L A B O R A T O R I E S F O R T H E 2 1 S T C E N T U R Y :

BEST PRACTICE GUIDE

O PTIMIZING L ABORATORY V ENTILATION R ATES

I ntroducti o n this guide highlights best-practice strategies focused

on reducing energy use, it does not specify how to set a
This Best Practice Guide is one in a series created by ventilation rate. Note that the terms good and better
the Laboratories for the 21st Century (Labs21) program, practices are used to describe options that improve stan-
a joint program of the U.S. Environmental Protection dard practices.
Agency and U.S. Department of Energy. Geared towards
architects, engineers, and facilities managers, these guides
provide information about technologies and practices to
use in designing, constructing, and operating safe, sus-
tainable, high-performance laboratories. Lighting
11%
Laboratories are highly energy intensive, often using Cooling
four to six times more energy per square foot than a typi- 22%
Plug
cal office building. Most existing labs can reduce their 23%
energy use by 30 to 50% with existing technology, which is
significant given their $12 billion annual energy costs in
the U.S. Nearly half of the electrical energy use in a typical
laboratory can be attributed to ventilation, and reducing a Ventilation
laboratorys ventilation needs can lower the cost to build 44%
and maintain a facility (see Figure 1).

The objective of this Best Practice Guide is to help

users optimize ventilation airflow and reduce associated Figure 1. Annual electricity use in Louis Stokes
energy use while maintaining or improving safety. While Laboratory, National Institutes of Health, Bethesda, MD.

U.S. Department of Energy

Energy Efficiency and Renewable Energy
Federal Energy Management Program
 2 L A B S F O R T H E 2 1 S T C E N T U RY

In the case of building codes for hazardous classifica-

Perspecti v e S ta n d a r d tions, ventilation rates are stipulated in terms of floor area
Practi ce

as cubic feet per minute (CFM) per square foot, while

ventilation guidelines from building design standards for
In standard laboratory-design practice, ventilation laboratories are based on the total volume of the space
rates are usually derived from guidelines, presented as and expressed as air changes per hour (ACH), which is
a range of values in design standards. When using a how many times the entire air volume in the laboratory is
guideline to determine a ventilation rate for a laboratory, replaced each hour.
the highest value from the range is often chosen because
the guidelines are highly generalized; however, design- Prevailing building codes and design standards
ers should be cautious when using these wide-ranging provide a context in which best-practice strategies can
recommendations. Design firms, or authorities having be implemented. Consider the ventilation guideline
jurisdiction, typically use a guideline without question- provided by OSHA 29 CFR Part 1910.1450, which calls
ing the true source and the reasoning behind its value. for a range of 4 to 12 ACH for a laboratory that often
Similarly, designing by only referencing past efforts, by has an occupancy classification of B. In contrast, the
whats in the drawer, limits energy efficiency and may International Building Code (IBC) (2004) calls for a rate of
even compromise safety. A simplified more is better 1 CFM/ft2 for an occupancy classification of H-5, which
design approach is not a substitute for due diligence. is considered to be a hazardous environment (see Table 1
Ventilation guidelines should only be applied as their for this and other examples). Note that for certain quanti-
authors intendedas ranges, and not as absolutes. ties of flammable liquids, these must be used in a con-
trol area in order for a building to have a B occupancy
Standard practice also entails the blanket adoption of classification.
ventilation guidelines as constant values, with the ventila-
tion rate rarely being dynamically controlled or otherwise As an example of how a codes ventilation rate (or
tailored to the occupancy or conditions of the site, or ambiguity) influences energy use, consider the following
optimized for energy efficiency or safety. Some publica- two scenarios. One laboratory space, whose occupancy
tions simply recommend 4 to 12 air changes per hour. The classification is IBC H-5, has a 10-foot-high ceiling and
result can be excessive (or inadequate) ventilation for the an exhaust airflow rate requirement of 1 CFM/ft2;
lab in question, causing unnecessary energy expenditures. the exhaust airflow from this lab will result in 6 ACH.
Facility owners also bear the consequences of requiring an However, another laboratory space, which has the same
unsubstantiated high ventilation rate, inadvertently forc- floor area but a 15-foot-high ceiling and a B occupancy
ing the engineer to design a potentially wasteful HVAC classification, follows the OSHA guideline of 12 ACH and
system. flows three times more air than the higher-hazard H-5 lab.
Applying fan-law energy calculations, the B lab will
Ven ti l a ti o n C o d e s a n d G u i d e lines consume more than three times the energy of the H-5
lab. In addition, the first cost of the H-5 labs smaller
After reviewing the users program, or design intent HVAC system will also be less. Note that even though the
document, the project architect determines the labora- airflow rate per unit floor area eliminates ceiling height
torys occupancy classification. For this resolved occu- as a determinant of air change requirements, neither the
pancy classification, the ventilation rate is provided in a volumetric ACH method nor the area-based CFM method
building code specified by the municipality or authority predicts the effectiveness of the ventilation.
having jurisdiction. The occupancy classification has a
significant impact on a buildings energy use. For exam- Codes
ple, for particularly hazardous occupancy classifications,
the ventilation rate is legally stipulated by the applicable Laboratory designers should study code require-
building code. However, for less hazardous occupancies, ments, understand each classification, be familiar with
design standards with a range of rates are used only as their potential energy impacts, and relate these findings to
guidelines for ventilation rates. Note that a design stan- the project design team.
dard may be adopted by the authority having jurisdiction
as a code requirement.
 L A B S F O R T H E 2 1 S T C E N T U RY 3

