mechanical properties (AISI 1998). Maintaining a high level surface quality has been a big hurdle in many demonstration projects (Opalka 1999). Additional technical work is needed on mold level control, mold cooling, deformation, and wear, surface roughness of the roll, and resistance of components to liquid steel, and atmospheric and surface oxidation (Kuster 1996, de Beer 1999). A much tighter control on upstream operations and flows are needed so as to ensure that the caster does not bottleneck the process (Kuster 1996, Worker 1998). There is also the issue of many mills having invested considerable resources into existing more conventional casting technologies. Finally, as the DOE research shows, it is unclear as to whether thin strip cast steel can compete with cold rolled steel for high end markets such as automobiles and appliances (Kuster 1996). However, given the significant research efforts that are being undertaken on this technology by consortia in Europe, Japan, and Australia, to address technical concerns, we believe that the penetration rate for nonhigh end applications before 2015 is likely to be high, yielding potential savings of 9 percent of steel energy use. Our recommended next steps on this technology include further research and development to overcome remaining technical barriers and the use of small scale flat rolling demonstration projects.

100

Emerging Energy-Efficient Industrial Technologies

New EAF Processes (Steel-3)

The U.S. iron and steel industry is made up of integrated steel mills that produce steel using a blast furnace and the Basic Oxygen Furnace (BOF), and electric steel mills that produce steel from scrap steel or direct reduced iron (DRI) using an electric arc furnace (EAF). In the EAF scrap is melted and refined, using a strong electric current. DRI can be used to enhance product quality. Several process variations exist, using either AC or DC currents, and fuels can be injected to reduce electricity use. The majority of steel produced in the U.S. is from integrated steel mills, although the share of electric steel mills is increasing, growing from 15 percent of production in 1970 to 40 percent in 1995. Electric steel mills are located throughout the U.S., with some concentration in the South, near waterways for shipping and in areas with lower-cost electricity. In 1997 there were 85 electric steel companies operating 122 mills with 226 EAFs. These facilities are spread throughout 35 states, with the largest number of plants in Pennsylvania, Ohio, and Texas. The electric arc furnaces at these mills range in age from brand new to 74 years, with an average age of 24 years (Worrell et al. 1999). Total annual nominal capacity listed in 1994 was 55.6 million tons (50.5 Mt). Between 1995 and 1997 an additional 13 million tons (12 Mt) of EAF capacity was built. In electric steelmaking, energy is mainly consumed in the EAF and the rolling. On basis of the literature and statistics we estimated the energy consumption in different steps of electric steelmaking. In 1994, EAF based mills produced 39.6 million tons (35.9 Mt) of crude steel, consuming 23 TWh of electricity and 154 TBtu (162 PJ) of fuels (Worrell et al. 1999). Of this EAFs consumed 17 TWh and 5.7 TBtu (6.0 PJ) injection fuels (Worrell et al. 1999). The average rated power consumption is 436 kWh/ton (480 kWh/tonne) and fuel consumption for injection and preheating is estimated at 0.14 MBtu/ton (0.16 GJ/t). While modern EAFs are generally more energy efficient many technologies exist to improve energy efficiency in existing furnaces, such as process control, efficient transformers, oxy-fuel injection, bottom stirring, post-combustion, eccentric bottom-tapping and scrap preheating (Worrell et al. 1999). Several new EAF-designs are under development, which combine energy saving features like increased fuel and oxygen injection with scrap preheating (Greissel 2000, IISI 2000b). The aim is to produce a semicontinuous process with enhanced productivity through reduced resource use (e.g. refractories, electrodes) and reduced tap-to-tap times. At the same time increased product quality also demands increased feedstock flexibility (e.g. scrap, DRI or pig iron). Different developers are involved in new EAF-process design, the most important being the Twin Electrode DC (IHI, Japan), Comelt (Voest Alpine, Austria) and Contiarc and Conarc (SMS Demag, Germany). IHI (Japan) is currently developing a new process consisting of a shaft type preheater with twin electrode DC furnace (Takeuchi et al.1995, Jones 1997). By using two DC electrodes the heat flux is directed to the middle of the furnace, reducing the heat losses in the furnace walls. Process operation is fully automated. Two pilot/demonstration plants are in operation in Japan. The process parameters are estimated at an electricity consumption of 236 kWh/ton (260 kWh/tonne), a fuel consumption of 0.69 MBtu/ton (0.80 GJ/t), and an oxygen injection of 1165 cubic feet/ton (33 NM3/tonne steel) (Jones 1997). The capital costs are expected to be lower than that of conventional DC furnaces due to the higher productivity. The Contiarc process is being developed by Mannesmann Demag (Germany). The Contiarc process consists of a continuous scrap smelting process (instead of the current batch process) with a capacity of 1 Mt/year. The design aims to be energy efficient and low emission (Reichelt and Hofman 1996; Mllers et al. 1997). The designed and expected electric energy consumption is estimated to be 227-234 kWh/ton (250-258 kWh/tonne), while injecting 0.41 MBtu/ton (0.48 GJ/t) (Reichelt and Hofman 1996; Mannesmann 1998). The production costs are expected to be $9-13 lower per ton steel produced (Reichelt and Hofman 1996; Mannesmann 1998), or up to a 20 percent reduction. The first two orders for Contiarc were apparently placed in early 1999. The Comelt process (Voest Alpine, Austria) aims to develop a highly efficient semi-continuous process (Jones 1997). The process has four graphite electrodes and one bottom return electrode. The whole furnace is tilted to tap the heat. The position of the electrodes enables increased heat recovery as the shaft preheater can be located on top of the furnace. Electricity consumption is estimated to be 278 kWh/ton (307 kWh/tonne), natural gas use of 0.21 MBtu/ton (0.24 GJ/t, plus additional carbon use), with an electrode consumption of

101

LBNL and ACEEE

New EAF Processes Data Table

Units New EAF furnace processes steel-3 Advanced Electric Arc Furnaces Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Mtons

Iron and Steel Process Heating Electricity, fuel New, Replacement 58.0

331 Electric Arc Furnace to melt scrap into liquid steel Replacement at end of life existing furnaces

Electric arc furnace (average performance in 1994) ton/yr kWh MBtu MBtu 1 436 0.14 3.8 EAF annual capacities vary between 5,000 and 1.5 Million tons/year

Advanced Electric Arc Furnaces with Scrap Preheating and High Use of Oxygen kWh 240 MBtu 0.4 MBtu 2.5 Field Test 2000 Years 40 kWh/% MBtu/% MBtu/% 436 -0.3 1.4 Low 12% 100% -193% 36% No advanced EAFs in use, although some are very efficient designs Half of potential market between 2000 and 2015 Between 2000 and 2015 potentially 14 Mtons EAF capacity will be build

GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years %

3032 -2 23.9 4 Incremental -8 -0.02 27.40 -5.30 0.3 305% Additional costs over conventional AC-EAFs

Discount rate for all CCE calculations is 15%

Significant None Somewhat H,M,L Medium Technical, marketing High Field Test, Marketing Fair

Reduced tap-to-tap time, reduced electrode and refractory consumption Improved feedstock flexibility Reduced offgas volumes; easier to clean Promotional

H,M,L E,G,F,P

EIA, 1999 Worrell et al. 1999 Jones 1997; Worrell et all 1999; Mannesmann 1998 Worrell et al., 1999 Worrell et al., 1999 Jones, 1997; Mannesmann, 1998; Reichelt and Hofmann, 1996

102

Emerging Energy-Efficient Industrial Technologies only 3.6 lb./ton steel (1.8 kg/tonne) (Jones 1997). The capital costs of a large Comelt-unit are expected to be equal to that of a DC furnace (Jones 1997), and higher for small capacities. The production costs are estimated to be $8-10/ton lower than conventional DC or AC furnaces (Berger and Mittag 1995). Based on the projects discussed above, we assume for the year 2015 a new electric arc furnace to have an electricity consumption of 240 kWh/ton (265 kWh/metric tonne), fuel injection of 0.41 MBtu/ton (0.48 GJ/t) and oxygen injection of 1060 cubicfeet/ton (30 Nm3/metric ton). Energy consumption estimates are based on a 100 percent scrap charge. Increased use of DRI will increase power consumption, while hot metal charging will decrease power consumption. Oxygen production consumes approximately 0.68 kWh/Nm3 (IISI 1998). Total power consumption is estimated at 261 kWh/ton (287 kWh/t). The capital costs of a new concept electric arc furnace are lower than costs for a DC-furnace, but higher than capital costs of an AC furnace. The costs of an AC-furnace are approximately 10-15$/ton. We estimate the incremental capital costs at approximately $4/ton based on the additional capital costs of DC furnaces and scrap preheating systems as given in Worrell et al. (1999). The new furnace designs will result in lower operating costs due to reduced tap-to-tap time, lower electrode and refractory use, reduced air cleaning costs, as well as reduced energy costs. Based on the expectations of the various processes we estimate total production cost decreases at $8/ton. New furnace designs will be first applied in greenfield EAFs, followed by replacement of old EAFs in existing plants. Assuming an EAF-production of 58 Million tons in 2015 (EIA, 1999) we assume that an additional 8 million tons (7.3 Mt) of greenfield capacity will be constructed by the year 2015. Additionally 5.5 million tons (5.0 Mt) of existing capacity is likely to be replaced by 2015. We assume that half of this capacity will use the discussed EAF-concepts, or 7 million tons (6.4 Mt). Implementation barriers can be found in the perceived risks of the advanced technology, as well as higher capital costs. Although first orders are apparently placed for some of the processes, the technology needs fullscale commercial demonstration in the U.S. The suppliers of these technologies are well placed in the U.S. market with U.S. subsidiaries and manufacturing facilities.

103

LBNL and ACEEE

Low NOx Oxy-Fuel Burners in Steel Reheating Furnaces (Steel-4)

The iron and steel industry is one of the largest industrial energy consumers. The U.S. iron and steel industry is made up of integrated steel mills that produce pig iron from raw materials (iron ore, coke) using a blast furnace and steel using a basic oxygen furnace (BOF) and electric arc furnace steel mills that produce steel from scrap steel, pig iron, or direct reduced iron (DRI) using an electric arc furnace (EAF). In 1994, 2,180 TBtu (2,300 PJ) (about 11 percent of manufacturing primary energy use) was consumed in the production of about 100 Mton (91 Mt) of crude steel products (EIA 1997). Within steelmaking, the largest energy use is required to reduce iron ore in the integrated mills and to re-melt steel scrap in electric arc furnace mills. After the liquid steel is produced, it is cast and shaped. After casting, the shaped products are further rolled to produce sheet, strip, plate, and other structural products. In 1994, 88 Mtons (79.6 Mt) of steel was hot rolled with an estimated energy requirement of 245 TBtu (259 PJ) of fuel and 53 TBtu (56 PJ) of electricity, resulting in a primary energy intensity of 4.6 Mbtu/ton (5.3 GJ/t). Reheating furnaces consume approximately 2.2 MBtu/ton (2.6 GJ/t). In the reheating furnace steel is heated to temperatures between 1100 and 1300 oC. Hot rolled production in 2015 is estimated at 116.6 million tons (106 Mt) (EIA 1999). The high temperatures in the rolling furnace require high flame temperatures. However, high flame temperatures also lead to high NOx-emissions with standard furnace and burner designs. Existing efficient burners have aimed to recover the waste-heat to pre-heat the combustion air in the furnace, but this can lead to higher NOx-emissions. However, this is not necessarily the case since there is a lot of experience with the use of recuperative burners in the steel industry, and good designs do not lead to higher NOx-emissions (Flanagan 1993). An alternative way to increase efficiency is the use of oxy-fuel burners. Oxy-fuel burners are now extensively used in glass furnaces. Older designs of oxy-fuel burners for steel reheating furnaces led to higher NOx-emissions (Farrell et al. 1993). However, new designs provide close to the stoichometric amount of oxygen to the fuel, limiting the formation of NOx. The high velocities of the gases in the burner ensure that the fuel is completely combusted at a lower temperature zone of the flame. The high velocities also lead to a better heat distribution in the furnace, improving productivity in furnaces. Several manufacturers offer different designs of low NOx oxy-fuel burners for steel reheating furnaces. In the U.S. the main suppliers with experience in reheating furnaces are American Combustion (Atlanta, GA), Praxair (Tarrytown, NY) and Bricmont (Canonsburg, PA). Other manufacturers have developed oxy-fuel burners, but not yet used in steel reheating furnaces (e.g. Air Liquide, Air Products). Praxair has been experimenting and testing low-NOx oxy-fuel burners in steel reheating furnaces for the past decade. Their oxy-fuel burners have been tested at two steel furnaces: a furnace at Bethlehem Steel at Burns Harbor and at Auburn Steel (NY). The project at Bethlehem Steel is sponsored by the NICE3 program of DOE. Expected energy savings at Bethlehem steel were 35 percent, and actual energy savings were almost 50 percent (Selines 2000). At Auburn Steel no savings were achieved as the burners were primarily used to increase the production rate of a furnace, and the positioning of the burners did not allow them to achieve energy savings. In the original project up to 30 percent energy savings were expected (Valenti 1998). Older tests with oxy-fuel burners in continuous reheating furnaces demonstrated fuel savings of 28-39 percent (Farrell et al. 1993). Praxair is marketing the technology and talking to a couple of other prospective customers in the steel industry. American Combustions Pyretron burner is based on the similar concept, using high-velocities and selective oxygen supply to increase efficiency and reduce NOx-emissions. The technology has been used in reheating furnaces for many metals, including steel. In a pusher-type steel reheating furnace application of the new burner has led to increased productivity (25 percent) and fuel savings of 1.07 MBtu/ton (1.24 GJ/t), with an oxygen use of 0.5 million cubic feet/ton (or 14 Nm3/tonne) (American Combustion 2000). We estimate average 2010 fuel savings at 30 percent, or 0.66 MBtu/ton (0.77 GJ/t) with additional oxygen use of 0.3 mcf/ton (9 Nm3/tonne). Oxygen production consumes approximately 0.68 kWh/Nm3 (IISI 1998). Total power consumption is estimated at 6.1 kWh/ton (6.8 kWh/metric tonne). Net primary energy savings are estimated at 0.63 MBtu/ton (0.73 GJ/t).

104

Emerging Energy-Efficient Industrial Technologies Low NOx Oxy-Fuel Burners in Steel Reheating Furnaces Data Table
Units Notes Oxy-Fuel Burners in Reheating Furnaces Steel-4 Advanced Oxy0Fuel Burners in Steel Reheating Furnaces Improve Efficiency and Reduce NOx-Emissions Market Information: I ndust r i es SI C 33 I r on & St eel Enduse( s) R eheat i ng f ur naces i n hotr ol l i ng m i l l Pr ocess H eat i ng Ener gy t ypes N at ur algas,coke oven gas Fuel M ar ketsegm ent R et r of i t ,N ew Est i m at ed t hr oughputi n 2015 ( EI A,1999) 2015 basecase use 116. 6 Reference technology D escr i pt i on Aver age r eheat i ng f ur nace ( cont i nuous and bat ch) Thr oughputor annualop.hr s. Ton,capaci t y m ay r ange f r om a f ew t / ht o over300 t / h 1 El ect r i ci t y use kW h El ect r i ci t y use f orf ans i s ver y sm al l 0 Fueluse M Bt u 2. 2 Pr i m ar y ener gy use M Bt u 2.2 New Measure Information: D escr i pt i on R api df i r i ng oxyf uelbur ner s wi t h Low N O x com bust i on char act er i st i cs El ect r i ci t y use kW h 6 Fueluse M Bt u 1. 5 Pr i m ar y Ener gy use M Bt u 1. 6 C ur r entst at us NI C E3 pr oj ectofone oft he pr oduct s Fi el d Test / C om m er ci al i zed D at e ofcom m er ci al i zat i on 1998 Est .avg.m easur el i f e Year s 10 Savings Information: El ect r i ci t y savi ngs kW h/ % 6 90% Fuelsavi ngs M Bt u/ % 0. 7 90% Pr i m ar y ener gy savi ngs M Bt u/ % 0. 6 90% Penet r at i on r at e Low Feasi bl e appl i cat i ons % 30% Ot herkey assum pt i ons El ec svgs pot ent i ali n 2015 GW h 213 Fuelsvgs pot ent i ali n 2015 Tbt u 23 Pr i m ar y ener gy svgs pot ent i ali n 2015 Tbt u 21.2 Cost Effectiveness I nvest m entcost $ 2 Type ofcost Ful lcost C hange i n ot hercost s $ 0. 67 C ostofsaved ener gy ( el ec) $/ kW h 0. 03 C ostofsaved ener gy ( f uel ) $/ M bt u 0. 32 C ostofsaved ener gy ( pr i m ar y) $/ M bt u 0. 35 Discount rate for all CCE calculations is 15% Si m pl e payback per i od Year s 1. 2 I nt er nalr at e ofr et ur n % 82% Key non energy factors Pr oduct i vi t y benef i t s Pr oduct i on capaci t yi ncr ease by up t o 25% Si gni f i cant Pr oductqual i t y benef i t s N one Envi r onm ent albenef i t s N O x em i ssi on r educt i on ofup t o 7090% Si gni f i cant Ot herbenef i t s C ur r entpr om ot i onalact i vi t y H, M, L M ar ket i ng by pr oducer s,D O EOI T M edi um Evaluation M aj orm ar ketbar r i er s Econom i c Li kel i hood ofsuccess H, M, L Hi gh R ecom m ended nextst eps Di ssem i nat i on D at a qual i t y assessm ent E, G, F, P Fai r Sources: 2015 basecase EI A 1999 Basecase ener gy use W or r el letal . ,1999 N ew m easur e ener gy savi ngs Far r el letal . ,1993;Am er i can C om bust i on,2000,Sel i nes,2000 Li f et i me Fl anagan,1993 Feasi bl e appl i cat i ons C ost s D er i ved f r om W or r el letal . ,1999 Key non ener gy f act or s Far r el letal . ,1993;Am er i can C om bust i on,2000,Sel i nes,2000 Pr i nci palcont act s R on Sel i nes,Pr axai r( 914)3456457 Tol y Par nas,Am er i can C om bust i on ( 770)5644180 ( ext .251)

105

LBNL and ACEEE The oxy-fuel-burner can be installed in existing furnaces without rebuilding the furnace. Based on the costs of other burner systems we estimate the investment costs at $2.3/ton-capacity ($2.5/t-capacity) (Worrell et al. 1999). Assuming oxygen-supply is present, we only include the costs for oxygen delivery in the capital costs, and the production costs in the operation and maintenance costs. Oxygen costs are estimated at 0.04$/Nm3 (De Beer et al. 1998a). Application of oxy-fuel burners may result in productivity increases (up to 25 percent) and up to 70-90 percent reduction in NOx emissions. Although, oxy-fuel burners can lead to increased productivity, the value will depend on the utilization of the furnace. We assume that the productivity increase is on average valued at 1$/ton (derived from Worrell et al. 1999). The technology is likely to be successful. The applicability of the technology depends on the furnace design and the use of competing technologies. Competing technologies may be other low-NOx burner designs as well as recuperative burners. We assume that the technology could potentially be applied to 30 percent of the hot rolled steel production by 2015. Furhter dissemination of experiences with these burners to industry and air quality regulators is needed to increase the penetration of this technology in the market.

106

Emerging Energy-Efficient Industrial Technologies

Smelting Reduction Processes (Steel-5)

The iron and steel industry is one of the largest industrial energy consumers. The industry is made up of integrated steel mills that produce pig iron from raw materials (iron ore, coke) using a blast furnace and steel using a basic oxygen furnace (BOF) and electric arc furnace steel mills that produce steel from scrap steel, pig iron, or direct reduced iron (DRI) using an electric arc furnace (EAF). In 1994, 2,180 TBtu (2,300 PJ) (about 11 percent of manufacturing primary energy use) was consumed in the production of about 100 Mton (91 Mt) of crude steel products (EIA 1997). In 1997 there were 14 integrated steel companies operating 20 integrated steel mills with a total of 40 blast furnaces. These mills are concentrated in the Great Lakes region, near supplies of coal and iron ore and near key customers such as the automobile manufacturers. The blast furnaces in these mills range in ageaccounting for furnace rebuildsfrom 2 to 67 years, with an average age of 29 years. Production rates per plant vary between 0.6 and 3.4 million tons (0.5-3.1 Mt) per year. Total production of U.S. blast furnaces in 1997 was slightly over 59.5 million tons (54 Mt). Primary steel production using a basic oxygen furnace fluctuated between 44 and 83 million tons (40-75 Mt) over the period. Primary energy consumption for ironmaking is estimated at 780 TBtu (862 PJ), or equivalent to 45 percent of total energy use in the iron and steel industry (1994) (Worrell et al 1999). Pellet making at the mine-site adds 125 TBtu (132 PJ) (Margolis 1996), using 1.24 tons of pellet per ton of pig iron. Based on the AEO-2000 we estimate 2015 iron production at about 57 million short tons (52 Mt). Smelting reduction is the latest development in pig iron production. Smelting reduction processes currently under development combine coal gasification with the direct reduction of iron oxides. In this way, smelting reduction will abandon the need for coke, while future processes will also abandon ore preparation. The COREX process is the only commercial smelting reduction process. CCF, DIOS and HIsmelt are advanced research projects. Commercial COREX plants are operating in South Africa and South Korea, with new plants under construction in India and South Africa. First commercialization of the more advanced smelting reduction processes is expected around 2005-2010 (De Beer et al., 1998). In smelting reduction iron ore is pre-reduced by gases coming from a hot bath. The pre-reduced iron is then melted in the bath. The process produces excess gas, which is used for power generation, production of direct reduced iron (an alternative iron input for scrap), or as fuel gas. Due to the different reaction conditions and the full integration of iron and steel making, the theoretical energy demand of smelting reduction is lower than that of a blast furnace. Studies estimated the energy consumption to be 20-30 percent lower than that of the conventional blast furnace route. Currently operating plants already show energy consumption levels comparable to the blast furnace routes, but at much smaller scales. The second generation smelting reduction technology would reduce energy use in ironmaking by 30 percent relative to current processes. Smelting reduction has dramatically lower capital costs, and has other inherent advantages. It saves material costs, ( the process can use. less expensive coal than current metallurgical coal), allows for better pollution control, and shows favorable economics at smaller capacities compared to conventional technology. The main process developers can be found in Austria/Germany (Voest Alpine), Australia/Germany (CRA/Klockner), Japan (NKK and others) and The Netherlands/UK (Corus). Current smelting reduction plants are all greenfield plants. For brownfields (rebuilds of existing sites), smelting reduction processes will compete with extension of the lifetime of existing blast furnaces and with production of DRI (mainly as input into EAFs). Environmental issues could limit the operation of (older) coke ovens in the U.S. triggering interest in smelting reduction. While replacement of coke ovens could be a reason to invest in smelting reduction (Meijer et al., 1994), this has not been the case in the U.S. where producers have shown renewed interest in non-recovery coke ovens, as well as increased imports. There is some indication of change however. Geneva Steel (Utah) has shown interest in the COREX-process and recently Nucor has shown interest in the Hismelt process to replace the failed iron carbide process in Trinidad as a source of iron for their U.S. Plants (33Metalproducing.com 2000b). AISI was involved in the development of a bath smelting process, but abandoned the research. Coke making processes can have significant negative environmental impacts. Various emissions of environmentally hazardous compounds (e.g. sulfur compounds, poly-aromatic hydrocarbons) make

107

LBNL and ACEEE

Smelting Reduction Data Table

Units Notes Smelting reduction processes steel-5 New Production Route for the production of pig iron, replacing cokemaking, ore preparation and the blast furnace Market Information: Industries SIC 33 Iron and Steel End-use(s) Incl. Cokemaking, pelletizing, sintering and blast furnace Other Energy types Coal, oil, gas, electricity Market segment New Estimated 2015 production on basis of EIA 1999 2015 basecase use Mtons 57.0 Reference technology Description Blast Furnace Route, including cokemaking, pelletizing, sintering and the blast furnace Characteristics expressed per ton capacity, varies between 0.8 and 3.1 Mt/yr Throughput or annual op. hrs. short ton Electricity use kWh 57 Fuel use MBtu 15.5 Primary energy use MBtu 16.0 New Measure Information: Description Smelting Reduction Electricity use kWh 64 Fuel use MBtu 12.4 12.9 Primary Energy use MBtu Three COREX plants operating worldwide, others pilot-plant Current status Commercial COREX C-2000 in Sotuh-Korea, other processes expected around 2005-2010 Date of commercialization 1995 Est. avg. measure life Years 40 Savings Information: Electricity savings kWh/% -7 -12% Fuel savings MBtu/% 3.1 20% Primary energy savings MBtu/% 3.1 19% Penetration rate No U.S. plants Low average lifetime 70 years for blast furnace and one greenfield Feasible applications % 18% Other key assumptions Elec svgs potential in 2015 GWh -72 Fuel svgs potential in 2015 Tbtu 32 Primary energy svgs potential in 2015 Tbtu 31.6 Cost Effectiveness Smelting reduction investments: $227/ton, BF-route investments: $349/ton Investment cost $ -122 Type of cost Replacement Operating cost reduction may be between 12.6 and 14.5$/ton Change in other costs $ -13.6 Cost of saved energy (elec) $/kWh 4.57 Cost of saved energy (fuel) $/Mbtu -10.18 Discount rate for all CCE calculations is 15% Cost of saved energy (primary) $/Mbtu -10.38 Simple payback period Years -6.1 Internal rate of return % N/A. Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Somewhat None Significant H,M,L Low Reduced capital, lower fuel costs (fuel shift), lower operation costs Lower air and water emissions Limited interest with integrated steelmakers

H,M,L E,G,F,P

Marketing, integration Medium Commercial demonstration Good

Integration into brownfields, limited interest integrated steelmakers

EIA 1999;Margolis, 1996 Margolis, 1996; Worrell et al., 1999 De Beer et al. 1998a; IISI, 1998 Worrell et al., 1999 De Beer et al. 1998a De Beer et al. 1998a, IISI 1998; Meijer et al., 1994

108

Emerging Energy-Efficient Industrial Technologies extensive gas clean-up at the coke plant necessary. Coke making and ore preparation release large amounts of particulate matter. Inherent to the smelting reduction processing route is the absence of the formation of most of the problematic compounds in coke making, while the fuel gas produced has much lower sulfur content than coke gas. In smelting reduction, hydrocarbons are not condensed, but combusted at the high reactor temperatures. Integration, abandoning coke quenching, and reduced ore handling will reduce PM emissions. We base the description of the energy characteristics on the analysis on De Beer et al. (1998a) and IISI (1998). Smelting reduction plants generally have a higher coal input per ton product than current blast furnaces, but export larger quantities of fuel gas. The exported offgas of the COREX-process has a heating value of approximately 7 MJ/Nm3 (LHV) and is relatively clean (sulphur content of 10-70 ppm) (Phringer et al.1991). Net energy consumption of smelt reduction is therefore lower than that of the blast furnace route. U.S. integrated steel plants use on average 16.0 MBtu/ton pig iron (18.6 GJ/t hot metal (thm), including energy use for pellet making) (Worrell et al. 1999). Net primary energy consumption of smelt reduction process may vary between 11.4 and 13.4 MBtu/ton hot metal (13.3 and 15.6 GJ/thm) (De Beer et al. 1998). We will assume a 2010-2015 performance of 12.9 Mbtu/ton pig iron (15.0 GJ/thm). In the long term further reductions leading to a net specific energy consumption (SEC) of 11.4 MBtu/ton (13.2 GJ/thm) may be expected (De Beer et al., 1998). Primary energy use estimates depend heavily on the chosen use of the off-gas (i.e. power generation in a combined cycle or steam-cycle, or for production of DRI). Currently operating COREX-plants show energy consumption levels comparable to the blast furnace routes, i.e. 16.4 Mbtu/thm (17.3 GJ/ton hot metal) (IISI 1998), but at much smaller scales. Future improvements can be achieved by new process developments, increased capacities, optimization of the carbon monoxide/oreinteraction, and optimization of fuel gas use. Investment costs are lower compared to the conventional process route, as coke making and ore preparation may be abandoned. This will also result in lower operating costs. Capital costs of modern blast furnacebased plants are high are approximately $349/ton hot metal ($385/thm) (De Beer et al., 1998). The investments involve coke plants, ore preparation (sintering, pelletization) and the blast furnace. The investments required for the COREX-process are estimated to be around $227/ton hot metal (250 US$/thm) (excluding ore agglomeration plant) (Meijer et al.1994; De Beer et al. 1998). The capital required for a commercial sized CCF plant are estimated to be $136-163/ton hot metal (150-180 US$/thm) (Meijer et al.1994). The operating costs of a smelt reduction plant will depend on local conditions, but may be expected to be significantly lower due to the abandoned processes. The reduction of operation and maintenance costs for the CCF process in Western-European conditions is estimated to be $16/ton pig iron (18 US$/t) pig iron (Meijer et al. 1994). Coke plants consume high grade coking or metallurgical coal types, which are more expensive than steam coal. Steam coal costs are on average $5/ton ($5.5/t) lower than coking coal (De Beer at al. 1998). Smelting reduction technology makes it possible to use steam coal, thereby reducing fuel costs (compared to a blast furnace). The production costs of pig iron via the blast furnace route are estimated at $110-$145/ton hot metal ($122-160/thm) (De Beer et al. 1998). Smelting reduction processes will result in production costs of 5 percent to 35 percent below the costs of the conventional route (De Beer et al., 1998). For the period 2010-2015 we assume a cost reduction of 10 percent, or $11-15/ton hot metal ($12-16/thm). Next steps include the commercialization of the second-generation smelting reduction process through demonstration on a near-commercial scale. Due to the large costs involved this should preferably be a project shared by (various) steel companies, technology providers and other parties.

109

LBNL and ACEEE

Advanced Forming/Near Net Shape Casting (Alum-1)

The United States is the largest aluminum producer globally, with a combined production of primary and secondary aluminum in 1999 of 8 million short tons (7.3 Mt) (USGS 2000). Primary aluminum production, which is extremely energy intensive, accounts for roughly half of the production (3.8 million short tons (3.4 Mt)) and roughly 2 percent of primary energy use in U.S. manufacturing (USGS 2000, EIA 1997). Once the smelted aluminum is produced (either by the primary process or the secondary process), it is alloyed in a holding furnace and then cast into ingots or continuously cast in a rolling mill. The ingots or continuously cast aluminum is then hot rolled and cold rolled into plate, sheet, or foil. Estimates for the energy use for casting are 5.7 Mbtu/ton (6.6 GJ/t) and between 5.1 and 7.0 Mbtu/ton (5.9 and 8.1 GJ/t) for hot and cold rolling (Margolis 1997). On a national basis this means that casting and rolling accounts for roughly 60 TBtu (63 PJ) of energy use. Currently, the casting and rolling process is a multi-step process. In ingot casting, the more common of the casting processes, the aluminum is cast into ingots that are then transported, reheated, and rolled. The primary form of ingot casting is vertical direct chill where the molten metal is poured through a spout into a mold and then cooled. Other casting variations include horizontal direct chill and electromagnetic casting, where an electromagnetic field is used to hold and shape the metal in a special mold (Margolis 1997). Before hot rolling ingots are pre-heated in temperature controlled furnaces (walking beam furnaces are used to heat larger ingots), and for some applications, cold rolling is used to achieve desired thickness and finish. Near net shape or thin strip casting is a new technology that integrates the casting and hot rolling of aluminum into one process step, thereby reducing the need to reheat the aluminum ingot before rolling it. Instead of casting slabs in a thickness of 120-300 millimeters, slabs are cast much thinner, as low as 1-10 mm thickness. (Daaland et al. 1997, Erdman 1999, Opalka 1999). The aluminum is essentially cast and formed into its final shape without the reheating step. The first successful twin roll caster for aluminum was developed by Joseph Hunter in 1956 (Ertan et al. 1999). Because of the relatively low capital costs for casters they, are increasingly becoming more popular. Alcan developed a strip caster for painted sheet (that does not require a high level of surface quality) and there are other strip casters in operation with Barmet Aluminum and Vulcan Aluminum (Kuster 1996). In 1998, Kaiser was planning on starting up a micro-mill where the strip will be hot rolled into two strands for final gauge annealing (Kuster 1996). Aluminum Pechiney has also made significant improvements on thin strip casting technology for foil and beverage cans with its 3CM caster, as has Fata Hunter with its speed caster machines (Erdman 1999, Hamer 2000, Brooks 1997). The energy consumption of a thin strip caster is significantly less than the current process, since the preheating requirements are eliminated. We estimate fuel savings of 0.4 MBtu/ton (0.5 GJ/t) for hot rolled aluminum with electricity savings of 20 kWh/ton. A key driver for this technology is the potential for increased yield and reduced operational costs with thin strip technology. One rolling mill found a 15 percent productivity improvement over a two-year period after installing the new caster technology (Daaland et al. 1997). Thin strip casting is also expected to improve the quality of the cast aluminum (surface quality, center line segregation, geometrical tolerances) since thinner cast strips could be better controlled. Finally, the technology provides an opportunity to develop new aluminum products (Daaland et al. 1997). Dyllus et al. 1991 note that capital investment costs for continuous casting and hot rolling lines are generally lower than standard casting lines. Overall, capital costs range from $150-$200/ton, and incrementally are estimated to save $70-90/ton as compared to standard processes (Dyllus et al. 1991). Also, labor and maintenance costs are expected to be lower with casting machines; we estimate a $20/ton reduction (Dyllus et al. 1991). Assuming that the beverage can market can be accessible to thin strip casting, the next logical step in aluminum strip casting is to develop it for larger-scale applications such as auto body sheet and for applications of less than 2.5 mm thickness. However, the technical challenges increase significantly for this 110

Emerging Energy-Efficient Industrial Technologies

Advance Forming Data Table

Units Notes Advanced forming/near net shape technology alum-1 Thin strip casting Market Information: Industries Aluminum End-use(s) Process heating Energy types Fuels, electricity Market segment New Estimate of 2015 aluminum production 2015 basecase use Mton 13.6 Reference technology Description VDC ingot casting with hot and cold rolling Throughput or annual op. hrs. ton 1 Electricity use kWh Margolis 1997 193 Fuel use MBtu Margolis 1997 4.0 Primary energy use MBtu 5.6 New Measure Information: Description Thin strip casting Electricity use kWh Margolis 1997 174 Margolis, 1997 and Daaland et al., 1997 Fuel use MBtu 3.5 Primary Energy use MBtu 5.0 Current status Near commercial Date of commercialization Commercial for foil applications, not for harder alloys Est. avg. measure life Years 15 Savings Information: Electricity savings kWh/% 19 10% Fuel savings MBtu/% 0.5 13% Primary energy savings MBtu/% 0.7 12% Penetration rate Medium Feasible applications % Applies to sheet, foil, and plate 25% Other key assumptions Elec svgs potential in 2015 GWh 66 Fuel svgs potential in 2015 Tbtu 2 Primary energy svgs potential in 2015 Tbtu 2.3 Cost Effectiveness Investment cost $ -50 Type of cost Incremental Change in other costs $ -20 Cost of saved energy (elec) $/kWh <0 Cost of saved energy (fuel) $/Mbtu <0 Discount rate for all CCE calculations is 15% Cost of saved energy (primary) $/Mbtu <0 Simple payback period Years Immediate Internal rate of return % Key non energy factors Productivity benefits 15% productivity improvements shown Compelling Product quality benefits None Environmental benefits None Other benefits Current promotional activity H,M,L For foils and specific applications Medium Evaluation Major market barriers Technical Likelihood of success H,M,L High Recommended next steps R&D, demonstration Data quality assessment E,G,F,P Good Sources: 2015 basecase AEO, 2000 Basecase energy use Margolis, 1997 New measure energy savings Margolis, 1997 Lifetime Worrell et al. 1999 Feasible applications Hamers 2000 Costs Dyllus et al., 1991 Key non energy factors Daaland et al. 1997 Principal contacts Shaun Hamers, Fata Hunter. Shamers@aol.com Additional notes and sources

111

LBNL and ACEEE application and these have not yet been solved by the industry. One of the main reasons is that when casting harder alloys at thinner gauges, the rolling forces tend to separate the sheet (Erdman 1999, Hamer 2000, Ertan et al. 1999). Near net shape casting was identified as a high mid-term research priority by the aluminum industry in its aluminum technology roadmap process undertaken with the U.S. Department of Energy. It is unclear whether the technical challenges of operating thin strip casting machines with harder alloys will be overcome. However, significant opportunities exist in the near term with foil and thin sheet. Next steps involve further research, development and demonstration.

