managing experimental-data shortfalls for fair screening ...€¦ · in-situ reservoir conditions,...
Post on 06-Jun-2020
3 Views
Preview:
TRANSCRIPT
October 2016 • Oil and Gas Facilities 1
Managing Experimental-Data Shortfalls for Fair Screening at Concept Selection: Case Study To Estimate How Acid-Gas Injection Affects Asphaltene-Precipitation Behavior
Hideharu Yonebayashi, Katsumo Takabayashi, and Ryo Iizuka, INPEX Corporation, and Slavko Tosic, ExxonMobil Development Company
have complexity of technical evaluation. Such complexity might be encountered when assuming an emerging condition or when intro-ducing emerging technologies. In such cases, potential concepts are often difficult to evaluate fairly with existing technologies, but can possibly be evaluated with newly introduced or developing evalu-ation measures. However, these new and developing measures re-quire cost that can be justified at the matured stage of development, but that cannot be justified at the concept-screening stage. In the fu-ture, the exploration and production (E&P) industry will be required to access more emerging fields of lesser easy oil; thus, this case study will be an example engaging a similar situation.
IntroductionThe target oil field is a carbonate buildup with a high-permeability, karstified and fractured rim and relatively low-permeability platform interior. The reservoir pressure and temperature are extremely high. The field is assumed to contain sour gas (highly undersaturated light oil with large gas content). High-pressure miscible gas injection is one of the potential options for enhanced oil recovery (EOR), as well as sulfur management. In generally considering robust asphaltene-flow-assurance management, it is common to evaluate asphaltene-precipitation risk before applying gas injection. On the basis of the associated gas, the potential injection gas is assumed to have very high concentrations of H2S. In addition to its HP/HT condition, this makes laboratory measurement difficult because of safety concerns. This motivated us to consider an alternative evaluation approach by use of numerical models instead of laboratory measurements. In the future, it is anticipated that additional projects of this new nature will lead to sour/AGI projects; therefore, the numerical approach suggested in this paper will be valuable for managing the technical complexity of those projects.
Potential Development Concepts. Gas reinjection is the main sce-nario used in this study to manage produced sour gas. Many gas-reinjection projects have been implemented throughout the world, and available research on the topic was referenced to support this study (Darmentaev et al. 2010; Hopper et al. 2008). The general process involves reinjecting the associated acid gas back into the reservoir. The purpose of gas reinjection is to maintain the reser-voir pressure as much as possible, preferably at greater than the minimum miscibility pressure (MMP) for an extended amount of time. This can improve the oil-displacement efficiency. In addition, pressure maintenance at greater than the bubblepoint pressure can prevent premature gas breakthrough into the production wells. This not only increases the ultimate oil recovery, but also reduces the gas-processing volume, which is the key driver for reducing the amount of nonhydrocarbon byproducts (e.g., sulfur and CO2). For technical justification of gasflooding EOR compared with other op-tions, supplemental discussion was added as follows. Other EOR
Copyright © 2016 Society of Petroleum Engineers
This paper (SPE 170585) was accepted for presentation at the SPE Annual Technical Conference and Exhibition, Amsterdam, 27–29 October 2014, and revised for publication. Original manuscript received for review 26 November 2015. Revised manuscript received for review 7 May 2016. Paper peer approved 15 June 2016.
SummaryIn a carbonate field under high pressure and high temperature (HP/HT), a gas-injection scheme has been assessed to improve oil recovery through pressure maintenance and miscible displacement. The potential study assumed sequential application of several gas-injection concepts, including raw-gas injection (RGI) and acid-gas injection (AGI). Flow-simulation studies of these concepts revealed a variety of compositional changes to the in-situ fluid, depending on the injection scheme and composition of the injected gases. Fluid compositional change is a common trigger of asphaltene instability; therefore, to ensure a robust gas-injection development, it is impor-tant to evaluate the risk of asphaltene precipitation. Because of high hydrogen sulfide (H2S) concentrations of AGI fluid under HP/HT in-situ reservoir conditions, it is difficult to take an experimental approach for evaluating gas-mixed asphaltene-flow assurance at a normal laboratory. Hence, at the concept-selection stage, this paper focuses on an alternative approach for numerical-modeling analysis of the AGI scenario, and presents the way in which AGI impacts asphaltene-precipitation behavior. On the basis of the asphaltene model established by applying a cubic-plus-association (CPA) equa-tion of state (EOS), which was calibrated with the experimentally measured asphaltene-onset pressure (AOP), a new binary-interac-tion-parameter (BIP) correlation between H2S and hydrocarbons was incorporated to evaluate variation of the asphaltene-precipita-tion envelope (APE) with periodic compositional change observed from the AGI-flow simulation. In the AGI scenario, injection gas was assumed to be 90 mol% H2S and 10 mol% carbon dioxide (CO2). The original reservoir fluid contains 15 mol% H2S concentration. During the 3D reservoir-simulation study for the AGI scenario, H2S concentration in produced fluid was observed to increase up to 76 mol% at a well located near acid-gas injectors. In the APE sensi-tivity analysis that was conducted independently for each compo-sition of H2S and CO2, the asphaltene model revealed that the base APE decreased as the H2S concentration increased and expanded as the CO2 concentration increased. As a result, for the mixed com-positions, the opposing effects on the APE offset each other, and the acid-gas addition produced a subsequent decrease of the APE. In summary, this work supported a relative merit of AGI from a thermodynamic asphaltene-flow-assurance point of view, while verification is needed with experimental data in the next defined/detailed engineering stages.
From stage-gate process-engineering points of view, this case study is also worthy to appropriately estimate potential concepts that
SPE_OGF_170585_160008.indd 1 25/08/16 7:51 PM
2 Oil and Gas Facilities • October 2016
methods (i.e., chemical and microbial EOR) on the basis of wa-terflooding were considered difficult to be candidates because of corrosion risks in the notional reservoir condition of a sour envi-ronment. The notional oil in this study was very light and of low viscosity; therefore, thermal EOR had a generally lesser impact on the recovery enhancement of high-mobility oil and was also re-moved from consideration. Microbial EOR was not applicable in our case because of high reservoir temperature that exceeded the general condition of the microorganisms’ habitat. Finally, gas-in-jection EOR was considered as an appropriate measure for our case because of the expected miscible injection and a large amount of associated gas as a sufficient injection-gas source.
Raw-Gas Injection. It was assumed that soon after the start of production, partial reinjection of the produced gas would be used. Before injection, the associated gas is separated from the oil stream and then dehydrated to protect the injection compressors and pipe-lines from corrosion and to prevent hydrate formation. The RGI compressor unit receives the acid gas, and then increases the pres-sure of the gas stream for reinjection into the reservoir. One advan-tage of the RGI process is that it reduces the requirement for gas treatment and undesirable sulfur production.
Acid-Gas Injection. It was assumed that in future phases, AGI would be implemented to increase gas sales while also reducing sulfur production. An offshore gas plant is assumed to process all produced gas, generating sweet gas for export and fuel, plus a stream of acid gas requiring disposal. Acid gas is assumed to have a con-centration of H2S up to 90%, with the remaining 10% being CO2. This notional AGI scheme is assumed to include injector(s) located near the offshore gas-plant/acid-gas-pumping island. Fig. 1 shows a potential AGI process-flow diagram. An offshore gas plant includes a gas-treating solvent unit to sweeten raw gas and mole sieve units to dehydrate before further transportation to an onshore processing facility. Sour-oilfield development requires long-term surplus-sulfur management. Current typical managements are character-ized, in part, by temporary storage on land near the site, existing AGI schemes, and newly developed sulfur-consuming products to use large amounts of surplus sulfur produced from the oil field (Davis et al. 2008). Among these, the land-storage option is not effective for our notional offshore development, and even if it were possible to apply this storage option, it should be a temporary solution that is
resolved by another method. Until the final solution is determined, outside storage needs a preservation plan to avoid the weathering that might occur as an effect of the environment. The option of newly developed sulfur-consuming products also requires transportation of the surplus to a consumption location. Therefore, the AGI scheme was considered to be the most-practical application in our notional field from the sulfur-management-strategy point of view.
Economic advantage can also be expected by AGI. For exporting associated gas from a sour oil field, sulfur content must be removed by means of a sweetening process with solvent and H2S conversion process to sulfur. These processes are usually expensive. Wichert and Royan (1997) pointed out that acid-gas reinjection could elimi-nate need for those processes and have the advantage of minimal or no sulfur emissions to the atmosphere. Furthermore, it was notion-ally assumed that there was no water in the bottom of the field, and this might be an advantage to the AGI option in the aspect of mate-rial-corrosion resistance—namely, a proper planning of a dehydra-tion system was considered to mitigate corrosion risks satisfactorily.
Air Pocket in Stage-Gate Process for Project DevelopmentThe stage-gate process is widely used in many business develop-ments, and Fig. 2 shows a typical stage-gate process for project devel-opment in the oil industry. During the initiation phase, any concepts that have the potential to be beneficial are gathered. The number of concepts in this phase is the maximum throughout the entire project timeline, while each concept value is not matured. In the concept selection phase, those gathered concepts are roughly evaluated while incorporating any technology options. Furthermore, all available data are collected from site visits, experiments, and inquiry to ven-dors. To collect all information needed for rough concept selection, concepts are simulated for evaluation from various points of view. To conduct a fair comparison, simulation estimation is also used for compensating data-insufficient parts. After Gate 2, concept value is improved, while the number of concepts is screened (see Fig. 3). It is difficult to adequately control screen-mesh condition for concept selection at Gate 2 in the recent era of innovation progression. When technology innovation is active, the initial number of concepts devel-oped might be more than that of an innovation-stagnated era. In gen-eral, the innovation-activated situation is preferable for engineering because it offers more choices and potentials. However, the existing screen mesh, well-matured for evaluating the existing technologies, cannot work sometimes for newly innovated technologies. To catch up and adequately evaluate such new innovative technologies, the screen mesh should also be improved/adjusted, but this can be costly. Here, engineers have a dilemma because they do not want to miss a potential innovative technology, but it might be difficult to make cost justification for improving the current screen mesh (i.e., evalu-ating methods). In the case of large potentials, such improving costs are usually high. Management always requests experimentally veri-fied evidence to take such costs. However, concept value at this stage would not be counterbalanced to it yet. This is a type of air pocket in the stage-gate process for project development.
