SPE 172004-MS

Economics Of Steam Generation For Thermal EOR Marwan Chaar, GlassPoint Solar Milton Venetos, Wyatt Enterprises Justin Dargin, University of Oxford Daniel Palmer, GlassPoint Solar

Copyright 2014, Society of Petroleum Engineers This paper was prepared for presentation at the Abu Dhabi International Petroleum Exhibition and Conference held in Abu Dhabi, UAE, 10–13 November 2014. 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 The thermal EOR steam generation projects in Gulf oilfields are on such a large scale that they affect an entire country’s economic position. As such, the policies related to oilfield steam generation should be decided at the national level using the cost of the marginal fuel. This paper calculates the steam cost for three methods: 1) once-through steam generator (OTSG) 2) once-through heat recovery steam generator (OT-HRSG) and 3) solar steam generator (SSG). We have created detailed performance and economic models of the steam generation methods and used them to calculate the levelized cost of energy and the Fuel Break Even (FBE). We explore the environmental and economic burdens on the cost of steam generation. The effect of fuel price on the cost of steam is also analyzed with a focus on the marginal fuel price. The analysis shows that the fully burdened steam costs using $6/MMBTU fuel, for OTSG, OT-HRSG, and SSG are $27/ton, $20/ton, and $17/ton, respectively. The FBE for SSG vs. OTSG is $4.95/MMBTU when the OTSG is unburdened and decreases to $2.25/MMBTU when the environmental burden of Carbon Cost is added. The FBE for SSG vs. OT-HRSG is $7.70/MMBTU when burdened with Power Opportunity Cost and $4.50/MMBTU when the additional burdens of Carbon Cost and Water Opportunity Cost are accounted for. Finally, we analyze the limitations of OT-HRSG in an isolated oilfield where the electric:thermal demand necessitiates electricity-matched cogeneration. This limitation along with the steam cost at the marginal fuel price provides the decision-maker with a steam supply curve.

2 SPE 172004

1.1 Introduction/ Global Oil Outlook Of the remaining oil reserves in the world, only 30% is considered “conventional” or “light oil” (with API of 22 or lighter), while the remaining 70% is heavy1. According to the IEA, boosting oil recovery of these heavier crudes could unlock around 300 billion barrels of oil. There are three main categories of EOR: 1) thermal 2) miscible gas injection 3) chemical. Thermal methods are mainly applicable to heavier crudes, at shallower depths, and these thermal methods represent the majority of global EOR production, accounting for 2.3 million barrels per day in 20132. Some of the largest Thermal EOR projects in the world are in Canada, Russia, Venezuela, Indonesia, California, Oman, and soon to be Kuwait. The steam generated for Thermal EOR consumes 1.7 TCF per year of natural gas. Thermal EOR projects tend to be very long-term projects by oilfield standards. In California many of the oldest steam flood projects have been running for 40 or 50 years. The super giant heavy oilfields of the Middle East may produce for a century or more. With natural gas becoming increasingly constrained and expensive in many parts of the world, there is a need to better understand the economics of steam generation. As this paper discusses the combination of traditional sources of generating steam with solar steam generation, we will focus on countries with sufficient sunshine and constrained natural gas supply, such as Oman, Kuwait, Sauid Arabia, Bahrain, and Egypt. The EOR potential of these countries is estimated at 475 billion barrels of oil3. 1.2 Steam Generation For Thermal EOR We have considered three methods of steam generation:

1) Fuel-fired once through steam generator (OTSG) 2) Co-generation with a power plant using a once-through heat recovery steam generator (OT-HRSG) 3) Solar steam generator (SSG) using concentrating solar power (CSP)

The first method, OTSG, directly burns fuel to generate steam. OTSGs have the most flexible operations but are most dependent on fuel costs. The second method uses the gas turbine’s high temperature flue gas as “waste heat” to produce steam in a once-through heat recovery steam generator. The OT-HRSG’s steam production is linked to the GT’s power production. Operators sometimes add supplementary firing to the OT-HRSG, called Duct Burners. The steam produced from duct burning has the advantage of re-balancing the electricity vs. thermal demand but it is directly linked to fuel price. The third method, SSG, uses mirrors to concentrate the sun’s energy to generate steam. The three operating plants are: the 21Z (2011) and Amal SSGP (2012) projects use Enclosed Trough technology and the Coalinga project (2011) uses Tower technology. SSG has the highest CAPEX of the methods considered but consumes no fuel.