Table 1, below, lists typical design codes that are often Standards
used as ventilation-rate guidelines:

Table 2, below, lists common design standards con-

taining guidelines for a laboratorys ventilation rate:

Table 1. Common Laboratory Ventilation Rate Codes
Code Ventilation Rate Comment
IBC -2004 1 CFM/ft2 for H-5 Section 415.9.2.6 Sta nda r d Pr a ctice Mor e Is N ot
Necessa r ily B etter
IMC - 2004 1 CFM/ft2 Rate required for storage areas that While rules of thumb often dictate that more is
exceed maximum allowable quantities
better, i.e., that increased ventilation rates yield increased
of hazardous materials. (Section
502.8.1.1.2) safety, worker comfort, and research productivity, real-
UBC - 1997 1 CFM/ft2 for H-6 Uniform codes have been replaced by world experience shows that this is not the case. In fact,
international codes beginning in 2000. excessive ventilation can diminish safety conditions in
(Section 1202.2.5) labs that use hazardous and odorous materials as part of
their experimental studies (see sidebar on Ventilation
Dilution). Thus, best practices optimize rather than maxi-
mize ventilation, and consider the mixing factor of the
pollutant being removed from the lab.

Table 2. Common Laboratory Ventilation Rate Standards

Standard ACH Number Comment

ANSI/AIHA Z9.5 The specific room The latest version of the American National Standards Institute and the American Industrial
ventilation rate shall be Hygiene Association standard (ANSI/AIHA Z9.5-2003, Section 2.1.2) states that a method
established or agreed upon based on air changes per hour is not the appropriate concept for designing containment
by the owner or his/her control systems. Contaminants should be controlled at the source. ANSI/AIHA also states
designee. that the air changes per hour do not reflect actual mixing factors of a particular room.

NFPA-45-2004 Minimum 4 ACH According to the National Fire Protection Associations Standard NFPA 45, Appendix A:
unoccupied; occupied A8-3.5 (NFPA 45 2004), room air currents in the vicinity of fume hoods should be as low as
typically greater than 8 possible, ideally less than 30% of the face velocity of the fume hood. Air supply diffusion
ACH. devices should be as far away as possible from fume hoods and have low exit velocities.

ACGIHInd. Vent. 24th Ed.2001 The required ventilation This standard from the American Conference of Governmental Industrial Hygienists states
depends on the generation that Air changes per hour or air changes per minute is a poor basis for ventilation
rate and toxicity of the criteria where environmental control of hazards, heat, and/or odors is required. The
contaminant, not on the impact of the laboratorys ceiling height is identified as one reason why an air change
size of the room in which approach does not adequately address the required contamination control (Section 7.5.1,
it occurs. Air Changes).

ASHRAE Lab Guide2001 4-12 The ASHRAE Laboratory Design Guide includes suggestions relating to the following:

Minimum supply air changes

Minimum exhaust air changes

Minimum outdoor air changes

Recirculation considerations

OSHA 29 CFR 4-12 The Occupational Safety and Health Administration specifies a room ventilation rate of 4 to
Part 1910.1450 12 air changes per hour, which is normally adequate general ventilation if local exhaust
systems such as hoods are used as the primary method of control. This range is extremely
broad and provides a designer with little guidance.
 Studies of laboratory facilities have demonstrated hour (ACH) has a minimal effect on aerosol concentration
that the room air change rate has less effect than a room of microorganisms in the first few minutes after release.
air diffusing system or other ventilation characteristic on
environmental conditions. Designers need specifications Adjusting ventilation is not the only way to control
that are tailored to a laboratorys air circulation arrange- environmental conditions. For instance, Memarzadeh
ment, because many conventional design parameters (1999) has shown that controlling the humidity in animal
and recommendations should not be universally applied; rooms is more effective than using high air change rates in
for example, they may not relate to microenvironmental managing the production of ammonia from animal urine.
(e.g., cage) conditions in a laboratory (Zhang et al., 1992; This has allowed users to decrease the air change rate
McDiarmid, 1988). from 15 to as low as 5, while improving the welfare of the
animals.
Other studies show that air dilution or replacement
does not protect personnel from exposure to concentrated
bursts of aerosols in biological laboratories. For example, P re p a rat i o n P ro c e s s
O ve r vi e w
Crane (1994) quotes Chatigny and West (1976), who say that
increasing ventilation rates from 6 to 30 air changes per
Determining a laboratory facilitys ventilation rate is
not an exact science; therefore, a precise answer cannot
Ventilation Dilution* be the only goal of the ventilation system designer. More
The principal device used to contain harmful emissions from chemicals within important is the process carried out by the ventilation sys-
a laboratory is the chemical fume hood. Such hoods come in various sizes, but tem designer to resolve an appropriate ventilation rate. At
a typical internal working surface dimension is 66 inches wide and 26 inches
deep. If one assumes an 18-inch sash opening, and an average face velocity the initial (conceptual) level, the designer should consider
of approximately 100 fpm, approximately 850 cfm of air is induced through the the following four-step process:
hood by the laboratory exhaust system. Furthermore, for hoods designed in
accordance with the guidance provided in Appendix A.6.4.6, NFPA 45-2000,
a minimum of 300 cfm of dilution air is admitted when the sash is fully closed.
For laboratory ventilation systems designed in accordance with the guidance
Step 1 : R eview Design Intent
provided in Appendix A.6.3.5 of NFPA 45-2000, room air current velocities in Study and support the features embodied in the facil-
the vicinity of the hood should ideally be less than 30 fpm.
itys design intent document. The document should
Given the hood dimensions and ventilation system design guidance described include three main categories, which are followed by
above, consider an accidental one-liter spill of hydrogen fluoride or hydrogen
chloride in the vicinity outside of the fume hood. In calculating the concentra the designers actions:
tions resulting from either of these spills, it can be shown that the contaminant
threshold limit value IDLHImmediately Dangerous to Life and Healthis User programming:
exceeded regardless of sash position. Additional results are summarized below:
Characterize the buildings mission.
If room air current velocities in the vicinity of a hood are increased
above the minimum recommended level (as noted above), emission
Differentiate between needs and wants.
rates from the spill increase, as do the resulting concentration levels
of contaminants in the air flowing toward the fume hood.
Evaluate client statements.
If the volumetric airflow rate in the vicinity of the hood is increased,
as in the case of opening the sash of a variable air volume (VAV) List research goals.
hood, lower concentration levels result.
Unfortunately, ventilation airflow rates have to increase by at least Occupancy classification:
one or two orders of magnitude above those induced through the
fume hood to keep from exceeding life-threatening threshold levels. Interpret municipalitys building code.
Furthermore, ventilation systems designed to provide such airflow
rates are impractical because they are physically constrained by the Determine occupancy classification requirements.
size of the building.
Consequently, ventilation dilution is not the solution to pollution Analyze energy-use impacts.
resulting from an accidental spill of hazardous chemicals within a
laboratory. Relate findings to the project design team.
Finally, the above results point to the importance of handling chemi
cals safely before delivering them to chemical fume hoods. Essential building functions and systems:
Note: IDLH was established by the National Institute for Occupational Safety Identify architectural features.
and Health. It is defined as the concentration of airborne contaminant that
poses a threat of death, immediate or delayed permanent adverse health
Evaluate engineering approaches.
effects, or effects that could prevent escape from such an environment.

*Courtesy of John L. Peterson, P.E., Office of Facilities Planning and Itemize main design elements and characteristics.
Construction, University of Texas System
Determine boundaries for design.
Ste p 2 : I d e n t if y A u t h o r it y H a ving Programming
Start Here: Review Design Intent Document
Juri s d i c ti on Schematic
Design Codes or ?
Ensure that the authority is identified and involved,
Identify: Authority Having
and has a clear understanding of the difference between Jurisdiction. Adopted Standards or ?
codes and standards: Gather: Ventilation rate
required by the authority from:
Industry Standards
Codes:
Have force of law. Scope: Owners Priorities
Building occupancy class
Are restrictive. Determine design teams level Employee/user safety
of support in charrette by Buildings value
Require compliance. discussing: Sustainability design goals
Priority for optimizing vent rate
Standards: Availability of resources Budget: Value Engineering
Cost/benefit analysis
Are open to interpretation. Cost versus safety
First cost versus life-cycle cost
Have a wide span of acceptable values. Added engineering costs

