An Experimental Study of WettingBehavior and Surfactant EOR in

Carbonates With Model CompoundsYongfu Wu,* SPE, Patrick J. Shuler, SPE, Mario Blanco, Yongchun Tang, and William A. Goddard III, California

Institute of Technology

SummaryThis study focuses on the mechanisms responsible for enhanced oilrecovery (EOR) from fractured carbonate reservoirs by surfactantsolutions, and methods to screen for effective chemical formula-tions quickly. One key to this EOR process is the surfactant solu-tion reversing the wettability of the carbonate surfaces from lesswater-wet to more water-wet conditions. This effect allows theaqueous phase to imbibe into the matrix spontaneously and expeloil bypassed by a waterflood.

This study used different naphthenic acids (NA) dissolved indecane as a model oil to render calcite surfaces less water-wet.Because pure compounds are used, trends in wetting behavior canbe related to NA molecular structure as measured by solid adsorp-tion; contact angle; and a novel, simple flotation test with calcitepowder. Experiments with different surfactants and NA-treatedcalcite powder provide information about mechanisms responsiblefor sought-after reversal to a more water-wet state. Results indicatethis flotation test is a useful rapid screening tool to identify betterEOR surfactants for carbonates.

The study considers the application of surfactants for EORfrom carbonate reservoirs. This technology provides a new oppor-tunity for EOR, especially for fractured carbonate, where water-flood response typically is poor and the matrix is a high oil-saturation target.

IntroductionTypically only approximately a third of the original oil in place(OOIP) is recovered by primary and secondary recovery processes,leaving two-thirds trapped in reservoirs as residual oil. Approxi-mately half of world’s discovered oil reserves are in carbonatereservoirs and many of these reservoirs are naturally fractured(Roehl and Choquette 1985). According to a recent review of 100fractured reservoirs (Allan and Sun 2003), carbonate fracturedreservoirs with high matrix porosity and low matrix permeabilityespecially could use EOR processes. The oil recovery from thesereservoirs is typically very low by conventional waterflooding, duein part to fractured carbonate reservoirs (about 80%) being origi-nally less water-wet. Injected water will not penetrate easily into aless water-wetting porous matrix and so cannot displace that oil in place.

Wettability of carbonate reservoirs has been widely recognizedan important parameter in oil recovery by flooding technology(Tong et al. 2002; Morrow and Mason 2001; Zhou et al. 2000;Hirasaki and Zhang 2004). Because altering the wettability of arock surface to preferentially more water-wet conditions is criticalto oil recovery, alteration of reservoir wettability by surfactants hasbeen intensively studied, and many research papers have beenpublished (Spinler and Baldwin 2000). Vijapurapu and Rao (2004)studied the capability of certain ethoxy alcohol surfactants to alterwettability of the Yates reservoir rock to water-wet conditions.

Seethepali et al. (2004) reported that several anionic surfactants inthe presence of Na2CO3 can change a calcite surface wetted by aWest Texas crude oil to intermediate/water-wet conditions as wellas, or even better than, an efficient cationic surfactant. Zhang et al.(2004) investigated also the effect of electrolyte concentration,surfactant concentration, and water/oil ratio on wettability alter-ation. They reported that wettability of calcite surface can be al-tered to approximately intermediate water-wet to preferentiallywater-wet conditions with alkaline/anionic surfactant systems. Ad-sorption of anionic surfactants on a dolomite surface can be sig-nificantly reduced in the presence of sodium carbonate.

Xie and co-workers reported that after imbibition of reservoirbrine had ceased, immersion of a field core in surfactant solutioncan produce an additional 5 to 10% OOIP, and they ascribed thisadditional oil recovery to increased water wetness of the core(2004). Serrano-Saldaña et al. (2004) examined wettability condi-tions of solid/brine/n-dodecane systems at various surfactant con-centrations and different ionic strength. They concluded that thewettability in solid/oil/brine systems could be changed by diffu-sion, through the aqueous phase of surfactant species that wereoriginally present in the oil phase while the gradual adsorption ofthese molecules on the solid walls modifies the surface energy.

Standnes reported that spontaneous imbibition (SI) into carbon-ate porous medium can be improved when it is exposed to a waterphase containing cationic surfactants of the type CnTAB (alkyltrimethyl ammonium bromide), and developed a simple analyticalmodel to obtain quantitative information about SI rates of aqueoussurfactant solution (2004). Standnes and Austad (2002) studiednontoxic and low-cost amines as wettability alteration cationicsurfactants in carbonates. The mechanism for the wettability alter-ation using C10-amine is proposed to be desorption of stronglyadsorbed carboxylate groups from the carbonate surface by theformation of ion-pairs with the surfactant monomer.

Bryant and co-workers (2004) studied wettability alteration in-duced by adsorption and removal of amine surfactants of knownmolecular structure on mica surfaces that were exposed to decanesolutions of the surfactants. Low pH conditions that promote pro-tonation of the surfactant amine groups produced the greatest wet-tability alteration. Above a pH of 8 or 9, no adsorbed surfactantmolecule remained on mica surface (Bryant et al. 2004).

