The impact of connate water saturation and salinity on oil recovery and CO2 storage capacity during carbonated water injection in carbonate rock

Mahmood Shakiba ,Masoud Riazi ,Shahab Ayatollahi ,Mostafa Takband

1 Department of Petroleum Engineering,Amirkabir University of Technology,Tehran,Iran

2 Enhanced Oil Recovery (EOR) Research Centre,IOR/EOR Research Institute,Shiraz University,Shiraz,Iran

3 Petroleum Engineering Department,School of Chemical and Petroleum Eng.,Shiraz University,Shiraz,Iran

4 School of Chemical and Petroleum Engineering,Sharif University of Technology,Tehran,Iran

ABSTRACT Carbonated water injection (CWI) is known as an efficient technique for both CO2 storage and enhanced oil recovery(EOR).During CWI process,CO2 moves from the water phase into the oil phase and results in oil swelling.This mechanism is considered as a reason for EOR.Viscous fingering leading to early breakthrough and leaving a large proportion of reservoir un-swept is known as an unfavorable phenomenon during flooding trials.Generally,instability at the interface due to disturbances in porous medium promotes viscous fingering phenomenon.Connate water makes viscous fingers longer and more irregular consisting of large number of tributaries leading to the ultimate oil recovery reduction.Therefore,higher in-situ water content can worsen this condition.Besides,this water can play as a barrier between oil and gas phases and adversely affect the gas diffusion,which results in EOR reduction.On the other hand,from gas storage point of view,it should be noted that CO2 solubility is not the same in the water and oil phases.In this study for a specified water salinity,the effects of different connate water saturations(Swc) on the ultimate oil recovery and CO2 storage capacity during secondary CWI are being presented using carbonate rock samples from one of Iranian carbonate oil reservoir.The results showed higher oil recovery and CO2 storage in the case of lower connate water saturation,as 14% reduction of Swc resulted in 20% and 16% higher oil recovery and CO2 storage capacity,respectively.

Keywords:Carbonated water Connate water Carbonate reservoir Enhanced oil recovery Secondary recovery CO2 storage

1.Introduction

Enhanced oil recovery (EOR) technique is known as an attractive method to be responsible for more oil recovery in low oil price era.Among EOR methods,direct gas injection especially CO2injection is a well-established method in the oil industry.Besides the effectiveness of this method,it has some disadvantages such as[1-4]:(1) Premature breakthrough (BT) caused by gas overriding,(2) low vertical sweep efficiency because of large density contrast between CO2and resident fluids in reservoir,(3) low areal sweep efficiency due to viscous fingering phenomenon,and (4) high risk of gas leakage because of high mobility of CO2in the reservoir.The other CO2injection strategy e.g.water alternating gas (WAG)has been recommended to alleviate these problems.Although WAG can lessen drawbacks associated with direct gas injection,it has also some deficiencies.In WAG process,CO2can diffuse into the oil phase and results in more oil swelling and subsequently more oil recovery.WAG efficiency decreases if diffusion process is not completely accomplished.Water located between the oil and gas phases can be as an obstacle to gas transmission process,thus gas cannot be easily diffused [2,3].Hence,it is necessary to use another effective alternative strategy.Carbonated water injection(CWI)may be a suitable alternating technique,which can mitigate mentioned shortcomings somehow.At reservoir conditions,CO2-saturated water (i.e.carbonated water) can remain single phase.Besides,it is more effective than water flooding (WF) and gas injection individually[1-6].CWI can be effective in two ways:EOR process and CO2sequestration that have been investigated by Shakiba et al.both in secondary and tertiary modes [4].During CWI,CO2can transfer from water into the oil phase and subsequently results in oil swelling.It should be noted that unsaturated intrinsic water existing between initial oil in place and CW(i.e.CO2enriched water)could be a barrier in path of gas transition so-called water shielding or water blocking effect influencing the CWI performance.As a result,it can reduce ultimate oil recovery[2,3,5,6].Moreover,high salinity of connate water decreases the CO2solubility in the water phase [7,8].

