CO2-EOR and storage design optimization

International Journal of Greenhouse Gas Control 25 (2014) 79–92

Contents lists available at ScienceDirect

International Journal of Greenhouse Gas Control journal homepage: www.elsevier.com/locate/ijggc

CO2 -EOR and storage design optimization Amin Ettehadtavakkol a,∗ , Larry W. Lake b , Steven L. Bryant b a b

Department of Petroleum Engineering, Texas Tech University, United States Department of Petroleum and Geosystems Engineering, The University of Texas at Austin, United States

a r t i c l e

i n f o

Article history: Received 25 April 2013 Received in revised form 25 February 2014 Accepted 4 April 2014 Keywords: Coupled CO2 -EOR and storage Storage tax credit Integrated asset optimization

a b s t r a c t A partnership between oilfield operators and the federal government in the coupled CO2 enhanced oil recovery (EOR) and storage projects may bring long-term benefits for both. This paper describes the field-scale design optimization of coupled CO2 -EOR and storage operations from the viewpoint of oilfield operators. We introduce two categories of EOR-storage design optimization problems: the fixed storage requirement and integrated asset optimization. The first problem follows an environmental-driven objective by giving priority to the storage element of CO2 -EOR and storage; the second prioritizes the oil recovery to the storage and adjusts the storage volume commitment based on the economic benefits. Each problem is elaborated through a set of common but specific technical and economic assumptions, which are mainly based on the USA circumstances. The results quantitatively describe the relations between the design parameters, reservoir properties, asset size, and the economic parameters. With appropriate economic conditions and fiscal incentives, specifically in the form of storage tax credit, coupled CO2 -EOR and storage could be attractive for both the oilfield operators and the government. In addition, CO2 -EOR and storage is not the ultimate solution to the industrial-scale CO2 storage requirements. A medium-size industrial capture and storage for 25 years through EOR requires an asset size greater than the current field-scale EOR operations in the USA. Finally, the reservoir properties significantly affect the economic benefits for both the oilfield operators and the government. The proposed workflow may serve as a screening tool to rank the EOR-storage candidates and prioritize the most promising prospects for both the oilfield operators and the government. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Scientific evidence confirms a relation between the dramatic increase of greenhouse gases in the atmosphere and the global warming side-effects (Nakicenovic et al., 2000; Oreskes, 2004; Thomas et al., 2004). This leaves little doubt about the need for large-scale reductions in greenhouse gas emissions. Carbon capture, utilization and storage (CCUS), especially CO2 enhanced oil recovery (EOR) and storage, is a promising short- to medium-term solution to mitigate the fast growth of greenhouse gas emissions (IPCC, 2005). CO2 -EOR refers to the injection of super-critical dense CO2 into a reservoir in some stage of maturity, typically after a waterflood phase. CO2 may be continuously injected, altered with water in a water-alternating-gas (WAG) process, or simultaneously injected with water. The ideal CO2 quality, crude oil properties, and reservoir

∗ Corresponding author. Tel.: +1 806 834 6617. E-mail addresses: [email protected], [email protected] (A. Ettehadtavakkol). http://dx.doi.org/10.1016/j.ijggc.2014.04.006 1750-5836/© 2014 Elsevier Ltd. All rights reserved.

properties for CO2 -EOR and the successful CO2 -EOR practices are widely discussed (Lake, 1989; Pariani et al., 1992; Sebastian et al., 1985; Bergman et al., 1997; Kovscek, 2002; Jarrell, 2002; Riddiford et al., 2003; Van Bergen et al., 2004; Bryant and Lake, 2005). The synergy between CO2 -EOR and CO2 storage is because of the physical and geochemical trapping mechanisms that store a fraction of the injected CO2 in the reservoir. CO2 -EOR and storage has recently received considerable attention (Stevens et al., 2001; Bachu and Shaw, 2003; Moritis, 2003; Beecy and Kuuskraa, 2005; MIT, 2010; Kuuskraa et al., 2011) because of the following advantages: (1) increasing the domestic oil production and energy security enhancement (ARI, 2010; Kuuskraa et al., 2011; Steelman and Tonachel, 2010), (2) promoting the oilfield operators participation in the CO2 storage projects because of tax credits (Mandelker, 1992; IRS, 2009; Thompson et al., 2010), (3) compensating for a portion of the capture and storage costs by the EOR benefits, and (4) promoting the CCUS infrastructure deployment, especially the pipeline network (ICF International, 2009; ARI, 2010; Johnson and Ogden, 2010). An industrial-scale EOR-storage implementation faces several challenges. A considerable impact on carbon emission reduction

80

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

Fig. 1. CO2 -EOR and storage design optimization workflow. The first three stages are used to optimize the field-scale CO2 -EOR and storage process.

requires sustainable large-scale CO2 storage for decades. This requires that the pipeline network and infrastructure, subsurface uncertainties, regulatory framework, long-term liability, and public acceptance issues are all resolved (IPCC, 2005, 2007; Dooley et al., 2009; ICF International, 2009; Johnson and Ogden, 2010; Marston and Moore, 2008; Smith, 2009; Davidson et al., 2011). Moreover, CO2 -EOR and CO2 storage objectives are not aligned; this conflict often ends in favor of the EOR objective because the tangible economic benefits of EOR outweigh that of the storage (Kovscek and Cakici, 2005; Leach et al., 2011). Because of these challenges, CO2 -EOR and storage is not the ultimate solution to the long-term storage requirements. However, it has important benefits that should be best used toward enhancing the energy security and mitigating the greenhouse gas emissions. This study focuses on the economic optimization of EOR-storage projects. We show that the economic fate of CO2 -EOR and storage depends on three field-scale operating parameters, namely oil production performance, CO2 utilization, and CO2 recycle ratio, and three economic parameters, namely oil price, CO2 cost, and storage tax credit. The dynamic relation between these parameters have been investigated in several studies, including Bock et al. (2003), Gozalpour et al. (2005), and Davidson et al. (2011). These studies do not consider optimizing the field development sequence and schedule for a large-scale EOR-storage operation. Neither do they consider the important fact that the government may also gain benefits from the EOR-storage projects. We propose a workflow to find the optimum values for the field-scale storage capacity, field-scale oil production capacity, field development schedule and sequence. Ultimately, we calculate the optimum CO2 storage profits for operators and the government. This study considers the economic optimization from the oilfield operators viewpoint. The effect of storage tax credit for a potential partnership between the government and the operators is presented in other studies (Ettehadtavakkol et al., 2014).

2. Field-scale EOR-storage design optimization workflow An integrated asset model specifies the relation between the surface facility and wells, operational design, reservoir response, and the economic prospect of an oilfield development project. Fig. 1 presents the integrated asset development and optimization workflow. The boxes represent the major steps. We use this workflow to optimize the field-scale EOR-storage operation design:

1. Develop a representative reservoir model, perform simulations and match the main parameters of interest with the available field data. 2. Aggregate and upscale the simulation results to the field-scale and find the optimum field development sequence and schedule for the EOR-storage design candidates. 3. Develop the surface facility and economic models and integrate them with the field-scale model. The result is an integrated asset model that is used to find the optimum EOR-storage design. 4. If there are any uncertainties, include the available information on the uncertainties in the integrated asset model and use

a stochastic optimization method to find the optimum design under uncertainty. We describe the three steps for integrated asset development and optimization under deterministic conditions in the following. The reader may refer to Ettehad (2013) and Ettehad et al. (2010, 2011) for applications of uncertainty analysis methods in integrated asset optimization.