It is generally accepted that adsorption of polar compoundsonto rock surface has a significant effect on the wettability ofreservoirs (Tie and Morrow 2005; Morrow 1990; Anderson 1986;Madsen et al. 1996; Madsen and Lind 1998; Thomas et al. 1993;Zou et al. 1997; Legens et al. 1998; Zhang and Austad 2005). Inother words, the wettability of hydrocarbon reservoirs dependson the specific interactions in the oil/rock/brine systems. Naph-thenic acids are the products of extensive oxidation of crude oiland play an important role in wettability control of reservoirs.Carboxylic groups in naphthenic acids from the crude oil are themost strongly adsorbed material onto the rock surface, and theymay act as “anchor” molecules for other surface-active compo-nents present in the crude oil. However, there is only limitedknowledge of the influence of organic acids on the three-phasesystem of oil/brine/rock.

In this paper, we will present and discuss (1) adsorption ofnaphthenic acids (NAs) on calcite powder from n-decane (modeloil) at room temperature and the relationship between molecular

* Now with Dow Chemical Company

Copyright © 2008 Society of Petroleum Engineers

This paper (SPE 99612) was accepted for presentation at the 2006 SPE/DOE Symposiumon Improved Oil Recovery, Tulsa, 22–26 April, and revised for publication. Original manu-script received for review 11 February 2006. Revised manuscript received 20 August 2007.Paper peer approved 21 August 2007.

26 March 2008 SPE Journal

structure of naphthenic acids and their adsorption from non-aqueous media; (2) wettability of the calcite powder pre-treatedwith various naphthenic acids and the influence of molecular struc-ture of naphthenic acids; and (3) contact angle of water surroundedby air on the surface of a calcite crystal pretreated with variousnaphthenic acids. These data also provide insight into the influenceof molecular structure of naphthenic acids on calcite surface en-ergy. In addition, increasing the water wettability of the surfaces ofthe porous matrix by use of surfactant aqueous solution is alsopresented and discussed. Furthermore, data are presented for someselected surfactants on recovery of model oil from limestone corethrough spontaneous imbibition tests.

ExperimentalMaterials. Calcite crystals (Iceland Spar) used in our study formeasurement of contact angle were purchased from WARD’sNatural Science (Rochester, New York). Calcite powder for mea-surements of adsorption of naphthenic acids from non-aqueousphase and flotation test is purchased from Alfa Aesar Company(Ward Hill, Massachusetts) and are activated at 120°C for 2 hoursbefore being used for experiments. The powder has an averageparticle size of 5 �m, and a specific surface area of 1.67 m2/g.

NAs studied in this research come from Aldrich, Inc. (St.Louis, Missouri) and were used without any purification. Thenaphthenic acids investigated are: (1) cyclohexanecarboxylic acid;(2) cyclohexanepropionic acid; (3) cyclohexanebutyric acid; (4)cyclohexanepentanoic acid; and (5) trans-4-pentylcyclohexanecar-boxylic acid. Their molecular structures and related physical pa-rameters as reported by the vendor are listed in Table 1.

Surfactants investigated are mainly divided between a series ofcationic and nonionic chemicals. Table 2 lists these surfactantsand their structure and properties as reported by the respectivechemical vendors. Cationic surfactants are included because lit-erature reports these can have good oil recovery performance. Twocommercial Arquad samples are studied here because they couldbe practical commercial cationic surfactant products. Four purecationic surfactants are included because their defined chemicalstructure allows easier interpretation between their structure andbehavior/oil recovery performance. The nonionic surfactants areincluded in the study because they have been field tested success-fully in fractured carbonate reservoirs. The commercial nonionicproducts selected are low-cost, plus their known differences instructure allow interpretation of their chemical structure and oilrecovery performance. Finally, sodium dodecyl sulfate (SDS) isincluded as a benchmark material, being perhaps the most commonexample of an anionic surfactant. Additional anionic surfactantswere not considered because this was beyond the scope of thecurrent study.

Measurement of Adsorption of NAs on Calcite Surface FromNonaqueous Phase. (1) Prepare naphthenic acid solution in n-decane. Solutions were made from 0.005 to 0.067 M, which isequivalent to a total acid number (TAN) ranging from 0.45 to 5.1for the selected naphthenic acids. TANs were calculated as theamount (in mg) of KOH that would be required to neutralize NAin 1 gram of the oil. (2) Mix 10 ml naphthenic-decane solutionwith 0.5 g calcite powder in a test tube, and shake the test tube atroom temperature for 16 hours to establish adsorption equilibrium.(3) Separate the solution and calcite powder with a centrifuge.Remove the supernatant solution for analysis of the equilibriumconcentration of NAs via GC-MS (Hewlett-Packard HP-G 1800AGCD system). Naphthalene (C10H8) was used as internal standard.

Flotation Test for Wettability of Calcite Powder With Adsorp-tion of NAs. We developed a simple flotation test to demonstratethe relative change in the surface water-wetting conditions for acalcite surface caused by exposure to different NA chemicals. Thistest uses the concept that with a powdered calcite sample (initiallystrongly water-wet) that more of this material will float in water asit is rendered to become less water-wet.