112

Emerging Energy-Efficient Industrial Technologies

Efficient Cell Retrofit Designs (Alum-2)

The United States is the largest primary aluminum producer in the world, with a production of 3.8 million tons (3.4 Mt) in 1999 from 23 primary plants (USGS 2000, OIT 1999). In 1994, primary aluminum production in the U.S. consumed 53,552 million kWh of electricity, or 2 percent of primary energy use in U.S. manufacturing (EIA1997). Energy consumption represents 20-30 percent of total production costs. Average energy consumption in the U.S. for aluminum production in 1994 was 16 kWh/kg and was estimated at 15 kWh/kg in 1997 (EIA 1997, USGS 2000, Eisenhauer et al. 1997). Primary aluminum production is an energy intensive continuous process and involves the electrolytic reduction of alumina into aluminum. This process, known as the Hall-Heroult smelting process is accomplished in a series of cells or pots that are connected in long lines in buildings. In each cell the refractory material is overlaid with a carbon lining and a carbon cathode. The molten aluminum lies beneath a bath of cryolite that serves as the medium to dissolve incoming alumina and to conduct electricity from the anode to the aluminum. Alumina is fed into the top of each cell on top of a crust of cooler cryolite (which acts as an insulator), and the crust is periodically broken to allow the alumina to be stirred into the bath. When a strong electric current (50-225 kilo-Amperes) is passed through the electrodes into the bath, the alumina is reduced via a reaction to Al2F6 (which is easier to reduce then Al2O3), to produce molten aluminum. The voltage drop across a cell is 4-4.5V, while the voltage in a potline can exceed 1,000 volts (Gitlitz 1995). The molten aluminum is periodically siphoned off through the tops of the cell into a holding furnace and then poured into ingots or billets, which are then cast and shaped. While a new generation of aluminum smelting technologies are being developed (see Inert Anodes for example), there are also a series of retrofit technologies that could significantly improve cell operation and reduce electricity consumption. These retrofit options are mainly geared to improving the current efficiency of the cell (i.e. closer anode-cathode spacing) and reducing heat losses. Options include improved conductivity for anode materials, bottom heat recovery, increased insulation in furnaces, advanced controls, improved design of the electrical bus components, operation with a low-ratio AlF3 electrolyte chemistry (i.e. improvements in the chemical bath), and improved housekeeping, especially for anode changes. (Blok et al. 1995, Energetics 1990, Margolis 1997). Additional enabling technologies include improved modeling, neural network process controls, continuous sensors, and signal analysis of cell voltage (Eisenhauer et al. 1997). The U.S. aluminum industry has targeted as a goal the reduction of energy intensity of aluminum production to 13 kWh/kg in the near term using retrofit technology (Energetics 1997). This reduction is consistent with other studies that have documented potential reductions of 14-16 percent between best and worst practices in modern smelting cells (Moisan 2000). Investment costs for a retrofit project of this type can vary depending on the existing condition and layout of the facilities. A recent project reported at Aluminerie Lauralco (a subsidiary of Alcoa) noted a cost of $60 Million Canadian for a 264 pot retrofit, or a unit cost of around $200/ton ($220/t) (CADDET 1999d). Aside from energy improvements, upgrading existing cell technology can also significantly reduce production costs (by increasing yield by up to 30 percent and by lowering anode replacement costs), and reduce labor costs (Moisan 2000). We note an average reduction of $10/t aluminum in operation and maintenance costs for the new cell installations. Cell retrofit programs are underway among several potlines in the U.S., and the markets full potential is unclear. We believe that given the demonstrated effectiveness in retrofits elsewhere, there is a high likelihood of success that efficient retrofits will become increasingly desirable in the near term in the absence of the commercialization of advanced cell technologies.

113

LBNL and ACEEE

Efficient Cell Designs Data Table

Units Notes Efficient cell retrofit designs Alum-2 New decoating and furnace technology Market Information: Industries SIC 3334 Aluminum End-use(s) Process heating Energy types Fuels Market segment New AEO 2000, smelting output 2015 basecase use Mtons 6.1 Reference technology Description Hall-heroult cell, primary aluminum smelting Unit consumption presented. Smelters cell amperage range from 1 Throughput or annual op. hrs. tonne 175-300 kA Electricity use MWh EIA, 1997, USGS, 2000 16.2 Fuel use MBtu EIA, 1997, USGS, 2000 5.45 Primary energy use MBtu 143.4 New Measure Information: Description Efficient cell retrofits Electricity use MWh CADDET case study, Energetics, 1997 13 Fuel use MBtu 5.45 Primary Energy use MBtu 116.0 Current status Commercialized Date of commercialization Est. avg. measure life Years 15 Savings Information: Electricity savings MWh/% 3.2 20% Savings compared to 1994 baseline Fuel savings MBtu/% 0.0 0% Primary energy savings MBtu/% 27.4 19% Penetration rate Cost effective and easily integrated into upgrade cycles High Feasible applications % 30% Other key assumptions Elec svgs potential in 2015 GWh 5367 Fuel svgs potential in 2015 Tbtu 0 Primary energy svgs potential in 2015 Tbtu 45.6 Cost Effectiveness Investment cost $ Assume full cost. Exchange rate of $CAD 1.2/$US 217 Type of cost Full Change in other costs $ Moisan, 2000 -10 Cost of saved energy (elec) $/kWh 0.01 Cost of saved energy (fuel) $/Mbtu N/A. Discount rate for all CCE calculations is 15% Cost of saved energy (primary) $/Mbtu 0.99 Based on $2.6/Mbtu primary energy price for electricity in the aluminum industry Simple payback period Years 2.7 Internal rate of return % 37% Key non energy factors Productivity benefits Can significantly reduce production costs Significant Product quality benefits Environmental benefits Reduced emissions of fluorocarbons Somewhat Other benefits Current promotional activity H,M,L Evaluation Major market barriers Cost, risk perception Likelihood of success H,M,L High Recommended next steps Demonstration Data quality assessment E,G,F,P Fair Sources: 2015 basecase EIA, 1999 Basecase energy use EIA, 1999; USGS 2000 New measure energy savings Energetics, 1997 Lifetime Author estimate Feasible applications Author estimate Costs CADDET, 1999d Key non energy factors CADDET, 1999d Principal contacts Additional notes and sources

114

Emerging Energy-Efficient Industrial Technologies

Improved Recycling Technologies (Alum-3)

The United States is the largest aluminum producer globally, with a combined total production of primary and secondary aluminum of 8 million short tons (7.3 Mt) in 1999 (USGS 2000). Primary aluminum production, which is extremely energy intensive, accounts for roughly half of the production (3.8 million short tons (3.4 Mt)) and roughly 2 percent of primary energy use in U.S. manufacturing (USGS 2000, EIA 1997). Because of the lower energy and operating costs (recycled aluminum production uses 90 percent less energy), the share of recovered or secondary aluminum production has more than doubled since 1970. A potential limiting factor in the continued growth of the secondary aluminum market is the ability to continue to secure high quality aluminum scrap or to better purify low quality scrap so that secondary aluminum products can be competitive in key markets. The use of improved or advanced recycling technologies therefore can further expand the secondary aluminum market, generating significant energy savings. The demand for recycled aluminum products is strong and expected to continue. Average growth since 1970 is nearly 5 percent per year. The transportation sector is a particularly hopeful prospect; and already accounts for 30 percent of aluminum shipments (Plunkert 1997, Aluminum Association 2000). The aluminum content of US automobiles has doubled since 1991, and is now up to 246 pounds (112 kg) per vehicle, about two-thirds of which is from recycled metal. This content is expected to double by 2005 (Pawlek 2000, Pickens 2000). Traditionally, all scrap is sorted and shredded before being charged into a melting furnace. Contaminants are primarily removed using pyrometallurgical techniques (roasting, delaquering, sweating), but also may be removed using catalytic techniques (cryolite catalysts in a barrel furnace) or hydrometallurgical techniques (use of water) (Margolis 1997). The treated scrap is then charged in furnaces designed for relatively dirty scrap (high-emitting furnaces) or relatively clean scrap (low-emitting furnaces). Molten salts (NaCl and KCl) are sometimes added to standard reverberatory furnaces to help separate out impurities and improve furnace efficiency. The black dross is periodically removed, and itself contains aluminum (8-13 percent) which can be captured in dross or rotary furnaces, or more advanced plasma furnaces. Salt cake, a residual product after processing in rotary furnaces, has a metal content of about 4-6 percent. About 750,000 short tons (680,000 t) of dross/salt cake are generated annually in the US, most of which is landfilled and environmentally harmful given potential leaching of salts into the water table (Margolis 1997, Pawlek 2000, Pickens 2000). Because the demand for secondary aluminum continues to grow, finding cost-effective ways to further increase metals recycling will be highly valued. Several new technologies have emerged that help to improve the recovery or processing of scrap or reduce energy use in the preparing and melting of scrap. The New York State Energy Research and Development Authority, Energy Research Company (ERCo), Philip Services Co., and Stein Atkinson Stordy, Ltd developed a new decoating kiln (the IDEXTM) that reduces kiln energy use by 41 percent while also improving product quality and increasing metal yield by 1 percent (ERCo and Wabash Alloys 1998, CADDET 1996c, OIT 2000). ERCo is now developing a new generation of decoating and melting technology called the vertical flotation dryer (VFD) where high velocity gases strip oils from the scrap in a VFD cone, and the oils are subsequently combusted reducing energy requirements.18 As well as reducing energy use, this decoater increases productivity (run times per decoating batch are significantly reduced) (De Saro et al. 1999, OIT 2000). Other new melt designs include a universal melting plant (Nottingham metal recyclers) that pre-heats and decoats the scrap in a dry hearth furnace and then melts the scrap in a closed well furnace. This technology achieved a 25 percent energy savings and a 2-8 percent increase in metal recovery (CADDET 1993d).

18

The VFD is also being designed to be used as a melter and can replace existing side-well reverberatory furnaces and electric induction melters, while offering additional gains in efficiency. In this write up however, we do not address new melt technologies but focus on decoating technologies. We do recognize that these two steps are likely to become increasingly integrated (as the VFD/VFM design suggests) into new combined designs in the future.

115

LBNL and ACEEE

Improved Recycling Technologies Data Table

Units Notes Improved recycling technologies alum-3 Scrap decoating and new secondary furnace technologies Market Information: Industries SIC 333 Aluminum End-use(s) Process heating Energy types Electricity Primarily retrofit applications. Also greenfields Market segment Retrofit, new Estimate of secondary output based on AEO 2000 2015 basecase use Mtons 7.5 Reference technology Description Existing scrap preparation and secondary furnace systems Throughput or annual op. hrs. tonne Unit consumption presented. 1 Electricity use kWh De Saro, 1998 0.0 Fuel use MBtu De Saro, 1998 2.00 Primary energy use MBtu 2.0 New Measure Information: Description New recycling technologies Electricity use kWh 0 IDEX kiln consumption of 500 btu/lb Fuel use MBtu 1.00 1.0 Primary Energy use MBtu Current status Commercialized Date of commercialization Est. avg. measure life Years 15 Savings Information: Electricity savings MWh/% 0.0 n.a. Savings compared to 1994 baseline Fuel savings MBtu/% 1.0 50% Primary energy savings MBtu/% 1.0 50% Penetration rate High Feasible applications % Strong demand expected for improved decoating technology 30% Other key assumptions Elec svgs potential in 2015 GWh 0 Fuel svgs potential in 2015 Tbtu 2 Primary energy svgs potential in 2015 Tbtu 2.2 Cost Effectiveness Investment cost $ Estimated incremental cost 20 Type of cost incremental Change in other costs $ 0 Cost of saved energy (elec) $/kWh n.a. Cost of saved energy (fuel) $/Mbtu 3.42 Discount rate for all CCE calculations is 15% Cost of saved energy (primary) $/Mbtu 3.42 Based on $2.6/Mbtu primary energy price for electricity in the 4.5 Simple payback period Years aluminum industry Internal rate of return % 22% Key non energy factors Productivity benefits Somewhat Product quality beneifts Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources H,M,L Significant medium Cost, risk perception Medium Demonstration Good EIA, 1999 De Saro, 1999 De Saro, 1999 Author estimate Author estimate De Saro, 2000 De Saro, 1999 Robert DeSaro, ERCo (rdesaro@er-co.com); John Pickens, ALUMITECH (JohnPickens@Alumitechinc.com) Reduced emissions - eases compliance for environmental regulation

H,M,L E,G,F,P

116

Emerging Energy-Efficient Industrial Technologies In the area of dross/salt cake recovery, ALUMITECH began operation of a commercial closed loop dross/salt cake recycling facility in Cleveland which allows for additional aluminum recovery (between 4-8 percent increase over existing processes) and the processing of the non-metallic portion (NMP) of the dross into usable products such as calcium aluminate for steel refining (Pickens 2000). The recycling of byproducts nearly eliminates landfilling costs ($20-40/ton) (Pickens 2000). Similar approaches for metal and salt recovery are also taking place in Europe and Japan. One system manufactured by Altek increases the initial recovery of dross through a dross press to very high levels, and requires further dross processing (Pawlek 2000). The costs of recovery systems vary. A modern beverage can recycling facility can cost $180 to $360 per ton annual capacity ($200-400/tonne annual capacity) (Pawlek 2000). About 20 percent of all recovered aluminum is from beverage cans (Aluminum Association 2000). The incremental investment cost for IDEXTM kiln technology is expected to be greater than existing decoating kiln technology, but is highly dependent on the initial kilns sophistication (DeSaro 2000). Environmental regulation may also be a driver for adopting new decoating kiln technologies. Recent regulations directed at the secondary aluminum industry regulate the emissions of particulate matter and of hazardous air pollutants (HAP) by limiting total hydrocarbon emissions (as a proxy for HAP). (DeSaro 2000, Federal Register 2000). New decoating and melt technologies that reduce emissions (the IDEX eliminates nearly all volatile organic compounds and significantly cuts other emissions) have become increasingly attractive for that reason. Yet, even these new technologies may still require add-on controls to meet EPA emission requirements (Santiago 2000). This combination of drivers (productivity and environmental) promises for a rapidly growing market for advanced recycling technologies. Since its introduction as a commercial technology in 1999, there have been 2 IDEXTM kilns sold in the US market and 10 in other countries. Given standard uptake, this is fairly rapid (DeSaro 2000). These types of technologies should become standard for greenfield facilities since any increased installation costs are outweighed by the significance of environmental compliance as well as increased yield. Next steps include continued demonstration of recycling technologies.

117

LBNL and ACEEE

Inert Anodes/Wetted Cathodes (Alum-4)

The United States is the largest primary aluminum producer globally, with a production of 3.8 million tons (3.4 Mt) in 1999 from 23 primary plants (USGS 2000, OIT 1999). In 1994, primary aluminum production in the U.S. consumed 53,552 million kWh of electricity, or 2 percent of primary energy use in U.S. manufacturing. Average energy consumption in the US for aluminum production in 1994 was 16 kWh/kg and was estimated at 15 kWh/kg in 1997 (EIA 1997, USGS 2000, Energetics 1997). Energy consumption represents 20-30 percent of total production costs. Primary aluminum production is an energy intensive continuous process and involves the electrolytic reduction of alumina into aluminum. This process, known as the Hall-Heroult smelting process, is accomplished in a series of cells or pots that are connected in long lines in buildings. In each cell the refractory material is overlaid with a carbon lining and a carbon cathode. The molten aluminum lies beneath a bath of cryolite that serves as the medium to dissolve incoming alumina and to conduct electricity from the anode to the aluminum. Alumina is fed into the top of each cell on top of a crust of cooler cryolite (which acts as an insulator), and the crust is periodically broken to allow the alumina to be stirred into the bath. When a strong electric current (50-225 kilo-Amperes) is passed through the electrodes into the bath, the alumina is reduced via a reaction to Al2F6 (which is easier to reduce then Al2O3), to produce molten aluminum. The voltage drop across the cell is 4-4.5V, while the voltage in a potline can exceed 1,000 volts (Gitlitz 1995). The molten aluminum is periodically siphoned off through the tops of cell into a holding furnace and then poured into ingots or billets, which are then cast and shaped. Most US cells use the prebaked carbon anode technology in which multiple anodes are baked prior to consumption in the pots. As the anodes are consumed and periodically replaced every 14 to 20 days of operation, they produce carbon dioxide and carbon monoxide (about pound of carbon per pound of aluminum produced) (EPA 1995). Inert anode/wetted cathode technology is the next generation technology capable of significantly improving cell efficiency. The development of this technology is a high priority for the US primary aluminium industry (Margolis 1997). Inert anodes are made from materials that are not consumed during the electrolysis reaction. These new materials (such as cermets of nickel oxide or iron oxide, or copper) allow for a closer anode-cathode distance (thereby reducing electrolysis energy consumption), and eliminate the carbon anode production process as well as emissions of perflourocarbons. Wettable cathodes (or drained cell technology) refer to cell designs that use new cathode materials (such as titanium diboride (TiB2) that are wetted by aluminum. The cathode can be sloped. This design eliminates the use of metal pads found in existing cells thereby reducing magnetically induced turbulence (the turbulence causes power loss and production inefficiency). It also allows for reduced anode-cathode distance better aluminum drainage, and improved cell operation (Welch 1999, Margolis and Eisenhauer 1998, ICF 1998, ASME 1999). The use of bipolar electrode designs (the packing of cells with many electrodes) as opposed to existing monopolar designs not only have the potential to increase cell productivity, but also may be the only configuration that can effectively utilize the inert anodes to reduce anode-cathode distance (ASME 1999, Welch 1999). The combination of these designs point toward the need for a systems approach to anode and cathode designs (ASME 1999). Combined inert anode/wettable cathode technologies are estimated to reduce energy requirements by up to 25 percent or more from current levels, and the U.S. aluminum industry has set a long-term goal of 11 kWh/kg (Margolis and Eisenhauer 1998). As a separate retrofit, wettable cathodes can achieve savings of 2 kWh/kg (ASME 1999). Inert anode technologies are not yet commercially available, and research has been ongoing for several decades. The most recent technical assessment report on this technology indicated that while various pilot scale designs are being tested by companies such as Moltech, Brooks Rand Ltd., Alcoa, and others, to date no fully acceptable inert anode materials have been revealed (ASME 1999, Brumm 2000). However, Van Leeuwen (2000) believes that Alcoas inert anode development is near commercial. The wettable cathode technology appears to be near commercial as well. In the U.S., Reynolds (now Alcoa) and Kaiser (with U.S. DOE support) field-tested TiB2G material cathodes at the Kaiser Mead plant. These

118

Emerging Energy-Efficient Industrial Technologies Inert Anode/Wettable Cathode Data Table

Units inert anodes/wetted cathodes alum-4 Inert anode technology Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Mt

Aluminum Process heating Electricity Retrofit, new 5.5

SIC 3334

Primarily retrofit applications. Also greenfields AEO 2000, smelting output

Hall-heroult cell, primary aluminum smelting tonne MWh MBtu MBtu 1 16.2 0.00 137.9 Unit consumption presented. Smelters cell amperage range from 175-300 kA EIA, 1997, USGS, 2000 EIA, 1997, USGS, 2000

Efficient cell retrofits MWh 11 MBtu 0.00 MBtu 93.5 Not yet commercialized 2005-2015 Years 10 MWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 5.2 0.0 44.4 Medium 15% 3943 0 33.5 1000 Incremental -50 0.03 N/A. 3.36 4.0 25% Significant Somewhat Significant Somewhat High Technical Medium RD&D Good 32% N/A. 32%

Margolis et al., 1998

ASME 1999 Savings compared to 1994 baseline

Assume replacement of existing potlines (ASME, 1999) Incremental costs assumed Credit Suisee, 2000

Discount rate for all CCE calculations is 15%

Reduced costs for anodes; reduced material losses No CO2 emissions or perfluorocarbons Safety

H,M,L

Identifying appropriate materials, design and testing

H,M,L E,G,F,P

EIA 1999 EIA 1999; USGS 2000 Margolis et al. 1998; Van Leeuwen 2000 ASME 1999 Author estimate ASME 1999; Van Leeuwen 2000 Van Leeuwen 2000; ASME 1999 S. Dillich (OIT, U.S. DOE) sara.dillich@ee.doe.gov

119

LBNL and ACEEE materials proved too difficult to maintain required quality in their manufacture (ASME, 1999). However, industrial research laboratories (Alcoa, Pechiney, Comalco, Commonwealth Aluminum, Kaiser) have all been continuing to work on wetted cathode technologies. Comalco (New Zealand) has built 25 TiB2 wettable cathode commercial demonstration cells over the past decade (with government support) (ASME 1999). The Comalco technology combines TiB2 with pitch to layer the cell bottom, while Reynolds and Kaiser are pursuing an approach to develop TiB2 metal tiles for the cell bottom. The tile design might extend the wear life of the cell. The U.S. DOE is also supporting research with Northwest Aluminum Technology, Advanced Refractory Technologies, Material Modification Inc., Electrochemical Technology Corp., Brooks Rand Ltd., and Pacific Northwest Laboratory on wettable cathode development (DOE 2000c). Since neither the inert anode nor the wetted cathode technology is commercially available, it is difficult to estimate the investment costs for the technology. Van Leeuwen (2000) asserts that inert anodes can be used in existing cells with minimal refit costs $5-18/short ton ($6-22/metric tonne) while wettable cathodes would have no refit costs. The use of a combined advanced anode/cathode cell in a bi-polar combination would probably be more appropriate for a greenfield facility. Current costs for total aluminum plants are $3600-4000/short ton annual capacity ($4,000-4,500/t) capacity, (Brumm 2000, Margolis and Eisenhauer 1998). The reduction of the anode baking facility would reduce costs by $900/short ton ($1,000/t). While advanced cell manufacturing would clearly be more expensive, the reduction in the cost of the anode production facility would offset some of the increase. ASME 1999 estimates a wide range of capital investment cost of $3200-$6800/short ton ($3,500-7,500/t). We conservatively assume in our analysis an incremental cost of $900/ton ($1,000/t) capacity. In addition to energy savings, advanced cell technologies could have significant environmental benefits since they would eliminate the emissions of carbon dioxide and per-fluorocarbons, which are greenhouse gases (GHGs) with very high global warming potentials (Margolis and Eisenhauer 1998, ASME 1999). These environmental benefits could become significant in light of increasing concern about global warming and reduction of GHGs. However, just as important are several productivity benefits drive the development of this technology. Van Leeuwen (2000) estimates reduced costs in potroom labor and up to 20 percent increases in productivity, as well as improved health and safety. Also, product quality is likely to improve since less material is contaminated by anode dissolution into the metal (Margolis and Eisenhauer 1998). The barriers to the development of advanced cells incorporating inert anodes and wettable cathodes are primarily technical and economic. Appropriate anode materials have yet to emerge, and additional modeling and systems based research approaches are going to be needed to achieve commercialization. The American Society of Mechanical Engineers rated the likelihood of success of an inert anode breakthrough as low (ASME 1999). However, a recent evaluation of the potential of both the inert anode and wettable cathode technology from Credit Suisse/First Boston rated the likelihood of success of both technologies as high, with commercialization of the anode and cathode technology separately within five years (Van Leeuwen 2000). Even if inert anode and wettable cathode technologies emerge as retrofit options for existing potlines within the next 5 years, a combined advanced cell technology, more likely suited for greenfield facilities is a more distant reality and could limit rapid expansion into the domestic market. Additional research and demonstration are needed to move the market forward in for this technology.

120

Emerging Energy-Efficient Industrial Technologies

Continuous Melt Silicon Crystal Growth (Electron-1)

Semiconductor devices are primarily fabricated from monocrystalline silicon, which is produced from polycrystalline silicon. The most common process used to produce single crystals from molten silicon is the Czochkralski (CZ) method. In the CZ process, crushed polycrystalline silicon is doped with arsenic, boron, phosphorus, or antimony and melted at high temperatures in a quartz crucible. A pull rod with a small silicon seed at the end is lowered into the molten liquid and rotated in a clockwise direction. When the rod is slowly pulled from the melt, a surface tension between the seed and the molten silicon is created, thereby causing a small amount of the liquid to rise with the seed. This liquid cools because of the lower temperature above the melt and forms a single crystal silicon ingot that has the same structural orientation as the seed. The crucible is rotated in a counterclockwise direction to create an eddy current that carries contaminants away from the crystal. The crucible and other components are surrounded by a containment structure that is filled with argon gas. The purpose of the gas is to carry away oxygen, a contaminant, before it reaches the crystal at the melt surface. The ingot diameter is determined by the temperature of the melt pool and the speed at which the rod is extracted. Most ingots are produced in 150mm and 200mm. The length of the ingot is determined by the amount of molten silicon in the crucible. Siemens Solar Industries has developed a process to improve the production of silicon ingot. The project is expected to reduce energy consumption by 40 percent, reduce cycle times by 15 percent, and improve silicon quality. The key changes to the process include additional insulation in the walls of the crucible and at the top of the molten hot zone, the addition of a conical shield above the crucible, and the addition of a continuous recharge system. The additional insulation reduces heat transfer from the melt surface, improves control over the temperature gradients at the melt surface, and allows the rod to be pulled more rapidly. All silicon produced for both semiconductor and solar photovoltaic end-uses are produced in batch. A continuous recharge system would allow the introduction of material during the run and would permit the growing of longer silicon ingots. The challenges to continuous melt growth include maintaining the growing environment, maintaining acceptable temperatures and temperature gradients when introducing materials, insuring the uniformity of the melted material, avoiding distruance of the melt surface, and avoiding contaminating the silicon being drawn from the melt. The market for monocrystalline silicon for semiconductors and photovoltaics is projected to grow at about 11 percent per year (EIA 1999). There are currently seven major manufacturers in the US who produce semiconductor grade silicon. They are Wacker, SHE, Komatsu, Mitsubishi Silicon America, MEMC, Sumitomo-SiTiX, and Motorola. The market for semiconductor silicon was about 12,100 short tons in 1995 and is estimated to reach over 60,000 short tons by 2015. The market for solar photovoltaic silicon represents about 5 percent of the total market for silicon. The solar industry has been dominated by single crystal silicon, of which Siemens is a major player in the market. The other companies involved in the solar market include Solarex, BP Solar, Evergreen Solar, ASE Americas, Photowatt, Sharp, and Kayocera. There has been a continued demand for solar products that is expected to continue. Even though the current market for monocrystalline silicon is primarily for semiconductor manufacture, the continuous melt technology may bring down the costs of photovoltaics to a more competitive level and create an even larger demand for the material. This technology has a high likelihood of success.

121

LBNL and ACEEE

Continuous Melt Silicon Growth Data Table

Units Notes Continuous Melt Silicon Crystal growth Electron-1 Replace batch crystal growth - Czochkralski (CZ) method Market Information: Semiconductor Industries SIC 3674 Process heating, other End-use(s) Electricity Energy types New, replace on failure, retrofit Market segment 2015 basecase Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources tons 60309.00 Gross output increases 598% between 1995 (12,100 tons) and 2015

Czochkralski (CZ) method 1.00 tons 54540 kWh 0 MBtu 465.2 MBtu Continuous melt silicon crystal growth 27273 kWh 0 MBtu 232.6 MBtu Pilot plant 2003 7 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 27267 0.00 232.6 Medium 40% 657.8 0 5.61 629076 Incremental 125815 10.16 1191 -5.04 #DIV/0! Significant Somewhat Somewhat H,M,L High Dependent on markets High Research, scale-up Excellent 0.50 0.00 0.50

Reed, et al. 1999 OIT 1998

Personal communication with Greg Mihalik, 2000

Personal communication with Greg Mihalik, 2000 Personal communication with Greg Mihalik, 2000

Cost of modifications to the grower plus cost of recharge system Estimate 20 percent of incremental cost

Pot scrap has been reduced from about 8.8 to 2.2 lbs per run Reduced scrap

H,M,L E,G,F,P

Testing to create a truly continuous process

Reed, et al. 1999 EIA 1999 Personal communication with Greg Mikalik, 2000 Personal communication with Greg Mikalik, 2000 Reed, et al. 1999 Personal communication with Greg Mikalik, 2000

122

Emerging Energy-Efficient Industrial Technologies

Advanced ASD Designs (Motorsystems-1)

Motors consume over 60 percent of industrial electricity in the United States (Xenergy 1998). Adjustable speed drives (ASD) have revolutionized motor systems by allowing for affordable, reliable speed control using rugged conventional induction motors. ASDs work by varying the frequency of the electricity supplied to the motor, thus changing the motors speed relative to its normal supply frequency, which in the U.S. is 60 Hz. This trick is accomplished by rectifying supplied alternating current to direct current and then synthesizing an alternating current at another frequency. The current method of synthesization is accomplished using an inverter, which is a solid-state device in modern ASDs. Ideally, the waveform of this synthesized current should look like a smooth sine wave. Unfortunately, the three major kinds of inverters in use: voltage-source (VSI), plus-width modulation (PWM) and current-source (CSI), with PWM being the most common used in integral horsepower drives. All create an approximation of a sine wave, though with some distortion. This distortion creates losses in the motor due to heating of the conductors and vibration, which have the effect of shortening the life of the motor. Special inverter duty motors are made which use a higher rating of insulation that extends motor life. The ideal solution would however be to design an inverter that produced a smoother wave pattern (Nadel et al. 2000). A number of researchers are actively working on the development of different inverter topologies (Peng 2000, von Jouanne 2000). Most of these topologies fall into the category of soft-switching inverters, which significantly reduce the voltage spikes that characterize PWM inverters. Reductions in these spikes can dramatically increase the life of the attached motor (Kueck 2000). One example of this technology is the snubber inverter developed at Oak Ridge National Laboratory. ASDs using this technology have an efficiency of about 98 percent compared to a PWM drive at 96 percent efficiency, for drives operating in the 10-20 kHz range. These soft-switching inverters enable the design of faster switching devices, which can further improve the waveform of the output (Peng 2000). Several manufacturers, including Rockwell Automation and Allen Bradley, have begun to offer softswitched inverters as premium products for use in sensitive applications such as medical devices. While these advanced inverters require more complex control strategies than do PWN inverters, they allow the substitution of semiconductor devices for electronic components such as filters. In addition, the improved inverter efficiency will make thermal management in the drives easier, reducing the mass of heat sink required and allowing for more compact packaging of the drive. These tradeoffs are likely to reduce the cost to about the same level as PWM drives. In the long run, soft-switching inverters could displace PWM inverters in most applications if the costs can be brought down (Peng 2000). These drives face a number of barriers. The most significant appears to be the cost of these drives due in large part the manufacturers investment in existing technology. Another issue is that of intellectual property. While manufacturers have expressed interest in licensing the ORNL technology, they were unable to come to terms with the Lab. They have subsequently developed their own soft-switching technology (Peng 2000). While it is likely that this advanced drive technology is likely to eventually succeed in the market, continued research is needed to further the development of these devices and reduce their cost.