2.1. Reservoir simulation model We introduce two reservoir simulation models to predict the EOR-storage operation performance under various operating conditions. The first model has properties similar to a sandstone reservoir and therefore is referred to as “Sandstone” (S); the second has properties similar to a carbonate reservoir and therefore is referred to as “Carbonate” (C). Table 1 summarizes reservoir characteristics of Sandstone and Carbonate. The main differences between the two cases are the reservoir permeability, heterogeneity, and the well spacing. Appendix A provides a description of Sandstone and Carbonate models. The objective of complex reservoir model is to predict the EOR-storage operation performance with different reservoir characteristics and under different operating conditions. Three parameters are considered for an EOR-storage process design: gas injection rate, flood duration, and water-alternating-gas (WAG) ratio. The first two design parameters are self-explained by their names. WAG refers to the alternating injection of water and gas (e.g. CO2 ) to reduce the expensive CO2 usage and also to mitigate the unfavorable effects of viscous fingering and channeling. For a simultaneous water and gas injection, the WAG ratio is defined as the ratio of water injection rate to gas injection rate at reservoir conditions. Three measures of EOR-storage performance are considered: average oil production performance, net CO2 utilization ratio, and CO2 recycle ratio. Average oil production performance is defined as total oil production rate divided by the number of active producers. Net CO2 utilization is the amount of CO2 retained (stored) per incremental barrel of produced oil. We use CO2 utilization to refer to this measure and it should not be confused with the gross utilization ratio. CO2 recycle ratio is the ratio of the CO2 production rate to the fresh CO2 injection rate. These parameters determine the pace of field development, the number of wells, CO2 separation plant size, and the compression power requirements. For the EOR-storage simulations, we assume that: (1) the reservoir has been under a waterflood operation long enough to reach an economic limit before the EOR-storage phase begins, (2) water and gas are simultaneously injected at a desired WAG ratio; this slightly overestimates the oil production performance (Pariani et al., 1992), (3) the well-pattern design and the well-spacing do not change before or during the operations; this means no infill drilling or producer-injector switches occur, (4) all the producers and injectors continue to operate in the EOR-storage phase without conversions or shut-ins, i.e. temporarily shut-ins and maintenance operations on wells are not included (5) the producers and injectors are assumed to operate all the time and there are no redundant

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

81

Table 1 Sandstone and Carbonate properties at reservoir-pattern scale. Property

Unit

Sandstone (S)

Carbonate (C)

Reservoir pore volume Average horizontal permeability Average vertical permeability Areal Dykstra–Parsons coefficient Vertical Dykstra–Parsons coefficient Average porosity Active producers/injectors Well pattern Producer well spacing Watercut before EOR-storage phase Residual oil saturation before EOR-storage phase

MMbbl Darcy Darcy – – – – – acre – –

45 6.3 × 10−2 6.3 × 10−3 0.6 0.6 0.22 6p/6i Line-drive 33 0.87 0.32

Same 18 × 10−2 1.8 × 10−3 0.65 0.9 Same 8p/6i Same 25 0.95 Same

wells (6) permanent producer shut-ins because of excessive water production or gas–oil ratio are included.

pushes the oleic phase (mixture of oil–CO2 ) toward the producers (Walsh and Lake, 1989). As a result the CO2 utilization, and subsequently the storage capacity, would decrease with an increased permeability. If the CO2 is continuously injected, the ultimate storage capacity of a high-permeability reservoir would be more than a low permeability reservoir, assuming all other properties being the same.

2.2. Reservoir simulation results We perform simulations for 90 EOR-storage design candidates for each reservoir. Table 2 summarizes the general simulation inputs and outputs, and the specifications of three selected EOR-storage designs performed on Sandstone and Carbonate. Figs. 2 and 3 respectively show the oil production performance and CO2 utilization as functions of time and WAG ratio. Major observations follow: 1. Oil production rate and CO2 utilization are time-dependent. The oil production rate generally rises to a peak and then follows an exponential-like decline. CO2 utilization sharply decreases and then remains fairly stabilized. Therefore the CO2 storage rate in the reservoir pattern exponentially declines after the incidence of the peak oil production and the CO2 breakthrough. 2. The WAG ratio has a key role in determining the shape of the oil production performance and the CO2 utilization curves. By increasing the WAG ratio, the peak oil production rate decreases, the time to reach the peak is delayed, and CO2 utilization decreases. By adjusting the WAG ratio, the operator adjusts the paces of oil production and CO2 storage. 3. The reservoir properties affect the EOR-storage performance. The average oil production rate for Sandstone is significantly greater than Carbonate, and the CO2 utilization is less. This is mainly because the Sandstone is less heterogeneous and more permeable. With higher permeability a greater portion of the reservoir volume is available to storage. When the CO2 and water are injected both fluids find more pathways from the injectors toward the producers. The CO2 drags more oil out and the water

Similar analyses are performed on the effects of gas injection rate and pattern flood duration (Ettehadtavakkol, 2013). The overall analyses show that Sandstone has an optimistic prospect of oil production performance whereas Carbonate has a moderate prospect. The average oil production performances of Sandstone and Carbonate are about 100 and 50 stbd/well, respectively. Fig. 4 shows the histogram of average oil production performance of 105 U.S. miscible CO2 -EOR projects (Koottungal, 2012). The majority of current CO2 -EOR projects produce 50 stbd/well or less on average; many show low rates because of the shut-in wells. The upper 40% of the current EOR projects perform similar to or better than the Carbonate. Therefore, the economics of EOR-storage operation in the Carbonate is important to the overall assessment of commercial storage prospects with moderate oil production performance. 2.3. Field-scale and facility models The reservoir simulation models, Sandstone and Carbonate, consist of several producers and injectors. The size of these simulation models is greater than a well-pattern and therefore we refer to them as reservoir patterns. A field-scale CO2 -EOR and storage model consists of many reservoir patterns that are CO2 -flooded in several phases, depending on the CO2 availability and the available operational capacity. Fresh CO2 is available from the capture plant on a continuous basis. The available CO2 for injection at any time

Table 2 Simulation inputs and outputs for Sandstone and Carbonate. Out of 90 EOR-storage designs performed for each reservoir three designs are presented. Simulation input/output Sandstone (S) EOR-storage design name EOR-storage duration Average gas injection rate Average WAG ratio Average oil production Average CO2 utilization Average recycle ratio Carbonate (C) EOR-storage design name EOR and storage duration Average gas injection rate Average WAG ratio Average oil production Average CO2 utilization Average recycle ratio

Unit

Overall range 90 designs

Design specification

Year MMscfd CO2 /well – stb/day/well Mscf/stb –

15–25 0.12–4 0–5 30–110 2.8–12.2 0.3–3

SD1 25 4 0 90 12.2 2.7

SD2 25 1.15 1.5 93 4.7 1.6

SD3 25 0.5 3 77 3.4 0.9

Year MMscfd CO2 /well – stb/day/well Mscf/stb –

15–25 0.8–5 0–6 47–61 5.2–20 1.6–3

CD1 25 5.0 0 50 20.4 2.8

CD2 25 3.0 1.2 60 9.5 2.8

CD3 25 1.0 4.2 50 5.5 1.75

82

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

Fig. 2. Simulated oil production performance for three EOR-storage designs (a) Sandstone: SD1 , SD2 , SD3 , and (b) Carbonate: CD1 , CD2 , CD3 .

Fig. 3. Simulated CO2 utilization for three EOR-storage designs: (a) Sandstone: SD1 , SD2 , SD3 , and (b) Carbonate: CD1 , CD2 , CD3 .

consists of the fresh CO2 and the recycled CO2 . This constraints the number of patterns that can be CO2 -flooded and the pace of EOR-storage pattern development. As the EOR-storage operation matures and more CO2 is collected for the producers and recycled toward the injectors, new patterns are developed and a greater portion of the field is covered. This process continues until either the whole field is covered or the EOR-storage project life is over.

cost, however, is updated for 2012 with a 6% annual escalation factor. The total CO2 compression power requirement and CO2 and water pump power requirement and costs are calculated based on McCollum and Ogden (2006). The produced CO2 recycle plant data is based on the feasibility by for a natural gas plant with a 2 MMtonne/year CO2 capture capacity (Ettehadtavakkol, 2013).