In the definition phase, a systematic evaluation covering all competing technologies is vital to reaching the best concept and final investment decision (FID). Professional perspective from the disciplines of construction, operation, and cost-estimation is essen-tial for determining a suitable scheme and for improving construc-tability, operability, maintainability, and accurate cost estimation. Cost estimation can involve multiple project-execution strategies for better cost efficiency. Those strategies are needed for phased im-plementation to spread the investment because of limited funding plans. This definition phase is an important stage for more-detailed evaluation and optimization of the best concept by developing pro-cess-flow diagrams and risk-assessment reviews [hazard-identifi-cation (HAZID) studies], and selecting equipment. More-accurate cost estimation is performed in this phase on the basis of vendor quotes for equipment/items, site preparation, project schedule, and any utilities and support-services requirements.
OffshorePlatforms
Raw Fluid(Oil+Gas)
Raw
Gas
Acid
Gas S
weet G
as
Oil
AcidGas
(reservoir)(onshore facility)
Injectors
OffshoreProcess
Plant
OffshoreGas Plant
AG Pump
Fig. 1—AGI process flow. (AG = acid gas.)
SPE_OGF_170585_160008.indd 2 25/08/16 7:51 PM
October 2016 • Oil and Gas Facilities 3
In this case study, RGI and AGI schemes are potential concepts. The former scheme can be evaluated adequately by the conventional methods at the concept-selection phase; however, the latter scheme is not as simple. Hurdles in the evaluation of AGI depend on several factors, one of which is reservoir condition (reservoir temperature and pressure). At the HP/HT condition, experiments that use high-concentration H2S gas can be dangerous. Thus, there are few labora-tories that can handle such experiments. To do so, new experimental apparatus would need to be developed and the laboratory would need to formulate new safety measures. This can result in large costs that cannot be justified at this phase, but that might be justified at the phases of definition and/or detailed engineering. This case study is a
good demonstration sample to handle the potential concept of AGI without dropping into the aforementioned air pocket in the stage-gate process, in which we tried to compensate for the lack of experimental data by numerical-modeling analysis. In the future, the industry will need to access more emerging fields similar to our notional case be-cause of fewer easily accessible fields. For this reason, our study would be an example for handling such emerging occasions.
General Asphaltene Risk in Gas InjectionThe combination of asphaltic oil and components that have lower polarizability values is generally considered to decrease oil-solu-bility parameters. This increases asphaltene instability. Methane
• Define opportunity
• Selected concepts to be optimized• Equipment to be selected
• Cost estimation• Fine selection of concepts
Potential concepts
Planning
Initiation ConceptSelection Definition Detailed
Engineering EPCOperation
Execution
Concepts value
FID = Final investment decision.EPC = Engineering procurement, and construction.HAZID = Hazard identification.
Project Timeline
• HAZID
• Concept developments
• Concept evaluations• Technology options• Data collection• Rough selection of concepts to be further studied
Gate 1Gate 2
Gate 3
FID
Gate 4
Gate 5
Fig. 2—Stage-gate process for project development.
Screened conceptshave high potentials.
Technology Evaluation Sieve
Rou
gh S
elec
tion
Low-potentialconcepts into trash.
Some key technologies canbe adequately evaluated bythe existing method(acceptable to put in the currentsieve).
Others cannot be evaluatedby the existing method(cannot put in the current sieve).
Key Technologiesthat have large potentials to break throughcurrent E&P industry’s constraints.
Other Industries
E&P Industry
Can we develop a new evaluationsieve, even taking cost?
Fig. 3—Schematic of concept selection, introducing “key technologies from outside industries” or “E&P industry’s self-innovative one.”
SPE_OGF_170585_160008.indd 3 25/08/16 7:51 PM
4 Oil and Gas Facilities • October 2016
(CH4), CO2, and nitrogen (N2) gases are typical nonpolar mole-cules. They have low polarizability values that work as strong asphaltene precipitants. On the other hand, H2S is categorized as a polar molecule. According to van Gisbergen et al. (1996), its polar-izability value (26.34) is reported as greater than those of nonpolar molecular gases (CH4 = 18.19, CO2 = 17.72, and N2 = 12.27). Compared with those nonpolar gases, therefore, H2S is expected to have the contrary effect on asphaltene precipitation. In the case of mixing oil with acid gas, CO2 and H2S would tend to offset their individual effect on asphaltene precipitation. There are many cita-tions related to asphaltene-flow-assurance work in the case of gas mixing. To focus on acid gas, asphaltene-flow assurance is summa-rized individually for CO2 and H2S in the following subsections.
CO2 Gas Injection. A historical overview of the overall CO2-injection topic has reported the advantage of ultimate-oil-recov-ery improvement. However, from a recent citation, it is possible to find many references that discuss the negative phenomena of CO2 injection. According to a survey of literature related to CO2 injection, it clearly cautions about possible formation damage re-sulting from CO2-induced asphaltene precipitation (see Appendix A). Many discussions have been generated from laboratory experi-ments and numerical-modeling analyses. The properties of the oils in the literature show a wide range of °API value and asphaltene content. Here, it should be emphasized that asphaltene-induced formation damage was a concern, even in cases of very high °API value oil and little asphaltene content. Among the listed references, two publications—Adyani et al. (2011) and Alian et al. (2011)—used the highest °API value range of oils that are still lower, but closer to those in our study (see Table 1). Available composition data (Adyani et al. 2011) were compared with data from our study. Nonhydrocarbon composition (N2, CO2, and H2S) differs between oil in the Adyani et al. study (i.e., sweet) and this study (i.e., sour); however, the overall trend of hydrocarbons is considered similar in each. With respect to lithology and core features, notional assump-tions of tight carbonate reservoir in this study are not matched to those of the two reference papers. Consequent to similarity of °API value and asphaltene content, the findings from this literature could
be useful analogs. Adyani et al. (2011) mentioned the potential for asphaltene plugging when injecting gas with a high concentration of CO2. Alian et al. (2011) observed a maximum permeability re-duction of 22% after CO2 coreflood tests. Therefore, CO2 injection into light-oil reservoirs should be checked carefully in advance, when considered from the formation-damage point of view and ac-cording to the two reference cases, which revealed permeability reduction or potential of asphaltene plugging with high °API value, little asphaltene content, and sufficient permeability range.
H2S Gas Injection. While SGI has been widely applied and many studies are cited (van Vark et al. 2004), no literature was discovered that specifically argues that asphaltene-flow assurance was impact-ed by H2S injection. To begin with, most of the studies assumed deep saline aquifers for the acid-gas sequestration; therefore, the discussion covered systems between H2S and brine and/or rocks except for hydrocarbons (Bennion and Bachu 2008; Ghaderi et al. 2011). In the reactive-transport modeling study for CO2 storage in deep saline aquifers (Xiao et al. 2009), coinjection of CO2 with H2S showed quite similar impacts on gas/fluid/rock (carbonate and sandstone) interaction in comparison with injecting of CO2 alone. Thus, H2S itself is considered to be of little impact from the mineralogy point of view. Furthermore, carbonate-rock dissolution through acidification will yield the advantage of porosity increase.
Flow SimulationThe subsurface potential of the AGI scenario was assessed by use of 3D numerical simulations. A dynamic model applied dual-porosity/dual-permeability and compositional-formulation options to reproduce complex and heterogeneous reservoir characteristics and miscible gas-injection processes appropriately. An EOS with 11 components was used to treat acid components (H2S, CO2) and lighter hydrocarbons separately in this study. The heavier compo-nents of C6+ were grouped and split into three pseudocomponents. Main compositions of the total reservoir fluid are assumed to be 14.8 mol% H2S, 4.1 mol% CO2, and 47.8 mol% C1. Injection-gas (acid-gas) composition is assumed to be 90 mol% H2S and 10 mol% CO2.
Because carbonate reservoir rock is often heterogeneous, sweep efficiency is critical for improving oil recovery. The notional res-ervoir used for this study is a thick and large carbonate comprising two main development facies—a lower-permeability “platform interior” area and a “platform margin/rim” area with locally en-hanced permeability, as shown in Fig. 4. The notional development strategy assumes gas reinjection in the platform interior to avoid unexpected early gas breakthrough by means of high-permeability pathways in the fractured rim area. Fig. 5 shows a schematic of a potential injection pattern. The pattern includes three acid-gas injectors located in a cluster in the middle of the platform inte-rior to contain the acid gas. Various injection patterns were evalu-ated through dynamic reservoir-simulation study to optimize the balanced margin between injection potential and actual injection
Alian et al.
(2011) Adyani et al.
(2011)
enotsdnaS ygolohtiL(Berea core
used)
NA (reservoir core used)
Permeability (md) 162–165 56.1 9.12 32–81 )%( ytisoroP
Reservoir fluid component (mol%) N2 – 0.22 CO2 – 0.44 H2 00.0 – SC1 – 40.28 C2 38.4 – C3 37.5 – i-C4 45.1 – n-C4 – 2.76 i-C5 – 1.54 n-C5 – 1.44 C6+ – 41.22
00.001 – latoT 4.14 14 )IPA°( ytivarg IPA
Asphaltene content (wt%) 0.42 0.07
Table 1—Summary of analogs for taking risk of CO2-induced forma-tion damage into account.
Dissolution Karst Bodies
Fracture network
Platform InteriorPlatform Interior
Transition Zone
Platform Margin/Rim
Fig. 4—Schematic of field geological concept.
SPE_OGF_170585_160008.indd 4 25/08/16 7:51 PM
October 2016 • Oil and Gas Facilities 5
rate per well in the heterogeneous model, mimicking dual-porosity/dual-permeability features that captured the matrix section of the platform interior and the karstified/fractured section of the rim. Injectors were surrounded by platform-interior production wells, and gas breakthrough was observed in those that were near pro-duction wells. Further careful monitoring was performed so that gas breakthrough was controlled in production wells located in the outer circle of platform-interior producers. In addition, a gas-man-agement strategy was applied to producers to reduce the produc-tion rate when the gas/oil ratio (GOR) exceeded the specified limit.