123

LBNL and ACEEE

Advanced ASD Designs Data Table

Units Notes Advanced ASD Designs Motorsys-1 Replace existing ASD technologies with advanced designs Market Information: Industries Cross cutting End-use(s) Motors and drives Energy types Electricity Market segment New, replacement, OEM Direct replacement for conventional ASDs Motor systems consume approximately 60% of industrial electricity 2015 basecase GWh 7,825,322 Reference technology Description Conventional PWM inverter with 100 hp inverter-duty induction motor at 95.4% efficiency Throughput or annual operating hours hours Assumes 7 day per week/16 hour per day 6000 Electricity use kWh Based on 96% inverter efficiency and 60% of full load 88 Fuel use MBtu NA Primary Energy use MBtu 0.749 New Measure Information: Description Soft-switching inverter with 100 hp inverter-duty induction motor at 95.4% efficiency Electricity use kWh Based on 98% inverter efficiency and 60% of full load 86 Fuel use MBtu NA Primary Energy use MBtu 0.734 Current status Limited production, special order for premium applications Commercial Date of commercialization 1998 Estimated average measure lifetime Years 15 Savings Information: Electricity savings kWh/% 2 2% Fuel savings MBtu/% NA Primary energy savings MBtu/% 0.015 2% 45% penetration - Because of advantages, likely to displace some current ASD Penetration rate Medium technologies Feasible applications % Feasible in all application 45% Other key assumptions for savings Electricity savings potential in 2015 GWh 71,865 Fuel savings potential in 2015 TBtu NA Primary energy savings potential in 2015 TBtu 24.5 Cost Effectiveness Investment cost $ Assumes a 10% price premium with PWM cost at $80/hp 800 Type of cost Unlikely to be retrofitted incremental Change in annual costs (O&M/other benefits) $ Prorated replacement cost reflecting motor life extension from 5 to 7 years (301.367) Cost of conserved energy (electricity) $/kWh <0 Cost of conserved energy (fuel) $/MBtu NA Cost of conserved energy (primary energy) $/MBtu Discount rate for all CCE calculations is 15% <0 Simple payback period Years 1.11 Internal rate of return % 90% Key non energy factors Productivity benefits Improved motor reliability due to reduced stress as discussed in text Significant Product quality benefits Improved process control Somewhat Environmental benefits None Other benefits Current promotional activity H,M,L Focus is largely on R&D, with some specialty deployment by manufacturers Medium Evaluation Major market barriers Cost, more sophisticated design, intellectual property issues Likelihood of success H,M,L High Recommended next steps Continued R&D Data quality assessment E,G,F,P For all factors except for cost, which is a preliminary assumption. Good Sources: 2015 basecase Nadel et al. 2000, Xenergy 1998 Basecase energy use Nadel et al. 2000, Peng 2000 New Measure energy savings Peng 2000 Lifetime Peng 2000 Feasible applications Nadel et al. 2000, Peng 2000 Costs Nadel et al. 2000, Peng 2000 Key non energy factors Nadel et al. 2000, Peng 2000 Principal contacts Fong Peng, ORNL pengfz@ornl.gov Additional notes and sources

124

Emerging Energy-Efficient Industrial Technologies

Advanced Compressor Controls (Motorsystems-2)

About 9 percent of industrial electricity is used to produce compressed air (Xenergy 1998). Controls match the air supply from the compressors with system demand, regulating the pressure between two levels called the control range. They are one of the most important factors in determining the overall energy efficiency of a compressed air system. Most compressed air systems typically consist of several compressors delivering air to a common header. The objective is to shut off or delay starting a compressor until it is needed. To this end, the controls try to operate all units at full-load, except the one used for trimming (adjusting compressed air supply based on the fluctuations in compressed air demand). In the past, control technologies were slow and imprecise. This resulted in wide control ranges and higher compressor set points than needed to maintain the system pressure above a minimum level. Most systems were controlled using an approach known as cascading set points. The set points for each individual compressor would either add or subtract the compressor capacity to follow the system load. This approach led to wide swings in system pressure, as shown in figure below (DOE 1998).
105.0

100.0

Cascading Set Point Controls

95.0

90.0

85.0

80.0 Multi-master controls 75.0

Modern microprocessor-based technologies allow for much tighter control ranges as well as lower systempressure-control points. The largest benefits of these controls can be obtained in multi-compressor systems, which are much more complex and sophisticated. Controls for single compressors can be relatively simple. System controls coordinate the operation of multiple individual compressors when meeting the system requirements.

Two general kinds of system controllers exist: single-master (sequencing) controls and multi-master (network) controls. Impacts of controls on system pressure (DOE 1998). Multi-master controls are the latest technology in compressed air system control, coming in systems capable of handling four or more compressors (Perry 2000). They provide both individual compressor control and system regulation by means of a network of individual controllers. The controllers share information, allowing the system to respond more quickly and accurately to demand changes. One controller acts as the lead, regulating the whole operation. This strategy allows each compressor to function at a level that produces the most efficient overall operation. The result is a highly controlled system pressure that can be reduced close to the minimum level required (DOE 1998).
time

These controls match system demand with compressors operated at or near their maximum efficiency points, and allow the system pressure to be set lower (Figure 1). Every 2-psi of pressure difference produces about a 1 percent change in energy consumption, so for this example, the system pressure can be reduced 15 psi, thus yielding about a 7.5 percent energy reduction. Although costing $1000 per compressor more than other controls, these controls represent the most energy-efficient system available (Perry 2000). In addition, to energy savings, the application of controls can eliminate the need for some existing compressors, allowing extra compressors to be sold or kept for backup. Alternatively, capacity can be expanded without the purchase of additional compressors. The reduced operating pressure will reduce system maintenance requirements. Also, a more constant pressure level can enhance production quality control by providing more precise operation of pneumatic equipment (DOE 1998). In spite of the attractive return, there are two principal barriers to the use of this technology: higher first cost, and lack of appreciation of the importance of compressed air system efficiency. Educational efforts,

system pressure (psig)

125

LBNL and ACEEE such as the Compressed Air Challenge (CAC 2000), are key to the expanded deployment of these technologies. Advanced Compressor Controls Data Table
Units Notes Advanced compressor controls Motorsys-2 Use of microprocessor-based air compressor controls in place of conventional cascading setpoint controls Market Information: Industries Cross-cutting End-use(s) Motors and drives Energy types Electricity Market segment Retrofit, new Compressed air is approximately 9% of industrial electricity 2015 basecase GWh 1,173,798 Reference technology Description Multiple screw compressor system using cascading setpoint controls Throughput or annual operating hours cfm/hr Assumes 6,000 annual operating @ 30% of rated capacity 100 Electricity use kWh Based on average 22hp/cfm (DOE 1998) 16.4 Fuel use MBtu NA Primary Energy use MBtu 0.140 New Measure Information: Description Multiple screw compressor system using microprocessor-based, multi-master controls System control pressure reduced from 90 to 82.5 psi w/ 1% savings per Electricity use kWh 15.8 2 psi Fuel use MBtu NA Primary Energy use MBtu 0.135 Current status Commercial Date of commercialization 1992 Estimated average measure lifetime Years 15 Savings Information: Electricity savings kWh/% 0.6 3.5% Fuel savings MBtu/% NA Primary energy savings MBtu/% 0.005 3.5% Penetration rate 45% penetration in 2010 Medium Feasible applications % 23% Assumes half of energy use in large, multi-compressor systems Other key assumptions for savings Reducing system pressure will result in additional savings from reduced leak volumes discussed in motorsys-3 Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources H,M,L GWh TBtu TBtu $ 9,244 NA 3.2 150 Full 0 0.0002 NA 20.9 0.04 5824% Significant Somewhat None Significant Low Prorated cost per 100 cfm assumes controls ~$1000 per compressor for 4 or more compressors (Perry 2000)

$ $/kWh $/MBtu $/MBtu Years %

Discount rate for all CCE calculations is 15%

Improved pressure control increases available capacity and improves equipment operation Precise pressure control may improve equipment performance May avoid need for addition compressor purchase or allow retirement of existing compressor with resulting reduced O&M and salvage value Focus has been on compressed air system optimization

H,M,L

High first-cost; lack of appreciation of compressed air system savings. Medium Education programs E,G,F,P Good EIA 1999; Xenergy 1998; Nadel, et al, 2000 Nadel, et al, 2000 Nadel, et al, 2000 Nadel, et al, 2000 Nadel, et al, 2000; Xenergy 1998 Perry 2000 Nadel, et al, 2000 Neal Elliott, ACEEE rnelliott@aceee.org

126

Emerging Energy-Efficient Industrial Technologies

Compressed Air System Management (Motorsystems-3)

Compressed air (CA) systems consume about 9 percent of industrial electricity (Xenergy 1998). They are made up of an assemblage of components including the motor and drive, the air compressor itself, controls, air treatment equipment, piping, and, often, storage. Typical compressed air system wire-to-air efficiencies are around 10 percent (DOE 1998). Achieving peak compressed air system performance requires addressing the performance of individual components, analyzing the supply and demand sides of the system, and assessing the interaction between the components and the system. This systems approach moves the focus away from components to total system performance. System opportunities have been shown to be the area of greatest efficiency opportunity. At the system level, savings opportunities can be grouped into three general categories: leaks, inappropriate uses of CA, and system pressure level. The goal of a management plan is to minimize all three. Leaks can be a significant source of wasted energy, often accounting for 20-30 percent of compressor output. They can also contribute to other production problems. A drop in system pressure can adversely affect equipment performance and efficiency, and the increased compressor runtime needed to satisfy the leak will lead to increased equipment maintenance and unscheduled downtime. Leak detection and repair is a critical element of a compressed air system maintenance program. Typically, the worst leaks are in remote areas of the plant, such as abandoned equipment and roofs. Some leaks are inevitable, but a wellmaintained system can keep them under 10 percent. Unfortunately, even when leaks are identified and repaired, the job is not over. New leaks will develop over time. The best strategy to avoid further problems is to set up a prevention program that monitors the system for new leaks and fixes them as they develop (DOE 1998). Many leaks are intentional, because compressed air is clean and usually readily available. Many people choose it for applications without comparing it to more economical energy sources. Alternatives are available and should be considered for many CA applications. Examples include circuit box and personal cooling that could be done with a blower, or an open pipe that is used to remove dust from a product, when a mechanical brush would accomplish the same function with lower noise (DOE 1998). A systems pressure level should be set at the lowest pressure that meets all requirements of the facility. Lowering the compressed air header pressure by 10 psi reduces the air leak losses by approximately 5 percent and improves centrifugal compressor capacity by 2-5 percent. One element of this may be the application of controls, as is addressed in technology motorsys-2. Reducing system pressure also decreases stress on system components, lessening the likelihood of future leaks (DOE 1998). The optimization process is an approach that addresses these opportunities in a systematic way. Because of the experience required, a CA expert usually offers this analysis as a service. These experts have found that after they implement the measures identified in a thorough review of the system, either one or more compressors can be shut down or a compressor can be downsized, with energy savings frequently exceeding 40 percent. While a survey by an expert can be an important step in establishing the plan, it is necessary to implement an ongoing maintenance program by plant staff, which requires both awareness and technical training (DOE 1998). Reductions in wasted air due to inadequate maintenance, leaks, and inappropriate uses can save 20-30 percent of CA energy. Although costs will likely vary from near zero to more than 5 cents/kWh, depending on the measure and the facility, Suozzo and Nadel (1998) estimate an average cost of saved energy of 1.5 cents/kWh. Most of the barriers to improved compressed result from lack of awareness of the opportunity. The staff reductions that have become common in United States industry and a hesitation to use pay for outside consultants compounds this problem. The Compressed Air Challenge (CAC) has developed a CA management training program that is available for plant staff, and the Compressed Air and Gas Institute (CAGI) has developed a CA experts train 127

LBNL and ACEEE

Compressed Air System Management Data Table

Units Notes Compressed air system management Motorsys-3 Implement a management plan to minimize system energy requirements and reduce leaks and inappropriate uses Market Information: Cross cutting Industries End-use(s) Motors and drives Electricity Energy types Market segment Retrofit 2015 basecase GWh 1,173,798 Compressed air is approximately 9% of industrial electricity Reference technology Description Assumes CA system using screw compressors Throughput or annual operating hours cfm/hr 100 Electricity use kWh Based on average 22hp/cfm (DOE 1998) 1640 Fuel use MBtu NA 13.989 Primary Energy use MBtu New Measure Information: Description Survey eliminates 25% of CA demand by identification of leaks and inappropriate uses Electricity use kWh 1230 Fuel use MBtu NA 10.492 Primary Energy use MBtu Current status Commercial Date of commercialization pre 1980 Estimated average measure lifetime Years 1.5 Measure requires regular reviews and resurveying as discussed in text Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources H,M,L

kWh/% MBtu/% MBtu/% % GWh TBtu TBtu $ $ $/kWh $/MBtu $/MBtu Years %

410 NA 3.497 medium 23% 66,026 NA 563 0 Full 6.15 0.015 NA 1.8 0.38 undefined Significant Somewhat None Significant Medium

25% 25% 45% penetration in 2010 Feasible in 50% of compressed air system capacity

No equipment purchases usually required Fee and staff time cost ~$0.015 per kWh saved (Nadel et al 2000)

Discount rate for all CCE calculations is 15% Based on increase in O&M cost Since there is no capital cost, IRR undefined Improve system operation and increases pressure stability More precise pressure control may allow for improved equipment performance May avoid need for addition compressor purchase or allow retirement of existing compressor with resulting reduced O&M and salvage value CAC is providing training and limited trade marketing

H,M,L

Customer awareness and availability of trained CA system experts Medium Significant problems have been encountered with conveying message Continued education and awareness building Good EIA 1999; Xenergy 1998; Nadel, et al, 2000 DOE 1998 DOE 1998 DOE 1998 Nadel, et al, 2000; Xenergy 1998, DOE 1998 Nadel, et al, 2000 DOE 1998 Bill Scales 516-248-9096

E,G,F,P

128

Emerging Energy-Efficient Industrial Technologies

Motor Diagnostics (Motorsystems-4)

Motors consume about two-thirds of industrial electricity. Once placed in service, a motor will operate for years and receive minimal attention before it fails. At that point the motor is likely to be repaired and placed back in service. Mean time between failures is seven to ten years, and motors are typically repaired three times in their life. During a motors service life, many changes can take place that affect a motors performance. The loads that the motor is servicing are likely to change, frequently resulting in a mismatch between the motor and its new load. The motor itself can also deteriorate mechanically and electrically. These changes can reduce the efficiency and reliability of the motor. The most common problem is bearing wear, which can ultimately lead to failure that usually damages the windings of the motor (Nadel et al. 2000). A number of techniques have been used for many years to assess the performance of motors. These techniques have ranged from monitoring the temperature of bearings, monitoring vibration, and measuring the voltage and currents for the different phases, to extensive test bench evaluations for performance and efficiency. These tests can detect changes in motors that indicate that it should be resized for a changing load, repaired or replaced before it fails. However, in the past these test procedures have been labor intensive and expensive, often requiring that the motor be removed from service. As a result, these tests are infrequently used, and the motor is left in service until failure (Nadel et al. 2000). Over the past decade, a number of new diagnostic devices have been introduced that make in-service testing much easier. These tests make use of advanced sensors and on-board computing to measure temperature, voltage, current, harmonics and flux density. These data allow for various analyses such as current signature that can assess performance and efficiency and detect problems before they lead to an inservice motor failure, allowing them to be repaired during normal service cycles (Nadel et al 2000). While there may be some secondary energy savings, it is unclear that this family of technologies offers any direct energy savings. The primary benefit is reduced downtime (Boteler 2000). Some manufacturers have begun to offer built in diagnostics as an option on new motors. These motors are positioned for mission critical applications and are intended to be integrated into a plant-wide monitoring and control network. The network uses the motor data to provide conditioned-based monitoring of critical components. Conditioned-based is the successor to predictive maintenance, allowing for improved reliability, availability and continuous system optimization though monitoring of system conditions. Integrated retrofit modules are also available for some newer motors. Current cost per motor for these advanced motor diagnostic packages is about $1000, though with volume it is projected that the cost could fall to the $750 range (Boteler 2000). Conditioned-based monitoring of motors offers a number of significant non-energy benefits. By identifying motors prior to failure, additional damage resulting from the failure can be avoided, thus reducing repair costs and avoiding potential permanent damage to the motors (Nadel et al 2000). By preventing most in-service failures, system availability is significantly increased, thus increasing annual throughput. This additional production capability can avoid the need to make capital investments to expand production (Boteler 2000). The major barriers to the adoption of motor diagnostics are the first cost of the equipment and the need to implement management practices necessary to realize the benefits. Case studies and education of end-users on the benefits are the most important actions to encourage more rapid adoption of the technology. Several programs, such as those offered by Sacramento Municipal Utility District and the Northwest Energy Efficiency Alliance have already begun to development programs to build customer awareness of this technology (Nadel, et al., 2000).

129

LBNL and ACEEE

Motor Diagnostics Data Table

Units Notes Motor diagnostics Motorsys-4 Use of external or internal sensor and monitoring system to assess the operational status of motors Market Information: Industries Cross cutting End-use(s) motors and drives Energy types electricity new, OEM, replacement Market segment 2015 basecase GWh 7,825,322 Motor systems are approximately 60% of industrial electricity Reference technology Description EPAct TEFC 100 HP induction motor Throughput or annual operating hours hrs 6000 Assumes 7 day per week/16 hour per day Electricity use kWh 89.4 Assumes average 60% load, 94.1% efficiency NA Fuel use MBtu 0.762 Primary Energy use MBtu New Measure Information: Description Electricity use kWh 89.4 Fuel use MBtu NA 0.762 Primary Energy use MBtu Current status Commercial 1999 Date of commercialization Years Estimated average measure lifetime 15 Savings Information: Electricity savings kWh/% 0.0 0% Fuel savings MBtu/% NA NA Primary energy savings MBtu/% 0.000 0% Penetration rate 45% likely penetration medium Feasible applications % 11% Justified in 25% of motors Other key assumptions for savings Electricity savings potential in 2015 GWh 0 TBtu NA Fuel savings potential in 2015 0 Primary energy savings potential in 2015 TBtu Cost Effectiveness Investment cost $ 750 Cost per motor, independent of size, in 2015 Type of cost Incremental Change in annual costs (O&M/other benefits) $ -1500 Wide range of benefits based on specific application Cost of conserved energy (electricity) $/kWh NA $/MBtu Cost of conserved energy (fuel) NA Cost of conserved energy (primary energy) $/MBtu NA Discount rate for all CCE calculations is 15% Simple payback period Years immediate Internal rate of return % 200% Key non energy factors Compelling Productivity benefits Can increase uptime which results in increased annual production Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources H,M,L Somewhat None Somewhat Low Increased uptime reduce product fluctuations No direct benefits, though indirect benefits may be significant May be able to avoid plant capital expansions due to increased production Vendors just beginning to promote and some regional programs offering

First cost and lack of management infrastructure necessary to realize benefits Non-energy benefits are compelling H,M,L High Demonstration and education E,G,F,P Good EIA 1999, Nadel, et al 2000; Xenergy, 1998 Nadel, et al 2000 Boteler 2000 Boteler 2000 Elliott 2000 Boteler 2000 Boteler 2000, Nadel et al 2000 Rob Boteler, USEM 314-553-1179

130

Emerging Energy-Efficient Industrial Technologies

Motor System Optimization (Motorsystems-5)

Motor systems are made up of a range of components centered around a motor-driven device such as a pump or fan. Systems performance optimization focuses on optimizing the flows in motor-driven systems, principally fan and pump systems, to meet end-use requirements. The opportunity derives from the physical fact that the power consumption of fans and pumps varies as the cube of the speed, while output varies linearly. As a result, small changes in motor speed can yield large energy savings, so it is important to closely match output to end-use requirements. This concept is referred to as the systems approach. Because accomplishing this goal requires specialized analytical and design skills, the concept has been deployed as a service. The Performance Optimization Service (POS) is built on Canadian utilities Performance Optimization program, which focused on identifying applications for adjustable speed drives (ASDs). ASDs represent one means of matching motor speed when the end-use requirements vary. Field experience has shown that most loads do not vary significantly, and speed can be varied by changing fan pulleys or trimming pump impellers (Nadel et al. 2000). The Energy Center of Wisconsin (ECW) ran a POS program from 1994 to 1998. In the ECW POS program, customer service representatives identified candidate projects. A POS engineer then offered the customer a quick, free engineering walk through analysis of their systems. If substantial savings were projected, a feasibility-study proposal was prepared to determine what needed to be done to improve efficiency and performance, and how much it would save the customer. Once a proposal was accepted, a POS engineer collected system-load and operating data, and prepared a feasibility study report, which recommended a design strategy and details on the technical and economic impacts of the project (Nadel et al. 2000). In another program, DOE, EASA, PG&E, California Energy Commission (CEC) and local motor distributors deployed a POS program targeted at water and wastewater pumping, based on several previous demonstration projects. They organized O&M pumping workshops for Northern California American Water Works Association (AWWA) members. The workshops focused on how to choose motors and pumps, maintenance and operation practices, and motor and pump repair (Oliver 1999). The POS programs have evolved over time, and have come to focus on providing information and convenience for the customer. POS now gives customers a comprehensive proposal right after the initial walkthrough, outlining what needs to be done and what it will cost and save. This immediate feedback keeps up momentum and motivation. POS also provides technical expertise to customers throughout the process, which is a key factor in building customer confidence in the program. POS success is attributed to identifying opportunities throughout the process, which is facilitated by training customers. Credibility is enhanced by the objectivity of the service provider (Wroblewski 1996). Based on the ECW program experience, typical energy savings from fan, pump, or blower-system upgrades are estimated at 20 to 50 percent for systems identified as good candidates for POS. To be conservative, we have chosen the bottom of the range for our calculations. Based on known and estimated costs and energy savings for sites that are proceeding toward implementation, the average payback is 1.2 years. These estimates do not account for productivity gains known to exist at many of the sites (Hanson 1997). Experience with POS has found that it is difficult to promote for a number of reasons. First, the concept is complex and difficult to explain even to a technical audience. This barrier is more difficult with smaller companies, where the customer is less technically-versed than in larger industries. Engineering fees account for most of a projects cost, and customers have shown a reluctance to approve these expenditures. Since limited customer demand has been evidenced for these services, the engineering design community has been reluctant to develop the required specialized skills (Nadel et al 2000). While POS offers significant energy and cost savings and performance enhancements, customer awareness and demand for the service must be developed. Possible strategies include case studies, training and seminars, and partnering with the design engineering community to help them market this service to candidate customers.

131

LBNL and ACEEE

Motor System Optimization Data Table

Units Motor system optimization motorsys-5 Survey of operating conditions for an existing Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

pump or fan system, with the goal of matching output to process demand Cross cutting Motors and drives Electricity Retrofit 7,825,322

GWh

Motor systems are approximately 60% of industrial electricity

Existing 20 HP exhaust fan system with damper controls hrs 6000 kWh 5.8 MBtu NA 0.050 MBtu Reduce motor speed to match flow requirements by changing belts and pulleys kWh 4.7 MBtu NA 0.040 MBtu Commercial Application of standard engineering practice, formalized in early ~1980 1980s Years Systems tend to fall out of optimization due to facility changes 10 kWh/% MBtu/% MBtu/% % GWh TBtu TBtu $ 1.2 NA 0.010 Medium 11% 176,070 NA 1,502 410 Full 0 0.0117 NA 1.368 1.50 67% Significant Significant Somewhat Significant Low Cost of replacing belts and pulleys, plus two hours of engineering at $100/hour If fan speed is reducted, bearing life my be extended 20% Savings are not dependent upon existing equipment efficiencies NA 20% 45% penetration in 2010 Feasible in 25% of industrial motor loads

$ $/kWh $/MBtu $/MBtu Years %

Discount rate for all CCE calculations is 15%

Better matching of motor driven equipment to demand can improve process throughput, and may allow free up capacity for expansion, Better matching of motor driven equipment to demand can improve process control Reduced fan speed can reduce environmental noise Reduced fan speed can reduce worker noise exposure

H,M,L

Lack of knowledge, reluctance to pay engineering fees, lack of skilled providers H,M,L Medium Expanded end-user education, development of engineering training E,G,F,P Good EIA 1999, Nadel et al 2000, Xenergy 1998 Nadel et al 2000, Xenergy 1998 Nadel et al 2000, Wroblewski 1996, Martin 1999, Hanson 1997 Nadel et al 2000, Wroblewski 1996, Martin 1999 Nadel et al 2000, Martin 1999 Nadel et al 2000 Nadel et al 2000 Vern Martin, Flow Care 519-740-8733

132

Emerging Energy-Efficient Industrial Technologies

Pump System Efficiency Improvements (Motorsystems-6)

Pumps consume approximately 20 percent of industrial electricity. The selection of a pump for a given application requires the consideration of the flow requirements, required delivered pressure, and the system effects. While most engineers are trained to select pumps to meet requirements as specified in a design, many motor selection decisions are based upon estimates of operating conditions that may not be close to the true operating conditions. Once a system is placed in operation, the conditions may change further, moving the pump into a range of operation that is not only inefficient, but potentially even destructive. These changes result from changes in application, such as increases, or more frequently, decreases in the flow requirements. System resistance can increase as a result of fouling and/or scaling, and the pump impeller can erode, changing its effective system curve. Many of these changes are gradual and so may not be evident (Nadel et al 2000). To bring a pump system back into acceptable operation, an engineer needs to first assess what the process requirements are. This task can be as simple as taking some measurements, or as complex as performing a systems optimization analysis as described in Motorsys-5. Once the pumping requirements are determined, the existing equipment must be assessed. An analysis needs to be performed to determine if the existing pump can meet the current operating characteristics. Among the options available are slowing the pump, trimming or replacing the impeller, and replacing the pump. Frequently, the initial reaction is to slow the pump. This may not be a good choice if the pump is significantly oversized. If the pump is slowed dramatically from its design speed, its system curve will change and may have a very limited range of operation. In many cases it may be much better to select another pump (Nadel et al. 2000, Hovstadius 2000, DOE 1999d). The savings from right-sizing a pump can be dramatic. The 17 percent savings in the example used in the Data Table is reflective of the savings that are achievable (DOE 1999d). The system analysis is perhaps the most difficult and costly portion of a project. However, payback periods of 3 years are typical (Nadel et al 2000, Hovstadius 2000, DOE 1999d). Because large pumps frequently require the largest motors at a facility, downsizing the pump can frequently also achieve significant electricity demand savings, thus reducing demand charges paid by the facility. In addition to the electricity savings, right-sizing pumps can lead to more stable system operation. Pulsation and flow variations that often result from pumps operated outside of their system curve can disrupt processes. Correction of these problems can improve product quality, and in some cases increase the capacity of systems that depend upon the pump. Sometimes the downsizing of a pump can free up space that can offer additional options for process improvements. Frequently, these benefits will be the driving motivation for project implementation (Nadel et al. 2000, Hovstadius 2000). While the engineering associated with pump systems is well understood, many engineers are not experienced in conducting these analyses. Software tools, such as the pump system assessment tool developed by DOE and the Hydraulic Institute (DOE-OIT 2000b), provide a means of addressing this issue. Engineers need to be made aware of this and similar tools, and receive training in its application. Unfortunately a trained and equipped consulting community does not create demand for the service by users. The end-user community must be made aware of the opportunity and must be encouraged to seek out these services. However, there is a delicate balance between creating market demand, and developing the capability to deliver services in the marketplace. Both demand and supply need to grow in parallel.

133

LBNL and ACEEE

Pump Efficiency Improvement Data Table

Units Notes Pump efficiency improvement Motorsys-6 Appropriate selection of pump system components to optimize system operation and minimize system losses Market Information: Industries Cross cutting End-use(s) Motors and drives Energy types Electric Market segment Retrofit 2015 basecase GWh 1,541,589 pump systems are 20% of industrial motor electricity Reference technology 200HP pump rated at 4650gpm at 114 ft of head, but operating at 3612 gpm with 107 ft of head Description Throughput or annual operating hours 4000 Electricity use kWh 128 Fuel use MBtu NA 1.096 Primary Energy use MBtu New Measure Information: Description Replaced with smaller pump optimized to process requirements and new 200hp motor Electricity use kWh 107 Fuel use MBtu NA 0.910 Primary Energy use MBtu Current status Commercial Date of commercialization Application of standard engineering practice, formalized in early 1980s ~1980 Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors principal contacts Additional notes and sources Years kWh/% MBtu/% MBtu/% % GWh TBtu TBtu $ $ $/kWh $/MBtu $/MBtu Years % 10 21.8 NA 0.186 Medium 23% 58,860 NA 502 15,693 Full -1,800 0.010 NA 1.19 3.0 33% Significant Significant None Somewhat Moderate 17% NA 17% 45% penetration in 2010 Feasible in all half of pump systems Systems tend to fall out of optimization due to facility changes

Electric demand savings

Discount rate for all CCE calculations is 15%

More stable system operation More consistent flow, allows for more stable process operation Ability to downsize equipment and free up space DOE has distributed a design program, and has been working with the Hydraulic Institute to deploy an educational program

H,M,L

Lack of knowledge, reluctance to pay engineering fees, lack of skilled providers Medium Expanded end-user education, development of engineering training E,G,F,P Good H,M,L EIA 1999, Nadel et al 2000, Xenergy 1998 DOE 1999d DOE 1999d Nadel et al 2000, Martin 1999 Nadel et al 2000, Martin 1999 DOE 1999d Nadel et al 2000 Vern Martin, Flow Care 519-740-8733

134

Emerging Energy-Efficient Industrial Technologies

Switched Reluctance Drives (Motorsystems-7)

Motors consume about 60 percent of industrial electricity, and a number of types of motors are available to meet specific application needs in industry. Most applications make use of a constant-speed motor, while some applications require some degree of speed control. The most common motor type is the NEMAstandard poly-phase induction motor. For operations that require speed control, these motors are coupled with an adjustable speed drive (ASD). These motor/drive combinations are now reliable and cost-effective for many applications. The switched reluctance motor is an old concept for designing a variable speed motor that has advanced recently with progress in solid-state electronics and software. The switched reluctance (SR) drive itself is a compact, brushless, electronically-commutated AC motor with high efficiency and torque, and simple construction. Available in virtually any size, the SR motor offers the advantage of variable speed capability (very low to very high) and precision control. As for its design, the motor comes as a package integrated with a controller. This setup enables some models to operate at speeds as low as 50-rpm and as high as 100,000-rpm (Howe et al. 1999). The rugged rotor of a SR motor is much simpler than that of other motors, since it has no field coils or embedded magnetic materials. However, the coils and magnets attached to the rotor can be subjected to very high stresses (Albers 1998). Both torque and efficiency are, in general, higher in SR drives (motor and controls) than in induction motors with ASDs. The current generation of SR drives have relatively flat efficiency curves with maximum efficiencies around 93 percent in integral-hp models and the low- to mid-80 percent range in fractional-hp units (Albers 1998). Because of its simplicity, the SR motor in mass production should theoretically cost no more than, and perhaps less than, mass-produced induction motor/ASD packages of comparable size. But at this time, automating the manufacturing of integral horsepower and larger fractional horsepower SR motors is proving difficult and it is uncertain whether the hoped-for price reductions will materialize (Wallace 1998, Albers 1998, Boteler 1999). Currently, an SR motor and its associated controls, starter, and enclosure cost 50 percent more than comparably sized and equipped induction motors with variable speed controls (Wallace 1998, Albers 1998, Means 1997). This amounts to about a $2,000 premium for a 20-hp installation. For this analysis we assume that the price premium will be cut in half, to 25 percent (or $1,000 for a 20-hp motor), once SR motors are more widely adopted. Because of their precise and wide range of speed control and their ruggedness of design, SR motors are attractive in a broad range of commercial and industrial applications. Most SR research and application in the U.S. is in fractional-hp printer, copier, precision motion tasks and appliances. SR motors are now also being used in residential and commercial washing machines. Industrial applications include manufacturing equipment, process fans and pumps, and machine (servo) control (Wallace 1998). In addition, SR motors with control systems are competing to supplant induction motors with variable speed drives in a number of applications. For example, SR motors are most attractive in new and OEM (original equipment manufacturer) installations where the full benefits of their speed control can be realized. In the future, there may be some retrofit applications for both general-purpose applications and as replacements for DC drives in process equipment, but the availability and understanding of how to use these motors has not yet progressed to the point that this is feasible. SR motors could potentially replace 20 to 50 percent of the existing general-purpose motors in service today (Albers 1998, Motor Challenge Clearinghouse 1998). We assume the middle of this range (35 percent) as the level of feasible applications once the technology matures. The primary technical challenge facing SR motor technology is the fact that while the motor is simple conceptually, it is complex to engineer and manufacture (Wallace 1998). Unless the cost premium can be reduced, it will limit SR motors to applications that require the unique features of this motor. Noise has been an issue in some designs. The development and commercialization effort is primarily through manufacturers, OEMs, and EPRI-funded R&D. The motors recent introduction in the Maytag horizontal-axis clothes washer should help speed the SR motors market development (Nadel et al. 2000).