After the measurement of adsorption, the separated calcitepowder in each test tube was dried at 85°C to remove any n-decane. 10 ml of distilled water was added to the tube, followedby vigorous shaking for 2 minutes. The volume of calcite pow-der at the bottom (water-wet portion) and top (less water-wet por-tion) were measured. After sitting for 2 hours, a final reading wastaken for the volume percentage of calcite powder at the top andthe bottom.

Measurement of Contact Angle of Water on Calcite Surface.(1) Clean new calcite crystals by rinsing with heptane and toluene,and then dry the samples in an oven at 85°C for 1 hour. (2) Preparevarious naphthenic acid solutions in decane at 6.62×10−2 M, whichis equivalent to a TAN of 5 for all selected naphthenic acids. (3)Immerse the clean calcite crystals in each naphthenic acid solutionin decane for 24 hours at room temperature. Remove the crystalsand dry them in an oven at 85°C for 1 hour to remove all extrasolvent. (4) Measure the advancing contact angle through a dropletof water that sits on the pretreated calcite crystal surface sur-rounded by air and at room temperature with an Advanced Goni-ometer (Model 500, Rame-Hart, Inc.). The crystal sample wasplaced in an environmental chamber saturated with distilled water.The contact angle was recorded every minute until the change ofcontact angle is less than 0.2° within a 10-minute interval. (5)Break a large calcite crystal to small pieces in order to get a freshsurface. Measure advancing contact angle of water on the newsurface, using the same method as described in Step 4.

27March 2008 SPE Journal

28 March 2008 SPE Journal

Model Oil for Surfactant Performance Testing. Based on theresults of the experiments described previously, a model oil com-position was selected that makes a calcite surface much less water-wet. The model oil used for the remainder of the study that inves-tigated surfactant effects was selected: cyclohexanepentanoic acidat 1.48 wt% in –decane. This is equivalent to a TAN of 4.50.

Flotation Test of Wettability Alteration by Selected SurfactantAqueous Solutions. Aqueous surfactant solutions were added totest tubes at different concentrations (100, 50, and 25 ppm) con-taining powdered calcite pretreated with the model oil. Almost allof this powder floats in the absence of any surfactant (the water-only case). After shaking the test tube vigorously for 2 minutes, itsat for at least 2 hours. The volume of calcite powder at the bottomand top was measured. If there was foam at the top, the bubbleswere broken before taking a reading. The larger the fraction of thepowder that sinks, the better the surfactant’s performance in chang-ing the surface wettbility back to a more water-wet condition.

Interfacial Tension (IFT) Measurement. In order to study equi-librium phase behavior at the oil and aqueous solution interface,the model oil and surfactant aqueous solution were mixed in a testtube in 1:1 volume ratio. The test tubes were shaken at roomtemperature and left standing for at least two weeks to achievephase equilibrium. The IFT between the top oil layer and bottomwater layer was measured by a spinning drop interfacial tensiom-eter, Model 510 from Temco, Inc. An oleic phase drop (2 �l) wasplaced into a glass tube containing the aqueous phase, and spun athigh speed. Spinning continued until observing an equilibriumcondition (normally<2 hours), as indicated by no change in dropshape for 30 minutes at a temperature of 30°C.

Spontaneous Imbibition Test of Model Oil Recovery. The lastseries of experiments compares the ability of each of these differ-ent surfactants to recover the model oil phase from outcrop lime-stone 2.5 cm×5.1 cm cores. The air permeability of these cores isfairly low, ranging from 5 to 20 md. The limestone cores were firstdried at 120°C for 2 hours to remove adsorbed moisture. Aftercooling to room temperature, the cores were placed in a vacuumsystem for 4 hours and the model oil was introduced and allowedto saturate the cores overnight. Then the saturated cores wereplaced into Amott cells (Hirasaki and Zhang 2004; Standnes 2004)

containing the various surfactant solutions at a concentration of 0.4wt% in distilled water. Amott cells are glass beakers with a buretteattached to the top of the beaker. As the aqueous phase imbibesinto the core, oil is expelled from the core and floats into thevolumetric burette. The cells were maintained at room temperatureand the oil recovery was monitored vs. time.

Results and DiscussionAdsorption of Naphthenic Acids on Calcite Surface From N-Decane Media. Adsorption isotherms of selected NAs on calcitesurface from n-decane are shown in Fig. 1. In general, adsorptionof the NAs on calcite surface from n-decane media is in the order:cyclohexanepropionic acid>cyclohexanebutyric acid>cyclo-hexanepentanoic acid>cyclohexanecarboxylic acid>trans-4-pentylcyclohexane carboxylic acid. Because cyclohexanepropi-onic acid, cyclohexanebutyric acid, cyclohexanepentanoic acid,and cyclohexanecarboxylic acid are analogues, it indicates thatadsorption of the NAs decreases with increase of alkyl chainlength, with the exception of cyclohexanecarboxylic acid. Thismay be explained by the interaction between alkyl chain of NAand n-decane molecules. The longer the alkyl chain, the strongerthe interaction between acid and solvent molecules; this reducesthe NA adsorption. As to the exception of cyclohexanecarboxylicacid, the steric exclusion of cyclohexane ring directly connected tothe carboxyl group in its molecules reduces its adsorption. Stericexclusion also may explain the low adsorption of trans-4-pentylcyclohexane carboxylic acid. The adsorption layer is formedby orientation of carboxyl groups toward calcite surface becausethe surface carries positive charges (Legens et al. 1998).