After long-term simulation, the compositional change was in-vestigated for produced fluid at each production well. The largest compositional change was observed at Well P9. Fig. 6 shows the change in H2S and CO2 concentration over time, with H2S and CO2 maximum concentrations increasing up to 76 mol% (five times the original) and 6 mol% (one and a half times the original), respectively. The dynamic model was still considered to have geologic uncertain-ties because of the sparseness of drilled wells against the supergiant field size, but will be updated as more data are derived from newly drilled wells. At this time, this study focuses on the most-gas-break-through-progressed well (i.e., P9) to cover the most-pessimistic pre-diction result as conservatively as possible. Bachu and Shaw (2005) pointed out that reservoir heterogeneity should be considered care-
fully because these characteristics affect injectivity and, hence, the number of wells that would be needed to deliver the injection gas to the reservoir. From another viewpoint, reservoir heterogeneity may have a positive effect because it may mitigate the gravity segrega-tion effect by controlling the rise of injection gas to the upper layer and forcing it to spread laterally, resulting in better vertical sweep efficiency (Doughty et al. 2001). Therefore, more-detailed study will be performed to consider expansion of the injection area into the outer circle away from the central platform interior after more geological data are accumulated.
Because high-pressure miscible condition is expected experi-mentally by the notional RGI scheme, remaining oil saturation can be minimized after coming into contact with the injection gas. Mis-cibility has not been confirmed experimentally for AGI, but, in gen-eral, lower MMP is achieved by CO2 gas than by hydrocarbon gas because CO2 gas extracts much-higher-molecular-weight hydro-carbons from reservoir fluids. Therefore, the AGI system is further expected to decrease MMP because of more CO2 without hydro-carbon gas compared with the RGI system. In addition, Sayegh et al. (2006) measured effects of adding H2S on MMP in a CO2- injection system. The MMP was measured with a rising-bubble ap-paratus, and showed linear reduction of MMP as H2S composition
P1
P5I1I2I3P6
P10 P9 P8 P7 Y2
X3
P2P3P4
ProducerAcid gas injector
Appraisal well
Fig. 5—Well location map.
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 10 20 30 40 50 60
Co
nce
ntr
atio
n (
mo
l%)
Year
CO2
H2S
Fig. 6—Simulated change of H2S and CO2 concentration in Well P9 produced fluid.
Fluid Modeling Binary Interaction Parameters Between H2S and C15+
Basecase Asphaltene Modeling
Sensitivity Analysis
Feedback to gas injection plan
Flow Simulation
Input Data
Pseudocomponent
PVT Matching
Vary H2S and CO2 Concentration
Basecase Model generated
Input Data and Calibration
Appropriate BIP derived from Stamataki and Magoulas (2000) correlation
Reservoir Simulation
fluid composition
Using Multiflash (released from Infochem)
As modified thermodynamic colloidal modelData input (AOP: two points at different room temp)Calibration (with saturation pressure, AOP)
assuming natural depletion
assuming AGITo evaluate impact on Basecase APE
assuming AGImonitor periodicalvariation ofproduced fluidcomposition
decided sensitivityanalysis range forH2S and CO2
molecular weightGORsaturation pressureSARAreservoir pressure and temperature
heaviest component splitmore than 15 pseudocomponents
EOS : SRKcalibration with
saturation pressureoil densityoil viscosity
Fig. 7—Numerical-modeling work flow.
SPE_OGF_170585_160008.indd 5 25/08/16 7:51 PM
6 Oil and Gas Facilities • October 2016
increased. This gives us further expectation of lower remaining oil saturation after being flooded out by the notional acid gas, which included 90 mol% H2S. If gas recycling is performed continuously in such a reservoir environment, then acid-gas composition would concentrate more rapidly. This is one method of reaching H2S con-centration up to 76 mol% in our flow-simulation study.
Numerical Modeling of Asphaltene Risk Analyses To describe the manner in which compositional changes affect precipitation mechanisms, numerical-modeling analysis was applied. Because no experimental data exist for the high concen-trations of H2S used in this study, a numerical method is an appro-priate alternative to evaluate gas-mixed asphaltene-flow assurance.
Model and Work Flow. The APE was evaluated on the basis of the CPA approach with Multiflash (version 4.1; released from Infochem; Edmonds et al. 1999a, 1999b, 1999c). The Soave-Redlich-Kwong (SRK) cubic EOS (Soave 1972) was applied for fluid characterization in models. Additional terms were used to describe the association of asphaltene molecules and their solva-tion by resin molecules. It is well-known that asphaltene consists of polar compounds that are stabilized in oil by the presence of resin. The resin concentration has a major impact on asphaltene stabilization. If it is decreased by oil dilution from increasing light hydrocarbons in gas injection, a point would be reached at which the asphaltene stabilization is lost and it precipitates to form solid depos-
its. The resin-stabilizing action, working through the mechanism of polar interactions, becomes weaker so that precipitation may occur as the temperature increases. However, further increases in temperature make asphaltenes more soluble in oil. Thus, those two contradictory characteristics are built into the model. The model characterizes the interactions between asphaltenes and asphaltene resin by tempera-ture-dependent association constants, while the remaining heavy hy-drocarbons are treated as pseudocomponents with the usual BIPs. The modeling was done in accordance with the work flow shown in Fig. 7.
Binary Interaction Parameters Between H2S and C15+. Because of the lack of laboratory test data, BIPs between H2S and C15+ components were originally set to zero in the Multiflash database. This default assumption might have little impact on fluid modeling under an assumption of natural depletion for the notional reservoir condition used in this study. Because the oil in this study was assumed to be highly matured, very few heavy ends were contained; therefore, the interaction between H2S and such a low number of heavy ends might have less effect on fluid thermodynamics. However, when applying gas-injection schemes, the BIP data should be more ap-propriately assumed in the AGI case in particular because of the high concentration of H2S injected. Therefore, the BIPs for H2S/n-alkanes were developed by use of the following generalized ex-pression by Stamataki and Magoulas (2000), which was based on all the available experimental vapor/liquid equilibrium (VLE) data of H2S/hydrocarbon binary systems that were reviewed and col-
–0.1
–0.05
0
0.05
0.1
0.15
0.2
N2
H2S
CO
2C
H4
ET
HA
NE
PR
OP
AN
EIS
OB
UT
AN
EN
-BU
TA
NE
ISO
PE
NT
AN
EN
-PE
NT
AN
EC
6–10
C10
–13
C13
–16
C16
–19
C19
–22
C22
–25
C25
–29
C29
–32
C32
–36
C36
–38
R38
R39
R39
AR
40+
AS
PH
ALT
EN
E
BIP
Multiflash default
Correlation
–0.1
–0.05
0
0.05
0.1
0.15
0.2
N2
H2S
CO
2C
H4
ET
HA
NE
PR
OP
AN
EIS
OB
UT
AN
EN
-BU
TA
NE
ISO
PE
NT
AN
EN
-PE
NT
AN
EM
ET
HY
LCY
CLO
PE
NT
AN
EB
EN
ZE
NE
CY
CLO
HE
XA
NE
ME
TH
YLC
YC
LOH
EX
AN
ET
OLU
EN
EE
TH
YLB
EN
ZE
NE
M-X
YLE
NE
O-X
YLE
NE
C6-
11C
11-1
4C
14-1
6C
16-1
9C
19-2
2C
22-2
6C
26-3
0C
30-3
4C
34-3
7C
37-3
9R
39-4
5R
45R
45A
R46
+A
SP
HA
LTE
NE
BIP
Multiflash default
Correlation
(2) Y2 asphaltene model
(1) X3 asphaltene model
Fig. 8—Comparison of BIPs for H2S/hydrocarbons system between the Stamataki and Magoulas (2000) correlation and default in Multiflash.
SPE_OGF_170585_160008.indd 6 25/08/16 7:51 PM
October 2016 • Oil and Gas Facilities 7
lected by Stamataki and Magoulas (2000). More details regarding this equation are presented in Appendix B.
kij = −0 1029 0 1498. . �. ..........................................................(1)
As shown in Fig. 7, the BIPs between H2S and C15+ were cal-culated by Eq. 1, and the default data assumption was replaced with the newly estimated ones. Comparisons of BIPs between the correlation and default are shown in Fig. 8.
Base Models. Two models were produced for the fluids of Wells X3 and Y2. Well locations are shown in Fig. 5. Inputs for the assumed model are summarized in Table 2. Experimentally measured AOPs
were used for model calibration. AOPs were measured for single-phase bottomhole samples by isothermal depressurizing tests in which the laser-light-scattering technique (LST) was applied. Transmittance power of laser light was monitored continuously af-ter passing through the sample fluids in a glass cell during the mea-surement. AOPs were detected as sudden drops of the transmittance power because asphaltene solids started to intercept the laser light. In addition to this LST-applied measurement, appearance of asphal-tene solid particles was observed by high-pressure microscopy for validating the LST-based AOP. For more-accurate modeling, more AOPs are preferable, but this can increase cost as a result of expen-sive single-phase pressurized sampling. Here, we use a type of cost effectiveness between model accuracy and sampling cost. In this study, two AOPs are adopted for modeling the Well Y2 fluid, and one AOP was adopted for modeling the Well X3 fluid.
The Well Y2 fluid model was generated by calibrating two AOPs with measured data at 65 and 55°C, respectively. The most-conser-vative envelope was determined as a base model, and calibration pri-ority was given to the AOP at 55°C (see Fig. 9). The Well X3 fluid model was generated by calibrating one AOP measured at a no-tional reservoir temperature of 100°C (Fig. 10). Both envelopes have a similar shape even though their locations are not identical on the phase diagram. The APE boundary condition of Well X3 is closer to the reservoir condition than that of Well Y2, while no asphaltene pre-cipitation is observed for Well X3 fluid at the reservoir condition. The base APE of Well Y2 never intersects the reservoir temperature. This is consistent with the isothermal depressurizing experiment resulting in no AOP detection for Well Y2 fluid at reservoir temperature.