According to annual reports of United States Environmental Protection Agency (USEPA),among greenhouse gases,CO2has the most proportion in increasing the earth temperature [9].Nowadays,using a secure method in order to sequestrate CO2underground would be favorite.CWI is a safe technique to capture the CO2underground because in this method,CO2can be soluble both in water and oil rather than as a free phase [7,8].The main mechanisms taking place during CWI can be proposed as follows [4,10,and 11]:(1) Oil swelling due to CO2mass transmission,(2) wettability alteration,(3) reduction of oil viscosity as a result of CO2transfer from aqueous phase to the oil phase,(4) decreasing the Interfacial tension (IFT),(5) reconnecting and mobilization of the trapped oil ganglia,and (6) Local flow diversion.

During EOR techniques in the reservoirs,when a low viscous fluid (e.g.water) displaces a high viscous fluid (e.g.oil),the shape of the displacement front would be changed and became unstable and eventually leads to viscous fingering phenomenon.An increase of instability at the interface due to disturbances in the porous medium promotes fingers.Wider fingers can sweep a large area of reservoir,thus it would improve ultimate oil recovery.On the other hand,thinner fingers hinder areal sweep efficiency through premature breakthrough,leaving a large proportion of the reservoir unswept.Connate water makes viscous fingers longer and more erratic with numerous branches[12].Besides,connate water can play as barrier between oil and gas phases;hence,it decreases oil recovery efficiency during flooding process [2-6].

There are many research works in the literature reported the effects of connate water saturation on the oil recovery efficiency for different EOR methods.Dickey and Bossler investigated the role of connate water saturation on oil recovery during secondary mode.They reported that the presence of connate water could help residual oil to be recovered by gas drive [13].Kelley and Caudle accomplished some waterflooding tests using a high viscosity waterflooding.They indicated that connate water forms a bank of low viscosity water and this decreases the oil recovery compared to a case without water bank.Thus,the connate water saturation should be considered in viscous water flooding [14].Wylie and Mohanty surveyed the impact of connate water content on oil recovery during near-miscible gas injection and showed that high water saturations could adversely affect the ultimate oil recovery.At this condition,the oil ganglions will be isolated by water and as a result,the mass transfer rate between the CO2and oil will be diminished [15].Thibodeau et al.investigated the impact of connate water on oil recovery with directing towards the effects of connate water on viscous fingering patterns.They reported at low injection rates,the presence of connate water makes viscous fingers longer and thinner dividing into branches and finally it has a detrimental effect on improvement of oil recovery.These irregularities of viscous fingering patterns are dependent on the amount of connate water[12].The effect of connate water on alkaline flooding and vapor extraction (VAPEX) processes have been investigated by Thibodeau et al.and Etminan et al.,respectively[16,17].Mohammadi et al.investigated the effect of shale geometry and connate water saturation on the performance of polymer flooding in heavy oil reservoirs using an oil-wet micromodel.They observed that the presence of connate water could result in additional disturbances to the displacement interface and as a result,it can influence the fingering patterns by making more irregularity and subsequently decreases the ultimate oil recovery.The results of their simulation indicated that higher connate water saturations would intensify this effect [18].Fodicsh et al.used micromodel to investigate the effect of connate water displacement on chemical EOR processes.They suggested that the process of connate water replacement is miscible.In addition,they showed that when the injected aqueous phase contacted completely,the connate water is rapidly displaced [19].The research works on the brine salinity for enhanced oil recovery,known as low salinity water injection,have been well studied in the literature.Zhang and Sarma investigated the impact of brine salinity on CO2injection in both continuous CO2injection and CWI modes.They observed more oil recovery at low brine salinity,as 10% incremental oil recovery was as a result of a reduction of 150000 mg·kg-1in brine salinity[20].The effect of connate water composition on performance of low salinity water flooding in sand stone reservoirs has been investigated by Shehata et al.They used two cores(Buff Berea and Bandera sand stone) in order to study the role of cation composition,temperature and pore throat distribution on performance of low salinity water flooding.They performed different spontaneous imbibition and core flood experiments.They indicated that the composition of connate water has major effect on oil recovery rate.For instance,connate water containing divalent cations (Mg2+,Ca2+) leads to higher oil recovery compared to connate water containing only monovalent cations (Na+) [21].As the CWI is being increasingly considered for both EOR and CO2storage,it would be interesting to see the effects of different reservoir parameters including water content saturation and salinity on its efficiency.Generally,pore size and texture of reservoir rock can influence the amount of connate water saturation[22].As discussed,the previous studies have mainly focused on the impact of connate water on direct gas injection and chemical EOR processes,however,to the best of our knowledge,the effect of intrinsic water saturation on the performance of CWI has not been yet studied.