Note that if the data in Fig. 1 are expressed in more typicalengineering units of adsorption as mg of NA per g of calcitepowder vs. the TAN, the adsorption amounts still have the samesequence of highest to lowest.

Flotation Test of Calcite Powder Treated With DifferentNaphthenic Acids. The general procedure for this flotation testmethod has been described previously.

Photos of the flotation test of calcite powder treated with dif-ferent NAs are shown in Fig. 2. Tubes 1 to 5 contain powderedcalcite samples treated with the five selected NAs separately. Thevolume of the less water-wet (floating) powder for the five inves-tigated NAs is in the following order: trans-4-pentylcyclohexanecarboxylic acid∼cyclohexanepentanoic acid>cyclohexane butyricacid>cyclohexanepropionic acid>cyclohexane carboxylic acid. Itis almost in reverse order of adsorption on calcite surface. Thisindicates that their ability to alter calcite surface to become less

Fig. 1—Adsorption isotherms of naphthenic acids on calcite inn-Decane solution (23°C, 16 hours).

Fig. 2—Flotation test of calcite powder treated with NA at a TANof approximately 0.5. Liquid phase: distilled water (pH∼6). Pow-der treated with NA: (1) cyclohexanecarboxylic acid; (2) cyclo-hexanepropionic acid; (3) cyclohexanebutyric acid; (4) cyclo-hexanepentanonic acid; (5) trans-4-Pentylcyclohexane carbox-ylic acid; (6) without treatment.

29March 2008 SPE Journal

water-wet depends on their molecular structures only. For ex-ample, although its adsorption is smallest among the five investi-gated acids, trans-4-pentylcyclohexane carboxylic acid reduces thewater-wetting nature of the calcite powder the most. On the otherhand, for the blank sample, powdered calcite sample with no NAtreatment, completely sinks in the water (Tube 6), indicating thatthe calcite surface remains in its original strong water-wet condition.

The volume percentages of the powder that floats in the waterat different NA molar concentrations are shown in Fig. 3. It can beseen that the wettability alteration of calcite surface by selectedNAs increases gradually with an increased concentration for cy-clohexanepropionic and cyclohexanebutyric acids. The wettabilitychanges only very slightly for the other NAs.

Wettability of Calcite Crystal (Iceland Spar) Surfaces TreatedWith Different NAs. The contact angle through a droplet of wateron a solid calcite surface surrounded by air is another measure ofsurface wettability. This was done on the calcite crystal surfacepretreated with different NAs and the results are shown in Fig. 4.Note that after 50 minutes when equilibrium is reached, the contactangle of water on the calcite surface treated with the selected NAsis in the order: trans-4-pentylcyclohexanecarboxylic>cyclo-hexanepentanoic>cyclohexanebutyric>cyclohexanepropionic>cy-clohexanecarboxylic>fresh calcite surface (without treatment ofNA). This is exactly in the same order as the flotation results andso agrees with that method results.

The untreated calcite surface has the smallest contact angle forwater, which is 21°. The contact angle decreases with time for thevarious NAs. This may be because of trace amounts of the ad-sorbed NA layer on the calcite being transferred from the treatedsurface to the water phase because of solubility effect. As thisoccurs, there is less NA on the surface and so the contact anglegradually decreases until no more phase transfer occurs. This ex-planation is supported by the fact that the contact angle changesvery little for the blank samples.

Furthermore, these results indicate, as expected, that the degreeof induced decrease in water-wetting increases as the NA is morehydrophobic. For example, cyclohexanepentanoic acid (alkylchain with five carbons) increases water contact angle more thanthe cyclohexanepropionic acid (alkyl chain with only three car-bons) or cyclohexanecarboxylic acid (no alkyl chain). It is recog-

nized that such increase of water contact angle is caused by de-crease of surface energy of calcite surface. This calcite surfaceenergy can be evaluated by Neumann’s Equation-of-State (Li andNeumann 1992):

cos� = −1 + 2��SV

�LVe−���LV−�SV�2, . . . . . . . . . . . . . . . . . . . . . . . (1)

where �LV is surface tension of water, 72 mN/cm at 25°C, and �SV

is the interfacial tension at the interface of solid and vapor. In ourcase, it can be used for estimation of the surface energy of calcitesurface; � is a constant of 0.0001247 m4/mJ2. Although this con-stant was originally obtained with polymer surfaces, it has beenused successfully as an approximation for �SV calculation of natu-ral surfaces such as apatite crystals (Suzuki et al. 2004). Even if theactual value of this constant is a bit different for calcite, the cal-culated surface energy data (solid-vapor surface tension) of thecalcite surface treated with various NAs (Table 3) have the samerelative ranking. From these data, the fresh calcite surface has thehighest surface energy. It indicates again that calcite surface isoriginally water wet. For the surfaces treated with NAs, the sur-

Fig. 3—Oil-wet calicite in powder flotation test (v/v) vs. eq. con-centration (M) of naphthenic acid.

Fig. 4—Water contact angles on calcite surface.

