Original article
Effect of multi-component ions exchange on low salinity EOR:
Coupled geochemical simulation study
Ehsan Pouryousefy, Quan Xie
*, Ali Saeedi
Department of Petroleum Engineering, Curtin University, Western Australia, Australia
a r t i c l e i n f o
Article history:Received 23 March 2016 Received in revised form 26 May 2016
Accepted 27 May 2016 Keywords:
Enhanced oil recovery Low salinity waterflooding Geochemical reactions Numerical simulation
a b s t r a c t
The mechanism(s) of Low salinity waterflooding (LSWF) has been extensively investigated for 15 e20 years, as a cost-effective and environmentally friendly technique for improved oil recovery. However, there is still no consensus on the dominant mechanism(s) behind low salinity effect due to the complexity of interactions in the Crude oil/Brine/Rock (COBR) system. While wettability is most agreed mechanism of low salinity EOR effect. Nevertheless, the mechanism(s) behind the wettability change is debated between multi-component ion exchange (MIE) and double layer expansion (DLE) in sandstone reservoirs. This paper aims to investigate the effectiveness of MIE with a coupled geochemical-reservoir model using published experimental data reported by Nasralla and Nasr-El-Din [1].
We created core-scale numerical models with parameters identical to those used in the exper-iments. We simulated the low salinity effect using a commercial reservoir simulator, CMG-GEM, by coupling three chemical reactions: (1) aqueous reaction, (2) multi-component ion exchange, and (3) mineral dissolution and precipitation. We modelled the adsorption of divalent cations on the surface of the clay minerals during low salinity water injection. Simulation results were compared with the experimental results.
Simulation results show that the fractional adsorption of divalent cations (Ca2þ) increased almost 25% by injecting a 2000 ppm NaCl solution, compared to initial 10,000 ppm NaCl. Injecting a 2000 ppm of CaCl2solution, however, significantly increased the adsorbed Ca2þfrom 0.1 to 1, which
implies the complete saturation of mineral surface with divalent cations. Moreover, injecting 50,000 ppm of CaCl2solution also demonstrated the same effect as the 2000 ppm CaCl2solution
but with a faster rate.
Upon combining the simulation and experimental results, we concluded that the multi-component ion exchange is not the sole mechanism behind low salinity effect for two reasons. First, almost 10% additional oil recovery was observed from the experiments by injecting the 2000 ppm CaCl2compared with 50,000 ppm CaCl2solutions. Even though in both cases the surface
is expected to be fully saturated with Ca2þaccording to the geochemical modelling. Second, 6% incremental oil recovery was achieved from the experiments by injecting 2000 ppm NaCl solution compared with that of 50,000 ppm NaCl. Although 25% incremental adsorption of divalent cations (Ca2þ) were presented during theflooding of the 2000 ppm NaCl solution. Therefore, it is worth noting that the electrical double layer expansion due to the ion exchange needs to be taken into account to pinpoint the mechanism(s) of low-salinity water effect.
Copyright© 2016, Southwest Petroleum University. Production and hosting by Elsevier B.V. on behalf of KeAi Communications Co., Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
Oil and natural gas are expected to remain an important resource for future energy demand [2,3]. However, there is a pressing need to develop cost-effective techniques to enhance oil recovery in the period of low oil prices. Engineering the injected * Corresponding author.
E-mail address:[email protected](Q. Xie).
Peer review under responsibility of Southwest Petroleum University.
Production and Hosting by Elsevier on behalf of KeAi
Contents lists available atScienceDirect
Petroleum
j o u r n a l h o m e p a g e : w w w . k e a i p u b l i s h i n g . c o m / e n / jo u r n a l s/ p e t l m
http://dx.doi.org/10.1016/j.petlm.2016.05.004
2405-6561/Copyright© 2016, Southwest Petroleum University. Production and hosting by Elsevier B.V. on behalf of KeAi Communications Co., Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
water chemistry and using water wisely to enhance oil recovery is a novel and emerging research area, which is called low salinity water flooding [4e7], smart water flooding [8e12], designer waterflooding[13e16], or ion tuning waterflooding[17,18]. Many researchers found that low salinity water injection could achieve an incredible additional oil recovery in sandstone and carbonate reservoirs from both experiments andfield tests[19,20].
Changing waterflood salinity/chemistry can increase oil re-covery from sandstone and carbonate reservoirs by 5e25%, though why this happens is unclear [13,17,19e29]. Several mechanisms have been proposed to describe how the low salinity effect (LSE) increases oil recovery, including: (1) fines mobilisation[30], (2) limited release of mixed-wet particles[30], (3) increased pH and reduced interfacial tension (IFT) similar to alkalineflooding[31], (4) multi-component ion exchange (MIE), (5) expansion of the double layer, (6) salt-in effect[12], and (7) osmotic pressure[32].
Although the mechanism(s) of low salinity water is still debated, two major advantages of low salinity water flooding over other enhanced oil recovery (EOR) techniques are: a) cost-effective (i.e., based on the relatively low cost of acquiring low salinity water), and b) relatively environmentally friendly. British Petroleum (BP) applied Low Salinity Waterflooding (LSWF) to the Endicott oilfield in Alaska and achieved 26% additional oil re-covery. In 2016, BP, ConocoPhillips, Chevron, and Shell will initiate the largest offshore low salinity waterflooding project in Clair Ridge oilfield (in the North Sea). Clair Ridge is expected to produce more than 40 million barrels of additional oil at a relatively low cost[33]. LSWF is also a promising technique for mature oil reservoirs where the production has significantly decreased, and further recovery is uneconomic.