From the perspective of model accuracy, the most-critical infor-mation is onset condition at reservoir temperature because this point is usually the most-upstream location at which asphaltene precipi-tation will start on the entire flow stream, from reservoir to surface facilities. Therefore, the onset condition at reservoir temperature must be measured at least. Measurement of additional onset con-ditions at different temperature points that are usually lower than reservoir temperature is preferred, and these data cover the as-phaltene onset condition in the downstream section (i.e., tubing/surface-facilities/flowline operating conditions). Consequently, those additional AOPs contribute to improving the accuracy for the upper boundary trend of asphaltene onset condition on the ther-modynamic diagram. As shown in Fig. 9 for the Well Y2 model, a comparison between APE estimation and two AOPs could give us a certain confidence of model reliability. The Well X3 model is less accurate than the Well Y2 model because only one AOP was available for calibration purposes; however, a comparison with the Well Y2 model fluid could indirectly provide us with reliability in
00
2,000
4,000
6,000
8,000
10,000
12,000
50
Temperature (°C)
Pre
ssu
re (
psi
a)
100
APEPsat Line
Psat measuredY2, AOP at 65°CY2, AOP at 55°C
200150
Fig. 9—Base model of Well Y2 fluid.
00
2,000
4,000
6,000
8,000
10,000
12,000
50Temperature (°C)
Pre
ssu
re (
psi
a)
100
APE (Base Model)
Psat Line
Psat measured
X3, AOP at 100°C
150 200
Fig. 10—Base model of Well X3 fluid.
3X 2Y lleWSample Number 5 8
Component Monophasic (mol%)
Monophasic (mol%)
N2 0.82 0.86 H2 31.51 55.51 SCO2 3.98 4.37 C1 45.81 49.96 C2 94.6 70.7 C3 56.3 21.4 C4 86.2 19.2 C5 58.1 21.2 C6+ 30.51 26.71
20.001 00.001 latoTC30+ density (g/cm3) 1.010 1.000
)%tw( ARAS 5.36 7.47 etarutaS 7.33 55.12 citamorA 96.2 30.4 niseR
Asphaltene 0.28 0.17 Asphaltene onset pressure (psia)
At 100°C Not detected 9,900 – 091,4 C°56 tA – 000,8 C°55 tA
Table 2—Fluid composition and model inputs.
SPE_OGF_170585_160008.indd 7 25/08/16 7:51 PM
8 Oil and Gas Facilities • October 2016
the Well X3 model. In general, reservoir fluids collected from the unique field (i.e., on the basis of the same source rock and iden-tical maturity process) should share close asphaltene-precipitating behavior, which is the reason that both APEs have similar shapes. Therefore, Well X3 model validation was achieved by a combina-tion of two types of checks: (1) directly by calibrating with the most critical control point of actual measured AOP, and (2) indirectly by comparing with the Y2 model’s upper boundary APE trend.
Sensitivity Results. Through the compositional flow simulation, H2S varied from 15 to 76 mol%, and CO2 varied from 4 to 6 mol% at Well P9. Thus, fluid compositions were assumed to be investigated at H2S concentrations of 15, 30, 50, and 76 mol% and CO2 concen-trations of 4 and 6 mol% (see Fig. 11). To confirm each gas effect, a sensitivity study was conducted with single-component gas, which was added to both Y2 and X3 fluid; then, the representative gas com-position of each production period was used for the sensitivity study. For example, a fluid containing 30 mol% H2S and 5 mol% CO2 rep-resented a fluid for a production period of 6 years. Subsequently, sensitivities were studied for both components together.
H2S Sensitivity. A sensitivity of H2S compositional changes was evaluated, as shown in Figs. 12 and 13. For Fluids X3 and Y2, the results indicate that the area of APEs gradually decreases as H2S concentration increases. In addition, the APE boundary moves to the lower-temperature and -pressure side as H2S concentration increases. The increase of H2S concentration works to mitigate asphaltene-pre-cipitation risks. Saturation-pressure lines are gradually decreased by adding H2S gas, while gas addition is generally expected to increase saturation pressure. This is explained by the decrease of C1 concen-tration that is caused by an increase in H2S concentration. Because
the phase envelope of C1/hydrocarbons is generally larger than that of H2S/hydrocarbons with the same concentration of C1 and H2S, the relative reduction of C1 causes lower saturation pressure. This reduc-tion helps to locate APE apart from the reservoir condition.
CO2 Sensitivity. Figs. 14 and 15 show the results of the CO2 sensitivity study. Although the APE shapes were very similar, loca-tions were slightly different. Because CO2 is one of the strongest accelerators to increase asphaltene risk, it was expected to cause a large expansion of APE and a shift of the upper APE boundary to-ward reservoir condition. However, a small change in CO2 compo-sition had negligible effects on the APEs.
H2S and CO2 Sensitivity. Figs. 16 and 17 show sensitivity evaluations of H2S and CO2 compositional changes. The results are similar to those of the H2S-only sensitivity study because H2S concentrations are predominant over those of CO2. Specifically, the CO2 increase was offset by the H2S decrease. The area of APEs is smaller with time. The overall trends were similar between Y2 and X3 fluids. Compared with the locations of both APEs in each corresponding H2S and CO2 concentration, the X3 fluid has a higher asphaltene-precipitation risk. On the basis of this numerical study, the high H2S composition in AGI may reduce the asphal-tene-precipitation risk. The results shown in this paper have not yet been validated with laboratory measured AOP with H2S gas added to the reservoir fluid. Verification with experimental data is recommended.
Conclusion1. In a literature survey, no citation was found for H2S mixing
effects on asphaltene-precipitation behavior, while many papers addressed CO2-induced asphaltene precipitation.
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
N2
H2SCO2
C1
C2
C3
C4
C5
C6+
0
6
27
52
Concentration (mol%)
Pro
du
ctio
n P
erio
d (
year
s)
Fig. 11—Compositional change of Well P9 produced fluid.
00
2,000
4,000
6,000
8,000
10,000
12,000
50Temperature (°C)
Pre
ssu
re (
psi
a)
100
Soild Line = APE
Green = OriginalBlue = H2S 30 mol%
Purple = H2S 50 mol%
Red = H2S 76 mol%
Dashed Line = Psat Line
150 200
Fig. 12—Sensitivity of H2S concentration for APE of Well Y2 fluid.
00
2,000
4,000
6,000
8,000
10,000
12,000
50Temperature (°C)
Pre
ssu
re (
psi
a)
100
Soild Line = APE
Green = OriginalBlue = H2S 30 mol%Purple = H2S 50 mol%Red = H2S 76 mol%
Dashed Line = Psat Line
150 200
Fig. 13—Sensitivity of H2S concentration for APE of Well X3 fluid.
SPE_OGF_170585_160008.indd 8 25/08/16 7:51 PM
October 2016 • Oil and Gas Facilities 9
2. Asphaltene-precipitation models were generated with the BIPs of H2S/C15+ on the basis of the Stamataki and Magoulas (2000) cor-relation. The compositional change results in the flow simulation (maximum 76 mol% H2S and 6 mol% CO2 observed in produced fluid for AGI with 90 mol% H2S and 10 mol% CO2) were evalu-ated by the asphaltene-precipitation models as a sensitivity study.
3. The sensitivity study showed how acid gas affected asphaltene-precipitation risk.• Additional H2S gas reduced the risk of asphaltene precipitation.• Additional CO2 gas increased the risk of asphlatene precipitation.• Additional H2S and CO2 gas consequently reduced the risk of
asphaltene precipitation because the impact of lower CO2 con-centration was more than offset by higher H2S concentration.It was concluded that the asphaltene-precipitation risk
decreased in the AGI scenarios evaluated in this study, while verification is needed with comparison to experimental data.
4. The numerical method for sensitivity analysis was useful to briefly evaluate acid-gas impacts on asphaltene-precipitation behavior. The method could present an alternative to conducting expensive and dangerous laboratory measurements with high concentrations of H2S.
Nomenclature kij = interaction parameter in quadratic mixing rule between
component i (H2S) and j (hydrocarbon) Pi = initial reservoir pressure, psi Psat = saturation pressure, psi Ti = initial reservoir temperature, °C ω = acentric factor
AcknowledgmentsThe authors would like to thank INPEX and ExxonMobil management for permission to publish this paper.
ReferencesAbdallah, D. 2012. Impact of Asphaltenes Deposition on Completion
Design for CO2 Pilot in an Onshore Abu Dhabi Field. Presented at the Abu Dhabi International Petroleum Conference and Exhibi-tion, Abu Dhabi, 11–14 November. SPE-162190-MS. http://dx.doi.org/10.2118/162190-MS.
Adyani, W. N., Daud, W. A. W., Darman, N. et al. 2011. A Systematic Approach To Evaluate Asphaltene Precipitation During CO2 In-jection. Presented at the SPE Enhanced Oil Recovery Confer-
00
2,000
4,000
6,000
8,000
10,000
12,000
50Temperature (°C)
Pre
ssu
re (
psi
a)
100
Soild Line = APE
Green = OriginalRed = CO2 6.24 mol%
Dashed Line = Psat Line
150 200
Fig. 14—Sensitivity of CO2 concentration for APE of Well Y2 fluid.
00
2,000
4,000
6,000
8,000
10,000
12,000
50Temperature (°C)
Pre
ssu
re (
psi
a)
100
Soild Line = APE
Green = OriginalRed = CO2 6.24 mol%
Dashed Line = Psat Line
150 200
Fig. 15—Sensitivity of CO2 concentration for APE of Well X3 fluid.
00
2,000
4,000
6,000
8,000
10,000
12,000
50Temperature (°C)
100
Soild Line = APEDashed Line = Psat Line
150 200
Green = OriginalH2S 14.5 mol%, CO2 4.1 mol%
Blue = H2S 30.0 mol%, CO2 5.3 mol%Purple = H2S 49.5 mol%, CO2 6.2 mol%Red = H2S 76.2 mol%, CO2 5.1 mol%
Fig. 16—Sensitivity of H2S and CO2 concentration for APE of Well Y2 fluid.