30 March 2008 SPE Journal

face energy decreases with an increase of –CH2– groups in theNA molecules. For example, a calcite surface treated with cy-clohexanecarboxylic acid has almost the same energy as freshsurface because there is no –CH2– group in the molecule. On theother hand, the surface treated with trans-4-pentylcyclohexane-carboxylic acid has the lowest energy as it has one –CH3 and four–CH2– groups.

Wettability Alteration Using Surfactant Aqueous Solutions atDifferent Concentrations. In order to investigate the ability ofsurfactants to reverse the treated calcite surface to strongly water-wet conditions, eight surfactants were selected. Calcite powderwas treated with model oil (n-decane containing 1.48 wt.% cyclo-hexanepentanoic acid, TAN�4.50). The wettability alteration re-sults are listed in Table 4. This shows the percentage of the pre-treated calcite that can be made to sink (made more water-wet) bymixing with different candidate surfactant solutions. Of the sur-factants tested by this procedure, the commercial cationic surfac-tant, Arquad T-50, has the best performance, with just a 25-ppmconcentration, causing more than half of the floating calcite pow-der to now sink. Of the nonionic surfactants, the Igepal CO-530has the best performance, showing at 50 ppm approximately 95%of the floating calcite powder sink to the bottom of the tube. Theone anionic surfactant, sodium dodecyl sulfate (SDS), has essen-tially no effect on changing the wetting of the treated calcite powder.

One factor that could be important to the wetting reversal andimbibition performance of a surfactant is its ability to remove theNA compound from the carbonate surface. Presumably if thesecomponents are stripped away from the surface, then the carbonatewould have the desired water-wet condition. This is exactly themechanism proposed for the action of cationic surfactants(Standnes and Austad 2000; Barenblatt et al. 2002). These authorsspeculate that the cationic surfactants form ion-pairs with the dis-sociated NA anions in the aqueous phase, and that this actionprovides a means to transport the adsorbed NA from the surface.These same authors hypothesize that the mechanism for wettingreversal for nonionic and anionic surfactants is not the removal ofthe surface-absorbed NA species, but instead that these surfactantsco-adsorb on the carbonate surface. This so-called bilayer adsorp-tion, wherein the surfactants have strong hydrophobic-hydro-phobic interaction with the adsorbed NA species, leaves the polarhead group of the surfactants sticking out into the bulk solution.The hydrophilic groups of these adsorbed surfactants then providea water-wet layer near the surface.

Measurements were performed by GC-MS to examine the fateof the adsorbed NA on the calcite surface. The starting point is thetest tube samples from the flotation test described previously. Be-cause all of the NA initially is on the calcite powder, any NAdetected in the aqueous surfactant solution must representNA desorbed from the calcite surface. The GC-MS method iscalibrated by running samples of known concentration of cy-

clohexanepentanoic acid solution dissolved in n-decane. Un-known samples are taken from the aqueous surfactant solutions inthe previous calcite powder flotation test. Results are shown inTable 5. All of the samples show detectable amounts of NA aretransported into the aqueous phase. As expected, the cationic sur-factants desorbed more of the NA from the calcite surfaces, be-cause strong ion-pairing should have enhanced that process. Theanionic SDS surfactant appears to have removed a significantamount of the NA, but still had little success in changing thewetting as indicated by the flotation test. This is because strongadsorption of SDS itself on calcite surface reduces the water-wetcondition remarkably. This was demonstrated in a simple test: mix1 g of new calcite powder with 10 g of 1000 ppm surfactantaqueous solution in a test tube and shake it vigorously. For theSDS, all of the calcite powder floats on the aqueous phase. How-ever, for other cationic and nonionic surfactants, the powder sinksin the aqueous phase.

Spontaneous Imbibition Test of Porous Limestone Cores. As afollow-up to the flotation test, 12 surfactants were selected forspontaneous imbibition test to evaluate their ability to recover oilfrom porous limestone core material. Experimental procedureswere described previously. All surfactant solutions were preparedwith distilled water at 0.4 wt.% concentration and the test wasconducted at 25°C. The selected surfactants are seven ionic sur-factants, including one anionic and six cationic surfactants, andfive nonionic surfactants. Most of them are commercial products.Molecular structure, critical micelle concentration, HLB values,and IFT results, as well as cumulative oil recovery are listed inTable 2.

In general, the results show: (1) Oil recovery by the cationicsurfactants is between 40 and 60%, except for n-decyl trimethy-lammonium bromide (C10TAB), which can recover only 12% ofthe oil. (2) Oil recovery by use of nonionic surfactants is between10 and 20%, except for nonylphenoxypoly ethanol (Igepal CO-530), which has around 50% oil recovery. (3) Oil recovery by thesodium dodecyl sulfate (SDS) is low, with only 6.8% of the oil. (4)There is some rough correlation between the observed oil recoveryand the IFT. Surfactants with high oil recovery (>40%) showgenerally a low IFT (∼0.5 mN/m). However, for the cationic sur-factants, n-decyl triphenylphosphonium bromide (C10TPPB) andn-dodecyl triphenylphosphonium bromide (C12TPPB), the oil re-covery is higher than 50%, but the IFT is also relatively high, at 3.6and 2.0 mN/m, respectively. The low IFT cases may include grav-ity effects in their oil recovery, whereas a case with high IFT likelyhas oil recovery controlled by a uniform imbibition process. (5)There is no obvious relationship between oil recovery and criticalmicelle concentration (CMC). Molar concentrations of these sur-factants at 0.4 wt.% were calculated and are listed in the table.They all are higher than the CMC of the surfactants except forC10TAB and C12TAB. For the C10TAB, however, it may show less

31March 2008 SPE Journal

oil recovery in part because it is far below its CMC; if the mecha-nism relies on this cationic surfactant forming aqueous complexesbetween its monomers and the adsorbed NA, then not maximizingits monomer concentration could hurt its performance. For non-ionic surfactants, their molar concentrations are greater than theirCMC by two or three orders of magnitude.