Are subject to manipulation. Primary

Schedule: Available Time
Decision Point: Identified design resources
Design team experience
Adopted standards: Design Approach:
Critical path limitations
Project Teams
Support
May be based on sound judgment.
Could be biased or reflect entrenched doctrine. Constrained Moderate Full
Support Support Support
May be archaic and not reflect latest technology
or practices. Standard Good Practice
Practice Perform Preliminary
ventilation lab layout
Apply design
Step 3: Assess Priority and Resources standard
evaluation and use

for Optimizing Ventilation Rates Code-based

Apply design
refinements
Garner team support to optimize laboratory venti- determination

lation rate during design charrette. Better Practice

Review modeling
All stakeholders identify labs design goals and methods
issues. Decision
Point: Airflow
Determination
Promote fundamental impact of labs ventilation
Decision
rate on: Point: Modeling
Method
Continuous safety performance.
Build computer
Immediate HVAC first-cost. model (CFD)

Long-term energy use.

Consider: Build full-size or

Resolved: scale model
Scope: Owners priorities. Ventilation
Rate
Schedule: Available time.
Airflow Based Volumetric Air Refined Optimized
Budget: Value engineering. on Area Change Rate Ventilation Rate Ventilation Rate
(CFM/ft2) (ACH) (CFMft2 (CFM/ft2)
or ACH)

Step 4: Implement a Design Strategy

Design Lab Module
Once steps 1 through 3 have been completed, imple-
Design Development
ment the appropriate design strategy:
Finalize Ventilation
A design strategy can be chosen from one of the
following three options: Figure 2. Process flowchart.
 6 L A B S F O R T H E 2 1 S T C E N T U RY

Constrained design: Restricted, or constrained by factors), thereby reducing the guesswork involved in stan-

building code. dard practice, and ensuring that the facility will perform
well during emergencies.
Standard design: Conventional practice that
employs design standards guidelines.

Optimized design: See following details. G o o d Pr a ctice Str a tegies: C o n trol

If an optimized design strategy is chosen, a design R efinements
teams support must balance scope, schedule, and bud-
As with most pollution-abatement strategies, the most
get limitations with attitude and resources. The
effective strategies begin with source control, containment,
following should be considered by the design team:
and minimization. Ventilation is not a substitute for poor
Optimized design support requires a change in practices in handling hazardous materials, but the follow-
attitude: ing control refinements can help maximize the effective-
ness of ventilation while keeping energy costs in check:
Do not impose standard or usual policies and
methods. Occupancy control

Willingly provide information on experimental Demand control

procedures. Hazardous banding control
Remain open and flexible to new ideas and alter- Task ventilation control (localized exhaust ventila-
nate ways of doing science. tion, LEV)
Optimized design support needs to be quantified:
Occupancy Control Occupied versus Unoccupied
Use life-cycle costing as a measurement tool for
Ventilation Rates
decision making.
The differences in ventilation requirements between
Establish project schedule and project costs, and
occupied and unoccupied modes should be considered.
get designers concurrence.
The ASHRAE Laboratory Design Guide suggests that setback
control strategies can be used in laboratories to reduce air
changes hourly during unoccupied periods, e.g., at night
I mpl emen t a t i o n and on weekends. The NFPA 45 Standard recommends a

S trategi e s t o Op ti mi ze minimum ventilation rate of 4 ACH for unoccupied labora-

tories; some labs are designing for even lower rates.
L a b Venti l a ti o n
Demand Control Emergency Override Ventilation
There are many design factors to consider when opti-
mizing lab ventilation. These include the labs layout (e.g., Emergency override is a design refinement of the labo-
arrangement of equipment) and use (potential pollutants), ratorys supply and exhaust system to provide increased
control and removal of hazardous pollutants, and how to airflow and negative pressurization in an emergency. Such
achieve adequate ventilation while attending to cooling a design can reduce both energy use and first cost, unlike
load requirements. designs for continuous operation under rare worst-case
conditions. Emergency push-button overrides can be
The good and better practices outlined below located near fume hoods and at the labs entrance(s), and
begin with codes or standards as a starting point for should come equipped with indicator lights and audible
designs, while facilitating the adoption of ventilation spec- alarms. The facilitys automated control system can:
ifications that ensure safety and energy efficiency. Good Increase airflow through the lab during an

practices emphasize lab-specific operations and control emergency.

strategies, while better practices improve the ventilation

design process with advanced computer or physical mod- Notify the facilitys Environment, Health, and
eling techniques. These new techniques evaluate scenarios Safety (EH&S) staff.
in which the system will need to respond to critical condi- Discourage other workers from entering the

tions (e.g., hazardous material spills, pollutant mixing laboratory.