However, the major disadvantage of LSWF is that the mech-anism(s) by which it operates are uncertain. Several oil com-panies, such as BP [34], Shell [35], Saudi Aramco [36], and PetroChina [17,18,21,37]have conducted numerous lab experi-ments, i.e., Log-Inject-Log, Single Well Chemical Tracer Test (SWCTT), andfield scale trials. Their results have shown that, while in some cases LSWF has achieved an additional 5e25% oil recovery, in other cases, only minor incremental oil recovery was observed[38].
Even though the mechanism(s) of low-salinity effect is still uncertain, the most agreed mechanism of LSWF is wettability alteration[39]where again two main theories are proposed to justify wettability alteration: (1) multicomponent ion exchange (MIE) [40e43], (2) expansion of electrical double layer (DLE)
[1,18,44e47]. In this study we particularly focus on simulating the effect of multicomponent ion exchange on low salinity effect with a combination of geochemical modelling and experimental observations.
2. Model description
MIE theory is based on the water geochromatography where cation exchange is a normal phenomenon occurring on the sur-face of the minerals due to the different affinities of various cations towards the rock surface [48]. Meanwhile, crude oil/ brine/rock (COBR) system is intrinsically in a thermodynamic equilibrium [20,49,50]. Injecting low salinity with different electrolyte concentration than formation brine will disturb the equilibrium system and variations in ionic concentration results in the substitution of divalent cations by the monovalent cations [40]. Also, the presence of divalent cations in the formation brine can bridge the rock and the crude oil as shown inFig. 1 [40]. Theoretically, exchange with monovalent cations can release the crude oil from the rock surface, which alters the system to more water-wet thus increasing the recovery factor.
Fig. 1represents four different adhesion mechanisms of polar ends (e.g. COOH-) of the crude oil on the clay surface (precented as grey boxes) which can be affected due to the cations exchange [51]. Lager, Webb [40] hypothesised that the Van der Waal, Cation exchange and ligand bonds are strong and can lead into the detachment of organometallic complexes (seeFig. 2).
Fig. 1. Representation of the diverse adhesion mechanism occurring between clay surface and crude oil (Lager et al., 2008a).
3. Core-scale numerical model
This paper aims to elucidate the dominant mechanism(s) of low-salinity effect with coupling the geochemical reactions. We created a one dimension core-scale model (Fig. 4) with the identical properties to Nasralla and Nasr-El-Din[1]experiments. We also conducted a set of core-scale numerical simulations by injecting either NaCl solution or CaCl2under secondary mode using various salinity. To create the core-scale numerical model, five steps were taken as following;
In step 1, we created a core-scale numerical model using block centred grids. In this study, the core plug was 5 inch in length and divided into 40 grids. In step 2, to simulate the capillary end effect, two additional blocks were placed at both sides of the core plug with 0.0001 porosity and high permeability of 1000 D[25]. In step 3, we use Corey correlation to history matching Ramez's experimental data, thus acquiring a set of representative relative permeability data.
In step 4, we defined the initial condition (e.g. saturation and pressure) based on Ramez's experimental data. The core plug was saturated using a dead oil with an apparent viscosity of 3.7 cp at 212F, and at the connate water saturation (Swc) at 32.6%. The composition of formation brine extracted from the experiment is shown inTable 1. In Step 5, the core wasflooded using LS water at constant injection rate 0.5 ml/min, identical to the experiment. To validate the model, the simulation results was history matched against recovery factor and pressure drop (Fig. 3). While we did not explicitly investigate the non-uniqueness of history matching, we can use the derived relative permeability curve to investigate the effect of multi-component ion exchange on recovery factor.
The common practice to simulate the effect of LSWF is incorporating wettability alteration by treating relative perme-ability curves from mixed or oil wet to more water wet[52]. Two sets of relative permeability curves should be defined to repre-sent HS and LS water effect. Also, an interpolant is used to interpolate between the two sets of relative permeability curves [48,53,54], although the correct interpolant is still an open research area. However, in this study, we aimed to examine the hypothesis that multi-component is a dominant mechanism behind low salinity effect. Therefore, we exclusively focused on geochemical reaction instead of oil recovery.
3.1. Geochemical model
Low salinity water injection can give rise to geochemical alteration due to the interaction of fluidefluid and fluid-rock
[55]. Understanding the geochemical variation is a key factor in pinpointing the mechanism(s) of the low salinity water effect thus constraining the uncertainties. The fully coupled geochemical and equation of state model werefirstly introduced by Nghiem, Sammon[56]. In this study, we use CMG-GEM to simulate the effect of multi-component ion exchange on low salinity effect. The Ion-exchange takes place along with geochemical reactions where geochemical reactions can occur in two main groups: (1) aqueous reactions, (2) mineral dissolution/ Precipitation reactions[48].
Aqueous reactions are the spontaneous type of reactions and represented as the chemical equilibrium reaction, whereas mineral dissolution and precipitation reactions are rate depen-dent[57]. In this study, we consider the following mineral re-actions in the numerical simulation, according to the water composition and mineralogy of the core plug (Table 2) published by Nasralla and Nasr-El-Din[1].