2.4. Field-scale model results

2.5. Economic model and results

Fig. 5 schematically presents the field-scale model development for CO2 -EOR and storage. The sequence of patterns development is an output of the field-scale model. In Fig. 5, the white patterns are CO2 -flooded first; after the CO2 breakthrough and the availability of additional CO2 , the light-gray patterns, and then the dark-gray patterns. We carry the calculations using a dynamic programming model. Suggested formulas by Mohitpour et al. (2007) and McCollum and Ogden (2006) and the data adapted from Bock et al. (2003) are used for the compressor and pump calculations. The Bock et al. (2003) data on an industrial CO2 compressor are adopted calculating the number of compressors. The compressor

The EOR-storage costs are estimated based on a 3 MMtonne/year CO2 storage capacity for 25 years and the assumptions presented in Table 3. The EOR-storage operation costs includes the asset acquisition, new surface facilities, well completions, infill drilling, operation, royalties, taxes and permits, and liabilities and monitoring costs. No cost model can perfectly consider all aspects of a large-scale CCUS system. However, transparency of the assumptions is necessary and appropriate cost adjustments should be made when an assumption changes (Rubin, 2012). A list of economic assumptions is presented in Table 3.

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

83

Table 3 Economic assumptions and estimations for EOR-storage projects. Parameter Taxes Government revenue tax and royalty Capital depreciation Investment tax credit and tax holiday Financials Contracting strategy Financial source Time periods Capital expenditure period Operational and economic analysis periods Capital costs Capital cost escalations Asset lease costs CO2 plume monitoring cost

Operating costs and revenues Oil price, CO2 cost and storage tax credit

Infill drilling costs Cashflow discount rate

Escalation of production revenue, expansion costs, O&M costs, and tax credit

Assumption, description, or value 28% of the gross revenue, 23% to the federal government and includes revenue tax and royalty and 5% for the state governments 25 years using the straight line method, $0 salvage value for wells/facilities, project may operate longer than 25 years 0% and 0 years Engineering procurement construction management (owner assumes project risks for performance, schedule, and cost) 100% financed from operator’s equity capital One year for the initial facility installation, and contingent upon the project expansion thereafter 25 years both 6% annual on the initial capital cost adjustments based on the 2009 collected capital cost data. 2012 is the reference year for capital calculations. 0% capital escalation annual thereafter for the expansions Operator owns the asset; therefore the asset acquisition cost is zero. The federal government owns the land, and the royalty is 12% of the gross revenue (included in the gross revenue tax rate). Approximated as 5% of total EOR-storage operational cost per reservoir pattern in the injection period. Capital costs of monitoring, especially for drilling the monitoring wells, long-term monitoring, liability, pore-space ownership, and post-injection plume monitoring are not included $60/stb oil price, $80/ton CO2 cost, and $40/ton storage tax credit are assumed by default. Conservative oil price selected because of zero infill drilling and zero acquisition cost assumptions and uncertainties in monitoring and liability costs None, the existing wells from the waterflood phase proceed to the EOR-storage phase 10% for the oilfield operator, 3.5% for the federal government. The relationship between real and nominal discount rates is (1 + R) = (1 + r)(1 + i), where r is the real discount rate, i is the inflation rate, and R is the nominal discount rate. The inflation rate is assumed to be 0%, therefore the real and nominal discount rates are equal. Relatively high discount rate for the operator is because of high exposure to operational risks 0%

The main body of the estimates for the cost elements are adapted from the Energy Information Administration (2012). All the initial costs and benefits are escalated from the reported 2009 dollars to 2012 dollars (base year) at 6%, and kept constant after 2012. The costs will increase, and probably will the benefits. This assumption equally treats the cost and income cashflows and avoids the complicated uncertainty analyses for future costs and prices at this stage. Sensitivity analyses may be performed to better understand the cost model, as we describe in the following. One important application of the cost model is the sensitivity analyses of the operation performance and the input economic parameters. Breakeven oil price (BEPoil ) is the main measure of sensitivity, defined as the oil price above which an EOR-storage operation is economically feasible. NPVEOR-storage (poil = BEPoil ) = 0, NPVEOR-storage (poil > BEPoil ) > 0. Fig. 6 presents two sensitivity analysis examples: Fig. 6(a) with CO2 utilization and oil production performance being the sensitivity parameters and Fig. 6(b) with CO2 utilization and CO2 cost. All other parameters are set to their default values as listed in the Fig. 6 caption. Increasing the CO2 cost and decreasing the average oil production performance both increase the breakeven oil price. The $80/ton CO2 cost is based on the NETL report on the cost and performance of a net 550 MW pulverized coal plant with 65% CO2 capture (Black et al., 2011). The pipeline rent cost is included as a part of the CO2 supplier responsibility, the CO2 compression and transport cost for a 62 miles (100 km) pipeline distance is assumed to be a part of the oilfield operational costs. The $40/ton storage tax credit covers the difference between the anthropogenic CO2 cost and the current natural CO2 price; and therefore, making the anthropogenic CO2 competitive to the natural CO2 . The sensitivity analysis of the cost model provides an insight to the breakeven oil

price. As a first-order estimate an average West Texas mature well may produce 40 stbd with an approximate 6 Mscf/stb utilization. With $80/ton CO2 cost, $40/ton storage tax credit, 6 Mscf/stb CO2 utilization, and 40 stbd/well oil production performance, oil price should be at least $63/stb to make EOR-storage operation economically feasible. We select the three estimates, i.e. $60/stb oil price, $80/ton CO2 cost and $40/ton storage tax credit as the underlying economic assumptions; however, similar to any other assumption, these values are subject to appropriate adjustments when necessary. Similar sensitivity analyses may be performed on the other cost elements to gain insights into the effect of other parameters (van’t Veld et al., 2012). CO2 -EOR and CO2 storage have different objectives that are not in the same direction. EOR aims to maximize the oil production performance with a limited CO2 slug while storage aims to maximize the storage performance in a limited pore volume. Fig. 6(b) shows that increasing CO2 utilization and CO2 cost both increase the breakeven oil price. The breakeven oil price continues to increase with CO2 utilization because more CO2 is stored in the reservoir to produce an incremental barrel of oil and the net benefit of CO2 storage (40–80=)−$40/ton is negative. Increasing the CO2 utilization enhances the storage capacity and is beneficial to the storage objective; however, it is detrimental to the EOR objective because it imposes an additional fresh CO2 purchase cost. Therefore, CO2 -EOR and CO2 storage may be co-optimized to maximize the economic benefits. 3. Results and discussion The goal of EOR-storage design optimization is to balance the CO2 -EOR and CO2 -storage objectives such that the overall economic benefit is maximized. This optimization problem is subject to the constraints on the field-scale response to the EOR-storage design and operation plan and the constraints on the

84

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

a 40 35

Frequency

30 25 20 Sandstone 15 Carbonate 10 5 0

10

20

40

60

80

100

Average Oil Rate per Producer (stbd/well)

b

25

Fig. 5. Development of a field-scale model. The building block of the model is a well pattern; a reservoir pattern consists of several well patterns. The optimum sequence of pattern development is a major output of the field-scale optimization. Colors qualitatively show the sequence of field development: white region is first developed, light-gray is developed after the CO2 breakthrough, and dark-gray is delayed until sufficient CO2 is available for reinjection.

Frequency

20

15

10

5

0

10 30 50 80

120

180

250

350

Number of Injectors Fig. 4. Statistical review of active U.S. CO2 -EOR projects as published in the Oil and Gas Journal, April 2, 2012. Distributions of (a) average oil production rate per well for different fields, and (b) number of injectors per field. Numerous shut-in wells and long EOR operation decrease the oil production performance in the mature fields. Average oil production rates greater than 100 stbd/well are lumped in the 100 stbd/well bin.

surface facilities. The main outputs are the total capital cost, total operating cost, total storage cost, government revenue tax and tax credit, and levelized profit of storage. 3.1. Field-scale CO2 -EOR and storage design optimization problem We define two economic measures for the EOR-storage economic evaluation: annual profit of storage (APOS) and levelized profit of storage (LPOS). APOS is defined as, APOS(i) =

total revenue(i) − total cost(i) total CO2 injectio(i) − total CO2 production(i)

and LPOS is defined as, LPOS =

T  APOS(i) i=0

(1 + r)i

where i is the year index, r is the nominal discount rate and T is the project life. Typical CO2 suppliers are operators of natural CO2 resources, power plants, refineries and gas plants. We select a medium-size pulverized coal plant as the supplier. Typical CO2 customers are the oil and gas field operators and operators of abandoned fields and aquifers. We select operators of mature oilfields as the customer. The federal government initializes the regulations, determines tax rules and monitors the suppliers, customers and the market. CO2 is a commodity and has a positive value in this study. We present and optimize two problem categories: fixed storage requirement and integrated asset optimization. These problems are optimized from the operator viewpoint, i.e. we find an optimum design that maximizes the operator benefits. The optimum tax credits from the government viewpoint and the operator– government cooperation benefits deserve a separate study that we leave for future.