135

LBNL and ACEEE

Switched Reluctance Drives Data Table

Units Notes Switched reluctance motor motorsys-7 Use of a switched reluctance motor with integral control package in place of an induction motor with adjustable speed drive Market Information: Industries Cross cutting End-use(s) Motors and drives Energy types Electricity Market segment Some retrofit applications may occur in the future New, OEM 7,825,322 2015 basecase GWh Motor systems consume approximately 60% of industrial electricity Reference technology Description 20 hp induction motor with adjustable speed drive Throughput or annual operating hours hrs Assumes 7 day per week/16 hour per day 6,000 Electricity use kWh Operating at 60% load with 90% efficiency including 1% penalty for losses 18 in the ASD Fuel use MBtu NA 0.152 Primary Energy use MBtu New Measure Information: Description 20 hp switch reluctance motor with controls Electricity use kWh SR motor at 93% efficiency operating at 60% load 17 Fuel use MBtu NA 0.148 Primary Energy use MBtu Current status Commercial is some sizes Commercial Date of commercialization 1994 Estimated average measure lifetime Years 15 Savings Information: Electricity savings kWh/% 0.6 3% Fuel savings MBtu/% NA Primary energy savings MBtu/% 0.005 3% Penetration rate 25% penetration in 2010 Low Feasible applications % Feasible in 35% of applications 9% Other key assumptions for savings Electricity savings potential in 2015 GWh Assumes 25% of motor applications become eligible, 30% of those use SR 6,626 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors principal contacts Additional notes and sources TBtu TBtu $ $ $/kWh $/MBtu $/MBtu Years % NA 2.3 1000 Incremental 0 0.049 NA 5.8 7.41 12% Somewhat Significant None None Medium

Discount rate for all CCE calculations is 15%

Precise speed control may allow for increased output for equipment More precise speed control may allow for reduced defect rate

H,M,L

R&D, demonstrations

H,M,L

Manufacturing difficulties, price premium, noise Medium continued R&D focusing on cost reductions E,G,F,P Good EIA 1999, Nadel, et al 2000; Xenergy, 1998 Nadel, et al 2000 Nadel, et al 1998, Howe, et al, 1999 Nadel, et al 1998, Howe, et al, 1999 Nadel, et al 1998, Howe, et al, 1999 Nadel, et al 1998, Howe, et al, 1999 Nadel, et al 1998, Howe, et al, 1999 Neal Elliott, ACEEE rnelliott@aceee.org

136

Emerging Energy-Efficient Industrial Technologies

Premium Lubricants (Motorsys-8)

Lubricants are a critical element of every motor-driven system, reducing the friction on equipment and minimizing component wear. Motor driven equipment accounts for about 60 percent of the electricity consumed by industry (Xenergy 1998). Traditionally, petroleum derivatives have been used for most applications. These lubricants trade off lubricating characteristics with increased resistance due to higher viscosity. This selection is complicated by the fact that the lubricant characteristics can vary widely over the range of operating temperatures that a piece of equipment experiences from startup to sustained operation. In addition, the characteristics of these lubricants tend to deteriorate with use, requiring frequent replacement. Replacing conventional petroleum-based oils and greases with synthetic, engineered lubricants can reduce energy consumption and equipment wear while extending lubricant life. Synthetic lubricants, introduced beginning in the 1970s, use modern chemical-manufacturing processes to build organic polymers tailored to the specific lubricating requirements of various applications. Because synthetic lubricants are optimized for their application, friction can be reduced significantly. While friction is a relatively small loss in motors themselves, friction can represent a large loss in mechanical equipment like compressors, pumps, and gear drives. Savings of 2 to 30 percent have been realized in gear reducers, compressors, pumps, and motors while using synthetic lubricants (Howe et al. 1999). While these lubricants cost between 1.5 and 3 times more than conventional products, they retain their lubricating properties longer, allowing the relubrication interval to be extended two to five times. This life extension results in both lubricant cost and maintenance labor savings. Extended service intervals are extremely important, especially in view of reduction in maintenance staff size in many industries. In addition, improved lubrication can reduce equipment wear, further decreasing operating and maintenance costs while improving equipment reliability. In many cases, the additional cost of the synthetics can be more than justified by longer lubricant life alone (Howe et al 1999). Synthetic lubricants have been slow to be adopted because of their higher initial cost and limited awareness by end-users. In addition, because of specificity of the different synthetic lubricants, customers may find correct selection intimidating. Penetration has been higher in industrial than in commercial applications, in part because of the greater proportion of technically-oriented staff in industry. While information on lubrication has been included in many electric utility motor-system programs, it has not been a major focus (Elliott and Pye 1997). Further documentation of the benefits of synthetic lubricants, and guidance on selection and use, may help to overcome the barriers to this product and expand its use. In fact, a new level of product guidance, lubrication services, represents an emerging trend in the lubrication industry. In out-sourced lubrication services, a provider, often affiliated with lubrication distributors, performs normal lubrication services for a facility. This value added service can be combined with other monitoring, such as predictive maintenance. This new service channel is well positioned, as well as motivated, to increase diffusion of this technology into the market (Nadel, et al. 2000).

137

LBNL and ACEEE

Advanced Lubricants Data Table

Units Notes Advanced lubricants Motorsys-8 Replace conventional petroleum lubricants with advanced synthetic lubricants Market Information: Cross cutting Industries End-use(s) Motors and drives Energy types Electricity Market segment New, Retrofit 2015 basecase GWh Motor systems use about 60% of industrial electricity 7,825,322 Reference technology Conventional compressor oil used in a 350 HP air compressor Description 6,000 Throughput or annual operating hours hrs Assumes 7 day per week/16 hour per day 383 Electricity use kWh Hourly consumption at an average 80% load NA Fuel use MBtu 3.27 Primary Energy use MBtu New Measure Information: Substitution of an engineered, synthetic lubricant Description Electricity use kWh Based on case study in E Source 1999 373 Fuel use MBtu NA 3.18 Primary Energy use MBtu Current status commercial Date of commercialization 1978 Estimated average measure lifetime Years 0.5 Savings Information: Electricity savings kWh/% 10 3% Fuel savings MBtu/% NA Primary energy savings MBtu/% 0.1 3% Penetration rate 45% penetration rate in 2010 medium Feasible applications % About half of industrial motors use is eligible for customer 23% lubrication since many smaller motors use sealed bearing which are not user serviceable Other key assumptions for savings Electricity savings potential in 2015 GWh 45,931 Fuel savings potential in 2015 TBtu NA 15.7 Primary energy savings potential in 2015 TBtu Cost Effectiveness Investment cost $ Cost is 1.5-2.5 x of conventional lubricant, but last 3-4 times 300 as long Incremental Type of cost Because lubrication is assumed to be a maintenance measure, it will bear only incremental measure cost Change in annual costs (O&M/other benefits) $ Maintenance savings from extended lubricant life (3,162) Cost of conserved energy (electricity) $/kWh <0 Cost of conserved energy (fuel) $/MBtu NA Cost of conserved energy (primary energy) $/MBtu Discount rate for all CCE calculations is 15% <0 Simple payback period Years 0.05 Internal rate of return % 917% Key non energy factors Productivity benefits Reduced equipment wear and extended service internals Significant with labor and downtime savings. Somewhat Product quality benefits Greater equipment reliability Significant Environmental benefits Reduced volume of spent lubricant to be disposed of Other benefits None Current promotional activity H,M,L Low Evaluation Major market barriers First cost, lack of market knowledge Likelihood of success H,M,L Economics attractive if barriers can be overcome Medium Recommended next steps Market education and encourage development of lubrication service companies Data quality assessment E,G,F,P Good Sources: 2015 basecase Xenergy 1998 Basecase energy use EIA 1999 New Measure energy savings Howe et al. 1999 Lifetime Howe et al. 1999 Feasible applications Nadel, et al 1998 Costs Howe et al. 1999 Key non energy factors Howe et al. 1999 Principal contacts Bill Howe, E-Source 720-548-5000 Additional notes and sources

138

Emerging Energy-Efficient Industrial Technologies

Advanced CHP Turbines (Utilities-1)

Combined heat and power systems generate electricity (and/or mechanical energy) and thermal energy in a single, integrated system. This contrasts with the more common practice where electricity is generated at a central power plant, and on-site heating and cooling equipment is used to meet non-electric energy requirements. Because CHP captures the heat that would otherwise be rejected in traditional separate generation of electric or mechanical energy, the total efficiency of these integrated systems is much greater than from separate systems. CHP is not a specific technology, but rather an application of technologies to meet end-user needs for heating and/or cooling, and mechanical and/or electric power. Steam turbines, gas turbines, combined cycles, and reciprocating engines are the major current technologies used for power generation and CHP (Elliott and Spurr, 1999). Collaborative research by government and industry has contributed significantly to the new generations of turbines and engines. Conventional electricity generation is inherently inefficient, converting only about a third of a fuels potential energy into usable energy. The significant increase in efficiency with CHP results in lower fuel consumption and reduced emissions compared with separate generation of heat and power. CHP is an economically productive approach to reducing air pollutants through pollution prevention, whereas traditional pollution control achieved solely through flue gas treatment provides no profitable output and actually reduces efficiency and useful energy output. The thermodynamic cycle associated with the majority of gas turbine systems is the Brayton cycle. In this cycle, atmospheric air is passed as the working fluid through the turbine. The thermodynamic steps of the Brayton cycle include compression of atmospheric air, introduction and ignition of fuel, and expansion of the heated combustion gases through the gas producing and power turbines. The developed power is used to drive the compressor and the electric generator. Since there are two or more usable energy outputs from a CHP system, defining overall system efficiency is more complex than with simple systems. The system can be viewed as two subsystems the power system and the heat recovery system. The efficiency of the overall system results from an interaction between the individual efficiencies of the power and heat recovery systems. We have considered the example of a 5MW gas-fired CHP turbine operating with a 73 percent overall efficiency and a power-to-heat ratio of 0.607. This system is compared to the example of purchased grid electricity with an on-site gas boiler. The electric grid is estimated to have an efficiency of 33.4 percent and the gas boiler has an efficiency of 68 percent. Installed costs for a 5MW combustion turbine currently cost about $1070/kW and are projected to cost $950/kW by 2020 (Onsite Sycom 2000). Operating and maintenance costs are currently $0.0059/kW and are expected to drop to $0.0049/kW by 2020 (Onsite Sycom 2000). The estimated technical potential for CHP at existing manufacturing facilities is approximately 132,000 MW (Onsite Sycom 2000). Approximately 44,000 MW of CHP capacity is already in place at existing manufacturing facilities, leaving a remaining CHP potential of 88,000 MW. Much of the remaining potential is found in those industries that have traditionally relied on CHP paper, chemicals, food, primary metals, and petroleum refining. Most CHP development to date has focussed on large systems (20 MW or larger) and 55 percent of the remaining CHP potential is in systems of this size. However, small systems represent a largely untapped market for CHP. Thirty two percent of the remaining potential is in system sizes of 4 MW or less (Onsite Sycom 2000).

139

LBNL and ACEEE

Advanced CHP Turbines Data Table

Units Advanced CHP Turbines Utility-1 Replace Grid Electricity combined with Natural Gas Boiler Market Information: Industries End-use(s) Energy types Market segment 2015 basecase MWh Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricty) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality beneifts Environmental benefits Other benefits Current promotional activity Eval uat i on Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors principal contacts Additional notes and sources Notes

Cross Cutting Utilities Natural gas Retrofit 132,000

Estimated total industrial technical potential- Onsite Sycom, 2000

68% efficiency natural gas boiler plus 33.4% efficiency grid electricity 6000 hours Typical annual operating hours for a CHP system kWh 1.000 Obtain 1 kwh from grid MBtu 0.00562 Obtain 0.005 MBtu from boiler based on Power/Heat of 0.607 0.01848 MBtu Energy required to produce the 1kwh of grid elec.and .005 MBtu of boiler steam 5 MWe CHP system operated at 80% load with 73% efficiency ; Power/Heat = 0.607 kWh 0 MBtu Overall efficiency of 73% 0.012 MBtu 0.012 Commercialized 1998 Years A refit must be done after 10 years. Cost of refit is 20% of the initial 10 capital cost kWh/% MBtu/% MBtu/% % kWh Mbtu Mbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 1.000 100% -0.007 -120% 0.01 33% high 60% Average 2015 grid electricity price is $0.039/kWh and natural gas price $3.38/Mbtu 79200 Decreased grid electricity -534.8 Increased fuel use 483.8 Net primary energy savings 1070 full -0.0096 0.01 (1.45) 1.60 6.9 9% Significant Significant Significant H,M,L High Onsite Sycom, 2000 $45/kWyear for CHP, $5/Mbtu for gas boiler

Includes refit cost of $214 at year 11 Fewer shutdowns due to grid outages Greater reliability Higher efficiencies mean better fuel utilization

H,M,L E,G,F,P

High Favorable tax policies Excellent EIA 1999 EIA 2000 Onsite Sycom, 2000 Onsite Sycom, 2000 Onsite Sycom, 2000 Onsite Sycom, 2000

140

Emerging Energy-Efficient Industrial Technologies

Advanced Reciprocating Engines (Utilities-2)

Reciprocating engines can be used to generate a portion of a plants needs onsite, reduce demand during peak periods, or support premium power applications (e.g., microelectronics manufacturing) (Elliott and Spurr 1999). Most industrial facilities have some sort of on-site back-up power requirements that can be met by advanced generation technologies. Reciprocating engines are even more efficient when operated as part of a CHP system, which can meet some of the facilitys thermal requirements as well. However, reciprocating engines will have to compete with microturbines and fuel cells to gain a stronghold in this market. Reciprocating engines (e.g., diesel engines) are used to generate electricity. These internal combustion engines convert fuel to shaft power, which then spins a generator. Diesel generators have long been used to generate small amounts of electricity at industrial, commercial, and institutional sites, either for continuous use or for backup in case of utility power failure. Recent developments in engine design have increased power efficiency (now approaching 50 percent) and reliability, while dramatically reducing the emissions of these engines. These new designs can use a variety of liquid and gas fuels, including natural gas. For emissions reasons, natural gas-fired engines have become dominant for continuous operation applications (i.e., not emergency generators). Advanced reciprocating engines compete against other distributed generation technologies, as well as gridsupplied electricity. However, conventional electricity generation is inherently inefficient, converting only about a third of a fuels potential energy into usable energy. An advanced reciprocating engine can obtain an overall efficiency of 65 percent. A primary energy savings of 49 percent can be obtained with this technology. Advanced reciprocating engines cost approximately $350/kW. At this price, they are not competitive with purchase grid electricity. However, for high-value and niche applications they may be more cost-effective. Reciprocating engines are the dominant independent generation technology for small installations, accounting for 47 percent of sites but only 2 percent of the power generation. In the industrial sector in 1995, reciprocating systems generated less than 1 percent of total cogenerated electricity but accounted for 5 percent of the installed systems with an average installed size of less than 1 MWe. This type of system is most commonly found in the food products industry (SIC 20) (EIA 1997). A number of market barriers exist in installing distributed generation technologies in addition to the technical issues. These barriers will need to be removed for this technology to achieve its full market potential. (Alderfer, Eldridge and Starrs 2000). Efforts are underway at both the national and state levels to address these barriers. While our analysis did not compete the various electricity generating technologies against each other, it is worth noting that advanced reciprocating engines are currently being installed in larger quantity than fuel cells or microturbines. Reciprocating engines can be manufactured, delivered, and installed very quickly (usually within a few months). The waits for microturbines and fuel cells can be as much as 18 months. To put this length of time into perspective, consider the product cycle of a microchip manufacturer. New product lines for microchips are installed approximately every 18 months. The wait for delivery of other types of on-site generation may be an entire product cycle for these manufacturers, which is simply too long.

141

LBNL and ACEEE

Advanced Reciprocating Engines Data Table

Units Advanced Reciprocating Engines Utility-2 Replace grid-supplied electricity Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources H,M,L Notes

GWh

Cross Cutting Utilities Natural gas Retrofit, new 1,304,220

All industrial electricity, AEO 2000 forecast

Grid supplied electricity at 33.4% delivered efficiency hours 6000 kWh 1 MBtu 0 MBtu 0.0102 800 kW reciprocating engine operated 6000 hours per year at 85% load with 65% efficiency kWh 0 MBtu 0.0052 MBtu 0.0052 Commercial 2000 Years 7 A refit after 7 years. Cost of refit is 50% of the initial capital cost kWh/% MBtu/% MBtu/% % kWh Mbtu Mbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 1.000 100% -0.005 0.0050 49% low 12% Assumes 20% of electric power demand classified as premium Average 2015 grid electricity price is $0.039/kWh and natural gas price $3.38/Mbtu 156506 Decreased grid electricity -821.5 Increased fuel use 777.3 Net primary energy savings 350 Full 0.0142 0.01 (1.33) 1.40 8.3 4% Onsite Sycom, 2000 Competes against capital cost embedded grid price $85.20/kWyear - Onsite Sycom, 2000

Includes a 50% of original capital cost refit charge at years 8 and 16

Significant Significant Limited Somewhat High

Improved reliability can offer increase up-time Improve power quality can improve product quality in sensitive applications Increases on-site emissions and it is unclear whether is cleaner than grid supplied electricity Can allow expansions without needing to upgrade utility service, and can allow for peak load shaving Both manufacturer and government R&D and demonstration

H,M,L

Market barriers to distributed generation Medium Continued R&D and demonstrations E,G,F,P Excellent EIA 1999 EIA 1999 Onsite Syscom Energy 2000 Onsite Syscom Energy 2000, EIA 1999 Onsite Syscom Energy 2000 Onsite Syscom Energy 2000, Bautista 2000 Elliott and Spurr 1999

142

Emerging Energy-Efficient Industrial Technologies

Fuel Cells (Utilities-3)

A fuel cell generates direct current electricity and heat by combining fuel and oxygen in an electrochemical reaction. This technology is an advancement in power generation that avoids the intermediate combustion step and boiling water associated with Rankine cycle technologies (e.g. steam turbines), or efficiency losses associated with gas turbine technologies. Fuel to electricity conversion efficiencies can theoretically reach 80-83 percent for low temperature fuel cell stacks and 73-78 percent for high temperature stacks. In practice, efficiencies of 50-60 percent are achieved with hydrogen fuel cells while efficiencies of 42 percent-65 percent are achievable with natural gas as a fuel (Blok et al.1996). A fuel cell consists of two electrodes separated by an electrolyte. Electrochemical reactions in the cell release electrons from one electrode and take up electrons at the other electrode. When these electrodes are connected to an external circuit, they produce useful electrical work (Blok et al. 1996). The five main types of fuel cells are alkaline (AFC), polymer electrolyte membrane (PEMFC), phosphoric acid (PAFC), molten carbonate (MCFC), and solid oxide (SOFC). For industrial sector combined heat and power applications, the most promising types of cells are the PAFC, MCFC, and the SOFC. Of these, the PAFC is the most developed (there are over 200 operating worldwide), but only has fuel electricity conversion efficiencies of 5 percent better than the most recent combined cycle technologies (Hydrogren Fuel Cell Investor 2000, Blok et al. 1996). We therefore focus on the MCFC and SOFC in this write up. MCFC, due to their high temperatures of operation, are good candidates for small scale CHP. While the MCFCs have slightly better heat rates than PAFC, their high material costs and high parasitic loads (they operate in a pressurized system) detract from the technologys long term potential. A small scale 2 MW MCFC technology demonstration project started in 1996 for the city of Santa Clara, California achieved a conversion efficiency of 44 percent (Hydrogen Fuel Cell Investor 2000). SOFC, while a less developed technology, has also progressed. Siemens Westinghouse Power Corporation announced the first demonstration of a tubular solid oxide fuel cell (SOFC) power generation technology fueled by natural gas in Norway (Fuel Cells 2000). Other demonstrations and proof of concepts are planned for 2001 and 2002, with commercial orders available after that point (Siemens 2000). Because of their lower materials cost, SOFC may be a better candidate technology than MCFC for high temperature applications in the long term (Freeman 2000). A particular variant of fuel cell technology that also offers promise in an industrial context is the family of fuel cell/microturbine technologies, or so-called hybrid technologies. Combining fuel cells and microturbines can further boost efficiency by utilizing waste heat to further generate power. The first SOFC fuel cell/gas turbine hybrid power system (220-kW capacity) is being readied for shipment, installation and operation at the National Fuel Cell Research Center at the University of California, Irvine. The microturbine is said to add an additional 12 percent efficiency with the turbine system (Fuel Cells 2000, DOE 2000). This year Fuel Cell Energy was selected by the Department of Energy for a $3.1 million program, including 20 percent cost sharing, to support the design of an ultra-high efficiency, fuel cell/turbine hybrid power plant. The system proposes an innovative combination of FuelCell Energy's Direct FuelCell with a turbine without requiring any combustion in the turbine, or pressurization of the fuel cell (DOE 2000). In our analysis we compare fuel cells to a base case of average purchased electricity from the grid of 34 percent. While fuel cell efficiencies can vary from 40-65 percent, we assume that technologies entering the industrial market achieve conversion efficiencies of 60 percent based on improvements to existing fuel cell technologies but not including hybrid systems. While high temperature fuel cells do offer a significant potential for the production of both heat and electricity for useful purposes, we assume that in the near to medium term, initial fuel cell markets will not be attractive for combined heat and power. We anticipate that these initial markets will primarily value the high quality electricity.

143

LBNL and ACEEE

Fuel Cells Data Table

Units Fuel cells Utilities-3 Replace grid-supplied electricity Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Mill kwh

Cross cutting Utilities Fuels, other New, retrofit 1,304,220

All industrial electricity, EIA 1999 forecast

Grid supplied electricity at 33.4% delivered efficiency 6000 kWh Assume a 200 kW load application 1 MBtu Electricity purchased from the grid. no fuel inputs on site 0.0 MBtu 0.0102 Install fuel cells for industrial use kWh 0 MBtu 0.007 MBtu 0.007 Pre-commercial 2005 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu 7 1.000 -0.007 0.003 low 5% 65211 -2 185 MCFC costs run roughly $5000/kW; $1,500/kw after long term production. We assume $70/kwyear based on Onsite Sycom, 2000 100% 0% 33% Niche applications in the industrial sector

Assume a 50% electricity conversion efficiency

Both MCFC and SOFC models. PAFC already commercial Estimated economic life. A refit must be done after 7 years. Cost of refit is 50% of the initial capital cost

$ $ $/kWh $/Mbtu $/Mbtu Years %

1500 Incremental 0.012 0.06 (8.18) 16.46 58.6 -14% Somewhat Somewhat Significant

Discount rate for all CCE calculations is 15%

Increased reliability could lead to cost savings Higher power quality Little to no NOx emissions Several fuel cell promotion organizations (see text)

H,M,L

High Technical, cost Medium Good

H,M,L E,G,F,P

At first in niche markets Cost data projected for full scale production EIA 1999 EIA 1997 See text Blok et al. 1996 EIA, 1997; judgement See text

144

Emerging Energy-Efficient Industrial Technologies In addition to the improvements in energy conversion, fuel cells provide improved power quality and reliability, and have the flexibility to be scaled according to a variety of industrial processes. In the near term fuel cells will be particularly attractive for those industries which value an non-interruptible supply of high quality power, (such electronics manufacturing) especially in non-attainment areas that are facing tough air quality regulation. While fuel cells definitely have significant potential, particularly in niche end-uses in the industrial sector, production costs are still high due to the lack of large scale production lines. MCFC installation costs for recent projects are running about $5,000/kW (FuelCell Energy 2000). Siemens plans on achieving an installation cost of $1,300-1,500/kW for their SOFC model after achieving mass production, but initial costs are much higher (Forbes 2000). Eventual electricity delivery costs are estimated to be less than $0.05/kWh (Hydrogen and Fuel Cell Investor, U.S. DOE 2000, Forbes 2000). We assume an increase in O&M costs of $70/kw-year based on (Onsite-Sycom 2000). Given the progress in mobile and small-scale stationary applications, it is more likely that these applications may first reach levels of mass production with medium temperature fuel cells with some of the technology know-how spinning off to larger scale, high-temperature industrial cell applications. There is a high level of activity surrounding the promotion of fuel cell technologies given their potential application in all end-use sectors19. While PAFC are already commercialized, SOFC and MCFC are in the demonstration stage under cost-shared arrangements between the U.S. government. The key U.S. manufacturer of SOFC is Siemens-Westinghouse, and of MCFC is FuelCell Energy, Inc.(www.fce.com). Other international companies are also involved in MCFC and SOFC fuel cell development in Canada, Europe, and Japan (see http://www.h2eco.org/links.htm, and www.h2fc.com). Full commercialization expected before 2005. Given the higher costs of new fuel cell systems, we assume that penetration will be limited in the near term to niche applications (e.g. electronics manufacturing) where power quality and reliability are at a premium.

19

Some of the main U.S. bodies include: the National Hydrogen Association (http://www.ttcorp.com/nha/), the U.S. Department of Energy and Department of Defense, fuel cell research programs operated by the Federal Energy Technology Center (www.fetc.doe.gov), the Fuel Cell Commercialization group (http://www.ttcorp.com/fccg/), the American Hydrogen Association (http://www.clean-air.org/), and the U.S. Fuel Cell council (http://www.usfcc.com/), the Hydrogen and Fuel Cell Investor (http://www.h2fc.com/) and others.

145

LBNL and ACEEE

Microturbines (Utilities-4)
As discussed in the introduction, a number of technologies are available to generate electricity on site, and compete against grid supplied electricity for both energy and reliability. Microturbines are a new class of small combustion turbine engines, ranging in size from 25 kW to 500 kW of electric generating capacity (DOE 1999a). Like the current class of industrial turbines, which were developed using jet engines as a model, these devices have derived from several types of turbo-machinery, including aircraft auxiliary power units (APU) and industrial gas compressors. Like their larger siblings, microturbines can run on a variety of liquid and gaseous fuels, with natural gas projected to be the most common. Microturbines are high-speed devices, usually rotating at over 40,000 rpm. They come in several physical configurations, which represent tradeoffs in cost and performance. The engine can be a single-shaft machine, which reduces cost, or a split-shaft machine, which is more complex but allows for direct drive of a generator, thus avoiding the need for an inverter. Another design consideration is choice of bearings. Air bearings, which have emerged as the technology of choice, reduce the cost of the microturbine, but oil bearings offer longer life and are more rugged. One configuration of microturbines, the simple-cycle machine, is less expensive, but is also less efficient than a recuperated and/or intercooled machine. Simplecycle microturbines are projected to have an efficiency of 26-30 percent (DOE 1999a). When heat recovery is implemented, the efficiency could approach 40 percent (DOE 1999b). When used in a cogeneration or combined heat and power (CHP) system, the fuel conversion efficiency can approach 80 percent (DOE 1999a). The first cost premium for CHP is approximately 40 percent compared to a simple cycle configuration (Bautista 2000). Four domestic manufacturers are currently in the microturbine market: Honeywell (2000) (formally AlliedSignal), Capstone (2000), Elliott (2000) and Ingersoll-Rand (2000) (formerly Northern Research and Engineering Company). Most units are at the commercial demonstration stage and should not be considered a fully commercial technology. A number of large companies, such as GE (2000) and ABB (van Trigt 1998) are exploring entering the market either through an acquisition or the introduction of a new unit. Thus far, the marketing of microturbines has focused on the commercial marketplace. However, these devices can address important needs in manufacturing as well, because the standard manufacturing establishment has an average electricity demand of just under 400 kW (Census 1996). Microturbines can be used to generate a portion of a plants needs onsite, reduce demand during peak periods, or support premium power applications (e.g., microelectronics manufacturing) (Elliott and Spurr 1999). Microturbines are even more efficient when operated as part of a CHP system, which can meet some of the facilitys thermal requirements as well. However, microturbines will have to compete with reciprocating engines and fuel cells to gain a stronghold in this market. While the first cost of microturbines appear attractive compared to fuel cells, they are less attractive when compared with reciprocating engines. A recent analysis for the Energy Information Administration (EIA) has estimated the current installed costs in CHP mode to be $1,970 per kW, dropping to $915 by 2010. This is compared with $1,390 per kW installed today and $990 per kW in 2020 for a comparable reciprocating gas engine. It is projected that the first costs would become competitive during this studys time frame (Onsite 2000). These microturbine cost estimates appear higher than current prices that manufacturers are quoting (GE 2000, Honeywell 2000, and Tanner 2000). As the Onsite (2000) study notes, these cost quotes frequently do not include all equipment necessary for a functional install (GE 2000). Also, as a recent new analysis points out, many manufacturers appear to be selling below cost to build market share (Kaplan 2000). One of the promises of microturbines is greater reliability and a lower operating cost than reciprocating engines. However, field experience with microturbines has been limited, and because the technology is evolving very rapidly, reliable information about performance is not readily available. Another area in need of improvement is environmental emissions, in particular NOx. While manufacturers have raised hopes that microturbine emissions would be much lower than emissions from other technologies, current rates are similar to those for low-emissions gas engines (Greene and Hammerschlag 2000). The

146

Emerging Energy-Efficient Industrial Technologies Microturbines Data Table

Units Microturbines Utilities-4 Replace grid-supplied electricity Market Information: Industries End-use(s) Energy types Market segment 2015 basecase Reference technology Description Throughput or annual operating hours Electricity use Fuel use Primary Energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Estimated average measure lifetime Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions for savings Electricity savings potential in 2015 Fuel savings potential in 2015 Primary energy savings potential in 2015 Cost Effectiveness Investment cost Type of cost Change in annual costs (O&M/other benefits) Cost of conserved energy (electricity) Cost of conserved energy (fuel) Cost of conserved energy (primary energy) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New Measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

GWh

Cross Cutting Utilities Natural gas Retrofit, new 1,304,220

All industrial electricity

Grid supplied electricity at 33.4% delivered efficiency hours 6000 kWh 1 MBtu NA MBtu 0.0102 100 kW microturbine operated 6000 hours per year at 85% load with 40% efficiency kWh N/A MBtu 0.0085 MBtu 0.0085 Commercial demonstration 2001 Estimated economic life. A refit must be done after 7 years. Cost of refit is 50% of the initial capital cost. Years 7 kWh/% MBtu/% MBtu/% 1 -0.00001 0.002 Low

17%

% GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years %

Assumes that 20% of electric power demand can be classified as premium, penetration rate 25% in 2010 5% Average 2015 grid electricity price is $0.039/kWh and natural gas price $3.38/Mbtu Decreased grid electricity 39,900 Increased fuel use -0.3 Net primary energy savings 67 641 Full 0.0125 (0.00) 95.18 (0.48) N/A Undefined Significant Significant Somewhat Somewhat High Bautista 2000. Competes against capital cost imbedded grid price

Discount rate for all CCE calculations is 15% Not cost effective against grid electricity

H,M,L

Improved reliability can offer increase up-time Power quality can improve quality in sensitive apps Increases on-site emissions and it is unclear whether is cleaner than grid supplied electricity Allows expansion without util. upgrade and peak shaving Both manufacturer and government R&D and demo

H,M,L

High first cost, lack of proven reliability, and market barriers to DG Medium Continued R&D and demonstrations E,G,F,P Good EIA 1999 EIA 1999 Onsite Syscom Energy 2000 Onsite Syscom Energy 2000, DOE 1999c Onsite Syscom Energy 2000 Onsite Syscom Energy 2000, Bautista 2000 Elliott and Spurr 1999 Bruce Hedman, Onsite Sycom bhedman@onsitesycom.com

147

LBNL and ACEEE Department of Energy has identified this as another important area for development, with the application of low emissions technologies developed as part of the advanced turbine system program (DOE 1999c). In addition to the technical issues, a number of market barriers exist in installing distributed generation technologies, which will need to be removed for this technology to achieve its full market potential (Alderfer, Eldridge and Starrs 2000). Efforts are underway at both the national and state levels to address these barriers. Current projections for operating costs are for $90 per kWyear falling to $75 by 2020 (Onsite 2000). These compare favorably with reciprocating engines today and appear even more attractive in the future. Realizing these performance goals require further product development with deployment of advanced materials and operating experience (DOE 1999c). In the configuration of electric generation only, microturbines appear unable to compete against the average industrial energy price for grid supplied electricity. In CHP mode the economics appears somewhat more efficient, but they are still not competitive. These assessments do not take into account variations in energy prices or valuation of ancillary services such as reliability. If a combination of these were to increase to twice the EIA 2015 estimate for electricity of $0.0039/kWh, the payback would fall to 2.9 years with an IRR of 34 percent in simple cycle mode and 3.4 year with an IRR of 29 percent in CHP mode. An ancillary benefit of $0.039 per kWh is perhaps at the low end for premium-power applications such as are currently seen in pharmaceutical and semiconductor applications (Elliott and Spurr 1999). The systems can offer high reliability and power quality as well as low noise. This analysis would indicate that the market for microturbines would likely be limited to premium power applications.

