SPE-173996-MS

Solar-Generated Steam for Oil Recovery: Process Integration Options, Net

Energy Fraction, and Carbon Market Impacts
J. O’Donnell, M.A. Heisler, and M. Chandra, GlassPoint Solar

Copyright 2015, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Western Regional Meeting held in Garden Grove, California, USA, 27–30 April 2015.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
Objectives/Scope: Review of the process integration alternatives for interconnecting solar steam
generators into existing steam facilities, and resulting facilities costs, energy contribution (solar fraction),
and net CO2 credits generated.
Methods, Procedures, Process: Siting and land footprints were chosen for several thermal recovery
projects. Three integration options were studied: water preheating, constant-rate steam with balanced
boiler turndown, and variable-rate steam distribution. Solar collector arrays were laid out and hourly
energy production modeled. Aspen Hysys studies were carried out and line sizes, pressure drops, and
facilities costs were estimated for the full path to the wellhead. Annual total contributions to steam
injected (solar fraction) were evaluated.
Results, Observations, Conclusions: Solar preheating of boiler feedwater required the least
changes to existing facilities, and was economical especially at fields with lower temperature boiler
feedwater; solar fraction was limited in most cases to under 6%. Constant-rate steam injection increased
practical solar fraction to above 15%, while requiring changes only to the steam generator setting.
Variable-rate steam injection involved facilities changes to wellhead rate control devices but allowed over
50% solar fraction with economical facilities costs.
Novel/Additive Information: In March 2014, the California Air Resources Board released an estimate
that if 30% of California oilfield steam were solar, 3.7 million tons/yr of credits would be generated in
the Cap & Trade program, and an additional 4.3 million tons/yr in the Low Carbon Fuel Standard
program. Facilities and land constraints associated with achieving that target are reviewed.

Background
Solar steam generation has emerged as a promising large-scale source of energy for thermal enhanced oil
recovery (EOR) operations, with several pilot projects now operating in California and the Middle East.
Shortages of natural gas in the Middle East and California’s emissions regulations make solar steam
potentially lower in cost than other available steam sources. The optimum solar configuration is
2 SPE-173996-MS

determined by several factors, including reservoir injectivity, land availability, and local environmental
conditions. A primary factor in facilities design, however, is the variability of solar energy.
Solar energy is intermittently available, with both day/night diurnal and seasonal variations in sunlight.
Solar collectors deliver energy at rates corresponding to incoming sunlight. This intermittency poses
potential challenges in both the integration cost of solar energy – the changes or cost increases to add solar
steam into a distribution network – and the solar fraction – the portion of total required steam that can
be supplied by solar energy.

Figure 1—Solar Steam Generator (Central Receiver)

Figure 2—Solar Steam Generator (Enclosed Trough)

Previous studies (van Heel et al. 2010, Sandler et al. 2012) have considered the impact of solar steam
injection in several formation types (sandstones and fractured carbonates), and reported results including:
1. Oil production rate and ultimate recovery are unaffected by day-night diurnal variation in steam
injection rate. A given daily mass of steam injected, whether all during an 8-hour daylight period
or spread across a 24-hour period, is equally effective. From a recovery standpoint, solar energy
could supply 100% of steam requirements.
2. Seasonal variability in solar energy, however, affects oil production rates if solar is the only source
of steam. Lower injection rates associated with winter or cloudy periods resulted in lower
production rates in later months. As seasonal variations in oil production rates are undesirable,
seasonal variation poses a practical limit to solar fraction.
Figure 3 shows average daily energy delivered, by month, for oilfield locations in southern Oman and
southern California. Results from the abovementioned studies suggest that for these locations, practical
SPE-173996-MS 3

values of 80% solar fraction in Oman, and 50% solar fraction in California, could be achieved without
changes in the oil production rate.