Aqueous reactions: CO2ðaqÞ þ H2O4 Hþ þ HCO3 (1) OH þ Hþ4H 2O (2)
While the core plugs that were testes in Ref.[58]were Berea sandstone, these samples contain 2.8 wt% calcite. It is worth
Fig. 3. History matched data (recovery factor vs. injected PV, left; pressure drop vs. injected PV, right).
Fig. 4. Simulated Core Plug. The colour code represents the water saturation dis-tribution at initial condition.
noting that calcite is a volatile mineral and can easily dissolve in the formation brine. Hence, the following mineral reaction has also been modelled to incorporate mineral dissolution and pre-cipitation effect. Mineral reaction: CalciteþHþ4Caþþþ HCO 3 (3)
Ion exchange is the key reaction and again considering the water composition and presented minerals, the main cations prevailing in our system would be Naþ, Hþand Ca2þ.
Ions exchange reaction;
Naþþ 12ðCa X2Þ4ðNa XÞ þ 1 2Ca
2þ (4)
3.2. Reaction kinetics
In the geochemical modelling, the heart of the model is the equilibrium system [59]. In GEM, the chemical reaction is in equilibrium if the forward reaction rate and backward reaction rates are equal Nghiem, Sammon[56], which can be replicated in the following equations;
Qa Keq;a¼ 0 ; a ¼ 1; …; Raq (5)
Qa¼ PnK¼1aq ðaKÞvka (6)
The Q
a
is the activity of the products and naqis the number of species in the intra-aqueous reaction; akis the activity of each species; Keq is the chemical constant of the reaction; R is the universal gas constant; T is thefluid temperature, and vkais the stoichiometry coefficient of a component in the chemical reac-tion [56]. The activity coefficient can be taken approximatelyequal to the concentration of the species (In Molality)[48]. In GEM the equilibrium constants for the above reactions are calculated with a fourth order polynomial where T is inC and followed byfive predefined coefficients a0 to a4 (equation(7)) (the coefficients are shown inTable 3)
log Keq
¼ a0þ a1Tþ a2T2þ a3T3þ a4T4 (7) As mentioned earlier mineral dissolution/Precipitation re-actions are rate dependant, hence other parameters such as the surface area and the activation energy are considered.
For the Ions exchange, the selectivity coefficient is imple-mented over the equilibrium constant to overcome the challenge of calculating the activity coefficient of Na-X and Ca-X compo-nents. The selectivity coefficient concept (the degree to which an ion selective electrode responds to particular ion with respect to reference ion) follows the GaineseThomas convention (1953) [60], and It also can be defined as the ratio of two equilibrium constant of each cations adsorbing to the surface of the substrate (Clays). For the ions exchange reaction (4), it can be expressed as equation(8) KNa=Ca0 ¼zðNa XÞ m Ca2þ0:5 ½zðCa X2Þ0:5m Naþ g Ca2þ0:5 g Naþ (8) where X denotes the clay mineral in the reservoir rock; zðNa XÞ and zðCa X2Þ are the equivalent fractions of Naþ and Ca2þon the exchanger respectively; m denotes the molality andY is the activity coefficient. It should be noted that equilibrium constant is a thermodynamic variable whereas the selectivity coefficient is an operational variable. Also, in the simulation, selectivity coef-ficient is implemented over the equilibrium constant to over-come the challenge of calculating the activity coefficient of Na-X and Ca-X components. After constructing the coupled geochemical model, the fractional adsorption of divalent cations during the coreflood with 2,000, 10,000 and 50,000 ppm solu-tions of NaCl and CaCl2was simulated, and were compared with the experimental data. The results are explained in subsequent section. The values for selectivity coefficient are based on experimental data and predefined in CMG-GEM.
4. Coupled numerical simulation results 4.1. Injecting NaCl solution
Naþis a typical monovalent cation in the injected brine and formation brine. The interaction between injected Naþ and original crude oil/brine/rock can provide implications of the multi-component ion exchange. Therefore, we simulated NaCl solution injection with three different salinity, 2,000, 10,000 and 50,000 ppm, respectively. The rock properties in the numerical model are consistent with the experiments. We particularly studied the variation of the Naþand Ca2þconcentrations in the solution and the Ca-X2 at the surface of the rock in a certain grid block (40 1 1).
First, we simulated 50,000 ppm NaCl solution injection. The concentration of Ca2þin the one dimension model can be seen in Table 1
Formation brine composition (Nasralla and Nasr-El-Din, 2014).
Ion type Concentration (ppm)
Naþ 54,400 Ca2þ 10,600 Mg2þ 1610 Cl- 107,000 HCO3- 176 SO42- 370 TDS 174,156 Table 2
Mineralogy of Berea core plug (Nasralla and Nasr-El-Din, 2014). Minerals BBS (wt%) Quartz 85.0 Feldspar 6.4 Calcite 2.8 Illite 2.0 Kaolinite 3.8 Table 3
Reaction Kinetics; constants used in eq(7)for mineral and aqueous reactions.