 L A B S F O R T H E 2 1 S T C E N T U RY 7

Demand-controlled ventilation (DCV) is an emerg- Control banding can be applied to laboratory

ing technology that utilizes pollutant sensors in order to chemical operations. For a specific process and associated

provide real-time variable-air-volume ventilation control. chemicals, the control band can specify what activities are
DCV differs fundamentally from typical practice, which permissible at a room air change rate, require local ventila-
blindly sets a fixed ventilation rate based on anticipated tion, and must be conducted in a fume hood at a particular
(but not verified) pollutant levels. Some key challenges in flow rate. (Chemicals with the highest risk are handled at
DCV design are correct sensor choice and placement. hood flows set for optimum containment, or performed
in a glove box.) A laboratory might optimize airflows for
A notable benefit of DCVin addition to energy work up to a prescribed control band, or designate specific
savingsis the introduction of monitoring equipment that hoods, based on airflow and contaminant containment,
can detect hazards and provide alarms and reporting. In for work within a certain control band.
addition to monitoring for spills and other accidents, DCV
can also help identify malfunctioning fume hoods or poor This new approach to classifying chemical hazards is
lab practice (e.g., chemicals left out of fume hoods) that being increasingly applied worldwide. For example, the
could otherwise go undetected. United Kingdom has incorporated control banding into
its recommended tools for compliance with regulations by
Hazardous Control Banding: Classifying Chemicals the Control of Substances Hazardous to Health.
and Hazards
During the design of a new lab building and retrofit of
Control banding is a strategy for classifying and han- an existing one, the University of Rochester (UR) recently
dling chemicals and hazards according to their associated used control banding to identify a hazard level for each of
health risks. A control band score is calculated by weigh- its labs. After performing a detailed review and analysis of
ing a chemicals level of toxicity, scale of use, and ability hazards being used in the universitys labs, the UR Health
to become airborne under certain conditions. The control and Safety Officer used control banding to create a new air
band score directs the user to appropriate control strate- change rate standard. Based on this approach, an A lab
gies. (See sidebar on Control Banding.) has 8 ACH when it is occupied, and 6 ACH when unoc-
cupied (8/6 ACH); a B lab has 6/4 ACH, and a C lab
has 4/2 ACH. Control banding can also be done on a basis
Control Banding for Optimizing Laboratory Ventilation Rates* of CFM/ft2.
Control banding is a means of classifying and grouping substances used in a
process or activity by health risk for the purpose of determining an appropriate In the example of the University of Rochester, the use
control strategy. Risk is most often described as a function of the likelihood of control banding is a step in the right direction, but it still
and consequences of an event. For control banding, chemical classification
has a similar risk basis. Toxicity (with consideration of the potential for skin
reinforces the conventional wisdom that more is better.
absorption) is a measure of the consequence of exposure. The scale of use As described later, better ventilation design and evaluation
(quantity) and the ability to become airborne (volatility for liquids, or dispersibil
strategies will provide greater protection from airborne
ity for solids) are measures of the likelihood of exposure. Combinations of the
different levels of toxicity, scale of use, and ability to become airborne under chemical hazards than simply increasing air change.
the conditions of use yield a score that equates to a control band. The control
band, combined with the tasks involved, directs the user to the appropriate
control strategies. Strategies are based on four key approaches: Task Ventilation Control
Employ good industrial hygiene practice; Special-purpose laboratories provide an opportunity
Use local exhaust ventilation; for designers to apply localized ventilation devices suited
Enclose the process; or for a labs particular use. Examples include animal labs
Contact a professional industrial hygienist. using cage ventilation as a task-specific ventilation or local
The control-banding concept can easily be applied to laboratory chemical
exhaust ventilation (LEV) strategy, electronic clean rooms
operations, where the chemical use quantities tend to be small, and chemical
toxicity and ability to become airborne vary widely with the chemicals of inter using mini-environments, or biomedical labs using bio-
est. For a specific process and associated chemicals, the control band might logical safety cabinets (BSCs).
specify activities permitted with various room air change rates, activities that
require local ventilation, and activities that must be conducted in a fume hood
at various flow rates, with the highest risk at hood flows set for optimum con In the case of animal labs, studies such as those by
tainment of airborne contaminants. A laboratory might have airflows optimized Memarzadeh (1999) have shown that increasing a rooms
to do work only up to a certain control band, or specific hoods might be desig
nated for work within a certain control band, based on airflow and contaminant
ventilation rate does not have a significant effect on cage
containment. ventilation. In addition, Riskowski et al. (1996) identified
Courtesy of John Piatt, Pacific Northwest Laboratory cage type as an important factor in determining the
ventilation rate in an animal facility, and Zhang et al.
 8 L A B S F O R T H E 2 1 S T C E N T U RY

(1992) found that providing a quality environment for Eliminating stagnant dead zones in which air

animal studies was more dependent on cage design, recirculates or there exist lazy airflow patterns.
room ventilation system design, and animal management
Examining numerous What if? scenarios.
practices than on room air exchanges.