The cumulative oil recovery curves for ionic and nonionic sur-factants are shown in Figs. 5 and 6, respectively. These plots usethe square root of the time (days) as the x-axis because this way thedata plot as a straight line where diffusion processes are involved(Barenblatt et al. 2002). Among the six cationic surfactants, duringthe early time (less than 5 days), the recovery rates are almost thesame. This indicates that early oil recovery is governed by imbi-bition of water near the surface and subsurface around the lime-stone core. Once the pores are filled by surfactant aqueous solu-tion, the surfactant molecules move to the next pores. As theprocess continues, the recovery rate will depend significantly ondiffusion of surfactant molecules. Faster diffusion aids in the pen-etration of the surfactant molecules, and hence oil recovery. For agiven surfactant and a porous core, higher surfactant concentrationand temperature should improve diffusion rates and hence oil re-covery performance (Standnes and Austad 2003).

One important point is that if diffusion is the major drivingforce toward oil recovery, then this surfactant EOR process faceslimitations when the matrix blocks are large. For example, it hasbeen calculated that surfactant penetration could require hundredsof years to penetrate as far as one meter into a matrix block(Adibhatla et al. 2005). Thus, in carbonate reservoirs with fewfractures and a relatively small fracture face area, the percent of oilrecovery may be limited.

Fig. 7 and Table 6 show that duplicate runs of several of thesurfactants show good reproducibility. As another check on theexperimental procedure, one imbibition oil recovery was per-formed with only water. As expected, this blank run with no sur-factant had very little recovery. This demonstrates the core soakedwith and containing the model oil has a low water-wetting condi-tion. This same behavior of minimal spontaneous imbibition bywater (no surfactant) was noted in additional tests by the model oiland also a West Texas crude oil with a moderate acid number (datanot shown).

The four cationic surfactants, C10TAB, C12TAB, ARQUADC-50, and ARQUAD T-50, are quaternary ammonium salts withdifferent alkyl chains. The shorter chain C10TAB has relativelypoor recovery as compared to these other three products. The othertwo cationic surfactants, C10TPPB and C12TPPB, show the bestperformance in this test series. These are quaternary phosphoniumsalts with a C10 and C12 straight alkyl chain, respectively. Be-cause of their very bulky hydrophilic head, their molecules cannot

Fig. 5—Oil recovery measured by spontaneous imbibition test(ionic surfactants, 0.4 wt.%, 25°C).

Fig. 6—Oil recovery measured by spontaneous imbibition test(nonionic surfactants, 0.4 wt.%, 25°C).

Fig. 7—Repeat oil recovery results by spontaneous imbibitiontest (nonionic surfactants, 0.4 wt.%, 25°C).

32 March 2008 SPE Journal

pack tightly at oil/water interface. Therefore, both of them do notproduce a low IFT at the interface. The mechanism responsible foroil recovery by this kind of phosphonium surfactants is currentlyunknown. But this, perhaps, indicates a new direction of candidateselection for EOR in fractured carbonate reservoirs.

The SDS is included as a benchmark anionic surfactant for ourtest program. This solution recovers only 7% of the oil. Its poorperformance may be attributable in part to strong adsorption ofSDS on the limestone surface because of a strong electrostaticattraction. Therefore, SDS molecules are prevented from diffusinginto the core pores and forcing oil recovery. The literature doesreport, however, reduced anionic surfactant adsorption and goodoil recovery if the aqueous surfactant solution is at high pH and hasan optimal salinity for the specific reservoir conditions (Hirasakiand Zhang 2004).

For the five nonionic surfactants used in our spontaneous im-bibition test, they are ethoxylated primary or secondary alcoholswith a linear or a branched alkyl chain as shown in Table 2.Among them, Tergitol 15-S-3, Tergitol 15-S-7, Tergitol 15-S-40,and Neodol 25-7 recover limited amounts of oil from the limestonecore. The ethoxylated nonyl phenol Igepal CO-530 (Rhodia, Inc.)has by far the best performance of these nonionic surfactants,recovering as much as 50% from limestone core. This is compa-rable to the oil recovery by the cationic surfactants, C12TAB,ARQUAD C-50, and ARQUAD T-50. Note that this observationis consistent with the result of the wettability alteration flotationtest discussed in the previous section of this paper. Another fea-ture of the Igepal CO-530 is it has almost the lowest IFT oftested surfactants.