Reaction Type a0 a1 a2 a3 a4
CO2(aq) H2O¼ (Hþ)þ (HCO3) Aqueous 6.54924 0.009 0.0001 2.76E-07 3.56E-10
Fig. 4. Simulation results show that injecting a solution with different electrolyte concentration than the one in the connate water will cause salinity buffering in the core sample, and consequently ion-exchange occurs at the surface of the pores. As mentioned before while ion-exchange is an instantaneous re-action[57]. Consequently, three distinct salinity fronts will be generated (Fig. 5). Thefirst front contains formation brine, and yet to be displaced by injection brine (red arrow). The second slug in the middle is followed by buffered brine (Green) at which the displacement is taking place, and concentration of Naþand Ca2þis changing from high salinity formation brine to the in-jection brine (green arrow). Then the trailing edge is followed by injected water where the composition is, to some extent, similar to the injected brine where in NaCL solutions (10,000 ppm), the concentration of Ca2þis almost zero (Blue arrow).
We also investigate the geochemical alteration in a certain grid block to elucidate the implication of low salinity effect and geochemical reactions. Grid block (40 1 1) was chosen to inves-tigate the concentration of Naþand equivalent fraction Ca-X2, as shown inFigs. 6 and 7. Simulation results show that concentra-tion of Naþ in the solution decreased first, and then become constant. The variation of Ca2þ concentration keep the same trend as Naþdoes, but the concentration of Ca2þdrop to zero. The decrease of Naþ is attributed to the lower salinity of the injected brine than the formation brine. Simulation results also show that the high concentration of Naþions will favour Naþ adsorption and remove Ca2þ cations at the surface initially, although Ca2þ has a higher affinity towards the clay surface
(Fig. 7). This is consistent with the multi-component ion ex-change theory. Also, this phenomenon was observed in core-flooding experiments conducted by Lager, Webb[40].
We then simulate the geochemical reaction during 10,000 ppm NaCl injection, as shown inFig. 8. Results show that the Ca2þconcentration drop to zero similar to 50,000 ppm NaCl injection, but the concentration of Naþdeclines dramatically due to the lower salinity of injected brine.Fig. 9also shows that Ca-X2 increases initially, and then drop to zero gradually. The in-crease of Ca-X2 is due to the fact that calcite in the core-plug start to dissolve which partially compensate the Ca2þreduction, hence the rate at which Naþconcentration drops is faster than Ca2þ concentration. Consequently, there will be a temporary increase in the Ca2þadsorption but eventually, the system will thermo-dynamically stabilize with a higher concentration of Naþ, and again a full desorption of Ca2þsimilar to 50,000 ppm injection (seeFig. 10).
Fig. 5. Salinity buffering during LSWF of Berea core plug (10,000 ppm NaCl).
Fig. 6. Cation concentration during 5% wt NaCl solution injection.
Fig. 7. Sorption of divalent cations during 5% wt NaCl solution injection.
Fig. 8. Cation concentration during 1% wt NaCl solution injection.
Geochemical reaction with invading of 2000 ppm NaCl was also simulated. Simulation results show the trend of concentra-tion variaconcentra-tion is in line with 50,000 and 10,000 NaClflooding. Naþ concentration decreases sharply, but the Ca-X2 increases initially, then it starts to drop.Fig. 11shows the concentration of Ca-X2 falls slower compared to the injection of 50,000 and 10,000 ppm NaCl, which is due to the smaller differential con-centration between Ca2þand NaþIons in the system.
4.2. Injecting CaCl2solution
Divalent cations are usually rich in the formation brines and injected brines. Also, divalent cations play a significant role in ion-exchange due to the strong affinity to adsorb at the surface of the rock [51]. Therefore, we have also simulated the CaCl2 flooding with various concentrations and investigated the geochemical reactions. Simulation results are given inFigs. 12 and 13. It is clear that the concentration of Ca2þ increases because the concentration of Ca2þwe injected (50,000 ppm) is much more than that (10,600 ppm) in the formation brine. However, it is worth noting that in all three cases (2000, 10,000 and 50,000 ppm CaCl2 solution) a spontaneous adsorption of Ca2þoccurs to the point that the entire ions exchange sites are saturated by divalent cations.
5. Discussion
5.1. Predicting recovery based on MIE and coupled geochemical model
As mentioned earlier based on MIE theory, there should be an agreement between recovery factors and divalent cation adsorption/desorption. The higher is the adsorption, the lower the recovery should be.Fig. 14represents the Ca2þadsorption/
desorption during NaCl and CaCl2 solutions injection. We can make the following predictions based on simulation results and MIE theory.
1. 50,000 ppm NaCl solution should contribute the highest re-covery because almost all the Ca-X2 are substituted by Naþ, which leads to losing the ability to bridge between rock and crude oil.
2. Injecting 10,000 ppm NaCl has also resulted in the full desorption of divalent cations, hence, the recovery should be high and almost equal to 50,000 ppm NaCl solution. 3. Injecting 2000 ppm should result in lower recovery than
previous cases since the adsorption of divalent cations increased by 20%.
4. Injecting all three concentrations of CaCl2 should result in equal recoveries and it should be lower than NaCl solutions.