00
2,000
4,000
6,000
8,000
10,000
12,000
50Temperature (°C)
Pre
ssu
re (
psi
a)
100
Soild Line = APEDashed Line = Psat Line
150 200
Green = Original
H2S 14.5 mol%,CO2 4.1 mol%
Blue= H2S 30.0 mol%, CO2 5.3 mol%
Purple= H2S 49.5 mol%, CO2 6.2mol%
Red= H2S 76.1 mol%, CO2 5.1mol%
Fig. 17—Sensitivity of H2S and CO2 concentration for APE of Well X3 fluid.
SPE_OGF_170585_160008.indd 9 25/08/16 7:51 PM
Pre
ssu
re (
psi
a)
10 Oil and Gas Facilities • October 2016
sium on Colloid Chemistry in Oil Production, Asphaltenes and Wax Deposition, Huatulco, Mexico, 14–17 November.
Edmonds, B., Moorwood, R. A. S., and Szczepanski, R. 1999c. A Unified Framework for Calculating Solid Deposition From Petroleum Fluids In-cluding Waxes, Asphaltenes, Hydrates and Scales. Fluid Phase Equilibria 158–160: 481–489. http://dx.doi.org/10.1016/S0378-3812(99)00138-7.
Feng, G. X. and Mather, A. E. 1992. Solubility of Hydrogen Sulfide in n-eicosane at Elevated Pressure. Journal of Chemical & Engineering Data 37 (4): 412–413. http://dx.doi.org/10.1021/je00008a009.
Feng, G. X. and Mather, A. E. 1993a. Solubility of H2S in n-do-decane. Fluid Phase Equilibria 87 (2): 341–346. http://dx.doi.org/10.1016/0378-3812(93)85035-K.
Feng, G. X. and Mather, A. E. 1993b. Solubility of H2S in n-hexadecane at Elevated Pressure. Can. J. Chem. Eng. 71 (2): 327–328. http://dx.doi.org/10.1002/cjce.5450710221.
Ghaderi, S. M., Keith, D. W., Lavoie, R. et al. 2011. Evolution of Hydrogen Sulfide in Sour Saline Aquifers During Carbon Dioxide Sequestra-tion. International Journal of Greenhouse Gas Control 5 (2): 347–355. http://dx.doi.org/10.1016/j.ijggc.2010.09.008.
Gonzalez, D. L., Mahmoodaghdam, E., Lim, F. H. et al. 2012. Effects of Gas Additions to Deepwater Gulf of Mexico Reservoir Oil: Experi-mental Investigation of Asphaltene Precipitation and Deposition. Pre-sented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, 8–10 October. SPE-159098-MS. http://dx.doi.org/10.2118/159098-MS.
Gonzalez, D. L., Vargas, F. M., Hirasaki, G. J. et al. 2008. Modeling Study of CO2-Induced Asphaltene Precipitation. Energy Fuels 22 (2): 757–762. http://dx.doi.org/10.1021/ef700369u.
Hayashi, Y. and Okabe, H. 2010. Experimental Investigation of Asphaltene Induced Permeability Reduction. Presented at the SPE EOR Con-ference at Oil & Gas West Asia, Muscat, Oman. 11–13 April. SPE-129271-MS. http://dx.doi.org/10.2118/129271-MS.
Hopper, B. L., Baldassarre, L., Detiveaux, I. et al. 2008. World’s First 10,000 psi Sour Gas Injection Compressor. Proc., Thirty-Seventh Tur-bomachinery Symposium, Houston, 8–11 September, 73–95.
Hu, Y.-F., Li, S., Liu, N. et al. 2004. Measurement and Corresponding States Modeling of Asphaltene Precipitation in Jilin Reservoir Oils. Journal of Petroleum Science and Engineering 41 (1–3): 169–182. http://dx.doi.org/10.1016/S0920-4105(03)00151-7.
Huang, S. S. S. and Robinson, D. B. 1984. Vapor-Liquid Equilibrium in Selected Aromatic Binary Systems: m-Xylene-Hydrogen Sulfide and Mesitylene-Hydrogen Sulfide. Fluid Phase Equilibria 17 (3): 373–382. http://dx.doi.org/10.1016/0378-3812(84)85006-2.
Huang, S. S. S. and Robinson, D. B. 1985. Equilibrium Phase Properties of the Ethylcyclohexane-Hydrogen Sulfide and n-Propylcyclohexane-Hydrogen Sulfide Binary Systems. J. Chem. Eng. Data 30 (2): 154–157. http://dx.doi.org/10.1021/je00040a006.
Knapp, H., Döring, R., Oellrich, L. et al. 1982. Vapor-Liquid Equilibria for Mixtures of Low Boiling Substances. In DECHEMA Chemistry Data Series, Vol. VI, Part 1: Binary Systems. Frankfurt, Germany: Scho-lium International.
Kohn, J. P. and Kurata, F. 1958. Heterogeneous Phase Equilibria of the Methane–Hydrogen Sulfide System. AIChE J. 4 (2): 211–217. http://dx.doi.org/10.1002/aic.690040217.
Laugier, S. and Richon, D. 1995. Vapor-Liquid Equilibria for Hydrogen Sulfide + Hexane, + Cyclohexane, + Benzene, + Pentadecane, and + (Hexane + Pentadecane). J. Chem. Eng. Data 40 (1): 153–159. http://dx.doi.org/10.1021/je00017a034.
Leu, A. D. and Robinson, D. B. 1989. Equilibrium Phase Properties of the n-Butane-Hydrogen Sulfide and Isobutane-Hydrogen Sulfide Bi-nary Systems. J. Chem. Eng. Data 34 (3): 315–319. http://dx.doi.org/10.1021/je00057a017.
Leu, A. D. and Robinson, D. B. 1992. High-Pressure Vapor-Liquid Equilib-rium Phase Properties of the Isopentane-Hydrogen Sulfide and Neo-pentane-Hydrogen Sulfide Binary Systems. J. Chem. Eng. Data 37 (1): 14–17. http://dx.doi.org/10.1021/jr00005a005.
Magoulas, K. and Tassios, D. 1990. Thermophysical Properties of n-Alkanes from C1 to C20 and Their Prediction for Higher Ones. Fluid Phase Equi-libria 56: 119–140. http://dx.doi.org/10.1016/0378-3812(90)85098-U.
ence, Kuala Lumpur, 19–21 July. SPE-143903-MS. http://dx.doi.org/10.2118/143903-MS.
Alian, S. S., Omar, A. A., Alta’ee, A. F. et al. 2011. Study of Asphaltene Precipitation Induced Formation Damage During CO2 Injection for a Malaysian Light Oil. International Journal of Chemical, Molecular, Nuclear, Materials and Metallurgical Engineering 5 (6): 482–486. http://www.waset.org/publications/3833.
Bachu, S. and Shaw, J. 2005. CO2 Storage in Oil and Gas Reservoirs in Western Canada: Effect of Aquifers, Potential for CO2-Flood En-hanced Oil Recovery and Practical Capacity. In Greenhouse Gas Con-trol Technologies 7, Proceedings of the 7th International Conference on Greenhouse Gas Control Technologies, Volume 1: Peer Reviewed Papers and Overviews, Vancouver, Canada, 5–9 September, 361–369. http://dx.doi.org/10.1016/N978-008044704-9/50037-9.
Bennion, D. B. and Bachu, S. 2008. Drainage and Imbibition Relative Per-meability Relationships for Supercritical CO2/Brine and H2S/Brine Systems in Intergranular Sandstone, Carbonate, Shale, and Anhydrite Rocks. SPE Res Eval & Eng 11 (3): 487–496. SPE-99326-PA. http://dx.doi.org/10.2118/99326-PA.
Broad, T. J., Rais, T. A. M., Hussain, A. et al. 2007. Deposition of “As-phaltenes” During CO2 Injection and Implications for EOS De-scription and Reservoir Performance. Presented at the International Petroleum Technology Conference, Dubai, 4-6 December. IPTC-11563-MS. http://dx.doi.org/10.2523/IPTC-11563-MS.
Carroll, J. J. and Mather, A. E. 1995. A Generalized Correlation for the Peng-Robinson Interaction Coefficients for Paraffin-Hydrogen Sul-fide Binary Systems. Fluid Phase Equilibria 105 (2): 221–228. http://dx.doi.org/10.1016/0378-3812(94)02612-5.
Choiri, M. and Hamouda, A. A. 2011. Study of CO2 Effect on Asphal-tene Precipitation and Compositional Simulation of Asphaltenic Oil Reservoir. Presented at the SPE International Symposium on Oil-field Chemistry, The Woodlands, Texas, USA, 11–13 April. SPE-141329-MS. http://dx.doi.org/10.2118/141329-MS.
Correra, S., De Simoni, M., Bartosek, M. et al. 2010. Assessment of Asphal-tene Deposition Risk in an EOR Intervention Through CO2 Injection. Presented at the SPE Annual Technical Conference and Exhibition, Florence, Italy, 19–22 September. SPE-135001-MS. http://dx.doi.org/10.2118/135001-MS.
Dahaghi, A. K., Gholami, V., Moghadasi, J. et al. 2008. Formation Damage Through Asphaltene Precipitation Resulting From CO2 Gas Injection in Iranian Carbonate Reservoirs. SPE Prod & Oper 23 (2): 210–214. SPE-99631-PA. http://dx.doi.org/10.2118/99631-PA.
Darmentaev, S., Yessaliyeva, A., Azhigaliyeva, A. et al. 2010. Tengiz Sour Gas Injection Project. Presented at the SPE Caspian Carbonates Tech-nology Conference, Atyrau, Kazakhstan, 8–10 November. SPE-139851-MS. http://dx.doi.org/10.2118/139851-MS.
Davis, P., Marriott, R., Fitzpatrick, E. et al. 2008. Fossil Fuel Development Requires Long-Term Sulfur Strategies. Oil & Gas Journal 106 (29). http://www.ogj.com/articles/print/volume-106/issue-29/drilling-pro-duction/fossil-fuel-development-requires-long-term-sulfur-strategies.html.
Deo, M. and Parra, M. 2012. Characterization of Carbon-Dioxide-Induced Asphaltene Precipitation. Energy Fuels 26 (5): 2672–2679. http://dx.doi.org/10.1021/ef201402v.