OTSG OT-HRSG SSG

SPE 172004 3

The pros and cons of these three methods are summarized below: Method Pros Cons

OTSG - Low capital cost per ton of steam produced - Short construction time - Flexible and controllable steam output

- Cost of steam highly dependent on fuel price

OT-HRSG - Low capital cost per ton of steam produced - Increases system efficiency of simple-cycle power

plant4

- Linked directly to power generation - Indirectly consumes natural gas - Duct burning dependent on fuel price

SSG - Does not consume fuel - Does not produce GHG emissions - Can extend field life5

- High capital cost - Dependent on weather

1.3 Middle East Fuel Pricing The fuel price throughout the countries in the GCC and the broader Middle East vary greatly, but the common theme is its subsidization. While the current cost of production in the Gulf non-associated fields is approximately $5-8/MMBTU6 the price at which gas is sold to the end user is typically a fraction of that price, averaging just $1.50/MMBTU in the GCC region7. Justin Dargin, has written extensively on the topic of gas pricing in the region and has argued that price reform is an essential step to increasing availability of natural gas and improving energy efficiency in these countries. Another factor facing these countries is the price of the marginal fuel. Countries like Oman and Kuwait are gas-constrained and the marginal fuel is either imported LNG8 (or lower LNG exports) or diesel & other liquid fuels. For the reasons above, we have chosen to run our economic analysis using two tiers of gas prices; the first is representative of true gas production costs in the region for non-associated gas fields and is taken to be an average of $6/MMBTU. The second is the LNG market price (or opportunity cost9), taken to be $13/MMBTU10. In addition to pricing, the region is facing a shortage of fuel availability. 1.4 Economic & Environmental Concerns There are two significant concerns of fuel-fired steam generation. The first is the broader nationwide implication of diverting natural gas from steam generation to economic development. An increase in the amount of gas available for domestic use will allow for investments in industry and subsequent job creation. The second concern is the environmental impact of greenhouse gas emissions. These two concerns are addressed in the Economic Burden and the Environmental Burden sections below. 2.1 Economics Of Steam Generation The first step in analyzing the cost of steam generated from the various methods is to calculate each on a stand-alone basis. We have chosen to compare the methods using the real Levelized Cost Of Energy (LCOE) of the steam produced, as calculated by the Solar Advisor Model (SAM):

LCOE real = NPV(Total Cost Of Ownership) nominal discount rate / NPV(Total Energy Produced) real discount rate

The inputs to the numerator include cost data (CAPEX, OPEX, and fuel cost) as well as economic and environmental burdens (discussed in Sections 2.1.1 & 2.1.2 below). The input to the denominator is the steam produced, which is dependent on the performance of the method chosen (discussed in Section 2.2 below). A fair comparison between the three methods of steam production can only be achieved if all methods are “fully burdened.” The burdens considered are economic and environmental. 2.1.1 Economic Burden The economic burden considered for the OT-HRSG is the opportunity cost of the “waste heat” from the GT exhaust. The decision maker may assume that waste heat into the OT-HRSG is free. This is not accurate. In reality, the OT-HRSG is dependent on the price of natural gas by its direct connection to power generation. The waste heat has an economic value that is equal to the opportunity cost of producing more power and water in an optimized plant configuration.

4 SPE 172004

Fig. 1: Energy Flow For Different Scenarios Of Electricity Production

Energy flow for SC w. OT-HRSG Energy flow for CC w. Water desalination

Fig. 1 above illustrates the energy flow for two different scenarios of fuel use. On the left, a Simple Cycle (SC) power plant is connected to the OT-HRSG to produce steam. On the right, a Combined Cycle (CC) power plant produces more power for the same fuel consumption and it can also produce water with no additional energy use. The Power Opportunity Cost is calculated by comparing the economic value created with the more efficient CC power plant (PP) with the SC configuration using an OT-HRSG. This Opportunity Cost is a result of two factors: 1) the LCOE of power generated by a CC PP (LCOECC) is lower than that of a SC power plant (LCOESC) and 2) for the same amount of fuel, the CC PP will produce more electricity than SC, calculated by the inverse of the Heat Rate (1/HR). Thus, the formula for the Power Opportunity Cost is:

Power Opportunity Cost = { (LCOE SC – LCOE CC) * (1/HR CC – 1/HR SC) } * Fuel Consumed SC

The Water Opportunity Cost is calculated by comparing the Cost of Water (CW) produced by a combined water and power plant using a thermal desalination method, such as Multi-Effect Distillation (MED), with an independent water plant using an electrical desalination method, such as Reverse Osmosis (RO). We assumed that the MED plant will replace the condenser in the CC PP and is sized accordingly. We then compared the CW of the cogeneration MED plant with a stand-alone RO plant where its cost of electricity is LCOESC. The difference in the CW for the two configurations is then multiplied by the total potential water production in the combined water and power plant configuration to calculate the Water Opportunity Cost for the Fuel Consumed. Thus, the formula for the Water Opportunity Cost11 is:

Water Opportunity Cost = (CW RO – CW MED) * (Water Produced in Combined Water & Power Configuration)