3.2. Problem I: fixed storage requirement The fixed storage requirement problem follows an environmental-driven approach by giving the priority to the storage element of EOR-storage. The fixed storage requirement problem has one CO2 supplier (coupled power plant-capture plant) and one CO2 customer (oilfield operator). CO2 is captured at a fixed rate and sold to the oilfield operator. The oilfield operator is committed to purchase and store the captured CO2 . The government awards a storage tax credit and collects the revenue tax and royalty for the produced oil. We find the optimum design and operation plan such that the project NPV is maximized for the oilfield operator. The storage requirement is 3 MMtonne/year CO2 captured from a 550 MW net pulverized coal plant and the CO2 cost delivered to the oilfield is $80/ton (Rubin et al., 2007; Black et al., 2011). The storage tax credit is $40/ton and the oil price is $60/stb. We intentionally select a conservative oil price because the infill drilling, asset acquisition costs, operational cost escalation and inflation are not considered in the cost model.

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

the cost model and subsequently the objective function if reliable site-specific cost estimates are available.

a 160

Breakeven Oil Price ($/stb)

140

3.3. Fixed storage requirement optimization workflow and results We implement the following workflow to find the optimum EOR-storage design for the fixed storage requirement problem, see Fig. 7(a).

120

Average oil production rate (stbd/well) 100

25 80

40 60

100 40

20

2

4

6

8

10

12

CO2 Utilization (Mscf/stb)

b 160 140

Breakeven Oil Price ($/stb)

CO2 cost ($/tonne) 120

120

100

80 80

40 60

40

20

85

2

4

6

8

10

12

CO2 Utilization (Mscf/stb) Fig. 6. Sensitivity of the CO2 -EOR and storage cost model. Breakeven oil price with respect to (a) CO2 utilization and oil production performance, and (b) CO2 utilization and CO2 cost. The default inputs (bold lines) are: average oil production rate = 25 stbd/well, average field recycle ratio = 0.5, CO2 cost = $80/ton, storage tax credit = $40/ton. The breakeven price is calculated based on the after-tax revenue and a 10% discount rate.

The target fields are Sandstone with 150 patterns (900 producers and 900 injectors) and Carbonate with 150 patterns (1200 producers and 900 injectors). Selecting large asset sizes ensures sufficient storage capacity for a variety of EOR-storage designs. These assets may include several fields (see Fig. 4(b)). The economic optimization objective for the fixed storage requirement problem is the average NPV per reservoir pattern. The associated capital cost of a single reservoir pattern development is assumed to be constant through time. Therefore, the selected objective (NPV/pattern) is equivalent to the profitability index, defined as PI = NPV/PV(Capex). The cost model does not account for the capital costs of asset acquisition, infill drilling and monitoring/surveillance. If any of these capital cost elements are applicable and an estimate is available, the associated average pattern cost may be subtracted from the calculated NPV/pattern to get a new average NPV value. Similar adjustments may be made to

1. Select an EOR-storage design (fix the WAG ratio, injection rate, pattern flood duration). Retrieve the reservoir simulation responses on the fluid injection and production rates and calculate the oil production performance history, CO2 utilization, and the CO2 recycle ratio. 2. For the selected EOR-storage design, perform the field-scale calculations for the total fluid injection and production rates. 3. Calculate the main parameters of the surface facility model, namely the pump and compression requirements, CO2 -gas separation plant size, and power requirement. 4. Collect the field-scale results and the surface facility outputs into the integrated asset model and perform the economic calculations. The main outputs are two cashflow diagrams for the oilfield operators and the federal government. 5. Repeat Steps 1–4 for different EOR-storage designs. 6. Collect all the integrated asset results for all the EOR-storage designs. Select the design that yields the maximum NPV per reservoir pattern. We first present the results for Sandstone and then for Carbonate. Fig. 8(a) presents the optimum field development schedule for the three EOR-storage designs listed in Table 2. For example, the initial number of active patterns for EOR-storage design SD2 is 23. CO2 breakthrough occurs in the third year, and it allows for the expansion of the active patterns because the produced CO2 can be reinjected. After the breakthrough, the number of developed patterns continuously increases to provide sufficient injection capacity for the fresh CO2 feed and the produced CO2 . In the final year, year 25, the number of active patterns is 69, equivalent to 414 producers and 414 injectors. Fig. 8(b) presents the optimum development schedule for Carbonate. Fig. 9(a) and (b) shows the EOR-storage performance lines for Sandstone and Carbonate, respectively. The field-scale results for the two reservoir cases have several important implications: 1. The WAG ratio significantly affects the pace of patterns development. According to Fig. 8(a) and (b), the number of patterns required to store 3 MMtonne CO2 over 25 years in Sandstone at a 3 WAG ratio (SD3 ) is five times greater than a 0 WAG ratio (SD1 ); the number of required patterns in Carbonate is four times greater at a 4.2 WAG ratio (CD3 ) compared to a 0 WAG ratio (CD1 ). 2. The required asset size to store industrial-scale captured CO2 with the EOR process is likely to be larger than a single field (Fig. 4(b)). Providing sufficient injection capacity in a timely manner for EOR under a fixed storage requirement is therefore an important consideration. The WAG ratio should be adjusted according to the field-scale recycle ratio to ensure the long-term availability of both injection capacity and storage capacity. 3. The reservoir properties (permeability and heterogeneity in these examples) significantly affect the EOR-storage performance. The low permeability and high heterogeneity of Carbonate causes the oil production performance to be small and the CO2 utilization to be large, compared to Sandstone. A comparison of Fig. 9(a) and (b) shows the maximum cumulative oil production from Sandstone is at least two times greater than the maximum for Carbonate for the same cumulative CO2 storage.

86

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

Fig. 7. CO2 -EOR and storage design optimization workflows for (a) the fixed storage requirement problem, and (b) the integrated asset optimization problem.

4. CO2 -EOR aims to maximize the incremental produced oil with the available CO2 . CO2 -storage aims to maximize the CO2 storage in the available pore volume. According to Fig. 8(a) and (b), increasing the WAG ratio would increase the required pore volume for the 3 MMtonne/year storage; this is detrimental to the storage objective. According to Fig. 9(a) and (b), increasing the WAG ratio would increase the incremental oil production; this is beneficial to the EOR objective. The integrated asset model balances the EOR and storage objectives to maximize the overall economic benefit. Fig. 10 shows the integrated asset results for Sandstone (SD2 ) and Carbonate (CD3 ). This figure summarizes the field-scale economic calculations in the form of two cashflow diagrams: one for the oilfield operator (white bars) and one for the government (gray bars). The cashflow for the government results from the government’s investment in the form of storage tax credit and subsequently the revenue tax and royalty collection. The vertical axis is the discounted levelized profit that is calculated as the discounted annual revenue or cost divided by the annual CO2 storage requirement. The algebraic summation of discounted levelized profits over the project life (T) yields the levelized profit of storage (LPOS), and if multiplied by the sum of annual storage (Strg), yields the net present value (NPV).