Conclusions1. Adsorption of naphthenic acids on calcite surface in n-decane

media is in the following order: cyclohexanepropionicacid>cyclohexanebutyric acid>cyclohexanepentanoic acid. Be-cause these three naphthenic acids are analogues in terms ofmolecular structure, this indicates that adsorption of the NAsdecreases with increase of alkyl chain length. As to cyclohex-anecarboxylic and trans-4-pentylcyclohexane carboxylic acids,their adsorption is similar. Adsorption of trans-4-pentyl-cyclohexanecarboxylic acid is slightly less than that of cyclo-hexanecarboxylic acid because the former has a longer straightalkyl chain of five carbons.

2. For the flotation test, the trend in the reduction in water-wettingof calcite powder with different naphthenic acids is in the fol-lowing order: trans-4-pentylcyclohexane carboxylic acid∼cy-clohexanepentanoic acid>cyclohexanebutyric acid>cyclo-hexanepropionic acid>cyclohexanecarboxylic acid. It is almostin reverse order of adsorption on calcite surface. This indicatesthat their ability to alter calcite surface to become less water-wetis not related directly to their adsorption on calcite surface, butdepends on their molecular structures. As to calcite powderwithout treatment of naphthenic acid, it is originally stronglywater-wet.

3. Contact angle and novel flotation test results are consistent inranking the relative wetting condition. At equilibrium, the con-tact angle through water in air on the calcite surface pretreated

with naphthenic acids is in the following order: trans-4-pentylcyclohexanecarboxylic∼cyclohexanepentanoic>cyclo-hexanebutyric>cyclohexanepropionic>cyclohexanecarboxylic>fresh calcite surface. This is in agreement with the wettingchanges observed in the flotation test results.

4. Among the 12 surfactants studied, cationic surfactants are gen-erally more efficient in recovering model oil from limestonecore than the others, but one nonionic surfactant, Igepal CO-530, has also been found to be efficient for oil recovery. For thetwo quaternary phosphonium cationic surfactants, C10TPPB andC12TPPB, these phosphonium surfactants with bulky headgroups recovered the model oil in limestone cores most efficiently.

Nomenclature� � constant in equation (1), m4/mJ2

�LV � IFT between liquid and vapor, mN/m�SV � IFT between solid and vapor, mN/m

� � contact angle through water droplet degrees

AcknowledgmentsThis work was conducted at the California Institute of Technologyand financially supported by the Department of Energy Grant DE-FC26-04BC15521. The authors thank Chevron and Akzo Nobelfor their collaboration.

ReferencesAdibhatla, B., Sun, X., and Mohanty, K.K. 2005. Numerical Studies of Oil

Production From Initially Oil-Wet Fracture Blocks by Surfactant BrineImbibition. Paper SPE 97687 presented at the SPE International Im-proved Oil Recovery Conference in Asia Pacific, Kuala Lumpur, 5–6December. DOI: 10.2118/97687-MS.

Allan, J. and Sun, S.Q. 2003. Controls on Recovery Factor in FracturedReservoirs: Lessons Learned From 100 Fractured Fields. Paper SPE84590 presented at the SPE Annual Technical Conference and Exhi-bition, Denver, 5–8 October. DOI: 10.2118/84590-MS.

Anderson, W.G. 1986. Wettability Literature Survey—Part I: Rock/Oil/Brine Interactions and the Effects of Core Handling on Wettability. JPT38 (10): 1125–1144. SPE-13932-PA. DOI: 10.2118/13932-PA.

Barenblatt, G.I., Patzek, T.W., and Silin, D.B. 2002. The MathematicalModel of Non-Equilibrium Effects in Water-Oil Displacement. PaperSPE 75169 presented at the SPE/DOE Improved Oil Recovery Sym-posium, Tulsa, 13–17 April. DOI: 10.2118/75169-MS.

Bryant, E.M., Bowman, R.S., and Buckley, J.S. 2004. Wetting Alterationof Mica Surfaces with Polyethoxylated Amine Surfactants. 8th Inter-national Symposium on Reservoir Wettability, Houston, 16–18 May.

Hirasaki, G. and Zhang, D.L. 2004. Surface Chemistry of Oil RecoveryFrom Fractured, Oil-Wet, Carbonate Formations. SPEJ 9 (2): 151–162.SPE-88365-PA. DOI: 10.2118/88365-PA.

Legens, C., Toulhoat, H., Cuiec, L., Villieras, F., and Palermo, T. 1998.Wettability Change Related to the Adsorption of Organic Acids onCalcite: Experimental and Ab Initio Computational Studies. Paper SPE49319 presented at the SPE Annual Technical Conference and Exhi-bition, New Orleans, 27–30 September. DOI: 10.2118/49319-MS.

Li, D. and Neumann, A.W. 1992. Equation of state for interfacial tensionsof solid-liquid systems. Advances in Colloid and Interface Science 39(6): 299–345. DOI: 10.1016/0001-8686(92)80064-5.

Madsen, L. and Lind, I. 1998. Adsorption of Carboxylic Acids on Reser-voir Minerals From Organic and Aqueous Phase. SPEREE 1 (1): 47–51. SPE-37292-PA. DOI: 10.2118/37292-PA.

Madsen, L., Grahl-Madsen, L., Gron, C., Lind, I., and Engell, J. 1996.Adsorption of Polar Aromatic Hydrocarbons on Synthetic Calcite. Org.Geochem. 24 (12): 1151–1155. DOI: 10.1016/S0146-6380(96)00096-4.