Another indirect benefit of displacing natural gas from steam generation is the direct and indirect jobs it creates. This has not been considered here. No economic burdens have been applied to the OTSG. 2.1.2 Environmental Burden The environmental burden considered is the Carbon Cost and it is applied to the OTSG and OT-HRSG methods. The emissions are calculated using the Emission Factors as defined in (EPA AP42)12. For the OT-HRSG, since fuel is burned (and carbon emitted) in the GT, it is necessary to allocate the emissions fairly between power and steam. We assume that in the cogeneration plant (GT/OT-HRSG) the emissions allocated to electricity production are calculated using the emissions intensity of a more efficient CC PP (tons CO2 per MWh). Therefore, the emissions allocated to steam production are the difference between the total emissions of the SC PP and the calculation mentioned above. Thus, the formula for the OT-HRSG Carbon Cost is:

OT-HRSG Carbon Cost = { Emissions SC – (Power SC * Emissions Intensity CC) }* (Carbon Cost per Ton)

The OTSG Carbon Cost is simply the Emissions & Carbon Cost per Ton. We have used a carbon cost of $40 per ton/CO2, based on the range of internal carbon prices used by the oil majors (Total: $34, Shell & BP: $40, Exxon Mobil: $6013).

Waste!22%!

Steam!44%!

Electricity''34%!

Fuel!100%!

Waste!10%!

Water'!35%!

Electricity''55%!

Fuel!100%!

SPE 172004 5

2.2 Performance Models The denominator of the LCOE calculation is the total energy produced by the various methods. To accurately calculate the LCOE of the various steam generation methods, it was necessary to build performance models that take into account regional weather data. The combined cycle power plant and OT-HRSG systems analyzed in this paper both rely on heavy duty gas turbine engines from General Electric to obtain the waste heat they utilize to generate steam. Gas turbine performance is heavily dependent on the prevailing ambient conditions, particularly ambient temperature. The tables below show the impact of ambient temperature on the performance of the GE 9E.03 (PG9171E) and 9F.05 engines that were chosen for our analysis:

GE 9E.03 Performance Data 15C 40C 59F 104F Net Power 130,000.0 108,506.0 kW 130,000.0 108,506.0 kW

Heat Rate (LHV) 10,403.0 10,933.0 kJ/kWh 9,860.0 10,362.0 BTU/kWh

Heat Cons. (LHV) 1,352.4 1,186.3 GJ/hr 1,281.8 1,124.4 MMBTU/hr

Exhaust Flow 1,496.8 1,330.7 x10^3 kg/hr 3,300.0 2,934.0 x10^3 lb/hr

Exhaust Temperature 541.7 560.0 Deg. C 1,007.0 1,040.0 Deg. F

GE 9F.05 Performance Data 15C 40C 59F 104F Net Power 299,000.0 235,660.0 kW 299,000.0 235,660.0 kW

Heat Rate (LHV) 9,295.0 9,880.0 kJ/kWh 8,810.0 9,364.0 BTU/kWh

Heat Cons. (LHV) 2,779.2 2,328.2 GJ/hr 2,634.2 2,206.7 MMBTU/hr

Exhaust Flow 2,400.2 2,075.9 x10^3 kg/hr 5,292.0 4,577.0 x10^3 lb/hr

Exhaust Temperature 643.5 657.2 Deg. C 1,190.0 1,215.0 Deg. F The 9E.03 was chosen as the prime mover for the OT-HRSG due to its use in similar projects around the world and its ability to produce about 6,000 TPSD of 100 bar, 80% quality EOR steam. The 9F.05 was chosen as the prime mover for our reference 2 x 1 combined cycle power plant since it represents a power plant with a net LHV efficiency of just over 58%, which is comparable to the efficiency of recently built combined cycle plants in the GCC region. As a result of the impact that ambient conditions have on gas turbine power, heat rate, exhaust flow rate and exhaust temperature a model that can predict the response of the whole plant is needed to accurately assess plant performance on an annual basis. We used the EBSILON Professional heat and mass balance software14 from STEAG Energy Services in Germany in conjunction with the VTU Energy OEM Gas Turbine library add-in15 for EBSILON to build steady state heat and mass balance models of an OT-HRSG boiler and a 2 x 1 combined cycle power plant. These models were then run to predict fuel consumption, electricity output, steam production and other key parameters for every hour of a year using hourly TMY3 format ambient condition data for Amal, Oman. We chose Amal as the reference point because we have SSG performance data from that site and wanted to make sure all methods of steam generation are compared at the same location. The following key results were obtained from these model runs:

Annual Fuel Consumption 9,378,095 MMBtu/yr

Annual Steam Production 2,192,081 metric tons/yr

Annual Net Electricity Production 882,079 MWhe/yr

Net GT Average Annual Net LHV Efficiency 32.4 %

GT Average Net LHV Heat Rate 10,538 Btu/kW-hr

OT-HRSG Heat Balance Model Results GT @ 85% Load The results above are for 100% availability. In the LCOE model, we applied an availability of 92% bringing the total annual steam production down to 2,016,715 metric tons, or 5,525 tons per day. Also, the OT-HRSG model assumes an operation at 85% load, which is common practice in the oilfield. We also ran the OT-HRSG model at 100% load and the results are shown in Endnotes16. The OT-HRSG model is an unfired system with a low pressure economizer section and recirculation to keep the stack temperature above the sulfuric acid dew point. The design point was chosen so that the boiler could produce 75.7 kg/s of 100