NPV = LPOS ×

T 

Strg(i)

i=1

This applies to both the operator and the government cashflows. The reference of gain or loss is set from the operator’s viewpoint; therefore, the operator’s profit is reflected as upward (positive)

white bars and the government’s profit is reflected as downward (negative) gray bars. The cashflow diagrams indicate that all EOR-storage projects have at least two partners: the oilfield operator and the government. Several important implications follow: 1. The discount rate (i) for the operator is 10% and for the government is 3.5% because the operator exposure to an economic loss is greater than the government. According to Fig. 10, the LPOS values for the operator and the government for Sandstone (SD2 ) are 26 and $10/ton, respectively and for Carbonate (CD3 ) are 12 and $7/ton, respectively. 2. EOR-storage operation in Sandstone generates a positive NPV and indicates an optimistic economic prospect. The capital recovery duration for the initial investment is 5 years and the project maintains a positive cashflow up until 25 years. Carbonate has a moderate positive NPV, the capital recovery duration for the initial investment is 10 years and the project maintains a marginal positive cashflow until year 25. 3. With all conditions being equal for the two reservoirs, EOR-storage operation in Sandstone should be prioritized to Carbonate. This conclusion becomes important for the EOR-storage development planning in several fields with different characteristics: an integrated asset model can economically rank the EOR-storage prospects. We perform the integrated asset calculations for 90 different designs for each reservoir. Fig. 11 shows the final optimization results for Sandstone and Carbonate. The main measure of economic performance is the average NPV per reservoir pattern. Two solid curves separately connect the optimum average NPVs at any given WAG ratio for Sandstone and Carbonate. These curves help

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

Fig. 8. Optimum reservoir development schedule for the fixed storage requirement (a) Sandstone: SD1 , SD2 , SD3 , and (b) Carbonate: CD1 , CD2 , CD3 .

to better observe average NPV variation with respect to the WAG ratio. All the data points that fall below the curves are sub-optimal designs for the given reservoir. The observations and conclusions follow: 1. Increasing the WAG ratio decreases the CO2 utilization, accelerates the project deployment and subsequently the incremental oil production; however, the required asset size to store the same CO2 mass increases because water takes up the pore space instead of CO2 . This increases the upfront investment, delays and reduces peak oil production, and reduces the plateau production duration for individual patterns. Between the two extremes of continuous CO2 injection (WAG = 0) and a pure waterflood (WAG = infinity) an optimum WAG ratio exists that maximizes the average NPV per pattern. This conclusion applies to both Sandstone and Carbonate. For Sandstone, SD2 is the optimum EOR-storage design with 1.5 WAG ratio. For Carbonate, CD3 is the optimum with 4.2 WAG ratio. Other design parameters for SD2 and CD3 are listed in Table 2.

87

Fig. 9. EOR-storage performance for three designs for (a) Sandstone: SD1 , SD2 , SD3 , and (b) Carbonate: CD1 , CD2 , CD3 .

2. Increasing the pattern flood duration would increase the cumulative CO2 storage in the patterns and subsequently reduces the asset size. Long flood durations in the medium-term reduce the pace of expansion, increase the CO2 recycle ratio, and decrease the oil production rate. For the examined flood durations, i.e. 15, 20, and 25 years, it is better to continue the patterns flood at least up to 25 years for simultaneous WAG designs, and up to 20 years for continuous CO2 injection. 3. The average NPV for the optimum design in Sandstone (SD2 ) is $23 MM/pattern and for Carbonate (CD3 ) is $8 MM/pattern. The optimum average NPV for Sandstone is nearly 3 times greater than Carbonate and therefore Sandstone is clearly a better option; however, less than 10% of the EOR-storage candidates have an optimistic performance similar to Sandstone. It is more likely that the future EOR-storage candidates perform similar to Carbonate (see Fig. 4(a)). The fixed storage requirement problem imposes a constraint on the annual CO2 storage and assumes a very large asset size. Therefore, in the optimization procedure, we look into a wide range of field development scenarios that may satisfy the CO2 storage

88

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

Discounted Average Profit ($/tonne)

a

100 50 0 -50 -100 -150 -200

Discounted Average Profit ($/tonne)

b

Government, i = 3.5% Operator, i = 10% 0

5

10 15 Time (year)

20

25

100

Fig. 11. Field-scale design optimization results for the fixed storage requirement problem. 90 EOR-storage designs are investigated for each reservoir model. For Sandstone the optimum design is SD2 and for Carbonate the optimum is CD3 .

50 a CO2 contract that delivers a constant fresh CO2 feed for 25 years and up to 3 MMtonne/year CO2 is potentially available.

0

3.5. Integrated asset optimization workflow and results

-50 The solution method is slightly different from the fixed storage requirement problem because annual storage becomes a design parameter. We implement the following workflow (see Fig. 7(b)):

-100 -150 -200

Government, i = 3.5% Operator, i = 10% 0

5

10 15 Time (year)

20

25

Fig. 10. Integrated asset results for (a) Sandstone design SD2 , and (b) Carbonate design CD3 . White bars and gray bars respectively show the discounted levelized profits for the oilfield operator and the government. Positive white bars indicate profits for the operator and negative gray bars indicate profits for the government. The discount rates for the oilfield operator and the government are 10% and 3.5%, respectively.

constraint. This approach is useful when the environmental benefit of EOR-storage is prioritized and storage is the primary objective. In the following we investigate the integrated asset optimization problem in which the EOR element is prioritized. In this problem the asset size is fixed and the oilfield operator selects the commitment to CO2 storage based on the operational and the economic conditions. 3.4. Problem II: integrated asset optimization The integrated asset optimization problem prioritizes the oil recovery to CO2 storage. The goal of this problem is to find the optimum design, operational plan and the annual fresh CO2 slug for an EOR-storage project. Apart from the main supplier and customer, other suppliers and customers may provide (additional) CO2 or store (a surplus) CO2 . The oilfield operator is not obliged to store a pre-determined CO2 slug; however, once the CO2 contract is finalized, storage becomes mandatory. We consider multiple fields in the vicinity of the capture plant with overall 40 patterns. The oilfield operator is willing to negotiate

1. Select an EOR-storage design (fix the WAG ratio, injection rate, pattern flood duration). Retrieve the reservoir simulation responses on the fluid injection and production rates and calculate the oil production performance history, CO2 utilization, and the CO2 recycle ratio. 2. Select an annual fresh CO2 slug. We know the asset size for this problem is less than the fixed storage requirement example. Therefore, the upper bound for the optimum annual fresh CO2 is not greater than 3 MMtonne/year. The lower bound is normally zero, but it can be increased by investigation to narrow down the optimization interval. 3. For the selected EOR-storage design and annual fresh CO2 slugs, perform the field-scale calculations for the total fluid injection and production rates. Derive the optimum pattern development schedule and the EOR-storage performance curves. 4. If the total required producers and injectors for the selected EORstorage design and annual fresh CO2 slug are not greater than the total asset size, then proceed to the next step. Otherwise, discard the selected combination, return to Step-2 and select a smaller annual fresh slug. 5. Calculate the main parameters of the surface facility model, namely the pump and compression requirements, CO2 -gas separation plant size, and power requirement. 6. Collect the field-scale results and the surface facility outputs into the integrated asset model and perform the economic calculations. The main outputs are two cashflow diagrams for the oilfield operators and the federal government. 7. Repeat steps 2–6 for different fresh annual CO2 slugs. 8. Repeat Steps 1–7 for different EOR-storage designs. 9. Collect all the integrated asset results for all the feasible annual slugs and EOR-storage design combinations. Select the design that yields the maximum NPV per reservoir pattern.

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

89

The Carbonate is technically a feasible candidate for EOR but not economically feasible with an $80/ton CO2 cost. A tax credit, in this example $40/ton, decreases the project costs such that the oilfield becomes an economically feasible target for coupled EORstorage. About 50% of the current USA EOR projects perform similar to, or worse than Carbonate. Therefore, the presented workflow provides a method to quantify the effectiveness of storage tax credit on economic storage capacity. 4. Summary and conclusions We proposed a workflow for CO2 -EOR and storage design optimization and investigated the fixed storage requirement and integrated asset optimization problems for Sandstone and Carbonate reservoir models. The conclusions follow:

Fig. 12. Integrated asset optimization results. The optimum design for Sandstone is SD2 with 1.8 MMtonne/year fresh CO2 slug. The optimum for Carbonate is CD3 with 1.5 MMtonne/year fresh CO2 slug.