5.2. Comparison with experimental observation
Fig. 15shows the experimental observation of recovery fac-tors achieved in core flooding experiment via different brine compositions (2,000, 10,000, and 50,000 ppm)[1]. Nasralla and Nasr-El-Din [1] observed lower recovery factors by injecting CaCl2solutions compared to the NaCl at the same concentration, which can be considered as an agreement with MIE theory. However, injected different CaCl2 solutions produced different recoveries which are not consistent with the prediction as a result of multi-component ion exchange. Furthermore, geochemical simulation results from various NaCl concentration injection has no consistency with experimental data at all. Experimental results show that lower concentration of Ca2þhas Fig. 10. Cation concentration during 0.2% wt NaCl solution injection.
Fig. 11. Sorption of divalent cations during 0.2% wt NaCl solution injection.
Fig. 12. Cation concentration during 5% wt CaCl2solution injection.
a higher potential to improve oil recovery. This is also observed for various concentration of NaClflooding[1].
It is worth noting that water chemistry has a significant impact on the zeta potential for bothfluidefluid, and fluid-rock [1]. Double layer expansion plays a dominant role in the differ-ence of recovery factor as injected various CaCl2[1].
5.3. The impact of CEC on the recovery factor
CEC is an intrinsic parameters for clays, which can represent the ability of the surface of the clays to be adsorbed by the counter-ions. The magnitude of CEC ranges from 3 to 150, depending on the type of clays [61]. If MIE is the dominant mechanism, the CEC should presumably affect the flooding performance of LS. Hence, we simulated three extreme scenarios with CEC equivalent to 10, 50, and 150, respectively. Also, we monitored the adsorption/desorption of divalent cations to predict the recovery factor. Fig. 16 shows the adsorption of divalent cations by injecting CaCl2solution at concentration of 10,000 ppm injection. Simulation results indicate that the sur-face of rock is saturated with Ca-X2 after injected 1 PV CaCl2 solution for all three different CEC values. Also, CEC does not play a significant role to impact the surface Ca-X2 saturation, although CEC will influence the time to reach the full surface saturation with divalent cations. Hence again, If MIE is the dominant mechanism it is not expected to observe higher re-covery. However, it should be noted that the surface potential will affect the zeta potential and the repulsion forces (seeFig. 17).
5.4. Further discussion
In addition to cation bridging there are seven other mecha-nisms on which the organic materials can be bonded to the surface of the minerals[62,63], and three of those mechanisms; Cation Exchange, Ligand bonding, Anion bridging are highly affected by electrolyte concentration (Fig. 18). Furthermore Austad, Rezaeidoust [64] has also proposed other possible mechanisms where polar components in the crude oil can directly attach to the surface of the rock.
We simulation results suggest that MIE can affect the LSWF performance as one of the bonding mechanisms, but in this study, it has been shown that ion exchange cannot be the only mechanism of breaking those bonds. To break the bonds, there should be a repulsive force greater than the chemical bond strength, and the repulsive force can be explained by the concept of disjoining pressure[49]. The disjoining pressure is the direct function of electrical double layer (EDL), which can exhibit enough repulsion force for oil droplet to detach from the pore walls. However, there are still some challenges to confirm if DLE is the dominant mechanism. It was shown that the thickness of waterfilm can be a few angstroms at high salinity[65], and while hypothetically low salinity should result in expansion of EDL, but in some cases atomic force microscopy experiment was incon-sistent with this theory, and a higher disjoining pressure observed in high salinity[66,67]this was attributed to hydration and structural forces, which can eventually dominate in the low range (thin liquidfilm)[49]. However, wettability alteration is the most agreed mechanism during low salinity waterflooding. In this study, geochemical simulation results show that the MIE cannot be credited as the only mechanism of low salinity water Fig. 14. Comparison between sorption of divalent cations CaCl2vs NaCl.
Fig. 15. Recovery factors observed in coreflooding experiment (Nasralla and Nasr-El-Din, 2014).
Fig. 16. Effect of CEC on divalent cation sorption during 1% wt CaCl2solution
flooding, but double layer expansion, induced by multi-component ion exchange, needs to be included to elucidate the mechanism(s) of low-salinity waterflooding.
6. Condition for sequence of affinities
The relative affinity of metal cations to the clay surface depend on solutions, but the best approximation can be justified in terms of inner sphere, outer sphere and diffuse ion swarm [63]. Generally in the inner sphere, the electronic structure of the metal cation and surface function groups are dominant, and in the diffuse swarm layer only valency and surface charge are the governing factors. It should be noted that the solution pH can also extensively affect the extent of the diffuse swarm layer (Zeta potential), thus affecting adsorption process.
Our simulation results (Fig. 19) show that in all of the cases, the pH values were almost constant with different solutions. It is worth noting that the sequence of affinity will depend on charge density and the valency of the metal cations, thus following the below order;
Naþ< Mg2þ< Ca2þ< Hþ
Generally, adsorption isotherm will depend on the cation concentration, valance and hydration radius.
7. Conclusions
Core-scale numerical model was created to investigate the impact of geochemical reaction on low salinity effect using CMG-GEM with a combination of published experimental results. The variation of Naþand Ca2þin the solution was studied, and Ca-X2 absorbed at the surface of rock as injecting 2,000, 10,000, and 50,000 ppm NaCl and CaCl2 solutions. The simulation results were compared with the experimental results. Moreover, we discuss further the mechanism(s) of wettability change induced by low salinity waterflooding. Based on the results, the following conclusions can be drawn;
(1) The presence of divalent cations in low salinity water plays a great role in the adsorption of Ca-X2 to the surface of rock due to the strong affinity to minerals compared with monovalent ions.