Figure 3—Typical daily MMBTUs delivered, by month, vs annual average

These studies, however, did not consider facility limitations or integration cost limits, where the diurnal
variability of solar energy does create unique challenges. The balancing of fuel-fired steam and solar
steam, and the sizing of facilities which accept higher peak steam flows during daylight hours, could
increase costs or pose limits. The present work considers such costs and potential limits.
A San Joaquin Valley, California site was chosen as a model, with solar steam production based on
typical weather conditions at the site. The goal is to use energy from incident sunshine to heat up the
reservoir at the lowest possible cost, subject to different assumptions regarding facilities constraints. The
resulting solar fraction and integration costs are evaluated. Four cases are considered, ranging from 5%
to 50% solar fraction, with their facilities, cost, and carbon market implications.
Cases Considered
We consider a thermal EOR operation wth conventional water treatment, steam generation, and steam
distribution facilities as in common current practice. Solar energy is added to the project, replacing fuel
combustion, in one of four configurations.
A base fuel-fired steam generation and distribution case was chosen as per Figure 4 with the following
elements:

Figure 4 —Fuel-Fired Steam Generation and Distribution (Sarathi & Olsen 1992)
4 SPE-173996-MS

● 144,000 barrels of steam per day (cwe) average daily injection

● 30 Once-Through Steam Generators (OTSGs), each 85 MMBTU/h firing rate
● 300 injectors, each with flow control device
● 1000 psig, 75% quality min at wellhead (upstream of flow control)
● 100° F softened generator feedwater (per API RP-11T)
The setting of 30 OTSGs, each with 85 MMBTU/hr firing rate, which serve a distribution network and
well grouping, was chosen to be representative of typical California practice in steam operations. An
average flow rate per injector of 500 barrels of steam per day (bspd), was chosen to represent typical
California steam flood operations. Often California steamflood operations have a preceding cyclic steam
(CSS) operation period; during this startup period a rate of 1000bspd/injector is often used.
As solar energy is intertied to this facility, the existing fuel-fired OTSGs and distribution network
remain. Solar steam generators are positioned a distance away from the existing OTSGs, preventing
interference with current production and future potential drilling operations.
Solar energy reduces the amount of fuel fired in OTSGs for the same daily average steam produced.
Four different cases are considered with different operating modes, design considerations, and results.
Case 1 – Water Preheating
Solar energy is used to preheat boiler feedwater to the OTSGs as shown in Figure 5. Solar energy collected
during the day heats water in the softwater storage facilities, providing up to 24-hour increased steam
production. We assume 100°F water temperature exiting the Water Treatment Plant, and an upper limit
of 200°F into the storage tank. To mitigate line size and flow rates, pressurized water is returned at 446°F
and mixed down at the water storage facility. The solar field was sized to preheat 100% of daily 143,000
barrels of boiler feedwater on best summer days. Water preheating is sharply limited in its potential solar
fraction, both because of the limited enthalpy contribution of preheating 100°F to 200°F, and because
most thermal recovery projects experience rising produced fluid temperatures over the operating period,
which eliminate the opportunity for preheating.

Figure 5—Case 1: Solar Boiler Feedwater Preheating

Case 2 – Constant-Rate Balanced Steam

Solar steam generators are intertied to the steam distribution header. Hourly variations in steam produced
by the solar steam generators are balanced by adjustments to the fuel firing and pump rates of the OTSGs,
resulting in approximately constant steam on a 24-hour basis.
SPE-173996-MS 5

Figures 6 and 7 show the integration concept and balanced operation on a summer day. Fuel-fired
steam generators are turned down in firing rate based on current availability of solar steam, so that total
steam output is approximately constant. Fixed-bean choke flow control devices at the wellheads are
unchanged, as total flowrate remains within the sonic region and distribution pressures remain within
design limits. The solar field was sized to maximize solar fraction based on peak flow limit established
by the distribution network (10% higher peak vs average flow).

Figure 6 —Case 2: Constant-Rate Solar Steam

Figure 7—Case 2: Solar and Fuel-Fired Steam Flow By Hour

Cases 3 and 4 – Variable Flowrate Steam

In Cases 3 and 4, the peak and minimum rates of steam injection are allowed to diverge further. Due to
overnight lower flow rates, fixed chokes cannot reliably proportion steam among injectors. Instead, a
facilities design widely used in CSS operations is adopted, with per-injector motor operated valves and
flow sensors as shown in Figure 8, providing automatic steam flow proportioning across all injectors at
varying total injection rate.
6 SPE-173996-MS

Figure 8 —Cases 3 & 4: Variable-Rate Total Steam

In Case 3, the peak-hour steam flow rate is 1000bspd (42 bsph), with lower flow rate at other times of
day delivering the net 500bspd average as shown in Figure 9. Case 3 is an interesting design point to
consider, as the peak rate from the manifold to each well is unchanged from typical practice; however, the
distribution header carries a higher flowrate and is increased in size.