In this work,the effect of different connate water saturation conditions on CWI performance as the secondary mode in core samples of an Iranian carbonate oil reservoir has been investigated using heavy crude oil and the results are being presented here.Besides,the CO2storage capacity has been surveyed under different connate water saturations and analyzed using a dimensionless storage factor(Fsd)for different conditions performed in this experimental work.

2.Experimental Fluids

2.1.Saturation brine (brine #1)

At the first stage during this experimental work,the core samples were flooded using brine #1 in order to obtain fully brinesaturated rock.The brine composition (known as #1) is given in Table 1.To prevent clay swelling,which can influence the rock physical properties (such as rock permeability) and subsequently the experiment results,KCl was added into the brine [23].

Table 2 Crude oil properties at 40°C

Table 3 The composition of the carbonated water brine

2.2.The crude oil

Brine saturated samples were then inundated using crude oil provided from one of the Iranian carbonate oil field in order to achieve the initial reservoir conditions(i.e.initial oil and water saturation).The properties of crude oil sample are given in Table 2.

2.3.Carbonated brine (brine #2)

CW flooding process for all cores has been performed using a lower salinity brine (i.e.brine #2) compared to saturation brine(i.e.brine #1) in order to reduce possible corrosion in the experimental equipment.Besides,the lower salinity would increase the capacity of water to dissolve more CO2compared to connate water in the core samples [7,8].The composition of brine #2 is given in Table 3.

2.4.Carbonated water

CW utilized in this study has been prepared by mixing brine#2 and CO2with purity of 99.9%.Brine and CO2were fully intermingled using a recombination cell apparatus until reaching the equilibrium condition at pressure and temperature of 13789.52 kPa and 40°C,respectively.The solubility of CO2in distilled water was investigated by Wiebe and Chang et al.[24,25].They indicated that CO2solubility considerably increases as pressure rises up to 13789.52 kPa,however,above this pressure the ascendant trend of CO2solubility becomes moderate.Thus,pressure of 13789.52 kPa was considered as the optimum pressure condition in preparing of the CW for the experiments performed in this study.Any impurity in water can affect its capacity for gas solubility as a solvent of gases.It has been shown that salt composition and brine salinity influence the CO2solubility in brine.That is,higher salinity would decrease gas solubility in water.Duan and Sun and Duan et al.investigated CO2solubility in brine with different salinities and proposed a model predicting the CO2solubility in a brine with defined concentrations of different salts at any desired pressure and temperature conditions[7,8].CO2solubility for flooding brine used here(i.e.brine#2)was determined to be 1.2441 mol of CO2per 1 kg of brine at test conditions(i.e.at 13789.52 kPa and 40°C).The amount of CO2solubility in brine #2 used for CWpreparation has been measured via a gasometer apparatus working at standard conditions (i.e.at 101.352 kPa and 15.6°C).Measured quantity and the result of Duan et al.’s model indicate the same amount.

Table 4 Physical properties of carbonate core samples

Fig.1.Rocking cell apparatus.

3.Experimental Materials

3.1.Core samples

The core samples used in this study were taken out of an integrated outcrop of a southern Iranian carbonate oil reservoir.The properties of samples are given in Table 4.

The outcrop used here had a homogeneous fabric,thus it is expected the core samples to have the same mineral distribution and physical properties.As shown in Table 4,the permeabilities of core samples are almost in the same order.According to this matter,the results obtaining from coreflood experiments can be generally compared to each other.