(2) To observe the effect of LSWT there is expected to be a concentration difference between monovalent cations and divalent cations to enable monovalent cations substitute the divalent cations on the surface of the clay minerals. (3) Multi-component ion exchange (MIE) is not the only
mechanism to interpret the low salinity effect. Divalent Fig. 17. Effect of CEC on divalent cation sorption during 1% wt NaCl solution injection. Fig. 18 . Lef t different adsorp tion mechanism of organic mat erial right method proposed b y A ustad, R ezaeidoust [64] .
cations fully desorb at the surface of rock during 50,000 and 10,000 ppm NaCl solution injection, while in 2000 ppm solution, approximately 10% of available ex-change sites are saturated with Ca2þ. Based on MIE, it was expected to observe higher recovery with 50,000 and 10,000 ppm than 2000 ppm NaCl. It was not consistent with experimental data as 2000 ppm NaCl resulted in higher recovery.
(4) Double layer expansion needs to be considered to predict the performance of low salinity waterflooding, zeta po-tential test is recommended to conduct to preliminary assess the potential of low salinity water.
(5) The CEC of different clays may not have a significant impact on the ultimate surface saturation with divalent cations, but the surface potential may affect the repulsion forces and consequently the ultimate recovery factor.
References
[1] R.A. Nasralla, H.A. Nasr-El-Din, Impact of cation type and concentration in injected brine on oil recovery in sandstone reservoirs, J. Petroleum Sci. Eng. 122 (2014) 384e395.
[2] S. Shafiee, E. Topal, When will fossil fuel reserves be diminished? Energy Policy 37 (1) (2009) 181e189.
[3] H.J. Longwell, The future of the oil and gas industry: past approaches, new challenges, World Energy 5 (3) (2002) 100e104.
[4] W. Winoto, N. Loahardjo, S.X. Xie, P. Yin, N.R. Morrow, Secondary and tertiary recovery of crude oil from outcrop and reservoir rocks by low salinity waterflooding, in: SPE Improved Oil Recovery Symposium, Society of Petroleum Engineers, Tulsa, Oklahoma, USA, 2012.
[5] N. Morrow, J. Buckley, Improved oil recovery by low-salinity water-flooding, J. Petroleum Technol. 63 (5) (2011) 106e112.
[6] Pu H, Xie X, Yin P, Morrow N. Low-Salinity Waterflooding and Mineral Dissolution. Conference Low-salinity Waterflooding and Mineral Dissolution.
[7] N. Morrow, Low Salinity Waterflood Project EORI Commission Meeting, 2007.
[8] S. Rashid, M.S. Mousapour, S. Ayatollahi, M. Vossoughi, A.H. Beigy, Wettability alteration in carbonates during“Smart Waterflood”: underly-ing mechanisms and the effect of individual ions, Colloids Surfaces A Physicochem. Eng. Aspects 487 (2015) 142e153.
[9] H.S. Al-Hashim, H.Y. Al-Yousef, A. Arshad, A. Mohammadain, Smart Water Flooding of Carbonate Reservoirs: Core Flooding Tests Using a New Approach for Designing Smart Water, Society of Petroleum Engineers, 2015.
[10] A. RezaeiDoust, T. Puntervold, T. Austad, Chemical verification of the EOR mechanism by using low saline/smart water in sandstone, Energy& Fuels 25 (5) (2011) 2151e2162.
[11] S.J. Fathi, T. Austad, S. Strand,“Smart water” as a wettability modifier in chalk: the effect of salinity and ionic composition, Energy& Fuels 24 (4) (2010) 2514e2519.
[12] A. RezaeiDoust, T. Puntervold, S. Strand, T. Austad, Smart water as wetta-bility modifier in carbonate and sandstone: a discussion of similarities/ differences in the chemical mechanisms, Energy& fuels 23 (9) (2009) 4479e4485.
[13] A. Cense, S. Berg, K. Bakker, E. Jansen, Direct Visualization of Designer Water Flooding in Model Experiments, Society of Petroleum Engineers, 2014.
[14] S.C. Ayirala, E. Uehara-Nagamine, A.N. Matzakos, R.W. Chin, P.H. Doe, P.J. van den Hoek, A designer water process for offshore low salinity and polymerflooding applications, Soc. Petroleum Eng. (2014). Article number SPE-129926-MS.
[15] A. Cense, S. Berg, K. Bakker, E. Jansen, Direct visualization of designer water flooding in model experiments, in: SPE Enhanced Oil Recovery Conference. Kuala Lumpur, Malaysia, 2011.
[16] Armstrong R, Wildenschild D. Designer-wet micromodels for studying potential changes in wettability during microbial enhanced oil recovery. Conference Designer-Wet Micromodels for Studying Potential Changes in Wettability during Microbial Enhanced Oil Recovery, vol. 1. p. 1061. [17] Q. Xie, D. Ma, J. Wu, Q. Liu, N. Jia, M. Luo, Potential Evaluation of Ion Tuning
Waterflooding for a Tight Oil Reservoir in Jiyuan OilField: Experiments and Reservoir Simulation Results, Society of Petroleum Engineers, 2015. [18] Q. Xie, Y. Liu, J. Wu, Q. Liu, Ions tuning waterflooding experiments and
interpretation by thermodynamics of wettability, J. Petroleum Sci. Eng. 124 (2014) 350e358.