Doughty, C., Pruess, K., Benson, S. M. et al. 2001. Capacity Investigation of Brine-Bearing Sands of the Frio Formation for Geologic Sequestra-tion of CO2. Proc., First National Conference on Carbon Sequestra-tion, Washington, DC, 14–17 May.
Eakin, B. E. and De Vaney, W. E. 1974. Vapor-Liquid Equilibria in Hy-drogen-Hydrogen Sulfide—C9 Hydrocarbon Systems. AIChE Sympo-sium Series 70 (140): 80–90.
Edmonds, B., Moorwood, R. A. S., Szczepanski, R. et al. 1999a. Measure-ment and Prediction of Asphaltene Precipitation From Live Oils. Pre-sented at the Third International Symposium on Colloid Chemistry in Oil Production, Asphaltenes and Wax Deposition, Huatulco, Mexico, 14–17 November.
Edmonds, B., Moorwood, R. A. S., Szczepanski, R. et al. 1999b. A New Method To Give Quantitative Prediction of Asphaltene Deposition From Petroleum Fluids. Presented at the Third International Sympo-
SPE_OGF_170585_160008.indd 10 25/08/16 7:51 PM
October 2016 • Oil and Gas Facilities 11
Srivastava, R. K., Huang, S. S., Dyer, S. B. et al. 1995. Quantification of Asphaltene Flocculation During Miscible CO2 Flooding in the Wey-burn Reservoir. J Can Pet Technol 34 (8): 31–42. PETSOC-95-08-03. http://dx.doi.og/10.2118/95-08-03.
Srivastava, R. K., Huang, S. S., and Dong, M. 1999. Asphaltene Deposi-tion During CO2 Flooding. SPE Prod & Fac 14 (4): 235–245. SPE-59092-PA. http://dx.doi.org/10.2118/59092-PA.
Stamataki, S. and Magoulas. K. G. 2000. Prediction of Phase Equilibria and Volumetric Behavior of Fluids With High Concentration of Hydrogen Sulfide. Oil & Gas Science and Technology – Rev. IFP 55 (5): 511–522. http://dx.doi.org/10.2516/ogst:2000038.
Takabayashi, K., Maeda, H., Miyagawa, Y. et al. 2012. Do Asphaltene De-position Troubles Happen in Low Asphaltene Content of Crude Oil? Presented at the Abu Dhabi International Petroleum Conference and Exhibition, Abu Dhabi, 11–14 November. SPE-161489-MS. http://dx.doi.org/10.2118/161489-MS.
Takahashi, S., Hayashi, Y., Takahashi, S. et al. 2003. Characteristics and Impact of Asphaltene Precipitation During CO2 Injection in Sand-stone and Carbonate Cores: An Investigative Analysis Through Laboratory Tests and Compositional Simulation. Presented at the SPE International Improved Oil Recovery Conference in Asia Pa-cific, Kuala Lumpur, 20–21 October. SPE-84895-MS. http://dx.doi.org/10.2118/84895-MS.
Vafaie-Sefti, M., and Mousavi-Dehghani, S. A. 2006. Application of As-sociation Theory to the Prediction of Asphaltene Deposition: Deposi-tion Due To Natural Depletion and Miscible Gas Injection Processes in Petroleum Reservoirs. Fluid Phase Equilibria 247 (1–2): 182–189. http://dx.doi.org/10.1016/j.fluid.2006.06.019.
van Gisbergen, S. J. A., Osinga, V. P., Gritsenko, O. V. et al. 1996. Im-proved Density Functional Theory Results for Frequency-Dependent Polarizabilities, by the Use of an Exchange-Correlation Potential With Correct Asymptotic Behavior. Journal of Chemical Physics 105 (8): 3142–3151. http://dx.doi.org/10.1063/1.472182.
van Vark, W., Masalmeh, S. K., van Dorp, J. et al. 2004. Simulation Study of Miscible Gas Injection for Enhanced Oil Recovery in Low Perme-able Carbonate Reservoirs in Abu Dhabi. Presented at the Abu Dhabi International Conference and Exhibition, Abu Dhabi, 10–13 October. SPE-88717-MS. http://dx.doi.org/10.2118/88717-MS.
Verdier, S., Carrier, H., Andersen, S. I. et al. 2006. Study of Pressure and Temperature Effects on Asphaltene Stability in Presence of CO2. En-ergy Fuels 20 (4): 1584–1590. http://dx.doi.org/10.1021/ef050430g.
Wichert, E. and Royan, T. 1997. Technology Acid Gas Injection Eliminates Sulfur Recovery Expense. Oil & Gas Journal 95 (17): 67. http://www.ogj.com/articles/print/volume-95/issue-17/in-this-issue/production/technology-acid-gas-injection-eliminates-sulfur-recovery-expense.html.
Xiao, Y., Xu, T., and Pruess, K. 2009. The Effects of Gas-Fluid-Rock Inter-actions on CO2 Injection and Storage: Insights From Reactive Trans-port Modeling. Energy Procedia 1 (1): 1783–1790. http://dx.doi.org/10.1016/j.egypro.2009.01.233.
Ying, J., Sun, L., Sun, L. et al. 2006. The Research on Asphaltene Deposi-tion Mechanism and Its Influence on Development During CO2 In-jection. Presented at the SPE International Oil & Gas Conference and Exhibition, Beijing, 5–7 December. SPE-104417-MS. http://dx.doi.org/10.2118/104417-MS.
Zanganeh, P., Ayatollahi, S., Alamdari, A. et al. 2012. Asphaltene Deposi-tion During CO2 Injection and Pressure Depletion: A Visual Study. En-ergy Fuels 26 (2): 1412–1419. http://dx.doi.org/10.1021/ef2012744.
Zekri, A. Y., Shedid, S. A., and Almehaideb, R. A. 2009. Sulfur and As-phaltene Deposition During CO2 Flooding of Carbonate Reservoirs. Presented at the SPE Middle East Oil & Gas Show and Conference, Manama, Bahrain, 15–18 March. SPE-118825-MS. http://dx.doi.org/10.2118/118825-MS.
Appendix A—Citation Results of Asphaltene-Flow-Assurance Research During CO2 InjectionFrom a recent citation, many references that caution about risks of formation damage resulting from CO2-induced asphaltene precipi-tation were found. The literature survey results are summarized in Table A-1.
Memon, A., Qassim, B., Al-Ajmi, M. et al. 2012. Miscible Gas Injection and Asphaltene Flow Assurance Fluid Characterization: A Laboratory Case Study for Black Oil Reservoir. Presented at the SPE EOR Con-ference at Oil and Gas West Asia, Muscat, Oman, 16–18 April. SPE-150938-MS. http://dx.doi.org/10.2118/150938-MS.
Moysan, J. M., Paradowski, H., and Vidal, J. 1986. Prediction of Phase Be-haviour of Gas-Containing Systems with Cubic Equations of State. Chemical Engineering Science 41 (8): 2069–2074. http://dx.doi.org/10.1016/0009-2509(86)87123-8.
Ng, H. J., Kalra, H., Robinson, D. B. et al. 1980. Equilibrium Phase Prop-erties of the Toluene-Hydrogen Sulfide and Heptane-Hydrogen Sul-fide Binary Systems. J. Chem. Eng. Data 25 (1): 51–55. http://dx.doi.org/10.1021/je60084a020.
Nghiem, L. X., Kohse, B. F., Farouq Ali, S. M. et al. 2000. Asphaltene Precipitation: Phase Behaviour Modelling and Compositional Simu-lation. Presented at the SPE Asia Pacific Conference on Integrated Modelling for Asset Management, Yokohama, Japan, 25–26 April. SPE-59432-MS. http://dx.doi.org/10.2118/59432-MS.
Novosad, Z. and Costain, T. G. 1990. Experimental and Modeling Studies of Asphaltene Equilibria for a Reservoir Under CO2 Injection. Pre-sented at the SPE Annual Technical Conference and Exhibition, New Orleans, 23–26 September. SPE-20530-MS. http://dx.doi.org/10.2118/20530-MS.
Okabe, H., Takahashi, S., and Mitsuishi, H. 2010. Distribution of Asphal-tene Deposition in the Rock Samples by Gas Injection. Presented at the Abu Dhabi International Petroleum Exhibition and Confer-ence, Abu Dhabi, 1–4 November. SPE-138765-MS. http://dx.doi.org/10.2118/138765-MS.
Okwen, R. T. 2006. Formation Damage by CO2-Induced Asphaltene Pre-cipitation. Presented at the SPE International Symposium and Exhibi-tion on Formation Damage Control, Lafayette, Louisiana, USA, 15–17 February. SPE-98180-MS. http://dx.doi.org/10.2118/98180-MS.
Parra-Ramirez, M., Peterson, B., and Deo, M. D. 2001. Comparison of First and Multiple Contact Carbon Dioxide Induced Asphaltene Pre-cipitation. Presented at the SPE International Symposium on Oilfield Chemistry, Houston, 13–16 February. SPE-65019-MS. http://dx.doi.org/10.2118/65019-MS.
Pedersen, K. S., Shaikh, J. A., and Rasmussen, C. P. 2012. 1D Asphal-tene Deposition Simulator for Reservoir Fluids Undergoing Gas In-jection. Presented at the SPE EOR Conference at Oil and Gas West Asia, Muscat, Oman, 16–18 April. SPE-155447-MS. http://dx.doi.org/10.2118/155447-MS.
Reamer, H. H., Sage, B. H., and Lacey, W. N. 1951. Phase Equilibria in Hy-drocarbon Systems—Volumetric and Phase Behavior of the Methane-Hydrogen Sulfide System. Ind. Eng. Chem. 43 (4): 976–981. http://dx.doi.org/10.1021/ie50496a052.
Richon, D., Laugier, S., and Renon, H. 1992. High-Pressure Vapor-Liquid Equilibrium Data for Binary Mixtures Containing Molecular Ni-trogen, Carbon Dioxide, Hydrogen Sulfide and an Aromatic Hy-drocarbon or Propylcyclohexane in the Range 313-473 K. J. Chem. Eng. Data 37 (2): 264–268. http://dx.doi.org/10.1021/je00006a035.