6 SPE 172004

bar 80% exit quality steam with 5C ambient air and 55C produced water from the oil field. The EBSILON process flow diagram from the model is shown below:

Fig 2. EBSILON Model of EOR OT-HRSG The EOR OT-HRSG model incorporates assumptions listed in the Endnotes.17 An EBSILON heat and mass balance model of an unfired, 2 x 2 x 1 (2GTs/2HRSGs/1ST) combined cycle power plant was constructed based upon GE’s 9F.05 300 MWe class heavy duty gas turbine engines. The plant utilizes two, unfired, 3 pressure level (High Pressure (HP), Intermediate Pressure (IP) and Low Pressure (LP)) reheat Heat Recovery Steam Generators (HRSGs) to raise steam for a single, 305 MWe condensing steam turbine with once through sea water cooling. Basic design parameters for this plant were chosen from experience and some GE Reference (GER) papers18,19 on GE’s Steam and Gas (STAG) power plant product line.

Annual Fuel Consumption 42,689,075 MMBtu/yr

Annual Net Electricity Production 7,147,863 MWhe/yr

Plant Average Annual LHV Net Efficiency 57.12 %

Plant Average Annual Net LHV Heat Rate 5,974 Btu/kW-hr

2 x 1 9F.05 Based CCGT Heat Balance Model Results The plant’s process flow diagram is shown in the Endnotes20. The CCGT model incorporates the assumptions shown in the Endnotes21. The SSG performance model is built off GlassPoint’s proprietary model and operating experience at the SSGP plant in Amal, Oman22. 2.3 LCOE Models The LCOE models for the OTSG, OT-HRSG, and SSG have been built to provide a fair comparison of the methods. One factor affecting the LCOE is the size of the project. We have assumed the SSG and OTSG are built to match the size of the OT-HRSG, which, as discussed above, produces 5,525 TSPD, on an annual average. Also, the same macro-economic assumptions were used for all methods of steam generation, such as nominal discount rate (8%), inflation (3%), and project life (25 years).

SPE 172004 7

2.3.1 SSG Model To match the steam generated by the OT-HRSG, we require 32 blocks of GlassPoint solar steam generators. We assume an availability of 99% based on GlassPoint’s experience at SSGP. The CAPEX used is based on GlassPoint’s estimate of a project this size. The SSG steam LCOE was calculated to be $17 per ton of steam. 2.3.2 OTSG Model We have assumed each OTSG has a firing rate of 85MMBTU/h, which corresponds to 800 TSPD. To match the steam generated by the OT-HRSG and to account for an availability of 90%, we require 8 OTSGs. The OTSG efficiency is assumed to be 85%. The OTSG steam LCOE was calculated with and without the environmental burden of Carbon Cost and the results are displayed in Fig. 3 below:

Fig. 3: LCOE Of OTSG Steam Generation 2.3.3 OT-HRSG Model The OT-HRSG CAPEX was based on estimating software and a premium added for installation at an oilfield (due to higher costs of HSE and logistics). The OT-HRSG LCOE model is also affected by the power plant models generated for the SC and CC configurations (Power Opportunity Cost). For the power plants, heat rates were taken from the EBSILON Professional heat balance models built for this study. Overnight capital costs were from various press releases on similar projects and from EIA's April 2013 Capital Cost Estimates for Utility Scale Electricity Generating Plants Report23. Fixed and Variable Operations & Maintenance costs were also taken from the EIA report. CAPEX and OPEX for the SC/OT-HRSG plant were escalated to account for higher installation and operation costs at the oilfield. Capacity factors were from our models or EIA’s Annual Energy Outlook 201424. The PP LCOE calculated is the real LCOE calculated using the National Renewable Energy Lab (NREL) method25. This LCOE calculation neglects taxes, tax incentives, government subsidies and capital expenditures not directly related to the cost to procure and install the plant equipment. It represents the minimum price at which energy from the project must be sold in order for the project to cover its costs. It is a useful metric for fairly comparing one project to another. Also, the OT-HRSG LCOE is affected by the water model, described in the Water Opportunity Cost and whose input is taken from the Fichtner MENA Regional Water Outlook. The OT-HRSG steam LCOE was calculated with and without the economic and environmental burdens and the results are displayed in Fig. 4.

0

10

20

30

40

50

60

70

0 2 4 6 8 10 12 14 16 18 20

Cos

t of S

team

(LC

OE

- $/to

n)

Fuel Cost ($/MMBTU)

Unburdened + Carbon Cost SSG

8 SPE 172004

Fuel Price = $6/MMBTU Fuel Price = $13/MMBTU

Fig. 4: OT-HRSG Steam Generation Cost We also calculated the OT-HRSG burdened and unburdened steam costs across a range of fuel prices as shown in Fig. 5.