148

Emerging Energy-Efficient Industrial Technologies

Advanced Lighting Technologies (Lighting-1)

Lighting accounts for approximately 6.5 percent of industrial sector electricity demand. In 1994, more than 58,600 GWh of electricity was consumed by lighting in industrial facilities (EIA 1997). High-bay lighting, required to provide overall ambient lighting throughout manufacturing and storage spaces, is typically provided by high-intensity discharge (HID) sources, including metal halide, high-pressure sodium and mercury vapor lamps. HID accounts for approximately 60 percent of industrial lighting energy consumption (Johnson 1997). Supplementary lighting is used to provide low-bay and task-specific lighting for inspection, equipment operation, and fine assembly activities. Fluorescent, compact fluorescent and incandescent light sources are commonly used for task lighting needs and together account for approximately 40 percent of industrial lighting energy. A range of advanced lamp, ballast, fixture, and light pipe technologies can significantly reduce lighting energy consumption in industrial facilities. Electrodeless light sources, such as the induction lamps offered by Philips, Osram Sylvania, and GE, combine high quality light with high efficacy, long-lived lamps. These systems offer energy and maintenance cost reductions, particularly in spaces where lighting is hard to access and maintain. Remote-source lighting technologies, including fiber optics systems and light pipes using a variety of light sources such as sulfur lamps, LEDs, and hybrid artificial-natural lighting, offer numerous advantages in industrial settings. Benefits of remote-source lighting include: minimized heat gain in lit areas resulting in a lower cooling load; improved safety from elimination of lighting-related electrical wiring and equipment in wet or explosive areas; allowance for the use of more efficient and powerful light sources; more targeted and esthetically-pleasing light; and reduced installation and maintenance costs (Krepchin 1999). Another example is the replacement of HID lighting with high-intensity fluorescent lighting in high-bay applications. New high-intensity fluorescent lighting systems incorporate high-efficiency twin-tube or linear T5 fluorescent lamps, advanced electronic ballasts, and high-efficacy fixtures that maximize light output to the work plane. Each of the system components confers advantages over traditional HID fixtures. Advantages include: lower energy consumption; lower lumen depreciation over the lifetime of the lamp; better dimming options; faster start-up and restrike (virtually instant-on capability); better color rendition; higher pupil lumens ratings (translating into improved worker productivity and performance); and less glare (given fixture design and the more diffuse nature of the fluorescent light source) (Rogers and Krepchin 2000). Under similar operating conditions, high-intensity fluorescent replacements yield 50 percent electricity savings over standard metal halide HID. The use of dimming or on/off controls, which are impractical with most HID systems, can increase savings substantially (see Advanced lighting design). The first highintensity fluorescent systems suitable for high-bay industrial applications were introduced in 1996. Since that time, the number of fixture manufacturers has grown to more than a dozen and prices have dropped dramatically. In new facilities, high-intensity fluorescent and HID systems are comparable in cost. In retrofit applications, investment costs are approximately $185 per fixture ($150 fixture and $35 installation cost); the lamp costs are equal to HID (Rogers 2000, EBN 2000). Case studies have also found reduced maintenance costs resulting from the use of multi-lamp fixtures unlike HID fixtures, the fluorescent fixture continues to provide sufficient light in most applications even when one lamp fails. As a result, lamp replacement can be delayed until several lamps fail at which time the entire fixture (i.e., 4-6 lamps) is changed out (Rogers 2000). To date, promotional efforts have been focused predominately on commercial sector applications including large retail and warehouse spaces. The lack of readily available information targeted to industrial end-users and a lack of interest in upgrading facility lighting has prevented acceptance of the technology in the industrial sector. Furthermore, there has been reluctance on the part of contractors to share information on the technology and its benefits with their competitors. However, the potential for widespread application and large-scale energy savings in manufacturing facilities is beginning to spark an interest in greater promotion of the technology by utilities. Utilities in the Northeast are offering incentives in the form of custom rebates, but there has been some conflict with recent rebate programs for HID retrofits (Rogers and

149

LBNL and ACEEE

Advanced Lighting Technologies Data Table

Notes Units Advanced lighting technologies Lighting-1 High-intensity fluorescent replacements for high bay HID Market Information: Cross-cutting Industries End-use(s) Lighting Energy types Electricity Market segment New, retrofit Lighting electric is 7% of industrial electricity consumption 2015 basecase use GWh 91,000 Reference technology 400W metal halide lamp, ballast, bell-shaped spun-aluminum fixture Description Rogers and Krepchin 2000 Throughput or annual op. hrs. hours 5000 Rogers and Krepchin 2000; EBN 2000 2325 Electricity use kWh/yr N/A Fuel use MBtu 19.8 Primary energy use MBtu New Measure Information: High-intensity fixture w/4 T-5 lamps and electronic ballast Description Rogers and Krepchin 2000; EBN 2000 1170 Electricity use kWh N/A Fuel use MBtu 9.9 Primary Energy use MBtu Commercialized Current status Rogers 2000 1996 Date of commercialization 20,0000 hours at 5,000 hours/year 4 Est. avg. measure life Years Savings Information: 50% Electricity savings kWh/% 1155 N/A Fuel savings MBtu/% N/A 50% Primary energy savings MBtu/% 9.8 High Penetration rate 100% of industrial HID 60% Feasible applications % Other key assumptions 27,124 Elec svgs potential in 2015 GWh N/A Fuel svgs potential in 2015 Tbtu 30% primary energy savings. Lighting primary energy totals 231 Primary energy svgs potential in 2015 Tbtu 773.5 Tbtu in 2015 Cost Effectiveness Fixture cost $150; lamp costs equals HID; $35 install cost 185 Investment cost $ Full cost Type of cost Reduced maintenance cost -25 Change in other costs $ 0.03 Cost of saved energy (elec) $/kWh N/A Cost of saved energy (fuel) $/Mbtu Discount rate for all CCE calculations is 15% 4.05 Cost of saved energy (primary) $/Mbtu Based on $11.43/Mbtu average cost for electricity 1.3 Simple payback period Years 64% Internal rate of return % Key non energy factors Dimming capability/improved lighting quality increase worker Somewhat Productivity benefits performance Less glare/better color rendition improve lighting quality for Somewhat Product quality benefits product inspection None Environmental benefits Added savings with controls and sensors; faster start-up Significant Other benefits Demonstrations, incentives, supplier marketing, research High Current promotional activity H,M,L Evaluation Lack of info/user interest Some conflict with recent incentive programs for HID retrofits Major market barriers High Likelihood of success H,M,L Info dissemination; demos Rebates/incentives for retrofits Recommended next steps Substantial literature from industry and independent sources, field Excellent Data quality assessment E,G,F,P data Sources: EIA 1999 2015 basecase Rogers and Krepchin 2000; EBN 2000; E Source 1997 Basecase energy use Rogers and Krepchin 2000; EBN 2000 New measure energy savings Rogers and Krepchin 2000; EBN 2000; E Source 1997 Lifetime Rogers 2000 Feasible applications Rogers 2000; EBN 2000; E Source 1997 Costs Rogers and Krepchin 2000; EBN 2000 Key non energy factors Jim Rogers 978/256-1345; Nancy Clanton, Clanton & Associates Principal contacts 303/530-7229 Additional notes and sources

150

Emerging Energy-Efficient Industrial Technologies Krepchin 2000). In the Midwest, utilities are educating account representatives and customers about the products (Rogers 2000). and in California, several manufacturers and distributors of high-intensity fluorescent lighting products are expanding their marketing efforts and working with Southern California Edison to incorporate the technology into their new construction programs (Rogers 2000). Additional information dissemination, a broader range of demonstrations and case studies, and continued utility incentives and support would create further demand for the technology.

151

LBNL and ACEEE

Advanced Lighting Design (Lighting-2)

Advanced lighting design techniques that incorporate daylighting, lighting controls, and task lighting can substantially reduce lighting energy consumption in industrial facilities. Daylighting of facilities through the use of daylighting devices such as skylights, light shelves, and reflectors is most easily justified for new facilities. Existing facilities can take advantage of daylighting methods such as glazed windows and daylight pipes, which are installed through the roof and utilize a series of reflectors and a diffuser to direct sunlight to the interior space. Lighting sensors and controls adjust electric light levels to account for the level of natural light entering the space. Task lighting directs light to specific tasks being performed and to individual workers needs, thereby allowing for reduced ambient light levels throughout the facility. In order to reap the benefits of improved lighting design, the design must be integrated with compatible lighting technologies. For example, current lighting practice in the majority of manufacturing facilities relies on high-intensity discharge (HID) sources to provide overall ambient lighting. Dimming controls are impractical for use with most HID lamps because the long restart and warm-up times required for HID sources make on/off controls impractical. Additionally, unlike fluorescent sources, reductions in light output and energy consumption are not linear and overall energy savings can be quite small. Developments in HID lamp, ballast, and luminaire technology may help address these concerns, but it remains unclear whether HID technology will ever match the fast start and restrike, low lumen depreciation rates, and lighting quality achievable with fluorescent sources (Rogers and Krepchin 2000). Replacing low-bay fluorescent lighting with compact fluorescent task lighting specifically designed for the needs of individual workers and work stations can further improve lighting quality while reducing energy costs. One example of advanced lighting design for existing facilities that incorporates many of the energy-saving features discussed above is the replacement of HID light sources with high-output fluorescent lighting and installation of daylight pipes. (For more information on high-output fluorescent lighting, see Advanced lighting technologies.) High-intensity fluorescent lighting typically achieves a 50 percent reduction in lighting electricity compared to metal halide HID sources without the use of dimming or other control strategies. By incorporating dimming controls high intensity fluorescent lamps have full dimming capability and introducing daylight through the use of daylight pipes, overall lighting energy consumption can be reduced by approximately 80 percent. This system could effectively replace the majority of both HID and conventional fluorescent lighting found in most industrial facilities (Rogers 2000). The costs of such a system vary depending on the level of ambient light required and whether full daylighting (i.e., elimination of all electric lighting during daylight hours) is desired. For full daylighting of 300 to 400 square feet of floor space, replacement of two conventional 400W HID fixtures would cost approximately $1070 installed. This configuration would require two high-intensity fluorescent fixtures at $370 ($150 for each fixture and $35 installation cost per fixture), dimming controls including photosensors and control interface at $100, and a 21-inch diameter daylight pipe at $600 installed (Rogers 2000, EBN 2000, Miller 2000). Although the diffuser dome within the daylight pipe must be cleaned once or twice a year depending on the levels of dust generated in the facility, maintenance costs are expected to remain constant because of fewer lamp replacements. Additional benefits of the system include reduced HVAC loads resulting from lower lighting-related heat input and improved worker performance and productivity due to better lighting quality. More than a dozen manufacturers offer high-intensity fluorescent lighting systems (Rogers 2000). At least ten manufacturers produce daylight pipes (Krepchin 1999). This product was originally developed for residential use, but the manufacturers are increasing their focus on commercial and industrial applications. Demonstrations, utility incentives, and supplier marketing efforts are being employed to increase demand for these systems. To date, efforts have been more focused and more successful in the commercial sector, but interest among the industrial sector is growing. However, a number of remaining barriers including the high first cost for daylight pipes and a lack of detailed information on the costs and benefits continue to hinder acceptance of the technology. Detailed case studies to verify savings and benefits in industrial applications, broader promotion and incentives, and coordination among manufacturers of the system components could

152

Emerging Energy-Efficient Industrial Technologies improve the long-term market viability of these systems. In addition, studies on the impacts of daylighting on worker productivity are needed. The results of such studies in schools appear to have significantly increased interest in daylighting of school facilities and could have a similar effect in the industrial sector. Advanced Lighting Design Data Table
Units Notes Advanced lighting design Lighting-2 Daylighting with dimmable fluorescent replacement for HID Market Information: Industries Cross-cutting End-use(s) Lighting Energy types Electricity Market segment New, retrofit Lighting electric use is 7% of industrial electricity consumption 2015 basecase use 91,000 Reference technology Description 400W metal halide lamp/ballast/bell-shaped spun-aluminum fixture; 2 fixtures per zone Throughput or annual op. hrs. Hours 5000 Electricity use kWh/yr 4650 Fuel use MBtu N/A Primary energy use MBtu 39.5 New Measure Information: High-intensity fluorescent fixture w/4-T5 lamps/electronic ballast (2 fixtures per zone); dimming controls; daylight Description pipe 2000 hours 100% dimming; 2000 hours 50% dimming; 1000 hours no Electricity use dimming kWh 936 Fuel use MBtu N/A Primary Energy use MBtu 8.0 Current status Commercialized Date of commercialization 1996 Est. avg. measure life Years Fixtures; controls; light pipe 20 Savings Information: Electricity savings kWh/% 3714 80% Fuel savings MBtu/% N/A N/A Primary energy savings MBtu/% 31.6 80% Penetration rate Medium Feasible applications % 70% of industrial HID; 60% of industrial fluorescent 66% Other key assumptions Elec svgs potential in 2015 GWh 47,971 Fuel svgs potential in 2015 Tbtu N/A 53% primary savings. Lighting 2015 primary energy 773.5 Tbtu. Tbtu Primary energy svgs potential in 2015 408 Cost Effectiveness $370 for 2 fixtures w/installation; $100 controls; $600 light pipe Investment cost w/installation $ 1070 Type of cost Full cost Change in other costs $ 0 Cost of saved energy (elec) $/kWh 0.05 Cost of saved energy (fuel) $/Mbtu N/A Discount rate for all CCE calculations is 15% Cost of saved energy (primary) $/Mbtu 5 Simple payback period Years Based on $11.43/Mbtu average cost for electricity 2.97 Internal rate of return % 34% Key non energy factors Productivity benefits Dimming capability/improved lighting quality increase worker performance Somewhat Less glare/better color rendition improve product inspection Somewhat Product quality benefits None Environmental benefits Added savings w/ task lighting; reduced HVAC load; faster start-up Significant Other benefits Demonstrations, utility incentives, supplier marketing, research H,M,L High Current promotional activity Evaluation Major market barriers Cost of daylight pipe particularly high Lack of info; first cost Likelihood of success H,M,L Cost of daylighting must come down Medium Recommended next steps Info dissemination; demos Rebates/incentives for new construction and retrofits Substantial industry and independent literature, limited field data Data quality assessment E,G,F,P Good Sources: 2015 basecase EIA 1999 Basecase energy use Rogers and Krepchin 2000; EBN 2000; E Source 1997 New measure energy savings Rogers and Krepchin 2000; Krepchin 1999; Miller 2000 Lifetime Rogers and Krepchin 2000; Krepchin 1999; Miller 2000 Feasible applications Rogers 2000; Miller 2000 Costs Rogers 2000; EBN 2000; E Source 1997; Krepchin 1999; Miller 2000 Key non energy factors Rogers and Krepchin 2000; EBN 2000; Krepchin 1999; Miller 2000 Principal contacts Jim Rogers 978/256-1345; Greg Miller, Sun Pipe, 847/272-6977 Additional notes and sources

153

LBNL and ACEEE

High Tech Facilities HVAC Improvements (HVAC-1)

Within the manufacturing sector, a variety of high tech facilities such as laboratories and cleanrooms use a significant amount of energy to operate heating, ventilation, and air-conditioning equipment (HVAC). Much of this energy is used to ensure that production facilities are free from high levels of pollutants that could damage products. These facilities have energy intensities that can range from 5 to 50 (or more) times higher than typical commercial buildings, and HVAC loads account in some cases for 40-50 percent of total energy use (Tschudi 2000, Mills et al. 1996) High-tech facilities have grown dramatically, and are expected to continue to grow significantly. In the manufacturing sector, high-tech facilities are most commonly associated with the production of semiconductor-based integrated circuits and other electronic components, and with pharmaceutical and biotechnology derived products. Together these two industries account for 70 percent of the clean room square footage in the U.S. (Mills et al. 1996). Other industrial sectors that use cleanroom space include automotive, flat panel manufacturing, food, hospitals, medical devices, and other electronics. Research by McIlvane and Co. has identified a total space use of 14 million square feet (msf) in 1995, growing to 25 msf in 2000, a growth of over 10 percent annually (McIlvane Co. 1996). In California, which houses over 10 percent of the US cleanroom space, electronics- and computer- manufacturing sectors are the fastest growing energy users (Tschudi and Sartor 1999). There are no reported energy use estimates for high-technology facilities. Electricity intensities can range from 150-950 kWh/ft2 depending on the level of cleanliness required in the manufacturing environment. Based on detailed analysis undertaken of high-tech facilities in California, we estimate a primary energy consumption in 1995 of 60 TBtu (63 PJ), based on an average weighted energy intensity of 480 kWh/ft2. This represents about 0.3 percent of total 1994 manufacturing energy use. There are several HVAC technologies that have emerged recently which when combined, can achieve significant energy savings. Currently a large amount of energy is expended in heating, cooling, and filtering air that is then exhausted to the atmosphere. Minimizing exhaust flow reduces the amount of make up air that needs to be reconditioned. Ultra low fume hoods, a technology developed at Lawrence Berkeley National Laboratory, require 25 percent of normal exhaust flow. This technology is now being piloted in field trials (Tschudi 2000). Air re-circulation is another large HVAC energy user. If occupancy is reduced, then less airflow is required to maintain required cleanliness levels. Sensors and the use of laser-based particle counters are both technologies that can be applied to more efficiently moderate air flow. Additionally, more efficient air flow equipment that is near commercial (e.g. low face velocity fans, efficient duct systems, more efficient filter units) could be combined to further reduce recirculation requirements. Finally, new immerging filter technologies (HEPA/ULPA filters) offers the opportunity to significantly reduce filter energy use by reducing filter pressure drops (Tschudi 2000). While not emerging, there are several existing practices that can also be applied in conjunction with the above mentioned technologies that can further enhance energy savings, including right-sizing of exhaust systems, improved design guidance for ducting and other systems, and limiting the floor area that requires clean air flow to a smaller micro environment. Combined, these clean room HVAC technologies have the potential to reduce electricity consumption of the average clean-room facility by 25-30 percent, or an average of 145 kWh/ft2. Additionally, they are accompanied by several additional non-energy benefits including improved productivity and a reduction in emissions without sacrificing any product quality. When combined in a carefully optimized fashion, these measures can usually have a payback of 4 years or less, or an incremental cost of roughly $30 per square foot. More case studies are needed in order to improve the evaluation of costs and payback.

154

Emerging Energy-Efficient Industrial Technologies Hi-Tech Facilities HVAC Data Table

Units Hi-tech facilities HVAC HVAC-1 Improve HVAC systems in hi-tech industries M ar ketI nf or m at i on: Industries Cross cutting End-use(s) HVAC Energy types Electricity, gas Market segment New 2015 basecase use msf 37.9 R ef er ence t echnol ogy Description Existing hi-tech facility HVAC systems Throughput or annual op. hrs. Electricity use Fuel use Primary energy use N ew M easur eI nf or m at i on: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savi ngs I nf or m at i on: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 C ostEf f ect i veness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non ener gy f act or s Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sour ces: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources 1 kWh MBtu MBtu 480 0.9 5.0 Notes

Estimate of laboratory facility square footage in 2010

Normalize on a square foot basis for reference and new technologies Weighted average based on US cleanroom distribution

Efficient HVAC in hi-tech facilities kWh 336 MBtu 0.9 MBtu 3.8 Commercialized Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 144 0.0 1.2 Medium 30% 1637 0 13.9 20 Incremental 0 0.02 #DIV/0! 2.61 4.0 0% Somewhat None None Somewhat Medium Information Medium Consortiums, roadmap Fair 20 30% 0% 25%

Tschudi and Sartor, 2000

Some technologies also near commercial

High feasibility for new buildings

Based on Singapore wafer fab case study (Tschudi, 2000)

Discount rate for all CCE calculations is 15%

Improved worker safety with improved fume hoods

H,M,L

High perceived risk

H,M,L E,G,F,P

McIlvane, 1996 Mills et al., 1996 Tschudi and Sartor, 2000 Tschudi and Sartor, 2000 Tschudi and Sartor, 2000 Tschudi and Sartor, 2000 Tschudi and Sartor, 2000 William Tschudi, LBNL (WFTschudi@lbl.gov)

155

LBNL and ACEEE There are marketing obstacles to the promotion of this technology. These obstacles primarily are geared toward the need for better information to cleanroom designers, builders, and operators on HVAC efficiency opportunities through increased information and education, and the lowering of perceived risk given what would be a higher first cost. LBNL and other research institutions have taken an early role in attempting to promote collaborations with key hi-tech industry associations, but this work is still in its early stages. In addition to collaboration and engaging in a roadmapping process, other potential opportunities to encourage increased industry participation in HVAC efficiency include the development of a benchmarking tool to allow for inter-industry comparisons. We believe that in the absence of an existing consortium there is a medium likelihood of success of dramatically improving HVAC efficiency. A complicating factor is that the lifetime of cleanroom processes are significantly less than the lifetime of the buildings in which the measures are installed, so there is a possibility for increase in maintenance costs in the long term if HVAC equipment must be recalibrated. Next steps will require additional demonstration projects to improve acceptance by the industry.

156

Emerging Energy-Efficient Industrial Technologies

Anaerobic Wastewater Treatment (Other-1)

Industrial wastewater is typically treated by aerobic systems that remove contaminants prior to discharging the water. These aerobic systems have a number of disadvantages including high electricity use by the aeration blowers, production of large amounts of sludge, and reduction of dissolved oxygen in the wastewater which is detrimental to fish and other aquatic life (CADDET 1993e, CADDET 1996d). Anaerobic wastewater treatment is an alternative method for cleaning industrial wastewater which is based on the conversion of organic compounds in the wastewater into a biogas of methane, carbon dioxide, and hydrogen sulfide by bacteria in an oxygen-free environment (CADDET 1993e). The most widely used technology for anaerobic wastewater treatment is the Upflow Anaerobic Sludge Blanket (UASB) reactor which was developed in 1980 in The Netherlands (U.N. FAO 1997). Industrial wastewater is directed up through the UASB reactor, passing through a blanket that traps the sludge. Anaerobic bacteria break down the organic compounds in the sludge, producing methane in the process (CADDET 1993e, CADDET 1996d). This type of anaerobic wastewater treatment is currently used predominantly in the paper and food industries, but some industries such as chemical and pharmaceuticals have also used this technology and its use is growing for municipal wastewater treatment (Habets 2000, Motta 1998). Globally, there are approximately 1500 anaerobic wastewater treatment plants (80 percent are UASBs), of which approximately 150 are in the U.S. (Habets 2000). Energy savings from the use of an UASB come from displaced electricity as well as from the use of the produced biogas. Depending upon the type and size of plant, an anaerobic wastewater treatment facility will displace roughly 6,570 MWh of electricity used annually in an aerobic plant (Habets 2000). In addition, the biogas that is generated can be used in the manufacturing process or can be used by a steam boiler (CADDET 1993e, CADDET 1996d). An anaerobic wastewater treatment facility at a whey processing plant in the Netherlands saves 1,100 MWh/year of electricity through the reduced demand for aeration, although electricity use of 727 MWh/year is still required for pumping and aeration to remove residual organic compounds in the water (CADDET 1993e, Habets 2000). This same plant produces 344 m3/day of methane gas (CADDET 1996d). We estimate a total savings of roughly 1000 kWh/day and an additional generation of 70 Mbtu/day (74 GJ/day) of biofuel. The investment cost for the anaerobic wastewater treatment facility at the whey processing plant was $1.8M, which represents an additional investment beyond an aerobic plant of $274,000 (CADDET 1996d). Annual electricity and fuel savings are $167,000, savings from reduced sludge production are $294,000, and savings from reduced consumption of chemical compounds is $64,000. Costs of $17,000 are incurred for additional personnel. Thus, the overall savings are $508,000 per year, resulting in a payback period of less than 0.5 years for the incremental investment (CADDET 1996d).20 For use of this technology in a paper mill, the total capital cost of the anaerobic wastewater treatment facility was $2.28M (1993$). Savings from the production of natural gas are $80,000 per year (based on fuel prices in the U.K), sludge sales are worth about $50,000 per year, and savings of about $1,400,000 per year result from reduced effluent discharge costs. Operating costs are an additional $375,000 per year. Thus, total annual savings are over $1.5M, resulting in a payback period of about 1.5 years (CADDET 1993). The UASB technology is used around the world and several hundred facilities have been installed by the two leading UASB companies, Paques and Biothane (Heijnen n.d.). Evaluations of anaerobic wastewater treatment facilities in the UK, Netherlands, Canada and U.S. show a wide range of costs and energy savings, with payback periods ranging from 1.4 years to 3.7 years (CADDET 1993e, CADDET 1996d, CADDET 1997e, CADDET 1997f, CADDET 1997g, CADDET 1998b, CADDET 1999e). Currently, there are approximately 125 anaerobic wastewater treatment facilities in the U.S., which is about 0.5 plants per million residents. There is great potential to increase the number of anaerobic wastewater treatment plants; some countries have 3 to 5 plants per million people which implies that 750 to 1250 total plants could be installed in the U.S. (Habets 2000). For our analysis, we estimate that an additional 400 plants could be

20 These costs are calculated based on the cost of energy in The Netherlands. Using lower U.S. energy costs, the payback period is 0.8 years.

157

LBNL and ACEEE

Anaerobic Waste Water Treatment Data Table (Other-1)

Units Notes Anaerobic wastewater treatment other-1 Upflow anaerobic sludge blanket (UASB) wastewater treatment facility Market Information: Industries Cross-cutting End-use(s) Other Energy types Electricity For pumping Market segment New 2015 basecase use TWh 338 Electricity consumption in 2015 for paper, food, bulk chemicals (AEO, 2000) Reference technology Description Conventional aerobic wastewater treatment plant Throughput or annual op. hrs. m3/day 8760 annual operating hours (Habets, 2000) 2500 Electricity use kWh/day For aeration 3019 Fuel use MBtu/day 0 26 Primary energy use MBtu/day New Measure Information: Description Upflow anaerobic sludge blanket (UASB) wastewater treatment facility Electricity use kWh/day Load of 5kw/day for pumping, 78kw/d for aeration (Habets, 2000) 1992 Fuel use MBtu/day 0 17 Primary Energy use MBtu/day Current status Commercialized Date of commercialization 20 Est. avg. measure life Years Habets 2000 Savings Information: Electricity savings kWh/day 1027 34% CADDET 1996d Fuel savings MBtu/day 70 N/A. Generates a natural gas savings of 900,000 m3/year (CADDET 1996d) Primary energy savings MBtu/day 79 307% Penetration rate Medium Currently 5-10% of potential, should be able to triple by 2015 (Habets 2000) Feasible applications % 33% Weighted average for paper, food, and chemical industries Other key assumptions Elec svgs potential in 2015 GWh 150 Est. 375-500 new anaerobic plants in US in 2015 (Habets 2000; Richards 2000) 10.2 Fuel svgs potential in 2015 Tbtu 11.5 Primary energy svgs potential in 2015 Tbtu Cost Effectiveness 274000 Investment cost $ CADDET 1996d (full cost for plant is $1.8 million) Incremental Type of cost -341000 Change in other costs $ Reduced sludge handling & chemical consumption: added personnel costs Cost of saved energy (elec) $/kWh -0.79 Cost of saved energy (fuel) $/Mbtu -11.61 Cost of saved energy (primary) $/Mbtu Discount rate for all CCE calculations is 15% -10.33 Simple payback period Years 0.8 Internal rate of return % 125% Key non energy factors Productivity benefits None None Product quality benefits Significant Environmental benefits Reduced sludge production; in other applications, the biochemical oxygen demand (BOD) level can be significantly reduced (CADDET 1996d) Other benefits Somewhat Compact design (reduces area needed), simple design (CADDET 1996d) Current promotional activity H,M,L Low Mainly promoted by companies that produce the technology; DOE held seminars on the practice, discontinued in the early 1980s (Richards 2000) Evaluation Major market barriers Low energy prices, high cost of technology, lack of knowledge among potential users (Richards 2000) Likelihood of success H,M,L High Depends on the cost of energy, the cost of waste disposal, and the dissemination of anaerobic technology information (Richards 2000) Recommended next steps Government (DOE) sponsored demonstration projects, on-going targeted publicity Demonstration including data (Richards 2000) Data quality assessment E,G,F,P Good+ Sources: 2015 basecase Habets 2000 Basecase energy use CADDET 1996 New measure energy savings Habets 2000 Lifetime Habets 2000 Feasible applications Habets 2000 Costs CADDET 1996 Key non energy factors CADDET 1996 Principal contacts Leo Habets, Paques, NL (L.Habets@paques.nl); Dr. E.A. Richards, P.E., 414-5453629; fax 414-545-6094, (drer.execpc.com) Additional notes and sources

158

Emerging Energy-Efficient Industrial Technologies built by 2015. These plants can be used by a variety of industrial facilities, including papermaking, food processing, chemicals, pharmaceuticals, and distilleries. The market potential varies for these industries from 30 to 40 percent for the paper industry to 100 percent for processing of sugar, starch, and alcohol based on the size of the mills, types of mills, and their water consumption (Habets 2000). We estimate an average market potential of 33 percent of selected food, paper, and chemicals sectors based on a weighting of 1994 energy consumption. Adoption of this technology depends on energy costs as well as effluent controls and disposal costs. This technology is being rapidly adopted in Brazil, Japan, China, Mexico, and Europe. Adoption in the U.S. has been slow, especially in the paper industry. In the past, the U.S. Department of Energy held a number of seminars promoting this technology, but these were discontinued in the early 1980s. Currently, the only promotion in the U.S. is through the large companies that produce this type of system such as Biothane Corporation in Camden, New Jersey (Richards 1996). Market barriers include the low cost of energy, especially vis--vis the cost of the technology, as well as negative perceptions based on past experiences with less effective systems. Another barrier is the lack of information, including reliable data, available on this technology. Stricter effluent regulations combined with government-sponsored demonstration programs that provide real data on this technology, targeting marketing of this technology to potential users, and investment incentive programs could all help to promote increased adoption of anaerobic wastewater treatment (Richards 1996, Habets 2000).

159

LBNL and ACEEE

High-Efficiency/Low NOx-Burners (Other-2)

Recuperators and regenerators are the two major contributors to improved energy efficiency in combustion technology. These technologies preheat the combustion air. However, preheating leads to higher flame temperatures, which may lead to higher NOx emissions. Air quality regulation drives the demand for high efficiency but low emission (NOx, CO) emissions. Low NOx emissions can be achieved by reduce NOx formation in the combustion process or end-of-pipe catalytic removal. The costs of flue-gas removal are high. Integrated solutions in the combustion process are preferred due to the prohibitive costs of removal. NOx formation in the combustion process is reduced by reducing the amount of nitrogen in contact with oxygen at high flame temperatures. Available options are oxy-fuel combustion (e.g. in glass, metals industry), improved mixing of combustion air and fuel to maintain a stabile temperature profile of the flame, and near stoichometric conditions (reducing the amount of nitrogen in the flame) through staged combustion, as well as flue gas recirculation (FGR) (Berntsson et al. 1997). FGR is relatively expensive. Oxy-fuel burners were discussed for the steel and glass industry elsewhere. In this technology characterization we focus on high-efficiency low NOx burner designs using air as oxidant. We discuss burners for boilers, furnaces and direct heating. Note that the performance of a burner depends on the configuration of the boiler or furnace in which it is used. Hence, the savings may vary widely depending on the specific situation. In this description we try to separate the effects of improved burner design from furnace/boiler design. Boilers are used throughout industry and consume about 6.05 Quads (6.38 EJ) of fuels, or 37 percent of total industrial fuel use (Einstein et al. 2000). Because of their widespread use air quality regulation has historically been aimed at boilers. Natural gas is the dominant boiler fuel (40 percent), followed by biomass/wastes (38 percent) and coal (14 percent). In this description we focus on natural gas burners. Solid fuels are normally burned in stoker-boilers. In stoker boilers NOx emissions can be reduced by OverFire-Air (OFA, introduction of air not at the burner) and gas reburning. In oil systems OFA and FGR are the main methods to reduce NOx emissions. Research in Low NOx-burners is ongoing in many parts of the world. In the U.S. OIT is sponsoring a Crosscutting activity on combustion, while in many states research on low NOx-burners has been sponsored (e.g. in California and New York). The Vision (DOE-OIT 1998b) and Roadmap documents (DOE-OIT 1999a) aim at the development of low-NOx burners with increased efficiency. Efficiency goals are not determined for burners alone, but rather for the systems. For process heating systems the goal is and reduced fuel consumption of 20-50 percent, while reducing criteria pollutant emissions by 90 percent. Research to low NOx-combustion is done at research institutes (e.g. Gas Research Institute, Gas Technology Institute), national laboratories (e.g. Lawrence Berkeley National Laboratory) and universities (e.g. University of California at Irvine), as well as manufacturers. Most burner manufacturers also supply low NOx-burners, e.g. Alzeta, Bloom Engineering, COEN, Detroit Stoker, Hauck, and John Zink. Heat distribution and flux are important design features of furnaces to improve the efficiency. Burner concepts are developed that aim at improving the heat distribution by impulse firing (for heat treatment and intermittent kilns) and high velocity burners (as discussed in the roller kiln for ceramic products). In furnaces recuperators and regenerators are used to improve efficiency. A recuperator is a heat exchanger that extracts heat from the furnace waste gases to pre-heat incoming combustion air. Compared to furnaces without air preheating energy savings of 30 percent can be reached (Flanagan 1993). Development is aimed at higher temperature ceramic recuperators and so-called self-recuperative burners, while minimizing NOx emissions. In self-recuperative burners the recuperator is an integral part of the burner, which decreases costs, and might make it easier to retrofit existing furnaces. Regenerative burners are operated in pairs. While one is used to burn the fuel, another burner uses a porous ceramic bed to store heat. After a short period (minutes) the process is reversed, and the heat stored in the ceramic bed is used to preheat the combustion air. In this way about 85 percent of the heat in the flue gases is recovered, and the combustion air can be pre-heated to temperatures of only 150oC less than the furnace operating temperature (Flanagan 1993). Compared with cold air burners, regenerative burners can achieve fuel savings in excess of 50 percent (Flanagan 1993). However, potential high NOx-emissions may limit preheat temperatures and hence energy savings. Also, the full benefit of the burners depends on the integration in the furnace. For low to medium-temperature applications, we concentrate on burner designs that achieve low-NOx (<20 ppm) while improving energy efficiency. 160

Emerging Energy-Efficient Industrial Technologies

High-Efficiency Low NOx Burners Data Table

Units High-Efficiency Low-NOx Burners other-2 Low-NOx High Efficiency Burners Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Tbtu

Cross-cutting Boilers, Process Heating Natural Gas Retrofit 6758

Conventional burners in existing boilers or furnaces 1 Boilers and furnaces are available in any size 0 kWh 6758 TBtu 6758.0 TBtu Low-NOx High Efficiency Burners 0 kWh 6543.6 TBtu 6543.6 TBtu Commercial 1996 20 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $/MBtu $/MBtu $/kWh $/Mbtu $/Mbtu Years % 0 214.4 214.4 Low 10% 10% of industrial natural gas use is in non-attainment areas 0 21 21.4 7 Full -0.1 N/A 0.94 0.94 3.1 33% Somewhat None Significant H,M,L Medium 0% 3% 3%

Discount rate for all CCE calculations is 15%

Improved burner capacity could lead to higher throughput in specific cases Reduction of NOx emissions by 30 - 70%

H,M,L E,G,F,P

Medium Demonstration, Promotion Poor

Estimates based on few case-studies EIA 1999 EIA 1999 CADDET 1997h, COEN 2000, Berntsson et al. 1997 Author estimate Author estimate CADDET 1997h; Berntsson et al. 1997 Author estimate Steve Londerville, COEN Company (650) 697-0440

For example, the Pyrocore ceramic burner marketed by Alzeta Corp. (after development with the Gas Research Institute, US EPA and Southern California Gas Company) is used for direct firing applications in the food industry (with extremely low emissions) (CADDET 2000c) and for process heaters in the oil industry (CADDET 1989). Energy savings were not specified in either application. In natural gas boilers NOx is mainly generated through thermal processes. Advanced burner designs can reduce NOx emissions, while maintaining or improving efficiency. NOx emissions from standard industrial gas boilers can be between 60 and 200 ppm. Low NOx burners can reduce emissions to 20-30 ppm, while 161

LBNL and ACEEE ultra-low NOx burners (also used for direct heating applications) can reduce emissions to 5-9 ppm. COEN has installed low-NOx burners that comply with California air quality standards in the oil industry (Bakersfield, CA), textile plant (Vernon, CA), as well as heating plants (Sacramento, CA). In these plants NOx emissions between 9 and 26 ppm have been achieved, at high efficiencies. Efficiency gains are not always specified. In the case of a heating plant in Sacramento (CA), an efficiency gain of 3 percent was achieved (COEN 2000). We assume fuel savings of 2 percent for using high-efficiency (ultra) low-NOx burners. While the capital costs of the burner are comparable to those of standard burners, the total system costs would be lower, if SCR NOx-removal would need to be installed. Implementation of high-efficiency Low-NOx burners will be primarily driven by air regulation. We assume that implementation is limited to non-attainment areas. However, no industrial energy consumption data is readily available for nonattainment regions (STAPPA-ALAPCO 1999). In July 2000, 31 regions were non-attainment areas for ground level ozone (EPA 2000b). We estimate that 10 percent of all industrial boiler capacity is found in non-attainment areas, and would need NOx-reduction measures. Natural gas use for boilers in 2015 is estimated at 4118 TBtu (4344 PJ) (AEO 1999). For high-temperature applications NOx-emission reductions are limited by the necessary high flame temperatures needed. Still, modern burners designed to mix the combustion air and fuel well, reduce NOxemissions. The Gyro-Therm burner developed in Australia for the cement industry achieves reductions in NOx-emissions of 30 to 70 percent, while saving 5 percent on fuels in a clinker kiln in the cement industry. The stable flame reduced refractory wear. The technology has been applied in several cement plants around the world including the U.S. (e.g. Ash Grove, Durkee, OR). The payback period is around 2 years (CADDET 1997h). Stordy Combustion Engineering (United Kingdom) has developed a low-NOx regenerative burner that can achieve NOx emissions of 100-125 ppm at air-preheat temperatures of 1000oC (CADDET Newsletter 1999), resulting in fuel savings of 40 percent compared to conventional burners at near-stoichiometric conditions (Flanagan 1993). Potential applications are found in the metals industry, e.g. reheating furnaces, aluminum smelting, copper smelting. For high temperature applications, we assume that new burners can save 5 percent on average for natural gas burners, while maintaining low NOx-emissions, across various process heating applications. As implementation is driven by air quality regulation, uptake of the technology will be highest in nonattainment areas. Due to the lack of data on industrial energy use for process heating applications in nonattainment areas we assume that the technology can be used in 10 percent of natural gas fired heating applications. We exclude natural gas use in the chemical industry (where most gaseous fuels are not necessarily natural gas), glass industry and specific natural gas applications in the steel industry. Hence, natural gas use for process heating in 2015 is estimated at 2640 TBtu (2785 PJ) (AEO 1999). The costs will depend on the individual applications of the burners. The costs of low NOx-burners for large utility boilers is estimated at 10-20 $/kW (Berntsson et al. 1997). The costs are different for the often smaller industrial applications. Based on the case-studies we assume a simple payback period of three years for retrofit-situations. The reduced use of FGR in existing boilers may lead to reduced operating costs, as may reductions in NOx emissions offsets. For example, the ARCO Refinery boiler project in Bakersfield (CA) with a capacity of 62.5 MBtu/hr (65.9 GJ/hr) can result in a reduction $54,000 in reduced emission offsets (or 2.9 $/kW), while reduced use of FGR resulted in an additional saving of $40,000 at full capacity (COEN 2000). These cost-savings will be highly site-dependent. The main driving force for Low NOx-burners is air quality regulation. The relative low cost compared to options like SCR or FGR makes them attractive options. However, in the design of Low NOx burners energy efficiency should be an integrated part of the design. Future steps include the dissemination of information on Low-NOx burner technology to potential users and air quality regulators, as well as demonstration of burners in different applications, especially with respect to furnace applications.