Fig. 12 presents the final results. The WAG ratio and the fresh CO2 slug are the key design parameters. The WAG ratio effect is similar to the fixed storage requirement problem. Increasing the annual CO2 slug accelerates the project deployment and is beneficial to the economics; however, the asset size is limited and so is the storage capacity; increasing the WAG ratio occupies the available pore volume with water and subsequently constrains the CO2 storage capacity. Therefore, the annual CO2 slug and the WAG ratio should be simultaneously adjusted to optimize the pace of development and the dedicated CO2 injection and storage capacities. The optimum design for Sandstone is SD2 with 1.5 WAG ratio and 1.8 MMtonne/year fresh annual CO2 slug. The optimum for Carbonate is CD3 with 4.2 WAG ratio and 1.5 MMtonne/year fresh annual CO2 slug. The priorities of EOR and storage objectives significantly affect the optimum EOR-storage design. In the presented example, maximizing the economic benefits (EOR performance) is the primary objective and meeting a certain storage requirement (storage performance) is the secondary objective. As a result, the optimum EOR-storage design yields ($1057 MM NPV/40 patterns=)$26.4 MM/pattern NPV. If the two objectives are swapped, i.e. maximizing storage capacity is set as the primary objective and maximizing the economic benefits is set as the secondary objective, then Sandstone would store the maximum available CO2 (3 MMtonne/year) at a 0.5 WAG ratio, yielding a ($506 MM NPV/40 patterns=)$12.5 MM/pattern NPV. This reflects a ((26.4–12.5)/12.5 × 100=)53% decrease in the economic profitability. Performing a similar analysis for Carbonate shows that a cumulative 75 MMtonne CO2 storage at a 1.3 WAG ratio, yields a $-160 MM NPV which is economically infeasible. This important observation shows that both reservoirs technically have the storage capacity for 75 MMtonne captured CO2 . However, the economic conditions are not in favor of the storage objective. CO2 cost and storage tax credit affect the optimum annual CO2 slug and subsequently the economic storage capacity. It is more profitable to store larger CO2 slugs if CO2 cost decreases or storage tax credit increases. For example, a $20/ton decrease in CO2 cost enhances the economic storage capacity by 40% for Sandstone; a $25/ton decrease enhances the economic storage capacity by 50% for Carbonate. On the other hand, economic storage capacity decreases if the cost is too high and no storage tax credit is awarded.

1. Coupled CO2 -EOR and storage is an attractive storage option because of its potential to increase the domestic oil production from the depleted oilfields. An optimized EOR-storage process at $60/stb oil price and $40/ton storage tax credit may cover up to $80/ton of the levelized CO2 capture and transport costs while remaining profitable for the operators and the government. Therefore, the economic benefits of EOR-storage for operators and the federal government are significant. 2. Coupled CO2 -EOR and storage is not the ultimate solution for the greenhouse gas emissions control. The required asset size to store industrial-scale captured CO2 is a major challenge. The EOR-storage optimization results for Sandstone show that for the optimum 1.5 WAG ratio, more than 400 producers and 400 injectors in a mature oilfield are required to store 75 MMtonne CO2 over 25 years; few active USA CO2 -EOR projects have such a capacity and they may not be available near the anthropogenic CO2 sources as reported in Koottungal (2012) and illustrated in Fig. 4(b). Multiple large fields and a probably a deep saline aquifer should be considered as the primary and secondary storage targets under equal conditions to cover all the storage requirements for a medium-sized pulverized coal plant with 65% capture rate (Black et al., 2011). 3. Reservoir properties affect the field response to the EOR-storage process, including peak oil production rate, pace of recovery, average oil production rate, CO2 utilization factor, and CO2 recycle ratio. Sandstone has high permeability and moderate heterogeneity while Carbonate has low permeability and high heterogeneity. The oil production performance for Sandstone is nearly twice as much as Carbonate, whereas the CO2 utilization is about 40% less. The choice of a sandstone reservoir representing an optimistic performance and a carbonate reservoir representing a moderate performance should not be interpreted as a general rule. Reservoir characteristics, production history, and EOR-storage design ultimately determine the performance. 4. The proposed workflow can be used to economically categorize and rank the EOR-storage candidates. Sandstone potentially presents the top 10% of the candidates; the middle-range candidates are likely to be similar to Carbonate. Therefore, Sandstone should be prioritized to Carbonate under equal conditions. Preliminary studies estimate the regional EOR-storage potential of the U.S. oilfields (Kuuskraa et al., 2011). A nationwide survey may categorize and rank the EOR-storage candidates, and identify the economically attractive projects for early-stage EOR-storage deployment. 5. An EOR-storage prospect ranking should honor the equal comparative conditions. Appropriate cost adjustments should be considered before the ranking if necessary. For example, the acquisition, pipeline, and infill drilling costs are strongly casesensitive. These cost elements are defaulted to zero for both

90

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

Sandstone and Carbonate in the integrated asset model. In screening applications, the upfront costs of asset acquisition and pipeline construction should be included for fair comparison between the EOR-storage candidates. Acknowledgements This research has been financially supported by BP America, Center for Petroleum Asset Risk Management (CPARM) at the University of Texas at Austin and Industrial Affiliate Program (IAP) at the University of Texas at Austin. The suggestions and feedbacks from Dr. Christopher Jablonowski, Charles Christopher and Dr. Subhash Thakur are greatly appreciated. Appendix A: Sandstone and Carbonate reservoir models. The displacement of oil with CO2 and water in a permeable medium is a fairly complicated problem that requires appropriate tools for modeling multiphase fluid flow of compositional phases in a heterogeneous permeable medium. No closed-form set of analytical equations may predict the oil production response to the alternating injection of water and gas without making significant simplifying assumptions (McCoy and Rubin, 2008). We introduce two reservoir simulation models: Sandstone (S) and Carbonate (C). Both models are developed using CMG-GEMTM , a compositional reservoir simulator. The reservoir structure of both Sandstone and Carbonate is described by a 3D corner-point grid, consisting of 28 × 40 × 15 gridblocks in the x, y, and z directions respectively. The reservoir consists of two adjacent anticlines, each with a 150 ft approximate thickness, at a depth of 6100 ft. Fig. A.1 shows the 3D view of the reservoir, the oil saturation distribution in Sandstone at the beginning of the CO2 -EOR phase, and the well names and locations. There are 6 injectors and 8 producers in this reservoir pattern, and the well pattern is a line-drive with producers located on the surrounding and the producers located around the center. The majority of the residual oil is in the middle. All the 6 injectors remain active during the 25-year EOR operation and out of the 8 producers, 6 producers remain active. Two producers (P0826 and P1307) are shut-in early in the EOR phase because of excessive water production (a watercut greater than 0.99). For the Carbonate model, all the 8 producers remain active.

Fig. A.1. 3D view of the reservoir Initial oil saturation distribution at the beginning of EOR-storage for Sandstone and well locations. During the EOR-storage operation, 6 injectors and 6 producers remain active.

Fig. A.2. Tuning the reservoir model performance with real CO2 -EOR projects (a) Sandstone model and Kinder Morgan (San Andres) CO2 flood scoping predictions on oil production performance with EOR-storage design SD2 . A good agreement between the predictions of the two models is observed (b) Monthly production rate distribution of Sacroc (2000–2009) and field-scale production distribution of Sandstone model for EOR-storage Design SD2 . Both datasets have a log-normal distribution of production rate with a Dykstra–Parsons coefficient of about 0.55.