Fig. 5: OT-HRSG Steam Generation Cost vs. Fuel Cost

5.0

15.1

19.0 19.8 19.8

0

5

10

15

20

25

Cos

t of O

T-H

RSG

Ste

am (L

CO

E $/

ton)

5.0

24.9

28.8 30.5 30.5

0

5

10

15

20

25

30

35

Cos

t of O

T-H

RSG

Ste

am (L

CO

E $/

ton)

0

5

10

15

20

25

30

35

40

45

0 2 4 6 8 10 12 14 16 18 20

Cos

t of S

team

(LC

OE

- $/to

n)

Fuel Cost ($/MMBTU)

Unburdened w. Power Opp. Cost + Carbon Cost + Water Opp. Cost SSG

SPE 172004 9

2.4 Summary Of LCOE Results Fig. 6 shows a summary of the fully burdened LCOE for the three methods of steam generation and their relationship to the fuel price.

Fig. 6: LCOE vs. Fuel Cost What is clear in Fig. 6 is that the question of “which method of steam generation is best?” is answered only by another question “what is the fuel cost?” These questions should be considered at the countrywide level using the marginal fuel. As mentioned in Section 1.3 many countries in the Gulf region have multi-tiered and subsidized fuel costs, which do not reflect the true economic value. If a country is subsidizing gas to, say, $1.50/MMBTU and is also importing LNG at $13/MMBTU, the fuel price that should be used in decision making should be the marginal cost, which is LNG. From Fig. 6, we can see that at current market prices of diesel26, the fully burdened costs of steam from an OTSG and OT-HRSG are $58 per ton and $38 per ton respectively. At current market prices of LNG, these costs are $45 and $31 per ton for the OTSG and OT-HRSG, respectively. The cost of steam from SSG is independent of fuel price and is $17 per ton of steam. The LCOE for all scenarios are shown in the Table below:

Scenario LCOE ($/ton) SSG OTSG OT-HRSG

Fuel Price: $6/MMBTU

Unburdened

17

20 5 w. Power Opp. Cost 15 + Carbon Cost 27 19 + Water Opp. Cost 20

Fuel Price: $13/MMBTU

Unburdened 38 5 w. Power Opp. Cost 25 + Carbon Cost 45 29 + Water Opp. Cost 31

2.5 FBE Models A useful economic indicator that allows a decision maker to compare the fully burdened cost of steam from the mentioned methods is the Fuel-Break Even price, or FBE. This fuel price is where the total cost of ownership from the fuel-fired steam generation method (either OTSG or OT-HRSG) is equal to that of the SSG. The FBE allows a decision maker to compare the economics of the various methods of steam generation based on the marginal cost of fuel.

0"

10"

20"

30"

40"

50"

60"

70"

0" 2" 4" 6" 8" 10" 12" 14" 16" 18" 20"

Cost%of%S

team

%(LCO

E%/%$

/ton

)%

Fuel%Cost%($/MMBTU)%

OT-HRSG"OTSG"SSG"

Gas LNG Diesel

10 SPE 172004

The FBE for SSG when compared with a fully burdened OTSG is only $2.25/MMBTU and the FBE for SSG when compared with a fully burdened OT-HRSG is $4.5/MMBTU. The FBE for various burden scenarios is shown in the Table below:

Scenario FBE ($/MMBTU) SSG v. OTSG SSG v. OT-HRSG

Unburdened 4.95 w. Power Opp. Cost 7.70 + Carbon Cost 2.25 4.90 + Water Opp. Cost 4.50

Note: As discussed above, the unburdened OT-HRSG has no dependence on fuel price therefore it is not possible to calculate its FBE. Similarly, the OTSG does not have Power or Water Opportunity Costs so they were not calculated. 2.5.1 Factors Affecting The Economics Of Steam Generation The economic calculations above are all based on the specified assumptions. However, the economics of steam generation is dependent on various inputs ans sensitivities should be calculated in order to refine the output. Some of these factors include:

OTSG OT-HRSG SSG Marginal Fuel Price Solar Radiation

Carbon Price Project Size

3.1 Oilfield energy requirements Once the economics of the various methods of steam generation are calculated, it is necessary to determine the mix of the various methods of steam generation. To do so, the decision maker needs to understand the energy requirements and limitations at the oilfield. Mike Welch summarized the decision-making process for steam generation methods as follows:

…a gas turbine cannot exactly match the electrical load required and provide all the heat required. This gives the Operator a choice of whether to install a heat-matched system or an electricity-matched system. In a heat-matched system, the Gas Turbine is selected on the basis of its ability to provide all the heat required, which means that it is likely to generate far more electrical power than the production facilities themselves require. This necessitates the export of surplus electrical power to the local power network. In an electricity-matched system, the Gas Turbine is selected to provide just the power required by the production facilities, while the shortfall in steam is made up by installing additional conventional fired boilers.27