Robinson, D. B. and Bailey, J. A. 1957. The Carbon Dioxide-Hydrogen Sulfide-Methane System. Part I. Phase Behaviour at 100°F. Can. J. Chem. Eng. 35: 151–158.
Robinson, D. B., Lorenzo, A. P., and Macrygeorgos, C. A. 1959. The Carbon Dioxide-Hydrogen Sulfide-Methane System: Part II. Phase Behavior at 40°F and 160°F. Can. J. Chem Eng. 37 (6): 212–217. http://dx.doi.org/10.1002/cjce.5450370603.
Sayegh, S., Huang, S., Zhang, Y. P. et al. 2006. Effect of H2S and Pres-sure Depletion on the CO2 MMP of Zama Oils. Presented at the Ca-nadian International Petroleum Conference, Calgary, 13–15 June. PETSOC-2006-030. http://dx.doi.org/10.2118/2006-030.
Soave, G. 1972. Equilibrium Constants From a Modified Redlich-Kwong Equation of State. Chem. Eng. Sci. 27 (6): 1197–1203. http://dx.doi.org/10.1016/0009-2509(72)80096-4.
Srivastava, R. K., Huang, S. S., Dyer, S. B. et al. 1995. Quantification of Asphaltene Flocculation During Miscible CO2 Flooding in the Wey-burn Reservoir. J Can Pet Technol 34 (8): 31–42. PETSOC-95-08-03. http://dx.doi.org/10.2118/95-08-03.
SPE_OGF_170585_160008.indd 11 25/08/16 7:51 PM
12 Oil and Gas Facilities • October 2016
Cat
egor
y S
ourc
e Fi
eld
Sam
ple
Asp
hate
ne
Con
tent
(%)
AP
I G
ravi
ty
(°A
PI)
Pre
ssur
e (p
sia)
Te
mpe
ratu
re
(°C
)
Oil
Vis
cosi
ty
(cp)
D
ata
Ref
erre
d Fi
ndin
gs
M
Ngh
iem
et a
l. (2
000)
W
eybu
rn
(Can
ada)
–
≈4.0
–5.0
≈2
9–36
2,
321
59
– S
rivas
tava
et a
l. (1
995,
199
7)
The
impo
rtant
phe
nom
ena
asso
ciat
ed w
ith th
e dy
nam
ics
of a
spha
ltene
pre
cipi
tatio
n ar
e pr
oper
ly re
prod
uced
. M
Yi
ng e
t al.
(200
6)
Chi
na fi
elds
—
1.
49–1
.8
– 1,
537
(1,2
33–
5,07
6)*
– –
– D
ynam
ic c
hara
cter
istic
ana
lyse
s sh
ow th
at h
ighe
r ac
cum
ulat
ion
of a
spha
ltene
dep
osits
is m
ainl
y at
th
e in
let o
f the
cor
e.
– O
il 1
16.8
19
2,
176*
10
0 –
– W
eybu
rn
(Can
ada)
O
il 2
4.9
29.5
2,
321*
10
0*
– S
rivas
tava
et a
l. (1
995)
Ira
nian
fiel
d O
il 3
– –
Up
to
5,50
0*
97*
– –
M
Vaf
aie-
Sef
ti an
d M
ousa
vi-
Deh
ghan
i (20
06)
Irani
an fi
eld
Oil
4 –
– U
p to
5,
000*
97
* –
–
The
prop
osed
mod
el c
an p
redi
ct a
spha
ltene
de
posi
tion
in te
rms
of C
O2 f
lood
. As
the
amou
nt o
f re
sin
incr
ease
s, m
ore
CO
2 is
need
ed fo
r the
sam
e am
ount
of a
spha
ltene
dep
ositi
on.
– Li
ve o
il 1.
4 32
3,
256
147
– M
G
onza
lez
et a
l. (2
008)
So
uth
Amer
ican
S
TO
9 27
3,
256
147
– –
CO
2 ca
n in
crea
se o
r dec
reas
e th
e st
abilit
y of
as
phal
tene
s de
pend
ing
on th
e ch
arac
teris
tics
of th
e flu
id a
nd te
mpe
ratu
re.
M
Cor
rera
et a
l. (2
010)
Fi
eld
in
sout
heas
t Sic
ily
– –
10
4,73
5 10
2 40
(U),
300
(L)
– N
o si
gnifi
cant
risk
of a
spha
ltene
dep
ositi
on in
CO
2in
ject
ion.
–
Oil
1 16
.8
19
2,39
3*
20–1
90*
– M
C
hoiri
and
H
amou
da
(201
1)
Jilin
(Chi
na)
Oil
J2
1.82
–
2,90
1*
20–1
50*
– H
u et
al.
(200
4)
A m
odel
bas
ed o
n Fl
ory-
Hug
gins
pol
ymer
sol
utio
n an
d H
ildeb
rand
sol
ubilit
y co
ncep
t was
app
lied
for
desi
gnin
g EO
R b
y C
O2 i
njec
tion.
LB
_PC
M
N
ovos
ad a
nd
Cos
tain
(199
0)
Mid
ale
(Can
ada)
–
5.4
29.0
2,
320
65
– –
CO
2 gen
eral
ly d
esta
biliz
ed a
spha
ltene
.
Oil
A1
5.3
29.6
2,
321
59
4.2
Oil
A2
4.9
29.5
2,
321
59
4.6
LB_P
C
Sriv
asta
va e
t al.
(199
5)
Wey
burn
(C
anad
a)
Oil
A3
5.0
28.6
2,
321
59
4.6
– A
spha
ltene
floc
cula
tion
beca
use
of C
O2 w
as
inse
nsiti
ve in
the
oper
atin
g pr
essu
re. C
O2
conc
entra
tion
was
impo
rtant
.
Oil
W1
4.8
29
2,32
1 59
4.
2 O
il W
2 4.
0 36
2,
321
61
2.35
LB
_PC
LB
_CF
Sriv
asta
va e
t al.
(199
7, 1
999)
W
ybur
n (C
anad
a)
Oil
W3
4.9
31
2,32
1 63
3.
15
–
The
mai
n fa
ctor
on
asph
alte
ne p
reci
pita
tion
depe
nds
on C
O2 c
once
ntra
tion.
Vug
gy m
atrix
sho
ws
the
high
est a
spha
tlene
pre
cipi
tatio
n in
the
core
dur
ing
CO
2 flo
od.
LB_P
C
Parra
-Ram
irez
et a
l. (2
001)
R
angl
ey
(Col
orad
o)
– 2.
83
34
– 71
–
– M
ultip
le c
onta
cts
incr
ease
pre
cipi
tate
s fo
r the
sam
e C
O2 c
once
ntra
tion.
LB
_PC
LB
_CF
M
Taka
hash
i et a
l. (2
003)
M
iddl
e E
ast f
ield
–
3.9
– 3,
515
100
– –
Pre
cipi
tate
d as
phal
tene
incr
ease
s as
CO
2 inc
reas
es.
Mor
e de
posi
ts in
car
bona
te c
ore
than
san
dsto
ne
core
.
Oil
J1
1.68
–
2,90
1;
3,48
1;
4,20
6*
66
5.8
LB_P
C
M
Hu
et a
l. (2
004)
Ji
lin (C
hina
)
Oil
J2
1.82
–
2,90
1*
66
6.1
–
No
prec
ipita
tion
durin
g na
tura
l dep
letio
n, In
the
syst
em o
f CO
2 inj
ecte
d flu
id, p
reci
pita
tion
was
ob
serv
ed u
nder
ope
ratin
g co
nditi
on.
Sou
th A
mer
ican
S
A c
rude
9
27
8,70
2*
20–1
10*
– LB
_PC
M
V
erdi
er e
t al.
(200
6)
Mid
dle
Eas
tern
M
E c
rude
9
29
8,70
2*
20–1
10*
– –
Less
CO
2 is
requ
ired
to p
reci
pita
te a
spha
ltene
at t
he
high
er te
mpe
ratu
re.
LB_P
C =
labo
rato
ry e
xper
imen
t (P
VT
cell
test
). LB
_CF
= la
bora
tory
exp
erim
ent (
core
flood
test
). M
= m
odel
ing.
D
S =
dep
ositi
on s
imul
atio
n.*M
odel
ing/
expe
rimen
tal c
ondi
tions
in c
ase
of th
e ab
senc
e of
rese
rvoi
r pre
ssur
e an
d te
mpe
ratu
re.
Tabl
e A
-1—
Cita
tion
resu
lts o
f asp
halte
ne fl
ow-a
ssur
ance
rese
arch
dur
ing
CO
2 in
ject
ion.
SPE_OGF_170585_160008.indd 12 25/08/16 7:51 PM
October 2016 • Oil and Gas Facilities 13
Cat
egor
y S
ourc
e Fi
eld
Sam
ple
Asph
aten
e C
onte
nt (%
)
AP
I G
ravi
ty
(°A
PI)
Pre
ssur
e (p
sia)
Te
mpe
ratu
re
(°C
)
Oil
Vis
cosi
ty
(cp)
D
ata
Ref
erre
d Fi
ndin
gs
6.3 ;9.51 –
3.92 28.4
77H
SA
–LB
_CF
Okw
en (2
006)
–
C1/n
C10
0.
19
47.6
–
25; 6
5*
2.0;
0.9
–
Low
er o
il re
cove
ry o
btai
ned
from
AS
H77
(with
hig
her
asph
alte
ne c
onte
nt) b
ecau
se o
f res
ervo
ir po
re
bloc
kage
dur
ing
CO
2 flo
od.
LB_P
C
M
Bro
ad (2
007)
D
ukha
n fie
ld
(Qat
ar)
– –
35–4
0 –
– –
– A
spha
ltene
pre
cipi
tatio
n oc
curs
dur
ing
CO
2 inj
ectio
n an
d th
e pr
oper
ties
of th
e he
avy
end
fract
ion
are
sign
ifica
ntly
redu
ced.