The main differences between Sandstone and Carbonate are the reservoir permeability, heterogeneity, and the well spacing. Table 1 summarizes reservoir characteristics of Sandstone and Carbonate. Sandstone is moderately heterogeneous with an average 63 md areal permeability. Carbonate is highly layered with an average 18 md layer permeability. Areal and vertical Dykstra–Parsons coefficients are used as the main measures of heterogeneity. Moderate areal heterogeneity (VDP = 0.6) is assumed for both Sandstone and Carbonate (Lake, 1989). The vertical heterogeneity of sandstone is assumed to be fairly moderate (VDP = 0.65), while the Carbonate is vertically layered and highly heterogeneous (VDP = 0.9) with a cyclic permeability pattern (Kerans et al., 1994; Lucia et al., 2003; Lucia, 2007). The main advantage of using reservoir simulation as an EORstorage scoping tool is the ability to investigate the effects of reservoir heterogeneity, permeability, rock-fluid properties, and operational conditions. The reader is referred to Ettehadtavakkol

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

(2013) for further details on the reservoir and fluid properties of Sandstone and Carbonate models and a detailed discussion on the effect of reservoir properties on the EOR-storage performance. Here we provide a high-level summary of the major considerations in the development and tuning of the Sandstone and Carbonate models. 1. The reservoir properties and rock-fluid properties under the EOR-storage operation should be similar to the real cases. We select the West Texas CO2 -EOR data as the reference for tuning and comparison of the reservoir simulation results. These fields are widely studied and their information is available in the literature and the public databases (e.g. Drilling Info). CO2 storage is not the main objective of these projects; however, we may expect that the future mature reservoir candidates for EORstorage will have properties that are somewhat similar to these reservoirs. 2. Development of reservoir simulation models requires some background and knowledge of reservoir engineering, especially for CO2 -EOR. The Sandstone and Carbonate models are developed by integrating the available literature for CO2 -EOR screening (Winter and Bergman, 1993; Taber, 1994; Jarrell, 2002; Bachu, 2003) and the models for sandstone and carbonate reservoirs (Alston et al., 1985; Lake, 1989; Pariani et al., 1992; Martin and Taber, 1992; Kerans et al., 1994; Larsen and Skauge, 1998; Bossie-Codreanu and Le Gallo, 2004; Lucia et al., 2003; Ghomian, 2008; CMG, 2010). 3. The reservoir response to the WAG injection should be similar to the real CO2 projects. We select the published data in the Oil and Gas Journal (Koottungal, 2012) as a general source, the public injection and production data of Sacroc and the Kinder Morgan CO2 flood scoping spreadsheet (Kinder Morgan, 2002). Fig. A.2(a) compares the predictions of Kinder Morgan scoping spreadsheets for San Andres and Sandstone model. The EOR-storage design SD2 (Table 2) is provided as an input to both models. The oil production performance predictions are in a good agreement. A similar analysis is performed on the CO2 utilization. 4. Both production and injection rate distributions at field-scale should follow a log-normal distribution with a Dykstra–Parsons coefficient ranging between 0.5 and 0.7. Fig. A.2(b) compares the monthly production distribution of Sacroc and field-scale production distribution of Sandstone model for EOR-storage design SD2 . Both datasets have a log-normal distribution of production rate with a 0.55 Dykstra–Parsons coefficient. References Advanced Resources International (ARI), 2010. U.S. Oil Production Potential from Accelerated Deployment of Carbon Capture and Storage. Advanced Resources International, Inc., Arlington, VA. Alston, R., Kokolis, G., James, C., 1985. CO2 minimum miscibility pressure: a correlation for impure CO2 streams and live oil systems. SPE J. 25 (2), 268–274 (SPE-11959-PA). Bachu, S., 2003. Screening and ranking of sedimentary basins for sequestration of CO2 in geological media in response to climate change. Environ. Geol. 44 (3), 277–289. Bachu, S., Shaw, J., 2003. Evaluation of the CO2 sequestration capacity in Alberta’s oil and gas reservoirs at depletion and the effect of underlying aquifers. J. Can. Petrol. Technol. 42 (9), 51–61. Beecy, D.J., Kuuskraa, V.A., 2005. Basic strategies for linking CO2 enhanced oil recovery and storage of CO2 emissions. In: Proceedings of the 7th International Conference on Greenhouse Gas Control Technologies (GHGT-7), vol. I, September 5–9, Vancouver, Canada, pp. 351–360. Bergman, P.D., Winter, E.M., Chen, Z.-Y., 1997. Disposal of power plant CO2 in depleted oil and gas reservoirs in Texas. Energy Conv. Manage. 38, S211–S216. Black, J.B., Haslbeck, J.L., Jones, A.P., Lundberg, W.L., Shah, V., 2011. Cost and performance of PC and IGCC plants for a range of carbon dioxide capture. DOE/NETL report prepared by Research and Development Solutions, LLC. Report DOE/NETL-2011/1498. http://www.netl.doe.gov/energy-analyses/pubs/ Gerdes-08022011.pdf Bock, B., Rhudy, R., Herzog, H., Klett, M., Davison, J., De la Torre Ugarte, D., Simbeck, D., 2003. Economic evaluation of CO2 storage and sink options. Department of Energy Research Report.

91

Bossie-Codreanu, D., Le Gallo, Y., 2004. A simulation method for the rapid screening of potential depleted oil reservoirs for CO2 sequestration. Energy 29, 1347–1359. Bryant, S.L., Lake, L.W., 2005. In: Benson, S.M. (Ed.), Effect of Impurities on Subsurface CO2 Storage Processes, Carbon Dioxide Capture for Storage in Deep Geologic Formations – Results from the CO2 Capture Project. v.2. Geologic Storage of Carbon Dioxide with Monitoring and Verification. Elsevier, London, pp. 983–998. Computer Modeling Group Ltd. (CMG), 2010. GEM User’s Guide. Computer Modeling Group Ltd., Calgary, Alberta. Davidson, C.L., Dahowski, R.T., Dooley, J.J., 2011. A quantitative comparison of the cost of employing EOR-coupled CCS supplemented with secondary DSF storage for two large CO2 point sources. Energy Proc. 4, 2361–2368. Dooley, J.J., Dahowski, R.T., Davidson, C.L., 2009. Comparing existing pipeline networks with the potential scale of future US CO2 pipeline networks. Energy Proc. 1, 1595–1602. Energy Information Administration (EIA), 2012. Oil and gas lease equipment and operating costs 1994 through 2009. http://www.eia.gov/pub/oil gas/natural gas/data publications/cost indices equipment production/current/coststudy. html Ettehad, A., Jablonowski, C.J., Lake, L.W., 2010. Gas storage facility design under uncertainty, SPE Projects. Facilities Construct. 5, 155–165 (SPE-123987-PA). Ettehad, A., Jablonowski, C.J., Lake, L.W., 2011. Stochastic optimization and uncertainty analysis for E&P projects: a case in offshore gas field development. In: Paper OTC 21452 presented at Offshore Technology Annual Conference, Houston, 2–5 May. Ettehad, A., 2013. Storage compliance in coupled CO2 -EOR and storage. Greenhouse Gases: Sci. Technol. 4 (1), 66–80. Ettehadtavakkol, A., (Ph.D. dissertation) 2013. CO2 EOR-storage Design Optimization Under Uncertainty. University of Texas at Austin, Austin, Texas, USA, http://repositories.lib.utexas.edu/handle/2152/21470. Ettehadtavakkol, A., Lake, L.W., Bryant, S.L., 2014. Impact of storage tax credit on economic viability of CO2 storage with EOR. In: Paper SPE 169838-MS to be presented at SPE hydrocarbon Economics and Evaluation Symposium, 19–20 May, Houston, TX. Ghomian, Y., (Ph.D. dissertation) 2008. Reservoir Simulation Studies for Coupled CO2 Sequestration and Enhanced Oil Recovery. University of Texas at Austin. Gozalpour, F., Ren, S., Tohidi, B., 2005. CO2 -EOR and storage in oil reservoirs. Oil Gas Sci. Technol. 60, 537–546. ICF International, 2009. Developing a Pipeline Infrastructure for CO2 Capture and Storage: Issues and Challenges. Interstate Natural Gas Association of America Foundation, Washington, DC. Intergovernmental Panel on Climate Change (IPCC), 2005. In: Metz, B., Davidson, O., deConinck, H., Loos, M., Meyer, L. (Eds.), IPCC Special Report on Carbon Dioxide Capture and Storage. Cambridge University Press for the Intergovernmental Panel on Climate Change, Cambridge. Intergovernmental Panel on Climate Change (IPCC), 2007. Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge, United Kingdom and New York, NY, USA. Internal Revenue Service (IRS), 2009. Internal Revenue Bulletin: 2009-44, November 2, 2009. Notice 2009-83, Credit for carbon dioxide sequestration under Section 45Q. Internal Revenue Service, Washington, DC. Jarrell, P.M., 2002. Practical Aspects of CO2 Flooding. Henry L. Doherty Memorial Fund of AIME, Society of Petroleum Engineers, Richardson, TX. Johnson, N., Ogden, J., 2010. Transporting CO2 : independent pipelines for each source or organized regional networks. In: 9th Annual Conference on Carbon Capture & Sequestration, Pittsburgh, PA. Kerans, C., Lucia, F.J., Senger, R., 1994. Integrated characterization of carbonate ramp reservoirs using Permian San Andres Formation outcrop analogs. AAPG Bull. 78, 181–216. Kinder Morgan, 2002. CO2 flood scoping model for San Andres. http://www. kindermorgan.com/business/CO2/morscale.xls Koottungal, L., 2012. Special report: EOR/heavy oil survey: 2012 worldwide EOR survey. Oil Gas J. 110 (4), http://www.ogj.com/articles/print/vol-110/issue-4/ general-interest/special-report-eor-heavy-oil-survey/2012-worldwide-eorsurvey.html. Kovscek, A., 2002. Screening criteria for CO2 storage in oil reservoirs. Petrol. Sci. Technol. 20 (7–8), 841–866. Kovscek, A., Cakici, M., 2005. Geologic storage of carbon dioxide and enhanced oil recovery. II. Cooptimization of storage and recovery. Energy Conv. Manage. 46 (11), 1941–1956. Kuuskraa, V.A., Van Leeuwen, T., Wallace, M., 2011. Improving domestic energy security and lowering CO2 emissions with next generation CO2 enhacned oil recovery (CO2 -EOR). DOE/NETL report prepared by Energy Sector Planning and Analysis. http://www.netl.doe.gov/energy-analyses/pubs/storing %20CO2%20w%20eor final.pdf Lake, L.W., 1989. Enhanced Oil Recovery. Prentice Hall, Inc., New Jersey. Larsen, J., Skauge, A., 1998. Methodology for numerical simulation with cycledependent relative permeabilities. SPE J. 3 (2), 163–173 (SPE38456-PA). Leach, A., Mason, C.F., Veld, K.v.t., 2011. Co-optimization of enhanced oil recovery and carbon sequestration. Resour. Energy Econ. 33, 893–912. Lucia, F., 2007. Carbonate Reservoir Characterization. Springer, New York. Lucia, F., Kerans, C., Jennings, J., 2003. Carbonate reservoir characterization. J. Petrol. Technol. 55 (6), 70–72 (Paper SPE-82071). Mandelker, P., 1992. Tax Credit, bills may expand EOR opportunities in US. Oil Gas J., 69–72, http://rambiochemicals.com/tax incentives/EOR Tax Report.pdf.