3.2 Macro-Economic Planning The analysis of the economics of steam generation has to be considered in the broader macro-economic policy and planning of a government or its national oil company. The first step is to understand the decision-maker’s marginal fuel cost, which will specify the preferable method of steam generation. The next step is to understand the limitations to producing steam from the OT-HRSG. By definition, steam from cogeneration is linked to electricity production, which is constrained by the oilfield power demand or the ability to export power to the grid. It is necessary for the decision maker to analyze the combination of steam generation methods using two different lenses: 1) Isolated Oilfield or 2) Connected Oilfield. 3.2.1 Isolated Oilfield In this scenario, the ability to generate steam using an OT-HRSG is limited by the total power requirement in the oilfield. This necessitates an electricity-matched system. The isolated oilfield’s energy split between thermal and electrical demand is highly dependent on the type of field (how much heavy oil vs lighter crude) and its boundary (how many reservoirs are operated within the field boundary). For a stand alone heavy oilfield upstream electricity demand may be only 2% of total energy demand during a steam flood. In an electricity-matched system, the power plant is sized to deliver the power required by the field. The Electrical:Thermal energy split delivered by a cogeneration plant is approximately 1:1.528. To illustrate the limitations of the isolate oilfield, we have assumed

SPE 172004 11

that 10% of the field’s energy requirements is electric29. Fig 7 illustrates this scenario where the cogeneration plant is electricity-matched (delivers all electric energy demand) and delivers an additional 15% of total energy demand as thermal energy. Thus, the cogeneration OT-HRSG delivers 17% of thermal energy demand. The remaining 83% of thermal demand must be satisfied with another method of steam generation, either OTSG or SSG. Fig. 7: Electricity-matched Cogen System

3.2.2 Connected Oilfield In this scenario, we assume that the entire country (or region) is connected to the same power grid as the oilfield. The oilfield can use a heat-matched system and export the surplus power to the grid. Brandt and Unnasch have studied30 the energy requirements at various California thermal EOR fields and found that the average operator generates only 40% from its steam from cogeneration using OT-HRSG. The implementation of large-scale cogeneration for thermal EOR in a grid-connected oilfield has macro-economic implimcations that should be studied further and should be considered the topic of another paper. 3.2.3 Mukhaizna Case Study Fig. 8: Mukhaizna Is Located Far From Power Demand

The Mukhaizna oilfield in Oman, operated by Oxy, produces roughly 120,000 BOPD in one of the largest steamflood projects in the world. Steam is produced using a combination of OT-HRSGs and OTSGs. There are two types of OT-HRSGs at Mukhaizna: (Note: estimates are based on public domain data and may not be an accurate representation of the current field facilities and operations) 1) Two NEM OT-HRSGs, each connected to a GE Frame 9E GT, capable of delivering about 6,000 TSPD each from cogeneration, in addition to about 10,000 TSPD from Duct Burners31. These GTs deliver power to a central grid. 2) Approximately 10 En-Fab OT-HRSGs, each connected to a Solar Titan 130 GT, capable of delivering about 1,100 TSPD each32. The Solar Titan GTs have a total capacity of roughly 100MW, which is used within the field (no export of power). The Mukhaizna field has the ability to import surplus power from the grid. In addition to the OT-HRSGs, Mukhaizna also has over seventy OTSGs.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Demand Cogen OTSG

Fiel

d En

ergy

Dem

and

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Demand Cogen OTSG

Fiel

d En

ergy

Dem

and 83% of

Thermal

100% of Electric

17% of Thermal

/SSG

Thermal

Electric

350$km$

500$km$

Mukhaizna$

Muscat$

Salalah$

12 SPE 172004

Mukhaizna’s cogeneration steam production is limited to roughly ~30% of total steam demand because of the mismatch between thermal and electrical energy demand. Mukhaizna has the advantage of being connected to the grid, allowing it to use a higher fraction of cogeneration steam. If it was an isolated oilfield, the OT-HRSG steam would be limited to ~15% of thermal energy demand. 3.3 Hybrid Steam Generation Steam produced by SSGs is inherently dependent on solar radiation and will produce variable output. Two SPE papers out of Petroleum Development Oman33 and Stanford34 studied the effect of rate injection rate variation on oil production and concluded that the recovery rate and ultimate recovery is not affected. However, night-time steam is required for two primary reasons: 1) Health & Safety: to prevent backflow of H2S. 2) Maintenance & Lifetime: limit thermal cycling of well casings. Without the use of storage, the requirement of night-time steaming implies a limit to the total solar fraction, roughly 80% in the Gulf region. The decision-maker is faced with three different steam supply sources and has to select a combination of them to suit the energy needs of the oilfield. We can create a hypothetical example based on the discussion above for an electricity-matched Isolated Oilfield, where the marginal fuel cost is $6/MMBTU. The decision-maker will always select the most economic source of steam to fill as much of their supply as possible. At $6/MMBTU, the SSG is the lowest cost at $17/ton; next is the fully-burdened OT-HRSG at $20/ton; and, finally the OTSG is most expensive at $27/ton and will supply the remaining steam demand. This is shown in Fig 9.