[19] P.C. Myint, A. Firoozabadi, Thin liquidfilms in improved oil recovery from low-salinity brine, Curr. Opin. Colloid & Interface Sci. 20 (2) (2015) 105e114.
[20] J.J. Sheng, Critical review of low-salinity waterflooding, J. Petroleum Sci. Eng. 120 (2014) 216e224.
[21] Z. Zhuoyan, X. Quan, X. Hanbing, F. Jian, W. Feng, A. Juedu, et al., Evaluation of the potential of high-temperature, low-salinity polymerflood for the Gao-30 reservoir in the Huabei oilfield, China: experimental and reservoir simulation results, in: Offshore Technology Conference, 2015.
[22] P.V. Brady, N.R. Morrow, A. Fogden, V. Deniz, N. Loahardjo, Winoto. Elec-trostatics and the low salinity effect in sandstone reservoirs, Energy& Fuels 29 (2) (2015) 666e677.
[23] Sorop T, Parker AR, Chmuzh IV, Masalmeh SK, Suijkerbuijk B, Skripkin A, et al. Low salinity waterflooding at west salym: laboratory experiments andfield forecasts. Conference Low Salinity Waterflooding at West Salym: Laboratory Experiments and Field Forecasts. Society of Petroleum Engineers.
[24] H. Mahani, S. Berg, D. Ilic, W.-B. Bartels, V. Joekar-Niasar, Kinetics of Low-salinity-flooding Effect, 2014.
[25] S.K. Masalmeh, T.G. Sorop, B.M. Suijkerbuijk, E.C. Vermolen, S. Douma, H. van der Linde, et al., Low salinityflooding: experimental evaluation and numerical interpretation, SCA (2013) 2013e2022.
[26] P.V. Brady, J.L. Krumhansl, P.E. Mariner, Surface complexation modeling for improved oil recovery, in: SPE Improved Oil Recovery Symposium, Society of Petroleum Engineers, Tulsa, Oklahoma, USA, 2012.
[27] P.V. Brady, J.L. Krumhansl, A surface complexation model of oil-ebrineesandstone interfaces at 100C: low salinity waterflooding, J. Petroleum Sci. Eng. 81 (2012) 171e176.
[28] T. Austad, A. Rezaeidoust, T. Puntervold, Chemical mechanism of low salinity waterflooding in sandstone reservoirs, in: SPE Improved Oil Re-covery Symposium. Tulsa, Oklahoma, USA, 2010.
[29] G. Jerauld, K. Webb, C.Y. Lin, J. Seccombe, Modeling low-salinity water-flooding, SPE Reserv. Eval. Eng. 11 (6) (2008) 1000e1012.
[30] G.-Q. Tang, N.R. Morrow, Influence of brine composition and fines migra-tion on crude oil/brine/rock interacmigra-tions and oil recovery, J. Petroleum Sci. Eng. 24 (2e4) (1999) 99e111.
[31] McGuire P, Chatham J, Paskvan F, Sommer D, Carini F. Low salinity oil recovery: an exciting new EOR opportunity for Alaska's North slope. Conference Low Salinity Oil Recovery An Exciting New EOR Opportunity for Alaska's North Slope.
[32] Sandengen K, Arntzen O. Osmosis during low salinity water flooding. Conference Osmosis During Low Salinity Water Flooding.
[33] BP, Low Salinity Water Brings Award for BP, 2015.http://www.bp.com/en/ global/corporate/press/bp-magazine/innovations/offshore-technology-award-for-clair-ridge.html.
[34] J. Seccombe, A. Lager, G. Jerauld, B. Jhaveri, T. Buikema, S. Bassler, et al., Demonstration of low-salinity EOR at interwell scale, Endicott Field, Alaska, in: SPE Improved Oil Recovery Symposium. Tulsa, Oklahoma, USA, 2010. [35] B.M.J.M. Suijkerbuijk, T.G. Sorop, A.R. Parker, S.K. Masalmeh, I.V. Chmuzh,
V.M. Karpan, et al., Low Salinity Waterflooding at West Salym: Laboratory Experiments and Field Forecasts, Society of Petroleum Engineers, 2014. [36] A.A. Yousef, S.H. Al-Salehsalah, M.S. Al-Jawfi, New recovery method for
carbonate reservoirs through tuning the injection water salinity: smart waterflooding, in: SPE EUROPEC/EAGE Annual Conference and Exhibition. Vienna, Austria, 2011.
[37] Q. Xie, D. Ma, J. Wu, Q. Liu, N. Jia, M. Luo, Low Salinity Waterflooding in Low Permeability Sandstone: Coreflood Experiments and Interpretation by Thermodynamics and Simulation, Society of Petroleum Engineers, 2015. [38] K. Skrettingland, T. Holt, M.T. Tweheyo, I. Skjevrak, Snorre
Low-salinity-water Injectionecoreflooding Experiments and Single-well Field Pilot, SPE Reservoir Evaluation& Engineering, 2011 (04).