162

Emerging Energy-Efficient Industrial Technologies

Membrane Technology Wastewater (Other-3)

Water is used throughout industry for many applications. Daily industrial water use is estimated at 27,100 million gallons/day in 1995 (Solley et al. 1998), of which 85 percent is disposed after use. There is no information on water use by sector. Large water users are the food, paper, chemical and metal industries. Wastewater is produced in as many industries and may contain many different contaminants, ranging from bio-organic compounds to metal compounds. The water needs to be cleaned before it can be emitted or can be recovered for re-use in the plant. In 1995 only 110 million gallons/day were reclaimed and re-used by industry (Solley et al. 1998). Treatment with chemicals (sanitizing, flocculation), biological treatment, ozonation, ultraviolet treatment, gravity settling, flotation and screening are conventional methods used to clean water. The costs and energy use of wastewater treatment depends heavily on the facility, differences in flow, type of pollutants, as well as type of equipment used. For example, the wastewater treatment costs for printed wiring board manufacturers varied between 0.5$/1000 gallons to 35$/1000 gallons (EPA n.d.). Membranes can also be used to remove dissolved or suspended solids, microbes. The membrane types mostly used in water treatment are ultrafiltration (UF), nanofiltration (NF) and reverse osmosis (RO), while MF is mainly used to stabilize (pre-filter) the water for RO-treatment. The differences between the used membrane methods are the separation capabilities due to size and molecular weight. Membranes have been used for desalting of water for some time now, and facilities using membranes can be found in 13 states in the United States, with most in California and Florida. Some recently constructed plants in Arizona, California, and Hawaii use wastewater as feed. Membrane wastewater treatment plant design starts with the selection of the membrane. The type of membrane material used determines the contaminant rejection characteristics (i.e. chemicals removed from the water), durability and fouling characteristics (Jacangelo et al. 1998). Most membranes used today are polymer membranes, as these have lower costs. Ceramic membranes are more expensive, but can be used at higher pressures and with longer lifetimes (CADDET 1994e). Two membrane processes (e.g. MF and RO) can be combined to remove different contaminants. The main driver for membrane application is the cost of wastewater treatment, and not energy use, although membranes can reduce energy use when compared to evaporation. Koch Membrane Systems estimates the operating costs of an UF-system at 7.5$/1000 gallons compared to 22$/1000 gallons for chemical treatment (KMS 2000) for treatment of 5400 gallons/day of oily wastewater. Membranes have been used in many industries to clean wastewater before disposal and to recover water for internal re-use. Examples can be found in e.g. the metal industries, metal plating, food, paper, chemicals, and electronics. Tri-Valley Growers in Madera, CA installed an UF/RO-membrane system, with help of PG&E and DOE, to reduce wastewater discharge of an olive-oil plant. The system allowed the operation of the plant with zero discharges. The system reduced capital costs and energy costs compared to a biological wastewater treatment system. Gas use was reduced by 55 percent and electricity use by 30 percent, reusing up to 800,000 gallons of water per day (Fok and Moore 1999). The project was a technical success. Unfortunately, the olive oil plant closed recently in response to market changes in the olive oil industry. Replacement of polymer membranes by ceramic membranes in an UF-system to clean wastewater from an enameling plant reduced power consumption by 66 percent, due to the reduced silting of the system (CADDET 1994e). Reduced power and maintenance costs resulted in a simple payback period of 6-7 years, due to the high costs of the ceramic membrane. Barriers to implementation include the lack of information, relative high capital costs, as well as the need for specific membranes of specific applications. Major suppliers are APV (Denmark), Koch Membrane Systems (US), Osmonics (US), U.S. Filter (US). Research aims at new membrane materials and applications, more efficient and longer lasting membranes, and cost reduction of membrane costs. Federal research programs (e.g. ATP) support development of membrane technology, as well as development of specific applications (e.g. DOE, EPA, USDA).

163

LBNL and ACEEE

Membrane Technology Wastewater Data Table

Units Notes Membranes for Wastewater Treatment and Recovery other-3 Use of Membranes to Recover and Clean Industrial Wastewater Market Information: Industries Cross-Cutting Other End-use(s) Wastewater Treatment Fuel, Electricity Energy types New, Retrofit Market segment 23,035,000 2015 basecase use kGd 1000 gallons/day, 1995 industrial water discharge Reference technology Description Evaporation, Clarification 1 Throughput or annual op. hrs. 1000 gallons/day, 1995 industrial water discharge 18 Electricity use kWh/kG kWh/1000 gallons 0.4 Fuel use Mbtu/kG Mbtu/1000 gallons 0.6 Primary energy use Mbtu/kG Mbtu/1000 gallons New Measure Information: Use of Membranes to Recover and Clean Industrial Wastewater Description 45 Electricity use kWh/kG kWh/1000 gallons 0.0 Fuel use Mbtu/kG Mbtu/1000 gallons 0.4 Primary Energy use Mbtu/kG Mbtu/1000 gallons Current status Many applications commercial; new membranes under development Commercial, Research Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources 1990 10 -27 0.4 0.2 Medium 10% -19 276 117.8 30000 Full cost -6400 N/A. -1056.10 -2477.64 4.7 21% Somewhat None Somewhat H,M,L Medium Specificity, Unknown High Dissemination, R&D Fair, Poor Solley et al. 1998 Rough assumption based on process energy use Fok and Moore 1999 Wiesner and Chellam 1999 Rough estimate Koch Membrane Systems and current systems (Koch 2000; EPA n.d.) Case-studies (CADDET) -150% 100% 31% Actual savings depend on application Actual savings depend on application Actual savings depend on application Rough estimate, current use small Assuming daily use for 300 days/year Assuming daily use for 300 days/year

Years kWh/% MBtu/% MBtu/% % TWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years %

$/1000 gallons-day $/1000 gallons-day

Discount rate for all CCE calculations is 15%

Recover water and chemicals, reduced maintenance Reduced water use

H,M,L E,G,F,P

164

Emerging Energy-Efficient Industrial Technologies It is extremely difficult to estimate the potential energy savings from implementation of membranes for water treatment without a detailed study. For specific applications energy savings may be up to 40-55 percent of the energy needs for evaporation. Additional production savings are achieved through product quality, reduced water use, lower operation costs, which are site-specific. Energy use for other treatment may be very low (coagulation, coarse filtration), or high (evaporation). Energy use for evaporation is very high (i.e. 100 MBtu/1000 gallons water evaporated (106 GJ/1000 gallons)). Mechanical vapor recompression would reduce the heat demand, but has still a high power demand of 25-50 kWh/1000 gallons (Fok and Moore 1999). However, most industrial wastewater is probably pre-treated with physical, chemical or biological means before being disposed to the public sewer system or surface water. Large industrial facilities may need to evaporate water for sludge disposal. We assume that 50 percent of wastewater is treated using biological and chemical technologies (consuming 0.8 Mbtu/1000 gallons (0.8 GJ/1000 gallons) and 30 kWh/1000 gallons, derived from Fok and Moore (1999)), and 50 percent uses only coagulation and mechanical treatment (using 4 kWh/1000 gallons). Energy use for a membrane system is estimated at 15-40 kWh/1000 gallons (Fok and Moore 1999). We assume a two-step UF/RO-process with an average consumption of 45 kWh/1000 gallons. The investment and operating costs depend heavily on specific application, and are site-specific. However, for the purposes of this study we make a general estimate, noting that the costs may vary widely in practice. The capital costs of membrane systems are relatively high, but may be lower than alternative processes, as shown by the example of Tri-Valley Growers, CA. Generally, capital costs are expressed per unit of surface area, while about half of the capital costs are for the system components (e.g. pumps, piping) (Wiesner and Chellam 1999). Investment costs are estimated for polymer membrane plants for oily wastewater streams are estimated at 30$/gallon-day (Koch 2000), while annual operating costs are estimated at 5$/gallon-day. Operating costs for a chemical treatment plant are estimated at $10/gallon-day, and for evaporation at 16$/gallon-day (Koch 2000). We assume average operating costs of 11$/gallon-day for non-membrane equipment (EPA n.d.). Re-use of water will reduce the water purchase fees and discharge fees. The reduced costs are estimated at 0.4 cts/gallon-saved, although they may vary by location. Membrane life of a properly operated facility may easily exceed 10 years (Wiesner and Chellam 1999). We assume a lifetime of 10 years. The energy savings and cost estimates are rough. Given the large potential application area and potential energy savings, an in-depth study into membrane applications, energy savings, capital and operational cost benefits is warranted.

165

LBNL and ACEEE

Process Integration/Pinch Analysis (Other-4)

Process Integration (PI) refers to the exploitation of potential synergies that are inherent in any system that consists of multiple components working together. PI is the art of ensuring that the components are well suited and matched in terms of size, function and capability. Pinch Analysis is a tool for determining the optimum process integration strategy, generally applicable for manufacturing processes. Pinch Analysis takes a systematic approach to identifying and correcting the performance limiting constraint (or pinch) in any manufacturing process (Kumana 2000a). It was developed originally in the late 1970s at the University of Manchester in England and other places (Linnhoff 1993) in response to the energy crisis of the 1970s, and the need to reduce steam and fuel consumption in oil refineries and chemical plants by optimizing the design of heat exchanger networks. Since then, the pinch approach has been extended to resource conservation in general, whether the resource is capital, time, labor, electrical power, water, or a specific chemical species such as hydrogen. Energy Optimization. The critical innovation in applying pinch analysis was the development of composite curves for heating and cooling, which represent the overall thermal energy demand and availability profiles for the process as a whole. When these two curves are drawn on a temperature-enthalpy graph, they reveal the location of the process pinch (the point of closest temperature approach), and the minimum thermodynamic heating and cooling requirements. These are called the energy targets. The methodology involves first identifying the targets, and then following a systematic procedure for designing heat exchanger networks to achieve these targets. The optimum approach temperature at the pinch is determined by balancing the capital-energy tradeoffs to achieve the desired payback. The procedure applies equally well to new designs as well as retrofit of existing plants. The analytical approach to this analysis has been well documented in the literature (Kumana 2000b, Smith 1995, Shenoy 1994). Energy savings potential using Pinch Analysis far exceed that from well-known conventional techniques such as heat recovery from boiler flue gas, insulation, and steam trap management. Results from pinch case studies sponsored by various government and industry institutes (DOE, EPRI, GRI) over the last 15 years on thermal pinch compiled by Kumana are reproduced below. Energy Savings Potential Using Pinch Analysis Industry Oil refining Chemicals Food and Beverage Pulp and paper Textiles Iron and steel Source: Kumana 2000b. No. of plants 9 17 18 9 4 2 Percent cost savings 10-40 15-40 7-45 10-35 3-25 11-50 Payback range (years) 0.6-2.8 0.9-4.3 0.7-3.9 0.8-2.4 1.1-4.7 0.9-1.5

While opportunities for energy optimization still exist, even in the energy intensive process industries such as chemicals, petroleum refining, and pulp and paper sectors, the focus here is on emerging applications. Site-wide energy strategies or Total Site analysis. In general, direct heat exchange between hot and cold process steams is practical only within an individual process unit. For large complexes, involving multiple process units, direct heat exchange is usually not feasible, but there can still be opportunity for significant energy savings through indirect thermal integration, through the plant utility (steam and power) system. The technique applied here is to treat the residual heating and cooling requirements of each process unit (after all possible heat recovery has been accomplished within that unit) as if they were process streams, and then develop a set of heat source and sink curves representing the overall site. These curves reveal the optimum number and pressure for steam headers, and the optimum type and size of cogeneration projects. Savings in hot utility (i.e. using process heat to replace utility steam from boilers) of between 10 and 40 percent have been consistently demonstrated in over 70 projects (Eastwood and Kelfkens 1998). 166

Emerging Energy-Efficient Industrial Technologies Process Integration/Pinch Analysis Data Table

Units Process Integration (pinch analysis) other-4 Process integration emerging applications Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Tbtu Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality beneifts Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Cross cutting Process heating, process cooling Fuels, electricity New and retrofit AEO 2000. Based on EIA 1997 energy shares for specific industries 1871

Current energy consumption in existing processes applied to steam use and compression use in refining

TWh Tbtu MBtu

0.9 932.2 939.5

EIA, 1997; Xenergy, 1998; compression electricity use at refineries EIA, 1997. Indirect fuel use for boilers from SIC: 20, 22, and 20% of SIC 28 EIA, 1997

Process integration-emerging technologies TWh 25% compressor savings in refining 0.6 Tbtu 10% savings in various emerging applications 839.0 844.4 Tbtu Comercial Total site pinch, hydrogen pinch, batch processes Years TWh/% TBtu/% TBtu/% % TWh Tbtu Tbtu $/Mbtu-s $ $/kWh $/Mbtu $/Mbtu Years % 15 0.2 93.2 95.0 40% 0.1 37.3 38.0 5.0 Incremental 0 0.86 0.86 0.86 2.3 43% Somewhat None Somewhat H,M,L Information Medium Demonstration Fair EIA 1999 EIA 1997 Kumana, J. 2000b; Linnhoff, B.; Tainsh, B.; Wasilewski, 1999 Author estimate from general pinch literature Author estimate Kumana, J. 2000b; author estimate Kumana, 2000b Kumana & Assoc.-- jkumana@aol.com/Linhoff March (www.linhoffmarch.com) 25% 10% 10%

Discount rate for all CCE calculations is 15%

Can reduce bottlenecks in production lines

H,M,L E,G,F,P

167

LBNL and ACEEE Batch processes. While the methodology for application of pinch analysis to batch processes is not new (Kemp and Deakin 1989, Obeng and Ashton 1988), the market has not caught on, and is nowhere close to reaching its full potential. Two R&D projects carried out under the auspices of the Best Practice program in the UK on batch process integration identified an energy savings of 8 percent and 40 percent respectively. In the first case, in a resin factory, a key savings was the use of condenser heat to pre-heat the reactor fuel and material feeds. These case studies demonstrate that energy savings are not necessarily limited to energy intensive industries, but could have significant applicability to food, pharmaceutical, fine chemicals and other industries where batch processes dominate (ETSU 1999). The major benefit here is not necessarily energy, but productivity of capital and labor. The resource being conserved is processing time, through better scheduling and proper matching of equipment functionality, which means one can get more output from the same plant, or save capital when building a new plant for a given production rate. Water Pinch. While this application is not specifically geared toward energy savings we include it here because water conservation and wastewater minimization has a significant, albeit indirect, impact on energy, and in fact changes the optimum heat recovery strategy (Mann and Liu 1999). Only two companies Aspen Technology, Inc. and Linnhoff March, Ltd. - currently offer commercial-grade software for modeling/reconciling the water balance, and development of reuse/recycle options for wastewater minimization based on pinch principles. Reported savings have ranged between 15 and 60 percent depending on the industry sector (Aspen Technology 2000, Dhole et al. 1996, Linnhoff March Online 2000a, Tripathi 1996, Kumana 1996). Part of the reason that savings are so high is that water conservation has not received the attention it deserves, and so there is a lot of potential savings left on the table. Hydrogen Pinch. In certain process industries, (e.g. petroleum refining), high purity hydrogen gas is a critical (and very expensive) process raw material. During the process operation, the quality of hydrogen is degraded in terms of concentration and pressure. The impure hydrogen is unsuitable for reuse, and is typically burned as fuel. The goal of pinch analysis as applied to hydrogen management is to determine the optimum regeneration, reuse, and recycle strategy to minimize the total costs associated with capital investment in new hydrogen generation and upgrading facilities, and energy consumption (Linnhoff et al. 1999, Eastwood 2000). Reducing hydrogen demand/supply bottlenecks are particularly valued given the potential shortages that many refineries are facing. Using a pinch analysis approach for hydrogen systems has already demonstrated hydrogen feed savings of 25 percent and compression power savings of 35 percent (Linnhoff et al. 1999, Kumana 2000a). Some of the key measures include: the re-use of hydrogen rejected from one process directly in another process, the mixing hydrogen streams of different compositions to provide a stream suitable for re-use, compression and/or purification of reject hydrogen (e.g. using pressure swing adsorption), and process changes to improve hydrogen utilization (Linnhoff et al. 1999). Some of the early U.S. companies to explore the use of hydrogen pinch include Arco (recently purchased by BP Amoco) and Citgo/Lyondell (Kumana 2000a). Pollution Prevention. Pinch Analysis also has applications in pollution prevention. When less energy is consumed, the emission of combustion byproducts (CO2, NOx, SO2) is reduced. When freshwater is conserved, the flow of wastewater effluent is reduced. In addition, methodologies are under development that optimize the selection and sizing of specific pollution control technologies (Kumana and Rossiter 1994, Rossiter 1995, El-Halwagi 1997). However, the commercial experience and success rate so far has been limited. The strength of these new techniques is that they can be combined with existing thermal pinch analysis approaches to give particular process industries a broader array of analytical techniques to identify energy and capital savings. Whether in new plant designs or expansions, it is possible to reduce capital costs by 510 percent and to shorten the construction schedule when applying emerging techniques (Kumana 2000b). By comparison, the cost of undertaking a pinch study is relatively small ($25-250K) with ongoing software support costs of $15,000-25,000/year (Kumana 2000b). Based on the existing experience in the market, it is clear that the application of emerging pinch analysis techniques has the potential to cost-effectively save energy. We estimate 25 percent savings in compression energy use for refineries and 10 percent savings for the application of total site analysis, and batch analysis techniques to the food, textile, and specialty chemicals industries.

168

Emerging Energy-Efficient Industrial Technologies As noted above, careful planning utilizing pinch analysis techniques can reduce new plant construction costs as well. Based on payback times of 1-4 years for most thermal pinch projects, we estimate an average investment cost of $5 per MMBtu ($5.3/GJ) of energy saved. While traditional pinch analysis has been employed in certain industries, it is still underutilized. Kumana argues that this is primarily due to technical misconceptions by plant managers, who often believe that their processes are already optimized and that additional heat recovery projects will not be economical under todays fuel price regime (Kumana 2000b). The enabling technologies that accompany pinch projects are already available in the market; what is lacking is corporate commitment to conserving the earths rapidly dwindling resources of fossil fuels and clean water. Further demonstration projects are necessary to better prove the feasibility of these techniques in the marketplace.

169

LBNL and ACEEE

Process Control and Sensors (Other-5)

Energy currently has a low priority in industrial management practices. Energy management comprises a large variety of measures such as recognizing the importance of energy management, planning, monitoring, and implementing optimal control strategies. Generally, no or low initial costs are involved with these measures. We focus on process monitoring and energy management technologies. It is stressed that training and motivation are important, if not essential, measures in energy management, and should be an integral part of industrial energy management, as well as introduction of new technologies. A variety of process control systems are available for virtually any industrial process. A wide body of literature is available assessing control systems in most industrial sectors such as aluminum, chemicals, pulp and paper, iron and steel. The table provides an overview of classes of process control systems. Classification Of Control Systems and Typical Energy Efficiency Improvement Potentials. System Monitoring and Targeting Characteristics Dedicated systems for various industries, well established in various countries and sectors Improvement of overall economics of process, e.g. stocks, productivity and energy Typical energy savings (percent) Typical savings 4-17%, average 8% , based on experiences in the UK > 2%

Computer Integrated Manufacturing (CIM) Process control

Moisture, oxygen and temperature Typically 2-18% savings control, air flow control Knowledge based, fuzzy logic Note: The estimated savings are valid for the specific application (e.g. lighting energy use). The energy savings can not be added, due to overlap of the systems. Sources: (Caffal 1995, WEC 1995). Modern control systems are often not solely designed for energy efficiency, but rather at improving productivity, product quality and efficiency of a production line. Applications of advanced control and energy management systems are in varying development stages can be found in all industrial sectors. Control systems result in reduced downtime, reduced maintenance costs, reduced processing time, and increased resource and energy efficiency, as well as improved emissions control. For example, in cement kilns NOx emission reductions of 20 percent have been achieved through installing process controls (CADDET 1997i). Many modern energy-efficient technologies depend heavily on precise control of process variables, e.g. strip casting in the steel industry and process integration in the chemical industries. Hence, in estimating the potential energy savings double-counting should be avoided. For this characterization we exclude building energy management systems, lighting controls, as well as adjustable speed drives (discussed elsewhere in this study). Application of process control systems is growing rapidly, and modern process control systems exist for virtually any industrial process. However, still large potentials exist to implement control systems, and more modern systems enter the market continuously. For example, the journal Hydrocarbon Processing produces a semi-annual overview of new advanced process control technologies for the chemical and oil refining industry. Process control systems depend on information of many stages of the processes. A separate but important area is the development of sensors that are inexpensive to install, reliable, and analyze in real-time. Development aims at the use of optical, ultrasonic, acoustic, and microwave systems, that should be resistant to aggressive environments (e.g. oxidizing environments in furnace or chemicals in chemical processes) and withstand high temperatures. The information of the sensors is used in control systems to adapt the process conditions, based on mathematical (rule-based) or neural networks and fuzzy logic models of the industrial process.

170

Emerging Energy-Efficient Industrial Technologies Process Controls and Sensors Data Table
Units Notes Process Controls and Sensors other-5 Modern process control systems using advanced sensors and knowledge-based or fuzzy logic control systems Market Information: Cross-Cutting Industries Processes, utilities, other End-use(s) Excluding motor systems, lighting, HVAC Fuels, electricity Energy types Retrofit Market segment N/A 2015 basecase use Reference technology Energy consumption for applicable process (excl. motors, building energy use) Description Throughput or annual op. hrs. Electricity use TWh 337.1 TWh, based on MECS 1994 (or 31% of industrial power use) 337 Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources TBtu TBtu 12291 15156.4 12.291 Tbtu, based on MECS 1994 (or 89% industrial fuel use, excl. feedstocks)

Modern process control systems using advanced sensors and knowledge-based or fuzzy logic control systems TWh TBtu TBtu 327 11922 14701 Commercialized, research 1995 10 10 369.0 454.9 Low 30% 3 110.7 136.5 6 Full cost -1 0.001 0.20 0.20 2.0 50% Significant Significant Somewhat H,M,L High Technical High Fair Own estimates based on literature survey EIA, 1999 EIA, 1997; Xenergy, 1998 Conservative average of many case-studies Retrofit, costs are estimate of average costs, based on payback of 2 years Rough estimate value of average productivity benefits 3% 3% 3%

Depends on specific application

Years TWh/% TBtu/% TBtu/% % TWh TBtu TBtu $/Mbtu-s

$ $/kWh $/Mbtu $/Mbtu Years %

Discount rate for all CCE calculations is 15%

Reduced downtime and maintenance costs, improved yield Less off-spec production Reduced emissions, improved yield Process control is recognized as important measure

H,M,L E,G,F,P

Average of many case-studies Gensym Corp. (MA) (617) 547-2500

171

LBNL and ACEEE Neural network-based control systems have successfully been used in the cement (kilns), food (baking), non-ferrous metals (alumina, zinc), pulp and paper (paper stock, lime kiln), petroleum refineries (process, site), and steel industries (EAFs, rolling mills). New energy management systems that use artificial intelligence, fuzzy logic (neural network), or rule-based systems mimic the best controller, using monitoring data and learning from previous experiences. For example, improved process control using neural networks in electric arc furnaces in the steel industry can help to reduce electricity consumption beyond that achieved through classical control systems. Neural networks or fuzzy logic systems analyze data and emulate the best controller. For EAFs, the first fuzzy logic control systems have been developed using current, power factor and power use to control the electrodes in the bath (Staib and Bliss 1995). The average power savings are estimated to be up to 8 percent (or 38 kWh/t), with an average increase in productivity of 912 percent and reduced electrode consumption of 25 percent (Staib and Bliss 1995). The actual savings depend on the scrap used and the furnace operation. Furnace maintenance costs are reduced as well. In 1994, advanced control systems were installed at 16 furnaces in the U.S. (Kimmerling 1997), with a total capacity of 6.4 million tons (5.8 Mt) (equivalent to 9 percent of the U.S. EAF capacity in 1994). The capital and commissioning costs are estimated to be $250,000 per furnace, with annual costs savings at roughly $0.90/ton ($1/t) (Kimmerling 1997). The capital costs are expected to be $0.86/ton ($0.95/t) (Worrell et al. 1999). The measure is assumed to be applicable to 90 percent of the U.S. EAF capacity. Similar applications are found in the cement industry, where energy savings of up to 8 percent have been found, with a payback period between 1 and 2 years (CADDET 2000d). Process knowledge based systems (KBS) have been used in design and diagnostics, but are hardly used in industrial processes. KBS incorporates scientific and process information applying a reasoning process and rules in the management strategy. A recent demonstration project in a sugar beet mill in the UK using model based predictive control system demonstrated a 1.2 percent reduction in energy costs, while increasing product yield by almost 1 percent and reducing off-spec product from 11 percent to 4 percent. This system had a simple payback period of 1.4 years (CADDET 2000e). Although, energy management systems are already widely disseminated in various industrial sectors, the performance of the systems can still be improved, reducing costs and increasing energy savings further. For example, total site energy monitoring and management systems (Kawano 1996) can increase the exchange of energy streams between plants on one site. Traditionally, only one plant or a limited number of energy streams were monitored and managed. Research for advanced sensors and controls is ongoing in all sectors, both funded with public funds as private research. Several projects within DOEs Industries of the Future try to develop more advanced control technologies, and sensors and controls are also represented in a Crosscutting OIT-program. Outside the U.S., Japan and Europe also give much attention to advanced controls. The main opportunities can be found in further development of advanced controls and sensors, as well as the marketing of existing advanced controls. In our analysis we will assume ongoing development of energy management systems, resulting in improved performance through better control strategies and improved and real-time information as well as lower costs. We assume that on average energy efficiency savings of 3 percent are feasible. We exclude electricity use for motors and energy use for industrial buildings, as this is covered under other technologies (ASD, building management systems). By 2015 we assume that modern process control systems can be applied to an additional 30 percent of applicable industrial energy use. Estimating the specific costs of installing energy management systems is difficult. The pay back period of such systems is often not only influenced by energy savings, but more often by non-energy benefits which have a large impact (e.g. improved process throughput or product quality). Investment costs vary typically between 0.5 and 30$/MBtu-(0.47 and $28.40 per GJ) saved (ETSU 1994), with pay back periods mostly from 1 to 4 years (Caffal 1995) in industrial applications. We assume an average payback period of 2 years. The lifetime of EMS is dependent on the equipment for which it is used and progress in development of new controls. We estimate it to be 10 years on average for this study. Future steps include further development of new sensors and control system, demonstration in commercial scale, as well as dissemination of the benefits of control systems in a wide variety of industrial applications.

172

Emerging Energy-Efficient Industrial Technologies

Variable Mining Machine (Mining-1)

The coal mining industry mined over 1.1 billion tons of coal in 1998, of which 63 percent in surface mines and 37 percent in underground mines. Because the variable mining machine can be used underground, we focus on underground coal mining. Underground coal mines use different technologies including conventional, continuous, shortwall and longwall mining. Conventional and continuous mining are slowly decreasing, due to their relatively lower productivity. Longwall mining is the most productive underground mining method. In 1998, almost 200 million tons of coal were mined in longwall mines. Longwall mining has increased from 27 percent of 1983 production (EIA 1995) to 63 percent in 1998 of total underground production. Longwall mines can be found in Alabama, Colorado, Illinois, Kentucky, Maryland, Ohio, Pennsylvania, Utah, Virginia, West Virginia and Wyoming. These mines are operated by 24 companies, of which CONSOL Energy operates the largest number, followed by RAG American Coal Co., and Jim Walter Resources (Fiscor 2000). Fifty-nine longwall mining machines are operated in the U.S. mines, and it is expected that this will not change much in the near future. In the 1980s about 100 longwall installations were used in the U.S. (EIA 1995). The number has declined, but the average capacity has increased. Longwall mining has originally been developed in Europe to increase productivity of underground coal mines. Compared to conventional methods no pillars are needed, enabling to recover 25 percent more coal. In modern mechanized longwall operations, the coal is cut and loaded onto a face conveyor by continuous longwall miners called shearers or plows. The roof is supported by mechanized, self-advancing supports called longwall shields, which form a protective steel canopy under which the face conveyor, miners, and shearer operate. In combination with shields and conveyors, longwall shearers or plows create a continuous mining system with a huge production capacity. Two main longwall systems are widely practiced. The system described above, known as the retreating method, is the most commonly used in the United States (Britannica 2000). Longwall mining machines are marketed by various firms, e.g. DBT America (PA) and Longwall Associates (VA). Coal mining consumed about 13.5 TWh electricity and 88 TBtu in fuels to mine and wash the coal (AEO 1999). By 2015 EIA expects the coal industry to consume 15.6 TWh and 103 TBtu (AEO 1999). There is no information on the energy use by surface and underground coal mining separately. Underground mining is more energy-intensive. The average horsepower on the shearer was 1180 hp (Fiscor 2000). Energy use in coal mines on the depth, size and type of coal. Hence, it is difficult to estimate specific energy consumption for coal mining. We roughly estimate the energy use for the cutting of a longwall miner at 0.75 kWh/ton of coal (Kelley 2000). More energy is used for transport and ventilation. The variable wall mining machine is a variation on the longwall miner. It uses cutting heads that move vertically and sweep across the coal face. The important improvement is that the variable mining machine provides a separation between the coal face and the miners, providing a dual duct ventilation system. This would reduce the exposure of the miners to dust and methane, improve working conditions and safety considerably. The variable wall mining machine is developed by Kelastic Mine Beam Co., Greensburg, PA. The machine has been tested by the former U.S. Bureau of Mines and in a mine in Western Kentucky. The development has been supported by DOEs Inventions and Innovation Program in 1999. We have only been able to find data on energy savings in the cutting of the coal. The developers estimate the energy savings at roughly 20 percent, depending on mine conditions, of the energy use of longwall mining machine (Kelley 2000). Additional savings may be possible in ventilation, but were not quantified. We estimate the energy savings at 0.15 kWh/ton of coal-mined. Longwall miners are capital intensive machines, with a long lifetime. The longwall mining machine installed in 1994 in the Robinson Run Mine (West Virginia), owned by CONSOL Coal Group, consisted a 42 inch coal shearer and 172 hydraulic roof supports. The total capital costs were $15 million (EIA 1995), with an estimated production of 4.8 Million tons/year (1997) (Consol Energy 2000).