Figure 9 —Case 3: Solar and Fuel-Fired Steam Flow By Hour

In Case 4, the peak flow rate of steam injection is allowed to rise further, resulting in some increase
in sizing of the distribution lines to the wellhead. Case 4 achieves the highest solar fraction, and the lowest
blended cost of steam. Case 4’s minimum flow range is limited to prevent phase separation in the steam
distribution network and to avoid daily start-stop operation of fuel-fired steam generators.
Results Overview
For each of the study cases, a placement of the solar array was considered, annual energy contribution was
evaluated, and integration piping was simulated and costed. Figure 10 presents a summary of results,
which are discussed further below.
SPE-173996-MS 7

Figure 10 —Results Overview

Total daily steam injection is constant across all cases. The solar fraction is expressed as the portion
of total fuel replaced by solar energy. The integration cost – the cost of building the water and steam
distribution network – is expressed as a percentage of project cost. Land costs are not calculated, as land
cost is site-specific, but acreage (project area) is reported.
We find that for all the configurations, integration costs are small as a percentage of the total project
cost (cost of the solar facilities). As a result, the fundamental economics of the solar steam generators –
the net cost of steam, compared to steam generated from conventional fuel-fired sources – dominate
project economics.
Carbon Market Impacts
Solar steam generators displace steam that otherwise would have been provided by fuel combustion in
fuel-fired steam generators.
Winslow (Winslow 2012) reported that steam generation accounted for more than half of combustion
in California production operations, as shown in Figure 11. Separately, ICF (ICF 2015) reported that 90%
of California upstream energy use is the form of steam, and that 185 million MMBTU of fuel were used
for steam in 2012.

Figure 11—Combustion Sources in California Oil & Gas Production (Winslow 2012)

The emissions reductions associated with the operation of solar steam generators generate credits in
California carbon markets. Two separate credit market mechanisms apply to solar steam generation in
California oilfields. The credit calculation mechanisms differ slightly under each regulatory regime. This
section discusses the calculations and the net impact on the economics of operating solar steam generators
in California.
Cap and Trade
The California Greenhouse Gas Cap-and-Trade Program establishes an overall greenhouse gas (GHG)
emissions target – a “cap” – and a market mechanism for compliance instruments to allow industry to seek
the lowest cost emissions reductions.
8 SPE-173996-MS

All entities covered under Cap-and-Trade must report GHG emissions, and must submit Greenhouse
Gas Allowances corresponding to those GHG emissions. Operators of thermal EOR projects are granted
a portion of their required allowances based on the volume of oil produced; remaining allowances must
be purchased through a market structure. The free allowances are calculated under the regulation using an
Industry Benchmark output-based mechanism.
A simplified form of the calculation is

where
“A” is the number of California GHG allowances allocated;
“Oa” is the number of barrels of oil produced with thermal EOR;
“Ba” is the emissions efficiency benchmark (EEB) per unit of output, currently 0.0811 credits per barrel
of oil produced.
If the source of steam for an EOR project is an 88% LHV efficient OTSG, Ba of 0.0811 corresponds
approximately to a steam-oil ratio (SOR) of 3.36. Brandt and Unnasch (2010) report California produc-
tion-weighted average SOR of 5.13, indicating that the typical EOR operator is required to purchase
GHAs on an ongoing basis.
Steam provided by solar steam generators replaces fuel-fired steam one-for-one on a mass basis (van
Heel et al 2010; Sandler et al 2012). Each barrel of solar steam displaces one barrel of fuel-fired steam,
and thus eliminates the burner-tip combustion emissions associated with producing one barrel of fuel-fired
steam. In the 88% efficient OTSG case above, each barrel of solar steam eliminates 24,106 grams CO2
emissions at the oilfield, so every 41 barrels of solar steam produced eliminate the need to purchase one
greenhouse gas allowance.