Good practice therefore involves tailoring ventilation This virtual model can also be the basis of a full-scale
to a specific task, and to a location within a laboratory construction of a laboratory space.
equipped with LEV. When this is done, general ventila-
CFD simulation methods can help determine the labs
tion rates may be relaxed without compromising safety or
airflow characteristics, spill clearing performance, and
comfort at the location of the task. Note that LEV systems
mixing factors including removal of fugitive emissions,
can increase energy use if improperly designed, installed,
e.g., small continual releases from an evaporating solvent
or operated due to high ventilation system pressure drop
in an uncovered beaker outside a fume hood. Importantly,
requirements, leaking devices, and open unused LEV
CFD models can predict plume patterns of spill scenarios,
devices.
and the required clearing time following a spill before it
happens.

Be tte r Pra c t ic e S t r a t e g ie s : CFD modeling methods are useful for evaluating the
Si m u l a ti o n Me t h o d s dynamic effects of HVAC system features, layout, and
operation. Room geometry, HVAC system equipment,
In an effort to optimize ventilation system layouts diffuser placement, and laboratory equipment as well
and laboratory designs, better-practice strategies apply as operational procedures all influence air movement in
real or virtual laboratory models that permit airflow pat- the laboratory, particularly around the fume hood sash
tern simulations. These performance-based approaches opening. A CFD model simulates the interaction of all of
evaluate a simulated environments hazards, e.g., they these variablesas well as the turbulence caused by a
determine a chemicals clearing time by calculating the lab workers movementsto provide data that can be used
spaces mixing factors for a given spill scenario rather to understand a laboratorys temperature, air movement,
than simply applying a universal, prescriptive air change relative pressure, regions of turbulence, and contaminant
rate. This is an iterative process that accounts for facility concentrations. In addition, this modeled information
design features that influence one another. The following can be further analyzed to study fume hood containment
simulation methods may be applicable: capabilities, challenges to the hoods containment, e.g.,
CFD simulations supply temperature variations, residence time of air mov-
ing through the modeled lab, placement of ventilation
Tracer gas simulations inlets and outlets, and other factors.
Neutrally buoyant bubble simulations
Although costly, building a full-scale model of a
laboratory module can be justified when the module will
Computational Fluid Dynamics (CFD) Simulations be replicated many times in one facility or in multiple
For this better-practice approach, a geometric rep- facilities. Performance-design methods including pre-
resentation of the lab space is built within a computer. construction CFD modeling and full-scale lab modeling,
Then, a simulation of the airflow patterns inside the lab followed with in situ evaluations, can make laboratories
is modeled using a computational fluid dynamics (CFD) safer and more energy efficient.
computer program. Results from the model help design-
ers determine a labs airflow characteristics by: Tracer Gas Simulations
Developing answers to spill scenarios. Once a scaled or full-size mockup is built, a labs
ventilation system can be determined by using a tracer gas
Estimating residence time of a hazard. test, according to the ASHRAE Laboratory Design Guide.
Evaluating the placement of major design ele- The tracer gas is evenly distributed throughout the labora-
ments, such as hoods, benches, registers. tory, and the rate of decay in the tracer gas concentration
is used to calculate air changes per hour (ACH). To imple-
 L A B S F O R T H E 2 1 S T C E N T U RY 9

ment this strategy, sensors are installed in the room, a trac-

er gas is introduced, and ventilation rates are increased Meeting the Challenge

Pe rf o rm an c e E x am p l e s

until the desired rate of decay is obtained. (EH&S special-

ists typically determine the appropriate rate of decay.)

Neutrally Buoyant Helium Bubble Simulations

Pr inceton U niver sity, G uyo t H all
Using neutrally buoyant helium bubbles to study In this example (see Figure 4), CFD models were
airflow patterns in a laboratory space (see Figure 3) is a used to evaluate teaching and research labs at Princeton
relatively new method. Tiny helium-filled bubbles about Universitys Guyot Hall. Results from the CFD models
one-eighth of an inch (2 mm) in diameter are generated at allowed the design team to evaluate risks and reduce
the rate of approximately 400 bubbles per minute. These operating costs, and thus enhance the performance of the
bubbles quickly reach room temperature and follow the Princeton researchers laboratories. Using the CFD model
slightest air current in the room. They persist for up to two to study different ventilation rates provided a greater
minutes, providing designers an opportunity to study a understanding of the ability to remove airborne pollutants
labs ventilation system. Helium bubbles are also useful from these labs. Thanks to improved information pro-
for evaluating the efficacy and placement of supply dif- vided by the CFD analysis, the initial mandated rate of 10
fusers and return air grilles; their positions can be varied ACH was reduced to 8 ACH during occupied periods, and
during the test in order to mitigate areas of stagnant air. lowered to 6 ACH during unoccupied periods, while an
emergency rate of 10 ACH was designed into the HVAC
system.