92

A. Ettehadtavakkol et al. / International Journal of Greenhouse Gas Control 25 (2014) 79–92

Marston, P.M., Moore, P.A., 2008. From EOR to CCS: the evolving legal and regulatory framework for carbon capture and storage. Energy Law J. 29, 421. Martin, D., Taber, J., 1992. Carbon dioxide flooding. J. Petrol. Technol. 44 (4), 396–400. Massachusetts Institute of Technology Energy Initiative (MITei),2010. Role of enhanced oil recovery in accelerating the deployment of carbon capture and sequestration. In: Symposium report. MIT Energy Initiative and Bureau of Economic Geology at University of Texas at Austin. McCollum, D.L., Ogden, J.M., 2006. Techno-economic models for carbon dioxide compression, transport, and storage & correlations for estimating carbon dioxide density and viscosity. Institute of Transportation Studies, University of California at Davis. McCoy, S.T., Rubin, E.S., 2008. An engineering-economic model of pipeline transport of CO2 with application to carbon capture and storage. Int. J. Greenhouse Gas Control 2 (2), 219–229. Mohitpour, M., Golshan, H., Murray, M.A., 2007. Pipeline Design & Construction: A Practical Approach. American Society of Mechanical Engineers press, Fairfield, NJ. Moritis, G., 2003. CO2 sequestration adds new dimension to oil, gas production. Oil gas J. 101 (9), 39–44. Nakicenovic, N., Alcamo, J., Davis, G., de Vries, B., Fenhann, J., Gaffin, S., Gregory, K., Grubler, A., Jung, T.Y., Kram, T., 2000. Special report on emissions scenarios: a special report of Working Group III of the Intergovernmental Panel on Climate Change. Pacific Northwest National Laboratory/Environmental Molecular Sciences Laboratory (USA), Richland, WA, USA. Oreskes, N., 2004. The scientific consensus on climate change. Science 306 (5702), 1686. Pariani, G., McColloch, K., Warden, S., Edens, D., 1992. An approach to optimize economics in a West Texas CO2 flood. J. Petrol. Technol. 44 (9), 984–988 (SPE22022-PA). Riddiford, F., Tourqui, A., Bishop, C., Taylor, B., Smith, M., 2003. A cleaner development: the In Salah gas project, Algeria. In: Proceedings of the 6th International Conference on Greenhouse Gas Control Technologies (GHGT-6), vol. 1, 1–4 October, Kyoto, Japan, pp. 601–606. Rubin, E.S., Chen, C., Rao, A.B., 2007. Cost and performance of fossil fuel power plants with CO2 capture and storage. Energy Policy 35, 4444–4454.

Rubin, E.S., 2012. Understanding the pitfalls of CCS cost estimates. Int. J. Greenhouse Gas Control 10, 181–190. Sebastian, H., Wenger, R., Renner, T., 1985. Correlation of minimum miscibility pressure for impure CO2 streams. J. Petrol. Technol. 37, 2076–2082. Smith, C.E., 2009. Transportation-special report pipeline profits, capacity expansion plans grow despite increased costs. Oil Gas J. 107, 60. Steelman, J., Tonachel, L., 2010. Reducing Imported Oil with Comprehensive Climate and Energy Legislation. Natural Resources Defense Council, Washington, DC. Stevens, S.H., Kuuskraa, V.A., Gale, J., Beecy, D., 2001. CO2 injection and sequestration in depleted oil and gas fields and deep coal seams: worldwide potential and costs. Environ. Geosci. 8 (3), 200–209. Taber, J., 1994. A study of technical feasibility for the utilization of CO2 for enhanced oil recovery. In: The Utilization of Carbon Dioxide from Fossil Fuel Fired Power Stations (Appendix B). IEA Greenhouse Gas R&D Programme, Cheltenham, UK. Thomas, C.D., Cameron, A., Green, R.E., Bakkenes, M., Beaumont, L.J., Collingham, Y.C., Erasmus, B.F., De Siqueira, M.F., Grainger, A., Hannah, L., 2004. Extinction risk from climate change. Nature 427 (6970), 145–148. Thompson, J., Waltzer, K., Fowler, M., Chaisson, J., 2010. The carbon capture and storage imperative: recommendations to the Obama administration’s interagency carbon capture and storage task force. Report prepared by Cleanair Task Force. http://www.catf.us/resources/publications/files/The Carbon Capture and Storage Imperative.pdf Van Bergen, F., Gale, J., Damen, K., Wildenborg, A., 2004. Worldwide selection of early opportunities for CO2 -enhanced oil recovery and CO2 -enhanced coal bed methane production. Energy 29 (9–10), 1611–1621. van’t Veld, K., Mason, C.F., Leach, A., 2012. The Economics of CO2 Sequestration through enhanced oil recovery. http://www.biee.org/wpcms/wp-content/ uploads/Mason-TheEconomicsOfCO2SequestrationThroughEnhancedOil Recovery.pdf Walsh, M., Lake, L.W., 1989. Applying fractional flow theory to solvent flooding and chase fluids. J. Petrol. Sci. Eng. 2, 281–303. Winter, E., Bergman, P., 1993. Availability of depleted oil and gas reservoirs for disposal of carbon dioxide in the United States. Energy Conv. Manage. 34 (9–10), 1177–1187.