Low Carbon Fuel Standard

The California Air Resources Board (CARB) has proposed to readopt the Low Carbon Fuel Standard, a
performance standard which requires reductions in the carbon intensity (CI) of transportation fuels used
in California, with 10% reduction required for 2020 and subsequent years. The proposed regulation was
released December 2014 (CARB 2014) and a first board hearing on readoption was held February 19,
2015. Final adoption is scheduled for summer 2015. The “Illustrative Compliance Scenario” shows
compliance causing replacement of gasoline and diesel fuel with biomass-based liquid fuels. The proposal
allows “innovative methods” of crude oil production to generate LCFS credits in the petroleum supply
chain.
Of the potential innovative methods, the use of solar energy is the lowest-cost, lowest-risk, and
largest-scale opportunity to reduce the CI of petroleum fuels produced and used in California. Solar steam
generation used in thermal enhanced oil recovery (EOR) displaces natural gas combustion. CARB staff
released estimates that, net of technical and land use constraints, solar steam could replace 30% of oilfield
steam currently used in California.
ICF International carried out an economic impact study (ICF 2015) which compared the LCFS credits
created by solar steam generation versus total credits available from other potential compliance pathways
in 2020 (Figure 12). The range considered for solar steam corresponded to 5% and 30%, respectively, of
California oilfield steam supplied by solar steam generators. At the lower 5% figure, solar steam
generation would deliver more CO2 reductions than all electric vehicles combined; at the 30% figure, solar
steam generation would deliver more reductions than all cellulosic ethanol forecast to be available in
2020. ICF estimated that each 1 ton of credit delivered by solar steam generation would preserve
California market demand for between 120 to 130 gallons of diesel fuel, or roughly 564 million
gallons/year total.
SPE-173996-MS 9

Figure 12—Total LCFS Credits Available, 2020, By Pathway (after ICF 2015)

LCFS credits generated by solar steam are calculated using the methodology specified by CARB in the
regulatory package (CARB 2014, Appendix A and Appendix G):

where
Creditsinnov_SolarSteam is the amount of LCFS credits generated in metric tons by the volume of crude
oil produced and delivered to California refineries for processing;
VSteam is the volume in barrels of cold water equivalent of steam injected,
fsolar is the fraction of steam injected that was produced using solar energy;
Vcrude_produced is the volume (in barrels) of crude oil produced using the innovative method;
Vinnov_crude is the volume (in barrels) of crude oil produced using the innovative method and delivered
to California refineries for processing;
and C is the constant to convert from metric tons to grams (1.0*10-6 MT/gCO2e).
For oilfields which deliver all their output to California refineries – essentially all California production
operations – this simplifies to 29,360 grams per barrel of solar steam (CWE). The constant at the outset
of the equation, 29,360, is the “default score” in gCO2 per barrel of steam, for solar steam generated of
75% quality or greater. A second value, 28,011 is the default value for solar steam generated of 65%-75%
quality.
Note that the gCO2/barrel values are a higher value under the LCFS regulation than under Cap & Trade.
The LCFS program includes the assessed upstream emissions in producing natural gas in the lifecycle
analysis for solar steam at the oilfield, whereas Cap & Trade captures burner-tip emissions only; Cap &
Trade covers upstream emissions separately at the gas production site.
The 2011 LCFS regulation included provisions for solar steam, but required complex calculations of
net CI and only allowed credit monetization by refiners. Under the 2015 readoption, these matters have
been simplified. Default scores, shown above, simplify applications and record-keeping, and crude
producers may elect to bundle credits with the oil produced using solar steam, or to unbundle the credits
and sell oil and credits independently.
The credits market has experienced high volatility in recent years, primarily due to changes in the
near-term compliance curve. The market starting in 2018 is forecast to be tight, with market prices
reflecting alternate fuel production costs. Because of the relatively high cost of abating GHG emissions
with biofuels, the LCFS market is forecast to have high credit prices. The economic impact analysis
released by CARB (CARB, 2014, Appendix F) assumes a 2020 price of $100/credit. Due to concerns
10 SPE-173996-MS

about low-CI fuel availability, the system has a ceiling price set at $200/credit. ICF International’s study
for the Natural Gas Vehicle Coalition (ICF 2014) assumed 2020 prices in a range of $65/ton to $170/ton
(Figure 13).

Figure 13—Assumed LCFS prices based on marginal abatement (ICF 2014)

At such credit prices, solar steam would be the lowest cost source of steam for California operators.
When the value of credits are included, solar steam would be least-cost even when the value of natural
gas is assumed at $1/MMBTU.
Figure 14 provides an example of the value of the credits at two different price points. The value of
solar steam in 2020 is broken down by energy value (natural gas purchase avoided), cap & trade value
(credit purchase avoided), and LCFS value (credits for sale). At the nominal $100/credit 2020 price, the
value of the LCFS credits outweighs the direct fuel price and cap & trade value of solar steam. The
expected 2020 LCFS credit value is 60% of the total value of solar steam, assuming 2020 prices of
$3.33/MMBTU for natural gas and $17.33/credit C&T price (the floor price). If the credits market
proceeds as forecast, the net cost of energy to oilfield operators from solar steam will be well below the
cost of gas-fired steam.