W yeth R esea r ch a nd Developm ent

La bor a tor y
In this example (see Figure 5), the design team built
a CFD simulation of Wyeth Laboratorys new Viral and
Immunology Research Laboratory.1 The main concern
was determining the optimal air change rate, since avail-
able standards and guidelines only provide a wide range
of rates that are not immediately applicable to the specific
research mission of this new lab. The CFD evaluations
helped show that an air change rate of 8 per hour was
more effective in containing and removing pollutants than
the required rate of 12 ACH, which was accepted by the
authority having jurisdiction.

Various spill scenarios were modeled and evaluated

for transient concentration levels. The CFD model helped
designers characterize how an optimal layout of the ven-
tilation system can improve worker safety by effectively
removing spills. It also indicated that a lower air change
rate was potentially more effective than a higher ventila-
Figure 3. A laboratorys airflow pattern shown by neutrally tion rate in removing spills, thus potentially reducing
buoyant helium bubbles. annual energy costs by $280,000.

1 For additional information, see http://labs21.lbl.gov/DPM/Assets/e3_maine.pdf .

 10 L A B S F O R T H E 2 1 S T C E N T U RY

250 85.0

Windows

200
80.0

150
75.0
100 True North

70.0
50

0 65.0
Velocity [fpm] Temperature [F]
Short throw and large Effect of global air circulation
drop of supply air jets in the lab, flow towards windows Thermal stratification
(draft risk)

Effect of global air recirculation High air velocity, with lower

in the lab, flow away from windows temperatures, in the region Low air temperature
near the floor may cause discomfort in occupied zone

a) Velocity Field b) Temperature Field

Contours of Temperature and Velocity Profile Scale:

Teaching Lab, Scenario 1 N.T.S.
Interior Air Flow Analysis Date revised: RWDI
Guyot Hall Addition at Princeton University April 29, 2002

Figure 4. Guyot Hall CFD model.

CH3CL (mole)
Time = 1620.0
ppmv
125.000
109.375
93.7500
78.1250
62.5000
46.8750
31.2500
15.6250
0.00000

Figure 5. Wyeth Labs CFD model snapshot.

 L A B S F O R T H E 2 1 S T C E N T U RY 11

Perform a n c e As s u r a n c e precisely does the HVAC system airflow respond to a lab

fume-hood sash being operated, or How exactly is the

C omm i ss i o n i n g

lab room pressure differential maintained during a door-

opening sequence? In addition to possible safety breeches,
Commissioning is an optimization process that the inability of an HVAC system to respond to changes in
ensures the ventilation systems design intent is achieved a labs environment directly translates into energy being
in practice. According to Bell (2006), a key initial step in wasted, often in excess of 20%.
commissioning is to compare the expected use (at the time
of design) to the actual use (at the time of commission-
ing), and to the reasoning behind ventilation levels and C o n cl u si o n
strategies stipulated in the original design. Methods for
detecting faults in ventilation system operation range from Standard ventilation design practice optimizes neither
design review to functional testing of equipment, sen- safety nor energy efficiency. While predefined code- or
sors, and controls. A survey of commissioning experience standards-based designs are the most straightforward
nationwide found that ventilation systems were the most approaches, they do not optimize a laboratorys ventila-
frequent source of performance problems, and that labo- tion rate, or verify whether the intended levels of safety
ratories achieved greater benefits (higher energy savings and comfort have been achieved by the labs design. Good
and shorter payback times) from commissioning than any and better ventilation design practices, which involve
other building process (Mills et al., 2004). in-depth analyses of a researchers task, the location of a
task within a laboratory, and more coordinated exposure
Where new technology is applied in order to attain assessments to characterize risks, translate into higher
given performance levels, e.g., sensors and variable-speed energy efficiency, lower life-cycle cost, and, most impor-
drives for demand-controlled ventilation, the equipment tantly, enhanced safety.
itself also must be commissioned to ensure that it is per-
forming as intended.

Ventilation-related devices that should be checked for R e f e re n c e s

proper performance and accuracy are as follows: Barkley, W.E. Issues in laboratory ventilation for hazard
Supply air temperature sensors placed upstream control, Proceedings of the Symposium on Laboratory
and downstream of any reheat coil. Ventilation Hazard Control (Frederick, MD, 1976).

Volumetric airflow meter/devices. Bell, G.C., E. Mills, D. Sartor, D. Avery, M. Siminovitch,

M.A. Piette. A Design Guide for Energy-Efficient Research
Pressure differential monitor for lab space with Laboratories, LBNL-PUB-777, Lawrence Berkeley National
reference to cleaner space. Laboratory, Center for Building Science, Applications
Stack exit velocity meter. Team, September 19962003.
Bell, G. Retrocommissioning Laboratories for Energy
Supply and exhaust duct static pressure.
Efficiency, Laboratories for the 21st Century, Technical
Variable speed drive (VSD) readings. Bulletin, 2006.