LB
_CF
Dah
aghi
et a
l. (2
008)
Ku
pal (
Iran)
–
0.66
34
.3
6,00
0 71
–
– A
spha
ltene
pre
cipi
tatio
n de
pend
s on
CO
2 co
ncen
tratio
n an
d ra
pidl
y in
crea
ses
whe
n ex
ceed
ing
criti
cal v
alue
. Per
mea
bilit
y re
duct
ion
is
mai
nly
caus
ed b
y m
echa
nica
l ent
rapm
ent.
LB_C
F Ze
kri e
t al.
(200
9)
Fiel
d in
UA
E
Wel
l-A
fluid
0.
2 31
.7
4,00
0 12
1 5.
5 (a
t 40°
C)
– As
phal
tene
pre
cipi
tate
s by
CO
2 inj
ectio
n. L
ower
pe
rmea
bilit
y co
re s
how
s lo
wer
asp
halte
ne lo
ss a
nd
high
er o
il re
cove
ry th
an h
ighe
r per
mea
bilit
y co
re.
LB_P
C
LB_C
F H
ayas
hi a
nd
Oka
be (2
010)
–
– 0.
97
– 4,
154
129
– –
Per
mea
bilit
y re
duct
ion
by C
O2 i
njec
tion
is
appr
oxim
atel
y 20
% b
ecau
se o
f a la
rge
depo
sitio
n at
th
e up
stre
am o
f the
cor
e.
LB_C
F M
O
kabe
et a
l. (2
010)
M
iddl
e E
aste
rn
field
–
3.9
– 1,
715–
5,01
5*
70
– –
Asp
halte
ne is
sel
ectiv
ely
depo
site
d ne
ar th
e in
let o
f th
e co
re fo
r CO
2 inj
ectio
n.
LB_P
C
LB_C
F M
Ady
ani e
t al.
(201
1)
Fiel
d in
Sou
th
Chi
na S
ea
– 0.
07
41.4
2,
915
95.6
0.
26
– Th
ere
is a
pot
entia
l for
asp
halte
ne p
lugg
ing
whe
n a
high
con
cent
ratio
n of
CO
2 gas
is in
ject
ed.
LB_C
F Al
ian
et a
l. (2
011)
M
alay
sian
fiel
d –
0.42
41
(2
,015
; 2,
315;
2,
615)
*
98*
0.8
– Th
ree
core
flood
test
s sh
owed
16
to 2
2%
perm
eabi
lity
redu
ctio
n w
ith 6
to 1
9% p
oros
ity
redu
ctio
n be
caus
e of
asp
halte
ne d
epos
its.
LB_P
C
M
Mem
on e
t al.
(201
2)
Fiel
d in
Kuw
ait
– –
32.4
–
90
– –
Inje
ctio
n ga
s (C
O2/N
GL
mix
ture
) is
not c
ompa
tible
w
ith th
e or
igin
al re
serv
oir o
il.
Flui
d I
15.5
22
.5
8,02
9 97
.8
7.71
LB
_PC
G
onza
lez
et a
l. (2
012)
Fi
eld
in G
OM
Fl
uid
II 4.
0 28
.4
9,34
2 97
.8
1.05
–
The
effe
cts
of a
dded
gas
es o
n re
duci
ng a
spha
ltene
st
abili
ty in
the
live
oil a
re o
n th
e or
der o
f N
2>C
H4>
CO
2. Fi
eld
in S
outh
C
hina
Sea
Fl
uid
I 0.
10
41.4
2,
901
95.6
≈1
.0
Ady
ani e
t al.
(201
1)
DS
Pe
ders
en e
t al.
(201
2)
Jilin
(Chi
na)
Flui
d II
4.04
–
– 65
.9
– H
u et
al.
(200
4)
1D s
imul
atio
n of
CO
2 inj
ectio
n ca
uses
asp
halte
ne
prec
ipita
tes
in th
e tra
nsiti
on z
one
betw
een
gas
and
oil.
Thes
e de
posi
ts p
lug
the
rese
rvoi
r and
dec
reas
e pe
rmea
bilit
y.
– 8.21
7.41 –
dnoM-e-hu
KLB
_PC
Za
ngan
eh e
t al.
(201
2)
Gac
hsar
an
– 5.
3 31
(4
35;
870;
1,
450;
2,
031)
*
(35;
90)
* –
– C
O2 i
njec
tion
incr
ease
s as
phal
tene
dep
ositi
on in
all
pres
sure
rang
es.
LB_P
C
Deo
and
Par
ra
(201
2)
– –
2.8
34
3,00
0*
71*
– –
CO
2-in
duce
d pr
ecip
itate
s w
ould
be
enha
nced
be
caus
e of
the
mul
tiple
con
tact
s.
LB_P
C
LB_C
F Ta
kaba
yash
iet
al.
(201
2)O
ffsho
re fi
eld
in
UA
E
– –
– –
– –
– E
ven
thou
gh a
spha
ltene
floc
cula
tion
is o
bser
ved
in
the
PV
T ce
ll, n
o se
vere
per
mea
bilit
y re
duct
ion
occu
rs d
urin
g th
e C
O2 c
oref
lood
test
s be
caus
e of
lo
w a
spha
ltene
con
tent
. LB
_PC
M
Ab
dalla
h (2
012)
Fi
eld-
A in
UA
E
Res
ervo
ir-X
cru
de
0.25
–
– ≈1
82*
– –
AOP
is d
etec
ted
by m
ixin
g w
ith C
O2,
even
thou
gh it
is
not
det
ecte
d at
the
natu
ral d
eple
tion.
Pha
se
enve
lope
is p
redi
cted
to in
crea
se a
spha
ltene
un
stab
le a
rea
as C
O2 i
ncre
ases
. LB
_PC
= la
bora
tory
exp
erim
ent (
PV
T ce
ll te
st).
LB_C
F =
labo
rato
ry e
xper
imen
t (co
reflo
od te
st).
M =
mod
elin
g.
DS
= d
epos
ition
sim
ulat
ion.
*M
odel
ing/
expe
rimen
tal c
ondi
tions
in c
ase
of th
e ab
senc
e of
rese
rvoi
r pre
ssur
e an
d te
mpe
ratu
re.
Tabl
e A
-1 (c
ontin
ued)
—C
itatio
n re
sults
of a
spha
ltene
flow
-ass
uran
ce re
sear
ch d
urin
g C
O2
inje
ctio
n.
SPE_OGF_170585_160008.indd 13 25/08/16 7:52 PM
14 Oil and Gas Facilities • October 2016
Appendix B—Stamataki and Magoulas (2000) Correlation Between kij and ω in H2S/Hydrocarbon SystemStamataki and Magoulas (2000) reviewed literature and collected all the available experimental VLE data of H2S/hydrocarbon binary systems. Reference lists for available data include Kohn and Kurata (1958), Reamer et al. (1951), Robinson and Bailey (1957), Rob-inson et al. (1959), Knapp et al. (1982), Leu and Robinson (1989, 1992), Laugier and Richon (1995), Ng et al. (1980), Eakin and De Vaney (1974), Feng and Mather (1992, 1993a, 1993b), Huang and Robinson (1984, 1985), and Richon et al. (1992). As of their study in 2000, the heaviest hydrocarbon in the system was up to C20. The interaction parameter for each binary pair was established from all available VLE data. Acentric factors were obtained from Magoulas and Tassios (1990) for n-alkanes that have a carbon number below 20, and were predicted for larger n-alkanes. The correlation anal-ysis was conducted by minimizing the average absolute deviation in bubblepoint pressure by use of the modified Peng-Robinson EOS (Magoulas and Tassios 1990). The kij values of larger n-alkanes—H2S with nC26, nC32, and nC40—were obtained with the Gibbs
energy model. The derived correlation also showed a good agree-ment for these larger n-alkanes. The correlation was compared with those of Moysan et al. (1986) and Carroll and Mather (1995). As shown in Fig. B-1, the derived correlation is close to that of Carroll and Mather (1995).
Finally, through the regression analysis, the optimum kij values for H2S/n-alkanes yielded a good expression of Eq. 1 for kij as a function of ω.
SI Metric Conversion Factor °F (°F−32)/1.8 = °C psi × 6.894757 E+00 = kPa
Hideharu Yonebayashi is currently working for INPEX Corporation. He has more than 20 years of experience in the oil industry, and his ex-pertise includes microbial EOR, gas injection, steam-assisted gravity drainage, flow-assurance evaluation, and supervising field operations (acidizing, asphaltene removal, well test, production-logging tool). Yo-nebayashi’s main interests are EOR and related technologies. He deep-ened his knowledge of offshore carbonate field development during his secondment to Bunduq Co. Ltd. (1998–2001) and ExxonMobil Dev. Co. (2009–2014). Yonebayashi holds BS, MS, and PhD degrees from To-hoku University, Japan, all in petroleum engineering.
Katsumo Takabayashi is currently working for INPEX Corporation. He has more than 20 years of experience in reservoir engineering, espe-cially in CO2 flooding and in-situ combustion for conventional EOR. Tak-abayashi holds a BS degree in mining and reservoir engineering from Akita University, Japan.
Ryo Iizuka is currently working for INPEX Corporation. His work experience includes reservoir simulation, well testing, and gas EOR. Iizuka holds a BS degree in environmental and resources engineering from Waseda University in Japan.
Slavko Tosic is currently working for ExxonMobil. His work experience and research areas include reservoir simulation, well testing, multiphase metering, advanced production and reservoir analysis, fluid behavior, and field development. Tosic holds BS and MS degrees, both in petro-leum engineering, from the University of Belgrade and Texas A&M Uni-versity, respectively.
0 0.2 0.4 0.6 0.8
Moysan et al. (1986)Optimum
Carroll and Mather (1995)Stamataki and Magoulas (2000)
1 1.2 1.4 1.6 1.8–0.20
–0.15
–0.10
–0.05
0.00
0.05
0.10
0.15
Acentric Factor, ω
Inte
ract
ion
Co
effi
cien
t, k i
j
Fig. B-1—Correlated and predicted interaction parameters for the H2S/hydrocarbon binary systems (Stamataki and Magoulas 2000).
SPE_OGF_170585_160008.indd 14 25/08/16 7:52 PM
top related