Figure 14 —2020 Value Per MMBTU, LCFS credits at $50 and $100
SPE-173996-MS 11

Based on the land density achieved in Case 4, the total land used by solar facilities to deliver 4.3 million
credits/yr would be 1085 acres (9.6 square miles). Solar steam facilities deliver 701 tons CO2/acre/yr. The
highest land efficiency achieved by a volume production ethanol fuel appears to be sugar cane (CARB
2014, Appendix B, B-40), whose land efficiency is reported (Brazil Institute 2007) as 727 gal/acre/yr for
a net 2020 credit generation of 3.0 tons/acre/yr. This land efficiency ratio – 230:1 – is a striking result.
A second aspect of carbon market impacts deserves further study, as the financial value of solar steam
to the oilfield operator may be further increased by demand-related market factors.
Of the compliance pathways shown in Figure 12, all except Solar Steam and Solar Electricity involve
demand destruction for petroleum fuels in California. ICF’s economic impact analysis (ICF 2015)
identified refinery profitability contributions due to availability of solar steam credits; however the study
was not an energy model which could evaluate changes in the price of California crude.
The 564 million gallon prospective market impact could increase California producer crude prices, but
this has not been quantified.

Integration Study Details

The integration study assessed the interconnection and piping cost for integrating solar hot water or solar
steam into an oilfield operation, making conservative assumptions about pressure drop, erosional velocity
limitations, and solar field placement.
Integration costs are reduced when solar collectors are placed close to the point of use of the steam.
However, solar facilities must generally be placed outside the oil-bearing zone, to avoid adverse impacts
on future development. We selected a value of 1.25 miles of piping from the solar field outlet flange to
the intertie point, plus 10% additional pipe length for thermal expansion loops, as representative of typical
project placements.
As described above, four cases were considered:
1. Solar preheating of water 100°F-200°F at the softwater storage tank. Solar field sized to deliver
832 MMBTU/h as hot water (244 MWt), 5% solar fraction.
2. 80% steam delivered to the distribution network. Solar field sized to deliver 1,885 MMBTU/h as
steam (552 MWt), 15% Solar Fraction, 4:1 OTSG turndown, roughly constant 24-hour steam flow
(1.1:1 peak-to-average flow rate).
3. 80% steam, Solar delivers 3,985 MMBTU/h (1,141 MWT), 30% Solar Fraction, 2.1:1 peak-to-
average flow rate.
4. 80% steam, Solar delivers 6,394 MMBTU/h (1,873 MWt), 50% Solar Fraction, 3.24:1 peak-to-
average flow rate.

Modeling Approach
The hourly solar energy output for the entire year was derived using Coalinga, California Typical
Meteorological Year data provided by NREL’s Solar Prospector (NREL 2012).
Fuel-fired steam was considered to be provided by 85 MMBTU OTSGs with peak flow of 208 bsph
(4992 bspd), with 4:1 turndown capability. Turndown factor was selected based on discussions with
OTSG and burner vendors and field operators, including considerations of NOx emissions management.
A “dispatch model” was built which calculates hourly steam flow, based on available solar steam
output, OTSG control constraints, and facility constraints. The solar field sizing for each case was chosen
to optimize solar fraction and energy cost, after flow limitations were considered, using standard block
enclosed trough collectors as the unit.
Minimum steam injection pressure required at the wellhead surface upstream of wellhead flow control
device (fixed choke or MOV) is assumed to be 1000 psig with 75-80% minimum acceptable steam quality
in all cases.
12 SPE-173996-MS