Laboratory equipment process load and lighting Crane, J.T. Biological Laboratory Ventilation and
load. Architectural and Mechanical Implications of Biological
Safety Cabinet Selection, Location, and Venting, ASHRAE
Energy recovery systems, when present. Transactions, 100(1):12571265, 1994.

A recommended procedure in ANSI Z9.5-2003, the Maghirang, R.G., G.L. Riskowski, L.L. Christianson, L.L.
Ventilation and Environmental Quality in Laboratory
System Mode Operational Test (SMOT), should be part of
Animal Facilities, ASHRAE Transactions, 102(2):186194,
a good practice effort. This test procedure is a system-
1996.
wide functional verification that examines how well all of
the individual lab components and design features work Marshall, J.W. Health Care Ventilation Standard: Air
together as a whole. Adequate and measured responses Changes per Hour or CFM/Patient? ASHRAE Journal,
by the central HVAC system to normal lab operations are 38(9): 2730, September 1996.
evaluated during a SMOT. An HVAC system must respond McDiarmid, M.D. A Quantitative Evaluation of Air
directly to actions performed in a lab space, e.g., How Distribution in Full Scale Mock-Ups of Animal Holding
 12 L A B S F O R T H E 2 1 S T C E N T U RY

Rooms. ASHRAE Transactions, 94(1): 685693, 1988. (This

article also appears in Laboratory HVAC, 1995, 8994, A c k n o wl e d g m e n t s
ISBN 1-883413-25-7.)
A uthor
Memarzadeh, F. Of Mice, Men, & Research. Engineered
Systems, 16(4), April 1999. Geoffrey C. Bell, P.E., M. Arch.
Lawrence Berkeley National Laboratory
Mills, E., H. Friedman, T. Powell, N. Bourassa, D. Claridge,
T. Haasl, and M.A. Piette. The Cost-Effectiveness of
The author is very grateful to the following people for
Commercial-Buildings Commissioning: A Meta-Analysis of
their important contributions and thoughtful reviews:
Energy and Non-Energy Impacts in Existing Buildings and
New Construction in the United States. Lawrence Berkeley John Piatt, P.E., Pacific Northwest National Laboratory
National Laboratory Report No. 56637, 2004.
John Peterson, P.E., University of Texas System
Riskowski, G.L., R.G. Maghirang, W. Wang. Development
of Ventilation Rates and Design Information for Laboratory Otto Van Geet, P.E., National Renewable Energy
Animal Facilities. Part 2 - Laboratory Tests. ASHRAE Laboratory
Transactions, 102(2): 195209, 1996.
Paul Mathew, Ph.D., Lawrence Berkeley National
Zhang, Y., L.L. Christianson, G.L. Riskowski, B. Zhang, G. Laboratory
Taylor, H.W. Gonyou, P.C. Harrison. A Survey of
Laboratory Rat Environments. ASHRAE Transactions, Lou DiBerardinis, Massachusetts Institute of
98(2): 247253, 1992. Technology
 13 L A B S F O R T H E 2 1 S T C E N T U RY

For M ore In fo r ma t i o n

O n O p t i m i z i n g Ventilation S y s tem s
Geoffrey C. Bell, P.E., M. Arch.
Lawrence Berkeley National Laboratory
One Cyclotron Road, MS 90-R3111
Berkeley, CA 94720
510-486-4626
GCBell@lbl.gov

O n L ab o r at o r i e s for the 21s t C entur y

Dan Amon
General Engineer
OARM/OA/FMSD/Sustainable Facilities Practices
Branch
U.S. Environmental Protection Agency
1200 Pennsylvania Ave., N.W. (mail code 3204R)
Washington, DC 20460
202-564-7509
amon.dan@epa.gov

Will Lintner, P.E.

U.S. Department of Energy
Federal Energy Management Program
1000 Independence Ave., S.W.
Washington, DC 20585-0121
202-586-3120
william.lintner@ee.doe.gov

B e s t P r ac t i c e G uides on the Web:

http://www.labs21century.gov/toolkit/bp_guide.htm

Laboratories for the 21st Century

U.S. Environmental Protection Agency
Office of Administration and Resources Management
www.epa.gov/labs21century/

In partnership with the

U.S. Department of Energy
Energy Efficiency and Renewable Energy
Federal Energy Management Program
September 2008
www.eere.energy.gov/femp

Prepared at the Printed with a renewable-source ink on paper containing at least

National Renewable Energy Laboratory 50% wastepaper, including 20% postconsumer waste
A DOE national laboratory