173

LBNL and ACEEE

Variable Mining Machine Data Table

Units Variable Wall Mining Machine Mining-1 Variable Wall Mining Machine with Dual Duct Ventilation Market Information: Industries End-use(s) Energy types Market segment 2015 basecase use Reference technology Description Throughput or annual op. hrs. Electricity use Fuel use Primary energy use New Measure Information: Description Electricity use Fuel use Primary Energy use Current status Date of commercialization Est. avg. measure life Savings Information: Electricity savings Fuel savings Primary energy savings Penetration rate Feasible applications Other key assumptions Elec svgs potential in 2015 Fuel svgs potential in 2015 Primary energy svgs potential in 2015 Cost Effectiveness Investment cost Type of cost Change in other costs Cost of saved energy (elec) Cost of saved energy (fuel) Cost of saved energy (primary) Simple payback period Internal rate of return Key non energy factors Productivity benefits Product quality benefits Environmental benefits Other benefits Current promotional activity Evaluation Major market barriers Likelihood of success Recommended next steps Data quality assessment Sources: 2015 basecase Basecase energy use New measure energy savings Lifetime Feasible applications Costs Key non energy factors Principal contacts Additional notes and sources Notes

Coal Mining Other Electricity New 60

59 Longwall mining machines, total production 200 Million tons of coal

Modern Longwall Mining Machine (cutting only) 3,400,000 tpy Average size of longwall miners in US 2,540,000.0 kWh 0 MBtu 21590.0 MBtu Variable Wall Mining Machine with Dual Duct Ventilation 2,030,000 kWh 0.0 MBtu 17255.0 MBtu Prototype demonstration Currently no active development ongoing 25 Years kWh/% MBtu/% MBtu/% % GWh Tbtu Tbtu $ $ $/kWh $/Mbtu $/Mbtu Years % 510000 0.0 4335.0 Low 20% 6 0 0.1 200000 Incremental -10000 0.04 N/A. 4.83 10.6 7% Somewhat None None Significant Low Structure Mining Industry Low Commercial Demonstration Fair EIA 1999 EIA 1995a; EIA 1999; Kelley, 2000 Kelley 2000 Authors estimate Authors estimate Authors estimate Kelley 2000 J.H. Kelley, Kelastic Mine Beam Co. (724) 832 8832 Mike Plaha, RAG American Coal (410) 689-7500 Estimated costs of a new mining machine are $10.6 Million Improved working conditions and safety, improved productivity 20% N/A. 20%

Discount rate for all CCE calculations is 15%

Automated system may reduce costs compared to older longwall machine

Improved working conditions and safety

H,M,L

H,M,L E,G,F,P

174

Emerging Energy-Efficient Industrial Technologies The costs of a variable wall mining machine are likely to be similar or slightly higher to those of traditional longwall mining machines. The slightly higher costs may be due to the costs of the dual ventilation system. However, this may lead to operational cost savings. No specific cost data was available at the time of the study. We estimate the additional investments at $200,000 per machine, compared to the costs of a modern longwall mining machine of $10.6 million (EIA 1995). We assume reduction in annual costs due to improved automation and working environment safety. Currently, no further development work is going on, and implementation of the technology has not happened, due to the limited number of new longwall/variable mining machines installed in the U.S., as well as further concentration of equipment manufacturers and users. Demonstration of the technology on a close to commercial scale in collaboration with the coal mining industry would be necessary to demonstrate the potential benefits of this technology.

175

LBNL and ACEEE

REFERENCES
33Metalproducing.com Online. 2000a. Crawfordsville Eyed for Strip-Casting Plant. Industry News Briefs. Cleveland, OH: 33Metalproducing.com Online. Online at: www.33metalproducing.com. (Steel-2) -----. 2000b. Nucor Plans to Build U.S. Hismelt Plant. In 33 Metalproducing Online. June. Online at: http://www.33metalproducing.com/members/industry_news/2000_news/june.html#anchornucor. Cleveland, OH: 33Metalproducing.com Online. (Steel-5) Abrahamsson, K., G. Aly, A. Jernquist, and S. Stenstrom. 1997. Application of Heat Pump Systems for Energy Conservation in Paper Drying. In International Journal of Energy Research. vol. 21: 631-642. (Paper-5) [AEO] Energy Information Administration, U.S. Department of Energy. 1999. Annual Energy Outlook 2000. DOE/EIA-0383(2000). Washington, DC: U.S. Department of Energy, Energy Information Administration. [AISI]. 1998. Steel Industry Technology Roadmap. Online at: www.steel.org. February. Washington, DC: American Iron and Steel Institute. http://www.oit.doe.gov. (Steel-5)(Steel-2) Albers, T. 1998. Personal communication to Sam Wheeler. May. St. Louis, MO: U.S. Electric Motors Corporation. 941-746-3515. (Motorsys-7) Alderfer, R.B., M.M. Eldridge, G. Nakarado, and T. Starrs. 2000. Making Connections: Case Studies of Interconnection Barriers and their Impact on Distributed Power Projects. May. NreL/SR-200-28053. Golden, CO: National Renewable Energy Laboratory. (Utilities-2) (Utilities-4) Aluminum Association. 2000. Website: www.aluminum.org. (Alum-3) American Combustion, Inc. 2000. American Combustion: Rolling Mill - Reheat Furnace, Pyretron. Atlanta, Georgia: American Combustion, Inc. Online at: http://www.americancombustion.com/html/reheat_furnace.html. (Steel-4) [ANL] Argonne National Laboratory. 1998. Biodesulfurization of Gasoline A Technology Roadmap. Argonne, IL: Argonne National Laboratory. (Refin-1) [ASME] American Society of Mechanical Engineers. 1999. Technical Working Group on Inert Anode Technologies. Report prepared for the U.S. Department of Energy, Office of Industrial Technologies. Washington, DC: U.S. Department of Energy. (Alum-4) Aspen Technology, Inc. 2000. Website: http://aspentech.com (Other-4) Bautista, P. (Onsite Sycom Energy Corp.). 2000. Personal communication to Neal Elliott. August. Washington, DC. (Utilities-4) [BEA] Bureau of Economic Analysis, U.S. Department of Commerce. 2000. Summary Statistics from the BEA website. http://www.bea.doc.gov/. Washington, DC: U.S. Department of Commerce. (Overview) Belzer, D., D. Greene, J. Roop, and R. Sands. 1995. Energy Conservation Trends: Understanding the Factors Affecting Energy Conservation Gains and Their Implications for Policy Development. Washington, DC: U.S. Department of Energy, Office of Policy. (Overview) Berger, H. and P. Mittag. 1995. The Comelt Electric Arc Furnace with Diagonally Ordered Electrodes. In Stahl und Eisen 115 (9): 53-58 (in German). (Steel-3) Berglin, N., L. Persson, T. Berntsson. 1996. Energy System Options with Black Liquor Gasification. In Pulp and Paper Canada. 97(5). (Paper-1) Bernow, S., K. Cory, W. Dougherty, M. Duckworth, S. Kartha, M. Ruth. 1999. Americas Global Warming Solutions. Washington, DC: World Wildlife Fund and Energy Foundation. (Overview) Berntsson, T., P-A Franck, A. Asblad. 1997. Process Heating in the Low and Medium Temperature Ranges. CADDET Analyses Series, No. 22. Sittard, The Netherlands: IEA-CADDET. (Other-2) Betts, K. 1999. Ford Opens Door to Plastics Recycling. In Environmental Science and Technology News. August 1. See www.salyp.com. (Chemicals-8)

176

Emerging Energy-Efficient Industrial Technologies

Biddle, M. (MBA Polymers). 2000. Personal communication with Nathan Martin. August-September. (Chemicals-8) Blok, K., W.C. Turkenburg, W. Eichammer, U. Farinelli, T.B. Johansson (ed.). 1996. Overview of Energy RD&D Options for a Sustainable Future. European Commission, Directorate-General XII. Luxembourg: Office for Official Publications of the European Communities. (Utilities-3) Boerner, J., and D. Orloff. 1994. Effects of Basis Weight and Freeness on Sheet Permeability and Critical ImpulseDrying Temperature. In Tappi Journal. 77 (2): 163-168. Boteler, R. (U.S. Electric Motors). 2000. Personal communication to Neal Elliot. March, August. St. Louis, MO. (Motorsys-7) (Motorsys-4) Bott, T.R. Biofouling Control with Ultrasound. In the journal Heat Transfer Engineering. May/June. 21(3). Taylor and Francis, Inc. (Refining-2) Brooks, R. 1997. Norandal Speeds Delivery for thin-gage Aluminum. In 33Metalproducing Online. Cleveland, OH: 33Metalproducing.com Online. Online at: www.33metalproducing.com. (Alum-1) Brownell, K.A. 1998. Investigation of the Field Performance for Industrial Refrigeration Systems. M.Sc. Thesis. Madison, WI: University of Wisconsin. (Food-4) Brumm, J. 2000. Alcoas new smelter technology could reshape industry. Reuters news story, June 22nd. (Alum-4) [CAC] Compressed Air Challenge. 2000. Web site http://www.knowpressure.org/. (Motorsys-1) All of the following CADDET citations can be found on the CADDET website (www.CADDET-ee.org) by searching in the Register Database. Please see the first CADDET citation for full publication information. [CADDET] Centre for the Analysis and Dissemination of Demonstrated Energy Technologies. 1989a. Filtration and recycling of stenter exhaust in textile finishing. In CADDET Energy Efficiency Register. Technical Brochure R044. Article online at: www.CADDET-ee.org. -----. 1989b. The Pyrocore ceramic fiber-matrix burner. Technical Brochure 38. (Other-2) -----. 1990. Cooling system for fruit and vegetable storage plant. Technical Brochure R061. December. (Food-4) -----. 1992. Super compact heat exchanger in a corrosive application. Technical Brochure R100. (Chemicals-3) -----. 1993a. An ice pond system for industrial process cooling. Technical Brochure R135. June. (Food-4) -----. 1993b. Heat recovery at a Nitric Acid plant. Technical Brochure R166. (Chemicals-3) -----. 1993c. Energy-saving roller kiln. Technical Brochure R132. (Ceramics-1) -----. 1993d. Recuperative aluminium recycling plant. Technical Brochure R130. (Alum-3) -----. 1993e. Anaerobic digestion of paper mill effluent. Technical Brochure R157. Online at: http://www.CADDETee.org/techpdf/R157.pdf. (Other-1) -----. 1994a. Heat Recovery and Energy Saving in a Bakery. Project No. UK-90-079. (Food-2) -----. 1994b. Integrated Heat Recovery in a Malt Whisky Distillery. Project No. UK-94-509. (Food-2) -----. 1994c. Heat Recovery from a Soya Bean Toaster. Technical Brochure R184. (Food-2) -----. 1994d. Optimisation of drying-hood ventilation in the paper industry. Project number NL-94-541. Online at: http://www.CADDET-ee.org/register/dataee/CCE01994.htm (Paper-5) -----. 1994e. Ultra-filtration system uses ceramic filters. Technical Brochure R186. (Other-3) -----. 1994f. Heat recovery after catalytic combustion in printing. Technical Brochure R185. -----. 1995a. Integrated Heat Recovery in a Food Factory. Technical Brochure R221. (Food-2) -----. 1995b. Electrical impulse drying in the paper and pulp industry. Canadian Register, CA-95-503. (Paper-7) -----. 1996a. CADDET Register on Energy Efficiency Demonstration Projects. Sittard, The Netherlands: CADDET. (Food-3) -----. 1996b. Retrofit cogeneration system at milk processing plant. Technical Brochure R257. September. (Food-4) -----. 1996c. Demonstration of innovative equipment to decoat scrap aluminium. Technical Brochure R268. (Alum3) -----. 1996d. Anaerobic waste water treatment in a whey processing company. Technical Brochure D027. Online at: http://www.CADDET-ee.org/techpdf/R235.pdf. (Other-1)

177

LBNL and ACEEE

-----. 1997a. Bakery to Brewery Waste Heat Recovery for Process Hot Water Heating. Project No. US-90-138. (Food-2) -----. 1997b. High Efficiency Kiln Uses Indirect Fired Heaters and Air-to-air Heat Exchangers for Malt Drying. Project No. CA-87-016. (Food-2) -----. 1997c. Heat pipe saves energy in the baking industry. (Heat recovery without food contamination in a bakery.) March. Technical Brochure R277. Project No. AU-96-517. (Food-2) -----. 1997d. Mixture Preheating Installation for a Glass Furnace. Project No. NL-93-529. (Glass-1) -----. 1997e. Waste from a potato peeling company provides biogas for electricity production. Project no. NL-97-508. http://www.CADDET-ee.org/register/datare/CCR01915.htm. (Other-1) -----. 1997f. Waste water pretreatment recovery system uses anaerobic digestion to produce biogas. Project no. US90-141. Online at: http://www.CADDET-ee.org/register/dataee/CCE01403.htm. (Other-1) -----. 1997g. Simultaneous combustion of natural gas and biogas. Project no. NL-97-501. Online at: http://www.CADDET-ee.org/register/datare/CCR01908.htm. (Other-1) -----. 1997h. Revolutionary low-NOx high-efficiency gas burner. Technical Brochure R284. (Other-2) -----. 1997i. A cement furnace controlled by fuzzy logic. Project No. NL-92-018. (Other-5) (SCQ). Note: Project was completed in 1992. -----. 1997j. Energy efficiency and environmental benefits for a coffee roasting company. Technical Brochure R275. (Food-4) -----. 1998a. Concentration and desalination of whey in the dairy industry. Project No. NL-98-520. (Food-3) -----. 1998b. Anaerobic water purification in a brewery in the Netherlands. Technical Brochure 70. February. Online at: http://www.CADDET-ee.org/register/datare/CCR01990.htm. (Other-1) -----. 1999a. Heat recovery from blow down steam in the food industry. Project No. NL-99-501. (Food-2) -----. 1999b. Process integration in a brewery. Project No. NL-93-516 (Food-2) -----. 1999c. Membrane technology used to separate gaseous mixtures. Technical Brochure R362. (Chemicals-2) -----. 1999d. Optimising the electrolytic process in an aluminum plant. Technical Brochure R349. http://www.CADDET-ee.org/techpdf/R349.pdf (Alum-2) -----. 1999e. Energy from a flour mill waste water. Project no. CA-90-105. Online at: http://www.CADDETee.org/register/dataee/CCE00132.htm. (Other-1) -----. 2000a. Membrane filtration in the production of light and near beer. Project No. NL-99-511. (Food-3) -----. 2000b. Natural gas engine-driven low temperature NH3 refrigeration. Project No. US-93-535. (Food-4) -----. 2000c. Industrial hot air burner. Project No. US-97-502. (Other-2) -----. 2000d. Long term success with advanced controls. Project No. UK-00-503. (SCQ) (Other-5) -----. 2000e. Energy saving using model-based predictive control. Technical Brochure R371. (SCQ) (Other-5) CADDET Newsletter. 1997. Pertraction: a promising new technology for cleaning industrial waste water. In CADDET Newsletter. June, No.2. Online at: http://www.CADDET-ee.org/listings/nl_eus.htm. -----. 1997. Advanced control techniques improve drying energy performance. No. 4. (Food-2) -----. 1997. Reclaiming cleaning water using ultrafiltration and double pass reverse osmosis. June, No. 2. -----. 1997. Low temperature drying of food materials using energy efficient heat pump dryers. No.4 -----. 1999. 21st Century combustion technology now available. March, No. 1. pp.7-8. (Other-2) -----. 1999. Energy recovery from dust-laden gas with self-cleaning boiler. March, No. 1. -----. 1999. Heat recovery from flue gases with absorption cooling. March, No. 1. -----. 1999. Variable speed offers a new option in compressed air. and Electricity consumption of compressed air reduced by 60%. September, No. 3. Caffal. 1995. Energy Management in Industry. CADDET Analyses Series Report, No. AR_17. Sittard, The Netherlands: IEA-CADDET. (SCQ) Capstone, Inc. 2000. http://www.capstoneturbine.com/. (Utilities-4) [Census] U.S. Bureau of the Census. 1996. 1992 Census of Manufactures. Subject Series: General Summary. October. Report no. MC92-S-1. Washington, DC: U.S. Department of Commerce. (Textile-1) (Utilities-4) (Chemicals-2) (Chemicals-5) -----. 1998. 1996 Annual Survey of Manufacturers. Washington, DC: U.S. Department of Commerce. Available online at: http://www.census.gov. (Overview) -----. 2000. 1999 Current Industrial Report: Broadwoven Fabrics (Gray) (MQ313T(99)-9). Washington, DC: U.S. Department of Commerce. (Textile-1)

178

Emerging Energy-Efficient Industrial Technologies

Chemical and Engineering News. 1997. Chemical and Engineering News Production Statistics. Vol. 72 (27). (Chemicals-8) Christensen, T.S. and I.I. Primdahl. 1994. Improve Syngas Production Using Autothermal Reforming. In Hydrocarbon Processing 73 (3): 39-44, ASG-3. (Chemicals-7) [CMA] Chemical Manufacturers Association. 1996. U.S. Chemical Industry Statistical Handbook. Arlington, VA: Chemical Manufacturers Association. (Chemicals-7) COEN-Website. 2000. http://www.coen.com. (Other-2) Comb, L. 1995. Advances in Membrane Technology for Beverage Water Treatment. In Fruit Processing. September 1995. (Food-3) Conchie, S. 1993. Dryer hood optimization at Scott Northfleet (Cinderella comes of age!) In the journal Paper Technology. Pgs. 39-42. (Paper-5) Consol Energy. 2000. Website: http://www.consolenergy.com/operations/usmines.htm (Mining-1) Consonni, S., E.D. Larson, T.G. Kreutz, N. Berglin. 1998. Black Liquor Gasifier/Gas Turbine Cogeneration. July. American Society of Mechanical Engineers. (Paper-1) Czuppon, T.A. 1994. Advanced Ammonia Technologies. In Fertiliser Industry Annual Review. Vol. 17: 63-68. (Chemicals-7) Czuppon, T.A., S.A. Knez, and R.B. Strait. 1996. Commercial Experience of KAAP and KRES. In Proceedings 1996 AIChE Safety Symposium. Boston, MA. (Chemicals-7) Daaland, O., I. Alvestad, A.B. Espedal, and M.L. Nedreberg. 1997. Thin Gauge Twin-roll Casting, Process Capabilities and Product Quality. In Light Metals. Minerals, Metals, and Materials Society. (Alum-1) Daniels, E. (Argonne National Laboratory). 2000. Personal communication Nathan Martin regarding froth flotation. August. (Chemicals-8) De Beer, J. 1998. Potential for Industrial Energy Efficiency Improvement in the Long Term. Ph.D. Thesis. Utrecht, The Netherlands: Utrecht University. (Paper-2) (Paper-4) (Paper-7) De Beer, J., K. Block, M.T. van Wees, and E. Worrell. 1994. Icarus-3. The Potential of Energy Efficiency Imrovement in the Netherlands up to 2000 and 2015. Utrecht, The Netherlands: Utrecht University. (Paper-5) (Utilities3) De Beer, J., K. Block, E. Worrell. 1998a. Future Technologies for Energy-Efficient Iron and Steelmaking. In Annual Review of Energy and the Environment. Vol. 23: 123-205. (Steel-2) (Steel-4) (Steel-5) -----. 1998b. Long-term energy-efficiency improvements in the paper and board industry. In Energy. 23 (1): 21-42. (Paper-4) (Paper-7) De Canio, S. and J. Laitner. 1997. Modeling Technological Change in Energy Demand Forecasting. In Technological Forecasting and Social Change. 55 (3): 249-263 (Overview) DeGaspari, J. 1999. From Trash to Cash. In Mechanical Engineering Magazine. June. Online at: http://www.memagazine.org/backissues/june99/features/trash/trash.html (Chemicals-8) De Saro, R. (Energy Research Company). 2000. Personal communication to Nathan Martin regarding IDEX technology. August. (Alum-3) De Saro, R., W. Bateman, R. Jain. 1999. A Vertical Flotation Melter for Decoating and Melting Scrap Aluminum. Staten Island, NY: Energy Research Company. (Alum-3) Dhole, V., N. Ramchandani, R. Tainsh, M. Wasilewski. 1996. Make Your Process Wastewater Pay for Itself. In Chemical Engineering. Vol. 103 (1). (Other-4) Dimond, P. (Weyerhauser Corporation). 2000. Personal communication to Nathan Martin. July. (Paper-2)

179

LBNL and ACEEE

[DOE] U.S. Department of Energy. 1998. Improving Compressed Air System Performance: A Source Book for Industry. April. Washington, DC: Motor Challenge and the Compressed Air Challenge. (Motorsys-2) (Motorsys-3) -----. 1999a. Microturbines: What is a Microturbine? OITs Industrial Power Technologies. Washington, DC: U.S. Department of Energy. Online at: www.oit.doe.gov/power/micro1.html. (Utilities-4) -----. 1999b. Microturbines: Program Overview. OITs Industrial Power Technologies. Washington, DC: U.S. Department of Energy. Online at: www.oit.doe.gov/power/micro.html. (Utilities-4) -----. 1999c. Summary of the Microturbine Technology Summit: Orlando, FL. December 7-8, 1998. Draft Report prepared by Energetics, Inc. January. Oak Ridge, TN: U.S. Department of Energy, Oak Ridge National Laboratory. (Utilities-4) -----. 1999d. Optimized Pump Systems Save Coal Preparation Plant Money and Energy. Motor Challenge Technical Case Study. Washington, DC: U.S. Department of Energy, Office of Industrial Technologies. (Motorsys-6) -----. 2000a. Department of Energy Announces Worlds First Hybrid Fuel Cell-Turbine. April 17. (Utilities-3) -----. 2000b. Energy Department Awards $4 Million for Aluminum Research. Press release. March. (Alum-4) [DOE-OIT] U.S. Department of Energy, Office of Industrial Technologies. 1998a. Energy and Environmental Profile of the U.S. Petroleum Refining Industry. Washington, DC: U.S. Department of Energy, Office of Industrial Technologies. (Refin-1, table) -----. 1998b. A Vision by and for the Industrial Combustion Community. May. Washington, DC: U.S. Department of Energy, Office of Industrial Technologies. (Other-2) -----. 1999a. Industrial Combustion Technology Roadmap. April. Washington, DC: U.S. Department of Energy, Office of Industrial Technologies. (Other-2) -----. 1999b. Impacts: Turning Industry Visions into Reality. Washington, DC: U.S. Department of Energy, Office of Industrial Technologies. -----. 2000a. Energy and Environmental Profile of the U.S. Chemical Industry. Washington DC: U.S. Department of Energy, Office of Industrial Technologies. (Chemicals-1) (Chemicals-2) (Chemicals-5) (Chemicals-6) -----. 2000b. Decision Tools for Industry. PSAT (Pumping System Assessment Tool) CD-ROM. Washington, DC: U.S. Department of Energy and ORC Macro. (Motorsys-6) Drescher, S., J. Kozak, M. Okos, and N. Rao. 1997. A Review of Energy Use in the Food Industry. In 1997 ACEEE Summer Study on Energy Efficiency in Industry, Proceedings. Washington DC: American Council for an Energy-Efficient Economy. (Food-2) (Food-3) Dudley, D. 2000. Weyerhauser. High Consistency Forming. Personal Communication with Nathan Martin. September. (Paper-6) Dyllus, F.U., H. Feldmann, V. Groth, I. Gyngys, K.P. Hassel, D. Moritz, H. Muder, P. Rocholl, D. Weidemann, and R. Wrgler. 1991. A Technical and Economical Comparison of Conventional and Continuous Casting and Rolling Methods for the Production of Cold-Rolled Aluminum Strip. Frankfurt, Germany: Deutsche Gesellschaft fr Materialkunde e.V. (German Society for Materials). (Alum-1) Dziezak, J.D. 1990. Membrane Separation Technology Offers Processors Unlimited Potential. In Food Technology. September, pp.108-113. (Food-3) E Source, Inc. 1997. Technology Atlas Series, Volume I: Lighting. Boulder, CO: E Source, Inc. (Lighting-1) (Lighting2) Eastwood, A.R. and R.C. Kelfkens. 1998. Development of Site Wide Energy Strategies. Paper presented at Chemical Engineering Expo, Houston, TX. (Other-4) Eastwood, A.R. (Linnhoff March Online) 2000. Personal communication with Nathan Martin regarding Pinch Analysis. (Other-4) [EBN] Environmental Building News. 2000. T-5 Fluorescent Lighting Outshines HIDs. In Environmental Building News. July/August. p. 8-11. (Lighting-1) (Lighting-2) [EFMA] European Fertilizer Manufacturers Association. 1995. Production of Ammonia, Best Available Techniques for Pollution Prevention and Control in the European Fertilizer Industry. Brussels, Belgium: EFMA. (Chemicals-7) [EIA] 1995a. Longwall Mining. March. DOE/EIA-TR-0588. Washington, DC: U.S. Department of Energy, Energy Information Administration. (Mining-1) -----..1995b. Petroleum Supply Annual 1994 Volume 1. DOE/EIA-0340(94)/1. Washington, DC: U.S. Department of Energy, Energy Information Administration.

180

Emerging Energy-Efficient Industrial Technologies

-----. 1997. Manufacturing Consumption of Energy 1994. DOE/EIA-0512(94). Washington, DC: U.S. Department of Energy, Energy Information Administration. -----. 1999. Annual Energy Outlook 2000. DOE/EIA-0383(2000). Washington, DC: U.S. Department of Energy, Energy Information Administration. -----. 2000. Summary Statistics from the EIA website: www.eia.doe.gov. Washington, DC: U.S. Department of Energy, Energy Information Administration. (Overview) Eichhammer, W. 1995. Energy efficiency in industry: Crosscutting technologies. In Overview of Energy RD&D Options for a Sustainable Future. By K. Blok, W.C. Turkenburg, W. Eichhammer, U. Farinelli and T.B. Johansson. June. European Commission DG XII, Science, Research and Development. (Food-3) Einstein, D., M. Khrushch, E. Worrell. 2000. Industrial Sector Steam Systems - Energy Use Baseline and Energy Efficiency Measures. Berkeley, CA: Lawrence Berkeley National Laboratory. (Other-2) (Overview) Eisenhauer, J., N. Margolis, S. McQueen, and R. Scheer. 1997. Aluminum Industry Technology Roadmap Workshop. May. Document prepared for the Office of Industrial Technologies, U.S. Department of Energy, and the Aluminum Association, Inc. Washington, DC: U.S. Department of Energy. Online at: http://www.oit.doe.gov/aluminum/aluminum_roadmap.shtml#pperf. (Alum-2) (Alum-4) Eklund, R. (Weyerhauser Corporation). 2000. Personal communication with Nathan Martin regarding High Consistency Forming. July. (Paper-6) Elaahi, A. and H.E. Lowitt. 1988. The US Pulp and Paper Industry: An Energy Perspective. Washington, DC: US Department of Energy. (Paper-3) (Paper-5) (Paper-6) El-Halwagi, M.M. 1997. Pollution Prevention through Process Integration. San Diego, CA: Academic Press. (Other4) Elliott Company. 2000. Website at: http://www.elliott-turbo.com/. (Utilities-4) Elliott, R.N. and M. Pye. 1997. Analysis of Utility Motor-Systems Programs. Washington, D.C.: American Council for an Energy-Efficient Economy. (Motorsys-8) Elliott, R.N. and M. Spurr. 1999. Combined Heat and Power: Capturing Wasted Energy. Washington, DC: American Council for an Energy Efficient Economy. (Utilities-1) (Utilities-2) (Utilities-4) Elmi, G. 1993. Fast Firing Technology in the Brick and Tile Industry - A Technical Evaluation. In Ziegelindustrie 3. (Ceramics-1) Energetics, Inc. 1990. Aluminum Industry Profile. In Office of Industrial Technologies Industry Profiles, Final Report: Energy Profiles for U.S. Industry. Produced by Energetics Inc. Washington, DC: U.S. Department of Energy, Office of Industrial Technologies. (Alum-2) (Alum-4) Energy Innovations. 1997. Energy Innovations: A Prosperous Path to a Cleaner Environment. Washington, DC: Alliance to Save Energy, American Council for an Energy-Efficient Economy, Natural Resources Defense Council, Tellus Institute, and Union of Concerned Scientists. (Overview) Enneking, C.Q.M. 1994. Bewertung von Altglasscherben und Konsequenzen fuer den Galsschmelzprozess (Value of Recycled Glass Cullets and Consequences for the Glass Melting Process, original in German). In Glass Sci. Technol. 67 (6): N55-N58. (Glass-1) [EPA] U.S. Environmental Protection Agency. nd. Printed Wiring Board Pollution Prevention and Control Technology: Analysis of Updated Survey Results. Washington, DC: U.S. Environmental Protection Agency. (Other-3) -----. 1995. Profile of the Nonferrous Metals Industry. Washington, DC: Environmental Protection Agency. (Alum-4) -----. 2000a. EPA website at www.epa.gov. (Chemicals-8) -----. 2000b. USA Air Quality Nonattainment Areas. U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards. Online at: http://www.epa.gov/airs/nonattn.html. (Other-2) [ERCo] Energy Research Company and Wabash Alloys. 1998. Demonstrating an Indirect-Fired, Controlled Atmosphere Kiln to Decoat Oily Scrap. Final report prepared for the New York State Energy Research and Development Authority. Washington, DC: U.S. Department of Energy. (Alum-3)

181

LBNL and ACEEE

Erdman, T. 1999. R&D, and Expansion, Keep Strip Casting Growing. In 33Metalproducing.com Online. June. Cleveland, OH: 33Metalproducing.com Online. Online at: www.33Metalproducing.com. (Alum-1) Erikson, D. and C. Brown. 1999. Operating Experience with a Gasification Pilot Project. Tappi Journal. 82 (9): 4850. (Paper-1) Ertan, S., A.S. Akkurt,Y. Birol, M. Dundar, S. Hamer, E. Ozden, C. Romanowski, K. Sarioglu, G. Yildizbayrak. 1999. The Effect of Casting Paramaeters on Twin Roll Cast Strip Microstructure. Technical Paper. Assan Aluminum Turkey and FATA Hunter Inc. (Alum-1) [ETSU] Energy Technology Support Unit. 1994. An Appraisal of UK Energy Research, Development, Demonstration and Dissemination. London, UK: Her Majesty's Stationery Office. (SCQ) -----. 1999. Design and operation of energy-efficient batch processes. Best Practice Program, Report Brochure. (Other-4) [FAO] Food and Agricultural Organization, United Nations. 2000. Statistical database online at: http://www.fao.org. (Paper-4) Farrell, L.M., T.T. Pavlack, and L. Rich. 1993. Operational and Environmental Benefits of Oxy-Fuel Combustion in the Steel Industry. In Proceedings of the 12th Process Technology Conference. Pittsburgh, PA, October 24-27. (Steel-4) Fisher, M. (American Plastics Council). 2000. Personal Communication with Nathan Martin. August. (Chemicals-8) Fisher, M. and F. Mark. 1999. The Role of Plastics in Automotive Shredder Residue (ASR): Characterization and Enviornmental Assessment. Society of Automotive Engineers (SAE) International Congress and Exhibition, SAE Technical Paper Series 1999-01-0664. (Chemicals-8) Fiscor, S. 2000. U.S. Longwall Census 2000. In Coal Age. February. pp.32-36. (Mining-1) Flanagan, J.M. 1993. Learning from Experiences with the Process Heating in the Metals Industry. CADDET Analyses Series, No. 11. Sittard, The Netherlands: IEA-CADDET. (Steel-4) (Other-2) Flemming, G. 1995. Personal communication with Nathan Martin. Dusseldorf, Germany: SMS Schloemann-Siemag Aktiengesellschaft. (Steel-2) Fok, S. and B. Moore. 1999. Zero-Discharge: An Application of Process Water Recovery Technology in the Food Processing Industry. In 1999 ACEEE Summer Study on Energy Efficiency in Industry, Proceedings. Washington, DC: American Council for an Energy-Efficient Economy. 595-603. (Other-3) Forbes, C. (Siemens-Westinghouse). 2000. Personal communication to Nathan Martin regarding SOFC Fuel Cells. June. (Utilities-3) Freeman, B. (Electric Power Research Institute). 2000. Personal communication with Nathan Martin regarding fuel cell technologies. June. (Utilities-3) Fuel Cells 2000. 2000. Website: www.fuelcells.org. Washington, DC: Fuel Cells 2000. (Utilities-3) FuelCell Energy. 2000. Website: www.fuelcellenergy.com. Danbury, CT: FuelCell Energy. (Utilities-3) Gach, G.J., D.J. Paulson, and D.D. Spatz. 2000. Crossflow Membrane Filtration Applications in the Beverage Industry. Minnetonka, MN: Osmonics, Inc. (Food-3) [GE] General Electric. 2000. Microturbines. http://www.gepower.com/en_us/expl/html/by_prod/microturbines.html (Utilities-4) Online at:

Gitlitz, J.1995. Aluminum. In Encyclopedia of Energy Technology and the Environment. John Wiley and Sons, Inc. (Alum-2) (Alum-4) Giraldo, L.B. and B. Hyman, 1994. Energy Consumption Patterns in the Paper and Paperboard Industry, prepared for Pacific Northwest National Laboratory, Mechanical Engineering Department, University of Washington, Seattle, WA, August 1994. (Paper-6)(Paper-7)