3.2.Rocking cell apparatus

In order to prepare the CW utilized in this study,a rocking cell was used.Fig.1 shows the rocking cell apparatus.The cell was shaken to fully mix the brine and CO2.It is capable of working at pressure and temperature up to 34473.8 kPa and 148.9°C,respectively.The cell has been made of stainless steel in order to impede the possible corrosion of carbonic acid formed during CW preparation process.Carbonic acid emerged in CW causes acidic pH throughout the solution that can modify the carbonate rock properties such as porosity and permeability.Any change in rock porosity and permeability can influence fluid flow and acting mechanisms in porous medium.Hence,it is important to investigate the changes in the rock physical properties.Correlation 1 could be used to determine the pH of carbonated water [26]:

where Khis a constant for the hydration equilibrium,KHis the Henry’s constant,Ka1is the dissociation constant [carbonic acid is a weak acid that dissociates into a bicarbonate ion (HCO3-) and a hydrogen ion (H+)]and PCO2is partial pressure of CO2obtaining from Eq.(2) [26]:

where CCO2is CO2concentration.Crawford et al.and Donald showed that the pH of CW fully saturated with CO2is remarkably lessened even at low CO2partial pressure[27,28].In this study,the pH of carbonated water was about 3 using Eq.(1) at pressure of 13789.52 kPa.Low acidic pH of carbonated water can dissolve carbonate rock minerals and accordingly it can modify rock permeability.Carbonic acid can dissolve carbonate minerals as following equations [29]:

Brownlee and Sugg observed precipitation of rock minerals due to CO2dissolution in water[30].Abbaszadeh et al.investigated the impact of CW during spontaneous imbibition mechanism using carbonate core samples.They reported that carbonated water can dissolve carbonate components and subsequently increases rock sample permeability [31].Using the similar core samples in the experimental trials performed in our research group in the advanced EOR research center of Shiraz University indicated that CW could change the permeability and porosity of the samples in a range of ±10% and ±8%,respectively after 300 h contact with pores surface of rock [31].

3.3.Coreflood rig

The main apparatus in this work is coreflood rig consisting of three accumulators (i.e.oil,saturation brine and CW).Coreflood set-up was equipped with an oven in order to supply a constant and uniform temperature condition for all tests.The produced gas was recorded using gasometer apparatus set at the outlet of coreflood rig.The coreflood set-up is schematically shown in Fig.2.

3.4.Injection pump

To inject different fluids (i.e.oil,saturation brine and CW) into the core samples,a low rate pump has been used.This pump can operate at pressures up to 34473.8 kPa and a flow rate range of 0-100 cm3·min-1.It can operate both at constant pressure and rate conditions.In this study,constant rate condition has been applied for all the tests.

4.Experimental Procedure

Before starting each test,saturation brine (i.e.brine #1) was injected into the core samples pursued by crude oil flooding to obtain the initial saturation conditions of reservoir fluids (i.e.oil and water).Rapoport and Leas investigated the capillary end effect on test results during coreflood experiments and proposed a scaling coefficient that alleviated this effect.Following formula expresses this scaling coefficient [32]:

Fig.2.Schematic of the coreflood set-up.

where L is core length in cm,U is Darcy velocity in cm·min-1and μ is the displacing phase viscosity in cp.In this work,all experiments were carried out at pressure and temperature of 13789.52 kPa and 40°C,respectively and displacement rate of 0.2 cm3·min-1in order to diminish the capillary end effect.Two sets of coreflood tests using carbonated water were performed here as secondary recovery method.The first experiment was SCWI at the condition of low initial water saturation.In this experiment,carbonated water was injected into the core sample S1with initial water saturation (Swc)of 36%.Carbonated water injection process was performed for about five PVs of the used samples to guarantee that no more oil is being produced.In the other test,SCWI was carried out using core sample S2at the condition of high initial water saturation (Swc=50%) with the same conditions of the first test.