An Aspen HYSYS model was created for each case. Peak hourly energy output, during the highest
hour-of-year energy production from the dispatch model for the case, was considered. Due to the variable
availability of solar radiation, peak flowrates are observed to occur less than 390 hours per year.
Considerations of pressure drop, erosional velocity and bulk fluid velocity limits drove pipe size selection.
Based on API-14 E standards, C value of 125 was selected, corresponding to non-continuous service, is
used for steam header erosional velocity calculations.
Other key assumptions include:
● Solar array to intertie point is 1.25 mile pipe run
● Mineral wool (K⫽0.036 W/m.K, 3.5⬙ thickness) is used as the insulating material.
● Ambient temp. of 68°F and air velocity of 11.5 ft/s is used for pipe thermal loss calculation.
● Enclosed trough solar steam generator sizing, energy output profile, and placement was consid-
ered, using a standard block 945ft X 590ft in size with 36 rows of receivers inside each glasshouse.
● Feed water line sizing assumed 100 psig and 100°F water supply at the water treatment plant.
● Direct piping, installation, and insulation cost estimates are based on latest Richardson’s con-
struction pricing for Fresno, California, USA. The costs of flanges, elbows, tees, valves, and other
fittings was approximated at 15% of direct installed piping cost, based on field experience.
● Interconnection cost excludes pipe costs within the solar field battery limit, as they are considered
part of the “base cost” of the solar collectors.
● Pipe material selected is carbon steel.
● All the steam headers and sub headers are assumed to be schedule 80. All the feed water lines are
schedule 40.
● Modified Beggs and Brills correlation is used for pressure drop calculations.
● 10% length is added for thermal expansion to all lines (including main solar array line).
● 30 psi pressure drop is assigned for control valves and fittings in the steam header.

Discussion, Case 1
The solar array has a relatively larger interconnection line in the water-heating case than in the
steam-generation cases. In the steam generation cases (Cases 2-4), the solar array converts boiler
feedwater to 80% quality high-pressure steam, adding approximately 1000 BTU/lb. In Case 1, the solar
field lifts the temperature of water by 346F, resulting in approximately 3x higher required water supply
flow rate in the feedwater line per unit of solar energy delivered. For study purposes, a tank of 4000 barrel
capacity close to the solar array was considered. Feedwater is delivered on a 24-hour basis via a 6⬙ supply
line. Heated feedwater is returned at varying temperature and flow rate, using a 12⬙ line (Figure 15).

Figure 15—Case 1 Model Geometry & Layout

SPE-173996-MS 13

Figure 16 —Case 1 Model Results

The 1.25 mile line length with larger line diameter for the higher water flow rate results in somewhat
higher integration costs of solar water heating, compared to steam generation. However, these integration
costs are offset by three considerations:
– Reduced costs within the solar array, due to simpler controls and lower piping pressure ratings
(order 5%)
– Increased energy output from the solar array of about 8%, due to reduced average collector
temperature and subsequent reduction in thermal energy losses
– Simplified control strategy for the solar-OTSG interconnection. Feedwater temperature variations
to the OTSGs are handled by conventional OTSG boiler control systems, and net boiler controls
(changes in fuel firing or water supply rate) are on the order of 10% rate variation.

Discussion, Case 2
In Cases 2 through 4, the solar array is generating 80% quality steam, intertied to the steam distribution
network via a 1.25 mile long steam line. During daily startup the interconnecting line is warmed at a
managed rate by a small amount of reverse flow. The heat and liquid from reverse flow and startup
operations is recaptured in the water storage tank. The solar array is brought to operating temperature and
steam quality before steam production begins. During the operation of the solar steam generator, control
signals are delivered to the OTSGs, reducing firing rates to balance total instantaneous steam flow; as a
result, steam flow through segment P2 is approximately constant on a 24-hour basis.
The HYSYS case uses a simplified network which captures total piping length as shown. Line sizing
so as to achieve 1000 psig at the farthest injector is chosen.
14 SPE-173996-MS

Figure 17—Case 2-4 Model Geometry & Layout

An OTSG Tie-in Pressure of 1420 psig is assumed. Fixed-bean chokes are installed at all the injector
wellheads.
A small amount of rate variation is accommodated in this case; peak steam flow rate is 1.1 times
average rate. This rate increase causes a rise in wellhead pressure. The maximum OTSG output pressure
is then calculated by combining the effect of larger pressure drop through the steam headers due to
increase in flow after solar integration and adding it with the pressure increase due to fixed bean chokes.
Based on below formula1,

Where
n ⫽ index of isentropic expansion,
Pc ⫽ Critical pressure needed for choked flow.
Injector Downstream Pressure, P1 ⫽ 500 psig

1
(Ref: http://www.engineeringtoolbox.com/nozzles-d_1041.html)
SPE-173996-MS 15

Injector Upstream Pressure, Pu ⫽ 1000 psig

Since Pu⫽Pc, we can assume that the flow in the fixed bean chokes are sonic. For choked flow
following formula is used to define the relation between mass flow rate, Mc and upstream pressure Pu,

Where
Ac ⫽ Area of nozzle,
␳u ⫽ upstream steam density
Pu ⫽ upstream pressure
For the mass flow increase resulting from solar integration, the formula above predicts 180 psi pressure
increase caused due to the fixed bean chokes which increase the OTSG output pressure to 1520 psig,
which is below the typical high pressure trip settings for typical OTSGs.