Fig. 9: Steam Supply Curves for $6/MMBTU fuel price 4.1 Solar For EOR Vs. Solar For Power Generation Several countries in the Gulf region have announced targets for solar power, for example Kuwait is aiming for at leat 15% of its power needs from renewable sources by 203035. Similarly, Saudi Arabia is aiming to install 25GW of CSP by 203236. CSP is more effective in producing steam than it is for power generation. The capital expenditure in solar EOR for each unit of gas saved is significantly lower than solar for power generation. This analysis should be considered in further works.

!"!!!!

!5!!

!10!!

!15!!

!20!!

!25!!

!30!!

Stea

m C

ost (

LCO

E $/

ton)

SSG 80% Limit

80% 100% 0% 50%

OT-HRSG Isolated Oilfield 17% Limit

SSG

OT-HRSG

OTSG

SPE 172004 13

4.1.2 Solar For Electricity Production

Fig. 10: Economics of power generation in the oilfield37 PV has benefits for power production in the oilfield. At current market prices for PV, its electricity generated is cost competitive with fuel prices as low as $5/MMBTU. 5.0 Conclusions Thermal EOR projects provide very good ultimate recovery and increase production. However the fuel gas consumption is a problem. The thermal EOR steam generation projects in Gulf oilfields are on such a large scale that they affect an entire country’s economic position. The field development may entail steam injection for many decades. Thus the power / gas requirements have an important impact at a national level. Proper consideration of full economic cost, and the future impact on gas consumption is essential. Furthermore, the costs of the three methods of steam generation considered should be fully burdened to accurately account for the true macro-economic implications. At a fuel price of $6/MMBTU, SSG is the cheapest method of steam generation at $17/ton. Next is the fully burdened OT-HRSG at $20/ton. The most expensvie source is the fully burdened OTSG, which is $27/ton. Finally, the limitations of OT-HRSG steam as a result of the oilfield’s electric:thermal demand ratio will determine the ideal steam supply curve based on the marginal fuel cost.

0"

50"

100"

150"

200"

250"

300"

350"

400"

0" 5" 10" 15" 20" 25"

LCOE%($/M

Wh)%

Fuel%Price%($/MMBTU)%

Diesel%LNG%Gas%

PV%

14 SPE 172004

References/Endnotes 1 Kovscek, A., Stanford Thermal Oil Recovery course, August 2013 2 Kokal, S. and Al-Kaabi, A., Enhanced oil recovery: challenges & opportunities, World Petroleum Council, 2010 3 Manaar Consulting, EOR and IOR in the Middle East, http://www.manaarco.com/images/presentations/Fleming%20Gulf%20Manaar%20EOR%20Abu%20Dhabi%20March%202013.pdf (accessed 30 October 2013) 4 A Simple Cycle power plant is ~30-35% efficient. When an OT-HRSG is added the system efficiency increases to ~75-80%. 5 Production from a thermal EOR reservoir doesn’t stop when there is no more oil. Instead, production is “shut-in” when the oil is no longer economic to produce. Since fuel for steam purchase is the largest component of operating costs in a heavy oil field, eliminating fuel costs extends the economic life of the field. Simulations show that the ultimate recovery fraction can be increased by 12% of Original Oil in Place (OOIP) if 20% of the steam for the field is produced from solar. 6 Dargin, J. 2013. Development and Industrialization in the Arabian Gulf Region. Harvard Journal of Middle East Politics and Policy 7 Dargin, J. and Vladimirov, M. 2012. Energy intensity: a time bomb for the Middle East? Energy in the Middle East. 8 According to the IEA, Kuwait imported as LNG 25% of its gas consumption in 2011. 9 The “opportunity cost” for countries such as Oman that are LNG exporters is the decrease exports that will result in a decrease in gas availability in the country. The assumed LNG export price is the Asia spot market. 10 Used Asia spot price from BG Group http://www.bg-group.com/480/about-us/lng/global-lng-market-overview-2013-14/ (accessed August 20 2014). 11 The assumptions for the CW calculations were taken from Fichtner, MENA Regional Water Outlook, Part II Desalination Using Renewable Energy, March 2011. Calculation Results: at $6/MMBTU fuel price, CWRO = $1.07/m3 and CWMED = $0.77/m3. 12 http://www.epa.gov/ttnchie1/ap42/ 13 CDP North America, Use of internal carbon price by companies as incentive and strategic planning tool, December 2013. 14 http://www.steag-systemtechnologies.com/ebsilon_professional+M52087573ab0.html 15 http://www.vtu-energy.com/1_PDFs_2009/ENVTUEnergyGasTurbineLibrary.pdf 16