5.Results and Discussion

5.1.Additional oil recovery

The main factor clarifying the performance of a coreflood test is ultimate oil recovery factor.Importance of oil recovery factor can be affected by the oil price in the global markets.In addition to high profit,applied EOR technique with low operating cost compared to the other methods would be more favorable.The oil recovery curves of both tests in terms of the injected CW have been shown in Fig.3:

As can be seen from Fig.3,for the first injected PV of CW,both of tests result in more or less the same oil recovery.At the early stage of CWI (up to a pore volume injection) in both tests,water content had an inconsiderable effect on CWI performance,while its impact became substantial as both tests continued.The difference of oil recoveries has increased in each subsequent stage and reached 20.3%at the end.Oil production shows a stage wise trend due to CO2transmission from CW as CO2source into the oil phase.Stage wise trend in oil production indicating formation of several small oil banks,was observed in all experiments performed in this study.Martin et al.and Shakiba et al.observed this stage wise trend during CWI tests on sand stone and carbonate rock samples,respectively [4,33,34].This sequential additional oil recovery during the experiments indicates that CWI should be applied for a long period of time to be more efficient.During the flooding process,oil phase is continuously being trapped.CO2transmission from CW into the trapped oil ganglia can coalesce the trapped oil and as a result,the reconnected trapped oil ganglia can move again,thus more oil would be produced.High in-situ water content could play as a barrier between the oil phase and the CO2source (i.e.CW).High connate water saturation (HCWS) hinders CO2transmission from CW into the oil phase compared to low connate water saturation(LCWS)condition.In other words,this water phase lessens the CWI performance.Besides this,higher Swcalso further dilutes the CW by mixing with the flowing CW.These two factors would delay oil swelling mechanism and result in reduction of ultimate oil recovery during the experiments.It should be mentioned that during CWI test,whatever more oil is produced in the next stages,CO2transmission and subsequently oil swelling and oil reconnection mechanisms diminish over the time.This decreasing trend of oil production (shown by double arrows in Fig.3) is as a result of reduction of the residual oil saturation within the core sample after each plateau stage as the flooding continues.Hence,at the end of the tests,the oil recovery factors of high and low Swcconditions were 68.1% and 88.4%,respectively.Fig.4 shows the volume of oil produced at each stage for both tests.In this figure,the reduction of oil production has been illustrated.

Fig.3.Oil recovery of SCWI at conditions of high Swc and low Swc in terms of injected CW.

Fig.4.Oil production at each plateau stage for both HCWS and LCWS tests.

In order that the available oil volumes for each core sample to be comparable,the initial oil volume of each core was divided by its PV (it can be considered as a dimensionless volume that are 0.49 and 0.64 for HCWS and LCWS,respectively).It reveals that in HCWS test,the total amount of recoverable oil in place is less than that of LCWS case because of high in situ water content.This indicates that the same CW can act more efficient (i.e.more efficient in oil production period) during HCWS case compared to LCWS test as in Fig.3;the initial stage of oil recovery of HCWS test reached the plateau faster than that of LCWS(the highlighted area by a rectangular in Fig.3).It is worth mentioning that oil viscosity reduction is one of the main mechanisms helping more oil production during CWI process.CW can significantly decrease oil viscosity due to gas transmission into the oil phase.In fresh core sample(i.e.sample with continuous oil phase throughout the core without any trapped oil ganglia in the first instances),high initial water content needs some time in order to impose its adverse effect (i.e.water shielding effect)on carbonated water performance.At these conditions,CWI with the same initial CO2concentration and injection rate would result in the same oil recovery (almost 40%) during the first injected PV for both cases.As flooding continued,in HCWS test,with respect to lower residual oil saturation within the core sample compared to LCWS case,CO2(from fresh CW as source of it) can dissolve into the less residual oil phase and subsequently reactivate oil swelling mechanism and produce oil faster than that of LCWS test,however water shielding effect gradually decreases oil recovery during each subsequent stages.This means that the period of oil production plateau during the stages can be shorter for high Swccase compared to low Swccondition.During the plateau period,no oil is being produced and water is the only effluent of system.Fig.5 shows produced water during both tests.