Figure 18 —Case 2 Model Results

Discussion, Case 3
In Case 3 the wellhead flow control devices are motor-operated valves (MOVs) with flow sensors.
Automatic proportioning of total steam flow is accomplished by the field control system. We retain the
assumption of 1000psig at the inlet to each MOV. Pressure at the OTSG tie-in point is slightly lower than
in Case 2, as the wellhead pressure increase considered in the fixed-bean example is not applicable.
Note that in Case 3 the peak flow rate per injector is 1000 bspd rate – the rate typically used during
CSS at the project start – although daily total steam injection is 500 bspd per injector, as typical in many
16 SPE-173996-MS

California steamfloods. As a result no subsurface / well sizing / injectivity changes are assumed versus
a base fuel-fired steam injection case.

Figure 19 —Case 3 Model Results

Discussion, Case 4
Case 4, like case 3, has per-well motor-operated chokes (MOVs) with flow sensors. The ratio of peak to
average flow is 3.24:1 – during summer months, solar steam is providing roughly 80% of total steam, and
on a year-round basis the solar fraction is 50%.
Case 4, with the highest rate variation and highest peak flow, requires the greatest study of field-
specific factors such as injectivity and injector well thermal integrity. Previous work (van Heel et al. 2010,
Sandler et al. 2012) has suggested that for many formation types, the assumptions in Case 4 are
achievable; shallow or low-permeability formations, however, may not be able to benefit from this
operating configuration without reductions to daily per-injector total steam flow.
SPE-173996-MS 17

Figure 20 —Case 4 Model Results

Larger Field Layouts

In fuel-fired steam operations, as in solar steam generation, it is desirable to minimize the total length of
steam lines, to reduce costs and thermal losses. Common practice in larger steam flood operations in
California and Oman is to organize the steam network as multiple smaller networks, each with a setting
of steam generators that provide steam to that subsection of the total field.
Solar steam generation integration with such a larger field would likely organize the solar collectors
into separate settings, each integrating with a specific setting of OTSGs, where each sub-grouping would
be similar to Cases 1 through 4 above, or (more likely) somewhat smaller in each setting.
Therefore, the results obtained in Cases 1 – 4 for integration cost as a percentage of total cost would
be expected to apply in such larger projects.
A conceptual layout for such a larger field is shown in Figure 21.
18 SPE-173996-MS

Figure 21—Larger Oil Field Layouts

Key Considerations
The following key points were observed during the study,
1. For Case 2 there is no need to replace fixed bean chokes at the injector well heads, although overall
pressure rating would be lower with MOVs. For Cases 3-4 Motor operated valves (MOVs) are
required to properly manage flow to each injector wells.
2. To minimize the cost of steam headers and daily thermal losses through the steam pipes, the solar
blocks should be placed as close to the injector wells as possible. The cost of steam headers per
feet far exceed the cost of feed water and power lines combined. Hence reducing the steam header
length should be higher priority than feed water and power distribution lengths. For an oil field like
Figure 21, the solar fields should be placed as close to OTSG banks as possible to minimize the
length of long steam and feed water lines.
3. Splitigators or Insert Tees could be used for maintaining good steam quality at flow splits. Pipe
diameter could be reduced to increase bulk fluid velocity at the Tee for more uniform quality
distribution.
4. Automatic balancing of the output of OTSGs versus solar steam generators can be achieved with
modern OTSG controls and simple signaling from a field management PLC.
5. The inclusion of direct solar radiation measurement and whole-sky cameras as well as the use of
historical data can minimize pressure fluctuations in the steam header during solar transient events.
6. Proper ANSI rating of flanges should be used to ensure allowable working pressure is always
higher than expected peak operating pressure.
7. The steam pressure should be monitored at different points in steam network to ensure steam
SPE-173996-MS 19

headers are operating within the design boundaries.