182

Emerging Energy-Efficient Industrial Technologies

Golove, W.H. and L. Schipper. 1996. Long-Term Trends in U.S. Manufacturing Energy Consumption and Carbon Dioxide Emissions. In Energy. 21 (78). (Overview) [GPI] Glass Packaging Institute. 2000. GPI Environmental Policy. Washington, DC: Glass Packaging Institute. Article online at: http://www.gpi.org/Envio.html. [sic] (Glass-1) Grady, P. (North Carolina State University College of Textiles). 2000. Personal communication to N. Elliott. July. Raleigh, NC. (Textile-1) Greene, N. and R. Hammerschlag. 2000. Small and Clean is Beautiful: Exploring the Emissions from Distributed Generation and Pollution Prevention Policies. In The Electricity Journal. June. 13 (5). (Utilities-4) [GRI] Gas Research Institute. 1997. Gas Engine-Driven Refrigeration System Reduces Energy Costs. September. GRI Periodical article. Chicago, IL: Gas Technology Institute. Article online at: http://www.gri.org. (Food-4) Greissel, M. 2000. The Power of Oxygen. In New Steel. April. Published online at: http://www.newsteel.com/2000/NS0004f2.htm Oak Brook, IL: Cahners Business Information, a division of Reed Elsevier Inc. (Steel-3)

Habets, L. (Paques BV). 2000. Personal communication to Lynn Price. August 2. (Other-1) Hamer, S. (Fata Hunter). Personal communication with Nathan Martin regarding thin strip casting. July. (Alum-1) Hanson, M. (Energy Center of Wisconsin). 1997. Personal communication to Neal Elliott. Madison, WI: 608-2384601. (Motorsys-5) Hekkert, M.P. and E. Worrell. 1997. Technology Characterisation for Natural Organic Materials. Report No. 98002. Utrecht, The Netherlands: Utrecht University, Dept. of Science, Technology, and Society. (Paper-2) (Paper4) Honeycutt, Crawford. (EIA, Washington, D.C.) Personal Communication with Author. August 17,2000. 202-5861420 Honeywell, Inc. 2000. Website: http://www.parallon75.com/ (Utilities-4) Howe, B., A. Lovins, D. Houghton, M. Shepard, and B. Stickney. 1999. Drive Power: Technology Atlas Series, Vol. IV. Boulder, CO: E Source, Inc. (Motorsys-8) (Motorsys-7) Huangfu, E. (U.S. Department of Energy). 2000. Personal communication Nathan Martin. August. (Refin-2) Hydrogen Fuel Cell Investor, The. 2000. Website: http://www.h2fc.com/defaultNS4.html. Copyright David Redstone, H2FC Editor. (Utilities-3) [IISI] International Iron and Steel Institute. 1996. Steel Statistical Yearbook, 1995. Committee on Economic Studies. Brussels, Belgium: International Iron and Steel Institute. (Overview) -----. 1998. Energy Use in the Steel Industry. September. Brussels, Belgium: International Iron and Steel Institute. (Steel-1) (Steel-5) -----. 2000a. Continuously-cast steel output, 1997 to 1999. Brussels, Belgium: International Iron and Steel Institute. Data online at: http://www.worldsteel.org/trends_indicators/figures_7.html. (Overview) -----. 2000b. EAF Technology State of the Art and Future Trends. March. Brussels, Belgium: International Iron and Steel Institute. (Steel-3) Ingersoll-Rand. 2000. http://www.ingersoll-rand.com/energysystems/. (Utilities-4) Inoue, K. 1995. The Steel Industry in Japan: Progress in Continuous Casting. In Energy-Efficiency Utilizing High Technology: An Assessment of Energy Use in Industry and Buildings. Appendix A: Case Studies, by M.D. Levine, E. Worrell, L. Price, N. Martin. London, UK: World Energy Council. (Steel-1) Iron and Steelmaker. 1999. 1999 Continuous Caster Roundup. In Iron and Steelmaker. November. 26 (11). (Steel-2) Isenberg-OLoughlin, J. 1998. Strip Casting Comes Alive with BHP Steels Breakthrough. In 33 Metal Producing Online. Online at: www.33.metalproducing.com. (Steel-2)

183

LBNL and ACEEE

Jacangelo, J.G., S. Chellam, and R.R. Trussell. 1998. The Membrane Treatment. In Civil Engineering. September. Article online at: http://www.pubs.asce.org/ceonline/sepfeat.html. (Other-3) Jaccard, M. and Associates and Willis Enterprise Associates. 1996. Energy Conservation Potential in Six Canadian Industries. Vancouver, British Columbia, Canada. (Paper-3) (Paper-7) Johnson, S. 1997. Lighting Research Strategic Plan. Draft report. June 22. Berkeley, CA: Lawrence Berkeley National Laboratory. (Lighting-1) [JNT] Japanese National Team. 1996. Safe, CFC-Free, Refrigeration System Using Hydrogen Absorbing Alloys. In CADDET Newsletter. No. 4. (Food-4) Jones, J.A.T. 1997. Electric Arc Furnace Evolution: In Search of the Optimal Design. San Francisco, CA: Bechtel Corporation. (Steel-3) Kaplan, K. 2000. California Dealin: Financing the States Emerging Companies; Microturbine Firm Hopes IPO Generates $115 Million. Los Angeles Times. May 8. (Utilities-4) Kawano, Y. 1996. Iron and Steel Works Achieves High Energy Efficiency. CADDET Newsletter. June. No.2: 7-10. (SCQ) Kelley, J.H. (Kelastic Mine Beam Company). 2000. Personal communication Ernst Worrell. Greensburg, PA, August 30th. (Mining-1) Kemp, I.C. and A.W. Deakin. 1989. The Cascade Analysis for Energy and Process Integration of Batch Processes, Part 3: A Case Study. In Chemical Engineering Research and Design. Vol. 65: 517-525. (Other-4) Kimmerling, K. (Neural Applications Corporation). 1997. Personal communication to Nathan Martin regarding reference list. August 26. Coralville, IA. (SCQ) Kincaid, J. (ed.). 1998. 1998 North American Pulp and Paper Fact Book. San Francisco, CA: Miller Freeman Publications. (Paper-1) (Paper-3) [KMS] Koch Membrane Systems. 2000. Online at: http://www.kochmembrane.com. (Food-3) (Other-3) Kobler, R. (Recovery Plastics International). 2000. Personal communication with Nathan Martin. August/September. (Chemicals-8) Kseoglu, S.S. and G.J. Guzman. 1993. Applications of Reverse Osmosis Technology in the Food Industry. In Reverse Osmosis. Edited by Z. Amjad. New York, NY: Van Nostrand Reinhold. (Food-3) Kozak, B. and J. Dzierzawski. 2000. Continuous Casting of Steel: Basic Principles. On the American Iron and Steel Institute web page: www.steel.org. (Steel-2) Krepchin, I. 1999. Remote-Source Lighting: A Brighter Future? ER-99-7. Boulder, CO: E Source, Inc. (Lighting-1, Lighting-2) Kueck, J. (Oak Ridge National Laboratory). 2000. Personal communication to Neal Elliott. April. Oak Ridge, TN. (Motorsys-1) Kumana, J. 1996. Water Conservation and Wastewater Minimization Through Process Integration. Paper No. 57m, presented at 5th World Congress of Chemical Engineering, San Diego, CA. July. (Other-4) -----. 2000a. Personal communication with Nathan Martin via email, regarding process integration. (Other-4) -----. 2000b. Pinch Analysis What, When, Why, How. Additional publications available by contacting jkumana@aol.com. (Other-4) Kumana, J. and Rossiter, A. 1994. Rank Pollution Prevention and Control Options. In Chemical Engineering Progress (CEP). 90 (2). New York, NY: American Institute of Chemical Engineers. (Other-4) Kuster, T. 1996. Advances in strip-casting carbon and stainless. In New Steel. November. Online at: www.newsteel.com. (Steel-2) (Alum-1)

184

Emerging Energy-Efficient Industrial Technologies

Laitner, John A. "Skip", 2000a. "Exploring the Climate Change Impacts of Emerging Industrial Technologies: Some Preliminary Insights," Proceedings of the Earth Technologies Forum, International Climate Change Partnership, et al, Washington, DC, October 30-November 1, 2000. Laitner, John A. "Skip", 2000b. "The Information and Communication Technology Revolution: Can It Be Good for Both the Economy and the Climate?" a working paper for the EPA Office of Atmospheric Programs, Washington, DC. Larson, E.D. and Raymond, D. 1997. Commercializing Black Liquor and Biomass Gasifier/Gas Turbine Technology. Tappi Journal. 80 (12): 50-58. (Paper-1) Larson, E.D., C. Brown, J. Frederick, K. Iisa, T.G. Kreutz, E.W. Malcolm, G.W. McDonald, W. Yang. 1998. A CostBenefit Assessment of Black Liquor Gasifier/Combined Cycle Technology Integrated into a Kraft Pulp Mill. Paper presented at the TAPPI International Chemical Recovery Conference, Tampa, FL, June 1-4. (Paper-1) Larson, E. D., S. Consonni, and T. G. Kreutz. 1998. Preliminary Economics of Black Liquor Gasifier/Gas Turbine Cogeneration and Pulp and Paper Mills, International. Paper presented at the Gas Turbine and Aeroengine Congress and Exhibition, June 2-5, Stockholm, Sweden. American Society of Mechanical Engineers (ASME). Larson, E.D., C. Brown, W.M. Frederick, K. Iisa, T. Kreutz, E. Malcolm, G. McDonald, and W. Yang. 2000. A CostBenefit Assessment of BLGCC Technology. Tappi Journal. 83 (6). (Paper-1) Larsson, H., and D. Orloff, D. 2000. Lab-scale Impulse Drying Experiments with Extension to Higher Impulse. Paper presented at the 12th International Drying Sympsium. August. Noordwijkerhout, The Netherlands. Lehtinen, J. 1995. Condebelt Drying of Paper and Paperboard for Optimizing Quality and Production for Many Grades. In the journal Drying Technology. 13 (8-9): 2049-2068. (Paper-2) Linnhoff, B. 1993. Pinch Analysis: A State-of-the-Art Overview. In Chemical Engineering. 71 (AS): 503-522. (Other-4) Linnhoff, B., B. Tainsh, M. Wasilewski. 1999. Hydrogen Network Management. Paper presented at The European Refining Technology Conference. November. Paris, France. (Other-4) Linnhoff March Online. 2000a. Website. http://www.linnhoffmarch.com. (Other-4) -----. 2000b. WaterPinch and WaterTracker technical brochures. Online at www.LinnhoffMarch.com (Water Technology). (Other-4) Lorentzen, G. 1993a. Application of natural refrigerants. In IIF/IIR Energy Efficiency in Refrigeration and Global Warming Impact, Proceedings of meetings of Commissions B1/2 of the International Institute for Refrigeration(IIR)/Institut International de Froide (IIF). Paris, France. (Food-4) -----. 1993b. Large heat pumps using CO2 refrigerant. In IIF/IIR Energy Efficiency in Refrigeration and Global Warming Impact, Proceedings of meetings of Commissions B1/2 of IIR/IIF. Paris, France. (Food-4) Lorson, H., D. Dean, M. Schingnitz, and V. White. 1997. Black Liquor Recovery by Pressurized Oxygen-Blown Gasification. Tappi Journal. 80 (12): 111. (Paper-1) Lubitz, G., 1999. Oxy-Fuel Melter with Batch and Cullet Preheater. In Glass Sci. Technol. 72(1): 21-24. (Glass-1) Luiten, E. (Utrecht University). 2000. Personal communication to Nathan Martin regarding Impulse Drying. June. (Paper-7) Maaskant et al.: Maaskant, W, M.H.V. Mulder and J.J.P. Tholen, 1995. De Nederlandse Membraangids vol.1, Amsterdam, The Netherlands (The Dutch Membrane Guide, Volume 1, in Dutch) Mann, J.G. and Y. Liu. 1999. Industrial Water Reuse and Wastewater Minimization. New York, NY: McGraw-Hill Inc. (Other-4) Mannesmann Demag. 1998. Contiarc, The Revolutionary Electric Arc Melting Furnace. Brochure. Pittsburgh, PA: Mannesmann Demag Corporation. (Steel-3) Margolis, N. 1996. Energy and Environmental Profile of the U.S. Iron and Steel Industry. July. Washington, DC: U.S. Department of Energy. (Steel-1) (Steel-5)

185

LBNL and ACEEE

-----. 1997. Energy and Environmental Profile of the U.S. Aluminum Industry. DOE97-IOFAl2. Washington, DC: U.S. Department of Energy, Office of Industrial Technologies. (Alum-1) (Alum-2) (Alum-3) (Alum-4) Margolis, N. and J. Eisenhauer. 1998. Inert Anode Roadmap: A Framework for Technology Development. Document prepared for the Aluminum Association and the U.S. Department of Energy, Office of Industrial Technology. Columbia, MD: Energetics. (Alum-4) Martin, N., E. Worrell, A. Sandoval, J. Bode, D. Phylipsen (ed.). 1999. Industrial Energy Efficiency Policies: Understanding Success and Failure, Workshop Proceedings. Berkeley, CA: Lawrence Berkeley National Laboratory. (Overview) Martin, N., N. Anglani, D. Einstein, M. Khrushch, B. Lehman, L. Price, and E. Worrell. 2000. Opportunities to Improve Energy Efficiency and Reduce Greenhouse Gas Emissions in the U.S. Pulp and Paper Industry. July. LBNL-46141 Berkeley, CA: Lawrence Berkeley National Laboratory, Environmental Energy Technologies Division. (Paper-1) (Paper-2) (Paper-3) (Paper-6) McCall, R., M. Cato and P. Grady. 1992. Application of Ultrasound in Textile Wet Processing. Raleigh, NC: North Carolina State University College of Textiles. (Textile-1) McCormack, D. (Heatric, Ltd.). 2000. Personal communication to Ernst Worrell. West Sussex, UK. (Chemicals-3) McIlvaine Company. 1996. World Cleanroom Markets 1996-2000. Northbrook, IL: McIlvaine Company. (HVAC-1) [Means] R. S. Means Company. 1997. Electrical Cost Data 1997. Kingston, MA: R. S. Means Company. (Motorsys-7) MECS 1994. See EIA 1997 Meijer, H.K.A., J. Bernard, R. Boom, G. Flierman, and C. Teerhuis. 1994. The Cyclone Converter Furnace. In Ironmaking 2000, 18th Advanced Technology Symposium Proceedings. October. Myrtle Beach, SC. (Steel-5) Mercer, A. 1994. Learning from Experiences with Industrial Drying Technologies. CADDET Analysis Series No. 12. Sittard, The Netherlands: IEA-Centre for the Analysis and Dissemination of Demonstrated Energy Technologies. (Food-2) Miller, G. (Sun Pipe Company). 2000. Personal communication to Jennifer Thorne. August 15. (Lighting-2) Mills, E., D. Avery, G. Bell, A. Chen, A. de Almeida, S. Greenberg, L.E. Lock, G. Marton, D. Sartor, and M. Siminovitch. 1996. Energy Efficiency in California Laboratory-type Facilities. LBNL-39061. Berkeley, CA: Lawrence Berkeley National Laboratories. (HVAC-1) Moisan, D. 2000. Best Practice: Its Importance for Primary Aluminum Smelters. In Light Metal Age. April. (Alum-2) Mock, G. (North Carolina State University College of Textiles). 2000. Personal communication to N. Elliott. July. Raleigh, NC. (Textile-1) Mllers, G., W. Reichelt, H. Vorwerk. 1997. New Technologies For Electric Steelmaking. In 12th Aachener StahlKolloquim, Proceedings. Aachen, Germany. (Steel-3) Morris, K.L. 1998. Dryer Section Rebuild at Bowater Calhoun Lowers Steam Use, Ups Machine Speed. In Pulp and Paper Magazine. October. San Francisco, CA: Miller Freeman Publications. (Paper-5) Motor Challenge Clearinghouse. 1998. Personal communication to Neal Elliott. May. Olympia, WA: Washington State University Energy Office. (Motorsys-7) Motta, E. 1998. A Comparison of Upflow Anaerobic Sludge Bed (USAB) and the Anaeric Biofilm Fluidized Bed Reactor (ABFBR) for the Treatment of Municipal Wastewater. Project description shown on the National Center for Environmental Research web page online at: www.epa.gov/ncerqa_abstracts/centers/uwmrc/lamotta.html (Other-1) Mottal, R. 1995. Heat pump technology and working fluids. In Conference Proceedings of the 19th International Congress of Refrigeration. Part IVb. The Hague, Netherlands: IIR/IIF. (Food-4) Nadel, S., R. N. Elliott, M. Shepard, S. Greenberg, G. Katz, and A. T. de Almeida. 2000. Energy-Efficient Motor Systems, 2nd Edition. Washington, D.C.: American Council for an Energy-Efficient Economy.

186

Emerging Energy-Efficient Industrial Technologies

Nadel, S., L. Rainer, M. Shepard, M. Suozzo, and J. Thorne. 1998. Emerging Energy-Saving Technologies and Practices for the Buildings Sector. Washington, DC: American Council for an Energy-Efficient Economy. (Overview) Nilsson, L., K. Gilbreath, A. Gupta, and E. Larson. 1995. Energy Efficiency in the Pulp and Paper Industry. Washington, DC: American Council for an Energy-Efficient Economy. (Paper-1) (Paper-3) Nomura, Y., T. Shimizu, and K. Wada. 1989. High-Consistency Sheet Forming - Part 3: Sheet Quality and Engineering Data. Tappi Journal. 72 (5): 187-192. (Paper-6) Obeng, E. and G. Ashton. 1988. On Pinch Technology Based Procedures for the Design of Batch Processes. In Chemical Engineering Research and Design. 66 (5). (Other-4) Almost all of the following documents are Project Fact Sheets, available on the Department of Energy, Office of Industrial Technologies website. Please see the first citation (below) for full reference information. [OIT] Office of Industrial Technologies, U.S. Department of Energy. 1998. Characterization and Conditioning of Tars Produced During Black Liquor Gasification. Forest Products Project Fact Sheet. September. Washington, DC: U.S. Department of Energy, Office of Industrial Technologies. http://www.oit.doe.gov/forest/condtars.htm (Paper-1) -----. 1999. Fractionation of Corn Fibers for Production of Polyols. Chemicals Project Fact Sheet. February. http://www.oit.doe.gov/factsheets/chemicals/pdfs/cornfiber.pdf (Chemicals-1) Ojala, T. (Valmet Corporation). 2000. Personal communication to Nathan Martin. March. (Paper-2) Okos, M., Rao Nishant, Sara Drecher, Mary Rode, and Jeannie Kozak. 1998. Energy Usage in the Food Industry. Washington, D.C.: American Council for an Energy-Efficient Economy. (Food-1) Oliver, J. (DOE Motor Challenge Program). 1999. Personal communication to Neal Elliott. January. Oakland, CA. (Motorsys-5) Onsite Sycom Energy Corporation. 2000. The Market and Technical Potential for the Combined Heat and Power in the Industrial Sector. Prepared for Energy Information Administration. Washington DC: Department of Energy, Energy Information Administration. (Utilities-1) (Utilities-3) (Utilities-4) Opalka, W.A. 1999. Direct Strip Casting Developments Quicken. In Iron and Steelmaker Magazine. February. 26 (2). (Steel-2) (Alum-1) Orloff, D. and J. Crouse. 1999. Impulse drying: Status of the Pilot Scale Research Program. In Tappi Journal. 82(9). (Paper-7) Orloff, D., P. Phelan, and C. Asensio. 2000. Impulse Drying of South African Furnish. 2000 TAPPI Engineering/Process and Product Quality Conference and Trade Fair. Technical Association of the Pulp and Paper Industry. (Paper-7) Orloff, D., P. Phelan, and J. Crouse. 1999. Impulse Drying of Board Grades: Pilot Production Trials. 1999 TAPPI Engineering/Process and Product Quality Conference and Trade Fair, September 12-16. Technical Association of the Pulp and Paper Industry. pp. 37-50. See http://www.ipst.edu. (Paper-7) Orloff, D., P. Phelan, and J. Challas. 1999. Impulse Drying of Board Grades: Converting Trials. 1999 TAPPI Engineering/Process and Product Quality Conference and Trade Fair, September 12-16. Technical Association of the Pulp and Paper Industry. pp. 51-67. See http://www.ipst.edu. (Paper-7) Panchal, C.B. (Energy Concepts Company). 2000. Personal communication with Nathan Martin on fouling minimization. Anapolis, MD. July. (Refin-2) Panchal, C.B and E. Huangfu. 2000. Effects of Mitigating Fouling on the Energy Efficiency of Crude-Oil Distillation. In the journal Heat Transfer Engineering. May/June. 21 (3). Taylor and Francis Inc. (Refin-2) Pawlek, R.P. 2000. International Aluminum Recycling Seminar. In Light Metal Age. February. (Alum-3) Peng, F. (Oak Ridge National Laboratory). 2000. Personal communication to Neal Elliott. August. Oak Ridge, TN. (Motorsys-1)

187

LBNL and ACEEE

Perry, W. (Kaeser Compressors). 2000. Personal communication to Neal Elliott. August. (Motorsys-2)

Fredricksburg, VA.

Pfromm, P. 2000. Personal communication Nathan Martin regarding Direct Electrolytic Causticizing. August. (Paper3) Pickens, J. 2000 (forthcoming). Assuring the Benefits of Aluminum Recycling: Engineering Economical Environmental Solutions to the Issues of Black Dross and Saltcake (draft). To be presented at the Alumitech Conference, Atlanta, GA. (Alum-3) Pieper, H., T. Platzer, and J. Becher. 1995. Comparison of Ecological and Economic Aspects of a Modern Regenerative End-Fired Furnace and the Second Generation Sorg LoNOx Melter. In Glass Sci. Technol. 68 (7): 241-245. (Glass-1) Pivko, I. 1999a. Technology in Search of Markets. Nonwovens Industry Story Archive. Article online at: http://www.nonwovens-industry.com/jan991.htm. (Paper-4) -----. 1999b. Technology in Search of Markets. Longboat Key, FL: Notabene Associates Inc. (Paper-4) -----. 2000. (Notabene Associates, Inc.) Personal communication with Nathan Martin regarding dry forming technology. August. (Paper-4) Plunkert, P.A. 1997. Aluminum. Washington, DC: U.S. Geological Survey. (Alum-3) Portner, D. 1999. Experiences with an Oxy-Fuel Container Furnace. In Glass Industry 80 (6): 25-28. (Glass-1) Phringer, O., R. Hauk, B.H.P. Havenga, W.L. Kepplinger, F. Wallner, and H. Wiesinger. 1991. Betriebserfahrungen mit dem COREX-Verfahren und dessen Entwicklungspotential. Stahl und Eisen 111 (9): 37-44. (Steel-5) Reay, D. 1999. Learning from Experiences with Compact Heat Exchangers. CADDET Analyses Series, No. 25. United Kingdom: IEA-CADDET. (Chemicals-3) Reed, J., N. Hall, and A. Oh. 1999. Market Progress Evaluation Report Silicon Crystal Growing Facilities. Arlington, VA: TecMRKT Works. (Electron-1) Reichelt, W. and W. Hofmann, 1996. Contiarc - An Energy Optimised and Environmental Scrap Melting Process. In Stahl und Eisen 116 (5): 89-92 (in German). (Steel-3) Retulainen, E. and A. Hmlinen. 1999. Three Years of Condebelt Drying at Stora Enso Pankakoski Mill in Finland. Paper presented at the Tappi Engineering Conference, September 12-16th, 1999. (Paper-2) Richards, E.A. 1996. Bioenergy from Anaerobically Treated Wastewater. In Brewers Digest. Text of article online at: http://www.execpc.com/~drer/anadoc.htm. (Other-1) Roberto, R. 1992. Low-Pollution Processing with Advanced Radiant Gas Burner. In Gas Research Institute Digest 15 (1). Online at: www.gri.org. Robinson, V. (U.S. Dept. of Energy, Office of Industrial Technologies). 2000. Personal communication to Nathan Martin. June. (Paper-1) Rogers, J. (Independent Consultant). 2000. Personal communication to Jennifer Thorne. August 16. (Lighting-1) (Lighting-2) Rogers, J. and I. Krepchin. 2000. New High-Intensity Fluorescent Lights Outshine Their HID Competitors. ER-00-1. Boulder, CO: E Source, Inc. (Lighting-1) (Lighting-2) Ronkainen, P. (Valmet Corporation). 2000. Personal communication to Nathan Martin. June/July. (Paper-2) (Paper-4) Rossiter, A. P. (ed.). 1995. Waste Minimization through Process Design. New York, NY: McGraw-Hill Inc. (Other-4) Rostrup-Nielsen, J.R. 1994. Production of Syngas. In Catalysis Today. Vol. 18: 305-324 (Chemicals-7) Sadowski, R., B. Kinstrey, S. Henderson, and P. Parthsarathy. 1999. Black Liquor Gasification: A Combination Approach. In Tappi Journal. 82 (11): 59-62. (Paper-1)

188

Emerging Energy-Efficient Industrial Technologies

Salyp ELV Center. Website. www.salyp.com. Details on auto shredder residue. (Chemicals-8) Santiago, J. (U.S. Environmental Protection Agency) 2000. Personal communication with Nathan Martin. August. (Alum-3) Schedler, M. (NAPCOR). 2000. Personal communication with Nathan Martin regarding froth flotation technology. June. (Chemicals-8) Selines, Ron (Praxair, Inc.). Personal communication with Ernst Worrell. Tarrytown, NY. (914) 345-6467. (Steel-4) Shenoy, U. 1994. Heat Exchanger Network Synthesis. Houston, TX: Gulf Publishing Company. (Other-4) Siemens Westinghouse. 2000. Program Summary, Siemens Westinghouse Solid Oxide Fuel Cell Program. One-page summary. Siemens Westinghouse. (Utilities-3) Skerrett, P.J. 1997. Food Irradiation: Will It Keep the Doctor Away? In Technology Review. November/December. Online at: http://www.techreview.com/articles/nd97/skerrett.html. (Food-1) Smit, R.W., J.G. de Beer, K. Blok, and E. Worrell, E. 1994. Long Term Industrial Energy Efficiency Improvement: Technology Descriptions. Utrecht, The Netherlands: Utrecht University, Department of Science, Technology and Society. (Chemicals-7) Smith, R. 1995. Chemical Process Design. New York, NY: McGraw-Hill Inc. (Other-4) Solley, W.B., R.R. Pierce, and H.A. Perlman, 1998. Estimated Use of Water in the United States in 1995. Denver, CO: U.S. Geological Survey. Staib, W.E. and N.G. Bliss. 1995. Neural Network Control System for Electric Arc Furnaces. In Metallurgical Plant and Technology International. 2: 58-61. (SCQ) STAPPA/ALAPCO. 1999. Reducing Greenhouse Gases and Air Pollution. October. Washington, DC: STAPPAALAPCO. (Other-2) Stelco. 1993. Present and Future Use of Energy in the Canadian Steel Industry. Ottawa, Canada: CANMET. (Steel-1) Stene, J. 1999. Natural Working Fluids A Safe and Energy-Efficient Alternative in Compression Heat Pumping Systems. In IEA Heat Pump Centre Newsletter. 17 (3): 24-26. (Food-4) [STFI] Swedish Pulp and Paper Research Institute. 2000. Web page: www.stfi.se. (Paper-7) Suozzo, M. and S. Nadel. 1998. Selecting Targets for Market Transformation Programs: A National Analysis. August. Washington, DC: American Council for an Energy-Efficient Economy. (Motorsys-3) Takeuchi, O., T. Fukaya, K. Ishida, K. Yamada, T. Yamada, and H. Yoshida. 1995. New Scrap Melting Furnace With Continuous Preheating System. Iron and Steelmaker 22 (4): 13-14. (Steel-3) Tanner, C. (Capstone Turbine Corporation). 2000. Microturbines: A Disruptive Technology. Cogeneration and Competitive Power Journal. Article is also online at: www.retailenergy.com/articles/microturbines.htm. (Utilities-4) Thayer, D., E.S. Josephson, A. Brynjolfsson, and G.G. Giddings. 1996. "Radiation Pasteurizatiopn of Food." Council for Agricultural Science and Technology. (Food-1) Tomasseti, G. 1995. The Manufacture of Bricks and Ceramic Tiles in Italy: Two Innovative Production Processes. Appendix in Energy Improvement Utilising High Technology. London, UK: World Energy Council. (Ceramics-1) (Steel-2) Trepp, Ch., P. Savoie, and W.E. Kraus. 1992. Investigation of the Performance Behaviour of a Compression Refrigerating Unit with Halogen Refrigerant Mixtures R22/R142b, R22/R114 and R22/R12. In International Journal of Refrigeration 14 (2): 101-111. (Food-4) Tripathi, P. 1996. Pinch Technology Reduces Wastewater. In Chemical Engineering. Vol. 103 (11). (Other-4) Tschudi, W. and D. Sartor. 1999. Energy Efficiency in California High-tech Facilities: Market and Opportunity Assessment. Berkeley, CA: Lawrence Berkeley National Laboratory. (HVAC-1)

189

LBNL and ACEEE

Tschudi, W. 2000. Draft Roadmap for High Tech Buildings. Internal memo at Lawrence Berkeley National Laboratory. May. (HVAC-1) [USAEP] The United States Asian Environmental Partnership. 1998. Clean Technologies in U.S. Industries: Focus on Textiles. Online at: http://www.usaep.org/reports/textiles.htm. Washington, DC: U.S. Agency for International Development. (Textile-1) [USCAR] United States Council for Automotive Research. 1998. USCAR Bubbling About Plastics Separation for Recycling. In Newsletter. Fall. Article online at: http://www.uscar.org/techno/flotation.htm (Chemicals-8) -----. 2000. Personal communication with with Susan Yester to Nathan Martin regarding U.S. car research consortium. August. (Chemicals-8) [USGS] United States Geological Survey. 2000. Mineral Commodity Summaries. Washington, DC: U.S. Geological Survey. Online at: http://minerals.usgs.gov/minerals/pubs/mcs/. (Alum-1) (Alum-2) (Alum-3) (Alum-4) Valenti, M., 1998. A Cradle for New Steel Technologies. In Mechanical Engineering. 120 (11): 60-64. (Steel-4) Van Deventer, H.C. 1997. Feasibility of Energy-Efficient Steam Drying of Paper and Textile Including Process Integration. In Applied Thermal Engineering. 17 (8): 1035-1041. (Paper-5) Van Leeuwen, T. 2000. An Aluminum Revolution. Zurich, Switzerland: Credit Suisse First Boston Corporation. (Alum-4) Van Trigt, P. (ABB Power Generation Limited). 1998. Personal communication to Neal Elliott. October. Baden Switzerland. (Utilities-4) Virta, R.L. 1998. Clay and Shale. In Minerals Yearbook 1998. Washington, DC: United States Geological Survey. (Ceramics-1) Wallace, R. 1998. Personal communication to Sam Wheeler. April. St. Louis, MO: Reliance Corporation. 941-746-3515. (Motorsys-7) Ward, D.T. 2000. Growth of Nonwovens Shows No Signs of Slowing. In International Fiber Journal. 15 (1). Article online at: http://www.ifj.com/issue/february00/34.htm (Paper-4) Wartena, R. (Institute of Paper Science and Technology). 2000. Personal communication with Nathan Martin. September. (Paper-3) [WEC] World Energy Council. 1995. Efficient Use of Energy Utilizing High Technology: An Assessment of Energy Use in Industry and Buildings. London, UK: World Energy Council. (SCQ) (Refin-2) Welch, B. 1999. Aluminum Production Paths in the New Millenium. In Journal of Metals 51(5). (Alum-4) Wechsler, R. 2000. Caststrip LLC. Personal Communication to Nathan Martin. October. (Steel-2) Whittemore, O.J. 1999. Energy Use and Efficiencies in Firing Ceramics, Melting Glass. In American Ceramic Society Bulletin. July. pp. 69-71. (Ceramics-1) Wiesner, M.R. and S. Chellam. 1999. The Promise of Membrane Technology. In Environmental Science and Technology. 33(17): 360-366A. (Food-3) (Other-3) Worker, C. Advances at billet and thin slab mills. In New Steel. May. Online at: www.newsteel.com. (Steel-2) Worrell, E. 1994. Potential for Improved Use of Industrial Energy and Materials. PhD Thesis. Utrecht, The Netherlands: Utrecht University. (Refin-2) (Chemicals-8) Worrell, E. 1995. Advanced Technologies and energy efficiency in the iron and steel industry in China. In Energy for Sustainable Development. November. (Steel-2) Worrell,,E, J-W. Bode, and J. De Beer,. 1997a "Analysing Research and technology Development Strategies: The 'ATLAS' Project, Energy Efficient Technologies in Industry", Prepared for Directorate General XVII of the European Commission, Dept. of Science, Technology & Society, Utrecht University (Report 97001), (Paper 1-7)(Steel-2)

190

Emerging Energy-Efficient Industrial Technologies

Worrell, E., K. Blok, M. Levine, N. Martin, L. Price, and R. Van Den Broek. 1997b. Potentials and Policy Implications of Energy and Material Efficiency Improvement. New York, NY: United Nations Division for Sustainable Development, United Nations. (Overview) Worrell, E., N. Martin, L. Price. 1999. Energy Efficiency and Carbon Dioxide Emissions Reduction Opportunities in the U.S. Iron and Steel Sector. July. LBNL-41724. Berkeley, CA: Lawrence Berkeley National Laboratory. Worrell, E., D. Phylipsen, D. Einstein, N. Martin. 2000. Energy Use and Energy Intensity of the U.S. Chemical Industry. Report No. LBNL-44314. Berkeley, CA: Lawrence Berkeley National Laboratory. (Chemicals-7) Wroblewski, R. (Energy Center of Wisconsin). 1996. Personal communication to Neal Elliott. Madison, WI: 608-2384601. (Motorsys-5) Xenergy. 1998. United States Industrial Electric Motor Systems Market Opportunities Assessment. Prepared for U.S. Department of Energys Office of Industrial Technologies and Oak Ridge National Laboratory. Burlington, MA: Office of Energy Efficiency and Renewable Energy. (Food-4) (Motorsys-8) (Overview)

191