In coreflood experiments,the pressure of system declines as fluid BT takes place.When CWI process starts,oil is the first phase being produced from the core outlet.After a continuous oil phase production,water phase produces.Thus,the system pressure decreases as the water phase is being produced.Upstream and downstream pressures were almost the same in both experiments at the beginning.The downstream pressure was kept constant via a back pressure regulator(BPR)at the core outlet.Figs.6 and 7 show simultaneously the system pressure drop(i.e.Pup-Pdown)and water production for both high and low Swcconditions,respectively.

Fig.5.Water production during both tests in terms of injected CW.

As the results show,the maximum pressure drop during HCWS condition is lower than that of LCWS case (the difference is about 5%).This difference is due to faster and easier dissolution of CO2into a high content of fresh water existing within the core sample in HCWS test,which restrain system pressure rising.Figs.6 and 7 show 8370.23 kPa and 7935.86 kPa as maximum system pressure for LCWS and HCWS,respectively.In addition to pressure drop,the experimental data can also indicate the BT time.As both of tests have been performed at the same conditions (such as injection rate,pressure and temperature);thus water breakthrough points of both tests were observed to be the same and took place after about 0.2 PV of CW injection.

5.2.CO2 storage

In order to evaluate the CWI process for CO2retention underground,total delivered and produced gas have been analyzed.A container mounted at the BPR outlet used to collect the produced oil and water.The amount of produced CO2has accurately been monitored and recorded using a gasometer apparatus in both tests.Fig.8 shows gas production of both tests versus the injected CW.

As can be seen from Fig.8,no gas has been produced at the beginning of both tests.At this period,dissolved CO2in water is transferred into the oil phase within the core sample.Gas dissolution gradually makes the oil phase saturated with CO2.Thus,CO2production has been started once the oil phase was saturated with CO2.It was observed that oil and CO2were simultaneously produced and faster than water.It likely can be due to CO2mass transfer phenomenon,which causes gas to surpass the injected front.This observation approves the CO2transmission into the oil phase as it is passing through the core sample.CO2storage can be estimated via a dimensionless storage factor(Fsd).Fsdfor a core sample is defined as the ratio of the difference between injected and produced CO2(i.e.stored CO2within the core) at the end of each plateau stage to one PV of the sample when it is totally inundated with CW at the desired temperature and pressure (i.e.40°C and 13789.51 kPa for this study).Both gas volumes in denominator and numerator of Fsdfactor have been measured using gasometer,which is working at standard condition.Fig.9 shows Fsdfactor of both tests versus the injected CW.

Residual volumes of different fluids within the core samples and the amount of Fsdfor HCWS and LCWS cases have been presented in Table 5.

In both tests,CO2has been dissolved not only into the oil phase but also into the water in place.The difference between gas storage capacities of both tests increased as the flooding continued.CO2needs a certain time to be completely dissolved in the water and oil phases and reaches its ultimate equilibrium condition.As discussed before,the core sample used for HCWS test (i.e.core S2)has lower PV compared to core sample S1.Thus,the required time in order to inject five PVs of CW into the core sample S1with the same injection rate is more than that of sample S2.This means that LCWS test has lasted much more than that of HCWS test.That is,the LCWS test has been carried out for 475 min while HCWS test was performed for a duration of 366 min.Thus,if oil recoveries data of both cases plotted in terms of time,the end points of each test would be characterized.Fig.10 depicts the end points of both trials.