Annual Fuel Consumption 10,598,238 MMBtu/yr

Annual Steam Production 2,332,579 metric tons/yr

Annual Net Electricity Production 1,040,334 MWhe/yr

Net GT Average Annual Net LHV Efficiency 33.8 %

GT Average Net LHV Heat Rate 10,104 Btu/kW-hr EOR OT-HRSG Heat Balance Model Results GT @ 100% Load

17 Feedwater for EOR OT-HRSG is always 55C. EOR OT-HRSG exit quality is always 80%. Water leaving the EOR OT-HRSG’s economizer section is recirculated to its inlet to maintain an economizer water inlet temperature of 135C in order to prevent sulfuric acid condensation on the economizer tubes. No preheating of incoming natural gas. Fuel to the gas turbine is sour natural gas with 4 PPM H2S and the following composition:

Constituent Mol % Methane CH4 79.0 Ethane C2H6 12.0 Propane C3H8 5.0 n-Butane C4H10 1.5 n-Pentane C5H12 0.35 n-Hexane C6H14 0.15 Nitrogen N2 2.0 Carbon Dioxide CO2 0.01

SPE 172004 15

18 http://site.ge-energy.com/prod_serv/products/tech_docs/en/downloads/ger3767c.pdf 19 http://site.ge-energy.com/prod_serv/products/tech_docs/en/downloads/ger3574g.pdf

20

G

WATER COOLED

CONDENSER

GTG

Stack

NATURAL GAS

AMBIENT AIR

DC

C

E

F

E

IPST LPSTHPST

HOT REHEAT FROM HRSG2

HP FROM HRSG2

F

FUEL PREHEATER

D

COLD REHEAT TO HRSG2

LP FROM HRSG2

COND TO HRSG2

text

SEA WATER

21 Main steam conditions of 1815 psia / 1050F or 125 Bara / 565C. Hot Reheat steam conditions of 390 psia / 1050F or 27 bara / 565C. Design point condenser pressure of 2 in. HgA or 68 mbar. Natural gas to gas turbines is preheated to 365F / 185C. Cooling water temperature for the CCGT plant's condenser was assumed to be ambient temperature + 10C unless that would put it over 35C or under 20C in which case it was pegged to either 20C or 35C. Design point cooling water temperature rise of 20F / 11C. Natural gas composition for the CCGT model was the same as that for the EOR OT-HRSG model. the plant performance at ISO conditions (15C, 60% relative humidity and 1.013 bara) is as follows:

Plant Net Output 890 MWe Plant Net Heat Rate (LHV) 5865 BTU/kW-hr Plant Net Efficiency (LHV) 58.2 %

22 Palmer, D. and O’Donnell, J. 2014. Construction, Operations and Performance of the First Enclosed Trough Solar Steam Generation Pilot for EOR Applications. SPE 169745-MS. 23 http://www.eia.gov/forecasts/capitalcost/pdf/updated_capcost.pdf 24 http://www.eia.gov/forecasts/aeo/electricity_generation.cfm 25 From http://www.nrel.gov/analysis/tech_lcoe_documentation.html: Levelized Cost of Energy (LCOE, also called Levelized Energy Cost or LEC) is a cost of generating energy (usually electricity) for a particular system. It is an economic assessment of the cost of the energy-generating system including all the costs over its lifetime: initial investment, operations and maintenance, cost of fuel, cost of capital. A net present value calculation is performed and solved in such a way that for the value of the LCOE chosen, the project's net present value becomes zero. 26 Diesel market price taken as $2.80/gallon based on NYMEX No. 2 Heating Oil as of September 9, 2014. 27 Welch, M. 2011. Greener EOR: Employing Cogeneration to Improve the Energy Efficiency of Thermal EOR projects. SPE 144224

16 SPE 172004

28 Assumption based on the Frame 9E/OT-HRSG modeled in this paper. 29 Sigworth, H.W., Horman, B.W., and Knowles, C.W. 1983. Cogeneration Experience in Steam EOR Applications. SPE 12196 30 Brandt, A. and Unnasch, S., Energy Intensity and Greenhouse Gas Emissions from Thermal Enhanced Oil Recovery, Energy Fuels, 2010 31 http://www.nem-group.com/EN/projects/hrsgs/2/mukhaizna_/9/ 32 http://www.en-fabinc.com/en/project_oman1.shtml 33 Van Heel, A.P.G., van Wunnik, J.N.M., Bentouati, S., and Terres, R. 2010. The Impact Of Daily And Seasonal Cycles In Solar-Generated Steam On Oil Recovery. SPE 129225-MS. 34 Sandler, J., Fowler, G., Cheng, K., and Kovscek, A. 2012. Solar-Generated Steam for Oil Recovery: Reservoir Simulation, Economic Analysis, and Life Cycle Assessment. SPE 153806 35 http://www.oxfordbusinessgroup.com/economic_updates/kuwait-new-renewable-energy-project-works 36 KA CARE, Saudi Arabia’s Renewable Energy Strategy and Solar Energy Deployment Roadmap 37 Based on 2014 US PV projects including adjusted to Gulf conditions weather and cost.