Conclusions
Solar steam generators can be integrated with existing facilities networks with relatively small integration
costs. Design alternatives allow solar energy to economically provide between 5% and 50% of total
primary energy needs for oilfield steam in California locations. The management of rate variation added
a maximum of 3% to solar project cost across all cases studied.
The economics of solar steam generation depend critically on 4 factors:
1. The capital and operating costs of the solar steam generator units themselves;
2. The annual solar radiation at the project location;
3. The value of the fuel which is replaced by solar radiation; and
4. The value of carbon credits provided by solar steam generation.
In California’s emerging regulatory environment, solar steam may offer significantly lower cost of
energy than other sources of steam. Solar EOR gives California operators the opportunity to reduce their
long term fuel demand and emissions, meet state emissions criteria, and reduce sensitivity to future natural
gas and cap and trade prices. Crude produced with solar energy will be priced at a premium to other
methods and will prevent demand destruction for petroleum in the California fuel mix.

References
Agarwal A., Kovscek A. “Solar-Generated Steam for Heavy-Oil Recovery: A Coupled Geomechani-
cal and Reservoir Modeling Analysis”, SPE Western Regional & AAPG Pacific Section Meeting,
2013 Joint Technical Conference, Monterey, California, USA, 19⫺25 April 2013 SPE 165329-MS
http://doi.org/z9z
Bierman B., O’Donnell J., Burke R., McCormick M., Lindsay W. “Construction of an Enclosed
Trough EOR System in South Oman”, SolarPACES 2013, http://goo.gl/T2MB7m http://doi.org/
z94
Bierman B., Treynor C., O’Donnell J., Lawrence M., Chandra M., Farver A., von Behrens P., Lindsay
W., “Performance of an Enclosed Trough EOR system in South Oman”, SolarPACES 2013,
http://goo.gl/4Rxnen http://doi.org/z95
Brandt, A.R., Unnasch, S. “Energy Intensity and Greenhouse Gas Emissions from Thermal Enhanced
Oil Recovery” Energy Fuels, 2010, 24 (8), pp 4581–4589 http://doi.org/fcqghn
Brazil Institute, “The Global Dynamics of Biofuels”, April 2007. http://www.wilsoncenter.org/sites/
default/files/Brazil_SR_e3.pdf
California Air Resources Board, Cap and Trade Regulation “Regulation for the California Cap on
Greenhouse Gas Emissions and Market-Based Compliance Mechanisms to Allow for the Use of
Compliance Instruments Issues by Linked Jurisdictions”, January 2015 http://www.arb.ca.gov/cc/
capandtrade/capandtrade/unofficial_c&t_012015.pdf
California Air Resources Board, “Low Carbon Fuel Standard Regulation”, December 30, 2014.
http://www.arb.ca.gov/regact/2015/lcfs2015/lcfs2015.htm
ICF International, “California’s Low Carbon Fuel Standard: Compliance Outlook & Economic
Impacts”, April 2014. http://www.cngvc.org/pdf/ICF-Report-Final.pdf
ICF International, “The Impact of Solar Powered Oil Production on California’s Economy”, January
2015. http://www.icfi.com/insights/reports/2015/solar-powered-oil-production-california-econ-
omy
National Renewable Energy Laboratory, “The Solar Prospector”, 2012 http://maps.nrel.gov/node/10
20 SPE-173996-MS

Palmer D., O’Donnell J., “Construction, Operations and Performance of the First Enclosed Trough
Solar Steam Generation Pilot for EOR Applications”, SPE EOR Conference at Oil & Gas West
Asia, Muscat, Oman, 31 March – 2 April 2014. SPE 169745-MS http://doi.org/z9x
Sandler J., Fowler G., Cheng K., Kovscek A. “Solar-Generated Steam for Oil Recovery: Reservoir
Simulation, Economic Analysis, and Life Cycle Assessment” SPE Western Regional Meeting,
Bakersfield, California, USA, 19 –23 March 2012. SPE 153806-MS http://doi.org/z92
Sarathi P., Olsen D., “Practical Aspects of Steam Injection Processes, A Handbook for Independent
Operators”, 1992, National Institute for Petroleum and Energy Research, http://www.netl.doe.gov/
kmd/cds/disk27/Niper-580.pdf
van Heel A.P.G., van Wunnik J.N.M., Bentouati S., Terres R., “The Impact of Daily and Seasonal
Cycles In Solar-Generated Steam On Oil Recovery”, SPE EOR Conference at Oil & Gas West
Asia, Muscat, Oman, 11–13 April 2010. SPE 129225 http://doi.org/10/bq2536
Winslow, D. R. “Impact of California’s Greenhouse Gas Law on the Upstream Oil & Gas Industry”
SPE Western Regional Meeting, Bakersfield, California, USA, 19 –23 March 2012. SPE
154722-MS http://doi.org/z93