Thus,in this study during HCWS experiment,the water and oil phases cannot completely reach their maximum available solubility.Besides,in HCWS test,because of higher initial water content with high salinity compared to the water used for CW,the capacity of connate water to dissolve the CO2is lower than that in LCWS experiment and as a result,gas storage would be lower compared to LCWS mode.At the end of the tests,the results showed 54%and 38%CO2storage capacities for LCWS and HCWS tests,respectively.According to the results,CWI as an EOR technique can be generally taken into account as a high CO2storage capable method as well.However,it is important to take into account the interaction between CW and in particular,carbonic acid being emerged during CWI process in the porous medium and carbonate rock minerals.Carbonic acid appeared during dissolution of CO2into the brine during CWI can react with calcium and magnesium in carbonate rocks.This acid can dissolve these minerals and as a result,dissolved minerals passing through the rock pores can be precipitated,which can considerably affect rock porosity and permeability.Any change in rock properties can influence the EOR efficiency [35,36].CO2dissolution into the water increases the water density hence,the dissolved CO2moves downwards,where far from the cap rock and reduces the risk of CO2leakage.This means that CWI method could be considered as a secure technique for CO2sequestration [37].In order to investigate the efficiency of the both trials for oil production,it is better to consider the amount of oil produced stepwise during each test.Fig.11 illustrates the ratio of cumulative CO2consumed in both experiments versus cumulative produced oil.

Fig.6.Pressure drop and water production of HCWS test versus the injected CW.

Fig.7.Pressure drop and water production of LCWS test versus the injected CW.

Fig.8.CO2 production during HCWS and LCWS tests versus the injected PV.

Fig.9.CO2 storage in form of Fsd factor for both tests versus the injected CW.

Fig.11 indicates that LCWS would be more efficient compared to HCWS test,as lower CO2consumption during LCWS compared to HCWS experiment,resulted in higher oil recovery.In order to precisely investigate the CO2utilization efficiency,the quantity of the injected CO2and produced oil during each oil production stage for HCWS and LCWS modes are presented in Table 6.

As can be seen from Table 6,LCWS test resulted in higher CO2utilization efficiency compared to HCWS mode.That is,at the same experimental conditions,LCWS test requires less CO2compared to HCWS mode in order to produce the same oil volume.

6.Conclusions

Based on the results of this study,the following conclusions can be drawn:

(1) SCWI at the condition of low Swcresulted in more oil recovery efficiency compared to that of high Swc.In the condition of high water saturation in place,the water barrier between the CO2source (i.e.CW) and oil blobs hinders the oil recovery.This water barrier,so-called water shielding or water blocking effect,delays the oil swelling mechanism in the case of HCWS.Higher connate water saturation also further dilutes the carbonated water as a result of mixing with the flowing CW phase.These two mechanisms result in the reduction of ultimate oil recovery efficiency.

(2) It was observed that the oil production in both experiments is in a sequential mode indicating formation of several small oil banks,however incremental oil production declined over time,which is due to the reduction of residual oil saturation at each subsequent stage.The low residual oil saturation results in decreasing the oil swelling mechanism and oil reconnection phenomenon,which subsequently reduces the ultimate oil recovery efficiency.

Fig.10.Oil recovery of SCWI at conditions of high Swc and low Swc in terms of time.

Fig.11.Comparison of injected CO2 and produced oil in both tests during each stage.

(3) The outcomes indicated that CO2and oil have been simultaneously produced.This observation clarifies that CO2is released from CW in gas form that could be diffused into the oil phase as it is being passed through the core.Hence,CO2can be produced faster than water.

(4) According to the results,the CO2storage in both HCWS and LCWS cases increases over the time.This clearly indicates the time dependency of the CWI,as the mass transfer phenomenon controls this EOR technique.The gas storage for low and high Swcwere eventually 54%and 38%,respectively.

(5) It was observed that CWI process in LCWS test can store a substantial amount of injected CO2(more than half of the total delivered CO2) within the sample.Gas storage at the field scale helps to reduce the greenhouse gases causing global warming,therefore CWI can assists more CO2storage underground.

Table 6 CO2 utilization efficiency in both LCWS and HCWS cases

Nomenclature

D core diameter,mm

Fsddimensionless storage factor

K permeability,mD

L core length,mm

PV pore volume,cm3

Soiinitial oil saturation

Swcconnate water saturation

Wt mass percent,%

φ porosity,%

Acknowledgement

The authors gratefully acknowledged the help provided by Mrs.Meratian,and Dr.Escrochi in all our experimental work.We also would like to thank National Iranian South Oil Company (NISOC)for supplying the crude oil used in the experiments.