[39] N. Morrow, J. Buckley, Improved oil recovery by low-salinity water-flooding, J. Petroleum Technol. (2011). Article number SPE-129421-JPT. [40] A. Lager, K.J. Webb, C.J.J. Black, M. Singleton, K.S. Sorbie, Low salinity oil
recovery - an experimental investigation, Petrophysics 41 (1) (2008). [41] A. Lager, K.J. Webb, I.R. Collins, D.M. Richmond, LoSal Enhanced Oil
Re-covery: Evidence of Enhanced Oil Recovery at the Reservoir Scale, Society of Petroleum Engineers, 2008b.
[42] S.Y. Lee, K.J. Webb, I. Collins, A. Lager, S. Clarke, apos, et al., Low Salinity Oil Recovery: Increasing Understanding of the Underlying Mechanisms, Soci-ety of Petroleum Engineers, 2010.
[43] J.C. Seccombe, A. Lager, K.J. Webb, G. Jerauld, E. Fueg, Improving Wateflood Recovery: LoSalTM EOR Field Evaluation, Society of Petroleum Engineers, 2008.
[44] R.A. Nasralla, H.A. Nasr-El-Din, Double-layer expansion: is it a primary mechanism of improved oil recovery by low-salinity waterflooding?, in: SPE Improved Oil Recovery Symposium, Society of Petroleum Engineers, Tulsa, Oklahoma, USA, 2012.
[45] R.A. Nasralla, M.A. Bataweel, H.A. Nasr-El-Din, Investigation of Wettability Alteration by Low Salinity Water, Offshore Europe, Aberdeen, UK, 2011. [46] M.B. Alotaibi, R.M. Azmy, H.A. Nasr-El-Din, Wettability Studies Using
Low-salinity Water in Sandstone Reservoirs, 2010.
[47] Ligthelm D, Gronsveld J, Hofman J, Brussee N, Marcelis F, van der Linde H. Novel waterflooding strategy by manipulation of injection brine compo-sition. Conference Novel Waterflooding Strategy by Manipulation of In-jection Brine Composition.
[48] C.T.Q. Dang, L.X. Nghiem, Z. Chen, P. Nguyen, Modeling Low Salinity Waterflooding: Ion Exchange, Geochemistry and Wettability Alteration, 2013.
[49] G.J. Hirasaki, Wettability; fundamentals and surface forces, SPE Form. Eval. 6 (2) (1991).
[50] N. Morrow, Wettability and its effect on oil recovery, J. Petroleum Technol. 42 (12) (1990).
[51] G. Sposito, Surface Chemistry of Natural Particles, Oxford University Press, Cary, NC, USA, 2004.
[52] E.W. Al-Shalabi, K. Sepehrnoori, A comprehensive review of low salinity/ engineered water injections and their applications in sandstone and car-bonate rocks, J. Petroleum Sci. Eng. 139 (2016) 137e161.
[53] Y.-S. Wu, B. Bai, Efficient Simulation for Low Salinity Waterflooding in Porous and Fractured Reservoirs, Society of Petroleum Engineers, 2009.
[54] L. Yu, S. Evje, H. Kleppe, T. Kårstad, I. Fjelde, S.M. Skjaeveland, Spontaneous imbibition of seawater into preferentially oil-wet chalk cores d Experi-ments and simulations, J. Petroleum Sci. Eng. 66 (3e4) (2009) 171e179. [55] L. Wei, Sequential coupling of geochemical reactions with reservoir
sim-ulations for waterflood and EOR studies, SPE J. 17 (2) (2012) 469e484. [56] L. Nghiem, P. Sammon, J. Grabenstetter, H. Ohkuma, Modeling CO2 Storage
in Aquifers with a Fully-coupled Geochemical EOS Compositional Simu-lator, Society of Petroleum Engineers, 2004.
[57] D.L. Sparks, Soil Physical Chemistry, second ed., Taylor& Francis, 1998. [58] R.A. Nasralla, H.A. Nasr-El-Din, Impact of electrical surface charges and
cation exchange on oil recovery by low salinity water, in: SPE Asia Pacific Oil and Gas Conference and Exhibition. Jakarta, Indonesia, 2011. [59] C. Bethke, Geochemical Reaction Modeling: Concepts and Applications,
Oxford University Press, 1996.
[60] C.A.J. Appelo, D. Postma, Geochemistry, Groundwater and Pollution, second ed., CRC Press, 2004.
[61] D. Carroll, Ion Exchange in Clays and Other Minerals: Geological Society of America Bulletin, 1959.
[62] T.S. Arnarson, R.G. Keil, Mechanisms of pore water organic matter adsorption to montmorillonite, Mar. Chem. 71 (2000,) 3e4.
[63] G. Sposito, The Chemistry of Soils, Oxford University Press, 1989. [64] T. Austad, A. Rezaeidoust, T. Puntervold, Chemical mechanism of low
salinity waterflooding in sandstone reservoirs, in: SPE Improved Oil Re-covery Symposium, 24e28 April, Tulsa, Oklahoma, USA, 2010.
[65] T. Hassenkam, J. Mathiesen, C. Pedersen, K. Dalby, S. Stipp, I.R. Collins, Observation of the Low Salinity Effect by Atomic Force Adhesion Mapping on Reservoir Sandstones, Society of Petroleum Engineers, 2012. [66] S. Basu, M.M. Sharma, Measurement of critical disjoining pressure for
dewetting of solid surfaces, J. Colloid Interface Sci. 181 (2) (1996) 443e455. [67] S. Basu, M.M. Sharma, Characterization of Mixed-wettability States in Oil