Morrow, N.R. 1990. Wettability and Its Effect on Oil Recovery. JPT 42(12): 1476–1484; Trans., AIME, 289. SPE-21621-PA. DOI: 10.2118/21621-PA.

Morrow, N.R. and Mason, G. 2001. Recovery of Oil by SpontaneousImbibition. Current Opinion in Colloid and Interface Science 6 (4):321–337. DOI: 10.1016/S1359-0294(01)00100-5.

Roehl, P.O. and Choquette, P.W. 1985. Carbonate Petroleum Reservoirs.New York City: Springer-Verlag.

33March 2008 SPE Journal

Seethepalli, A., Adibhatla, B., and Mohanty, K.K. 2004. PhysicochemicalInteractions During Surfactant Flooding of Fractured Carbonate Res-ervoirs. SPEJ 9 (4): 411–418. SPE-89423-PA. DOI: 10.2118/89423-PA.

Serrano-Saldaña, E., Domínguez-Ortiz, A., Pérez-Aguilar, H., Kornhauser-Strauss, I., and Rojas-González, F. 2004. Wettability of Solid/Brine/n-Dodecane System: Experimental Study of the Effects of Ionic Strengthand Surfactant Concentration. Colloids & Surfaces A: PhysicochemEng. Aspects 241 (1–3): 343–349. DOI: 0.1016/j.colsurfa.2004.04.025.

Spinler, E.A. and Baldwin, B.A. 2000. Surfactant Induced WettabilityAlteration in Porous Media. In Surfactant Fundamentals and Applica-tion in the Petroleum Industry, ed. L.L. Schramm, Cambridge, UK:Cambridge University Press, 159–202.

Standnes, D.C. 2004. Analysis of Oil Recovery Rates for SpontaneousImbibition of Aqueous Surfactant Solutions into Preferential Oil-WetCarbonates by Estimation of Capillary Diffusivity Coefficients. Col-loids & Surfaces A: Physicochem Eng. Aspects 251 (1–3): 93–101.DOI: 10.1016/j.colsurfa.2004.09.013.

Standnes, D.C. and Austad, T. 2000. Wettability Alteration in Chalk 2.Mechanism for Wettability Alteration from Oil-Wet to Water-Wet Us-ing Surfactants. J. of Petroleum Science & Engineering 28 (3): 123–143. DOI: 10.1016/S0920-4105(00)00084-X.

Standnes, D.C. and Austad, T. 2002. Non-Toxic and Low Cost Amines asWettability Alteration Chemicals in Carbonates. 7th International Sym-posium on Reservoir Wettability, Tasmania, Australia, 12–14 March.

Standnes, D.C. and Austad, T. 2003. Wettability Alteration in Carbonates:Interaction Between Cationic Surfactants and Carboxylates as a KeyFactor in Wettability Alteration from Oil-Wet to Water-Wet Condi-tions. Colloids & Surfaces A: Physicochemical & Engineering Aspects216 (1–3): 243–259. DOI: 10.1016/S0927-7757(02)00580-0.

Suzuki, T., Hirose, G., and Oishi, S. 2004. Contact Angle of Water Dropleton Apatite Single Crystals. Materials Research Bulletin 39 (1–3): 103–108. DOI: 10.1016/j.materresbull.2003.09.013.

Thomas, M.M., Clouse, J.A., and Longo, J.M. 1993. Adsorption of OrganicCompounds on Carbonate Minerals 1. Model Compounds and TheirInfluence on Mineral Wettability. Chemical Geology 109 (1–4): 201–213. DOI: 10.1016/0009-2541(93)90070-Y.

Tie, H. and Morrow, N. 2005. Low-Flood-Rate Residual Saturation inCarbonate Rocks. Paper SPE 10470 presented at the SPE InternationalPetroleum Technology Conference, Doha, 21–23 November. DOI:10.2118/10470-MS.

Tong, Z.X., Morrow, N.R., and Xie, X. 2002. Spontaneous Imbibition forMixed-Wettability States in Sandstones Induced by Adsorption FromCrude Oil. 7th International Symposium on Reservoir Wettability, Tas-mania, Australia, 12–14 March.

Vijapurapu, C.S. and Rao, D.N. 2004. Compositional Effects of Fluids onSpreading, Adhesion and Wettability in Porous Media. Colloids &Surface A: Physicochem. Eng. Aspects 241 (1–3): 335–342. DOI:10.1016/j.colsurfa.2004.04.024.

Xie, X., Weiss, W.W., Tong, Z., and Morrow, N.R. 2004. Improved OilRecovery from Carbonate Reservoirs by Chemical Stimulation. SPEJ10 (3): 276–285. SPE-89424-PA. DOI: 10.2118/89424-PA.

Zhang, P. and Austad, T. 2005. The Relative Effects of Acid Number andTemperature on Chalk Wettability. Paper SPE 92999 presented at the

SPE International Symposium on Oilfield Chemistry, Houston, 2–4February. DOI: 10.2118/92999-MS.

Zhang, D.L., Liu, S., Puerto, M., Miller, C.A., and Hirasaki, G. J. 2004.Wettability Alteration and Spontaneous Imbibition in Oil-Wet Carbon-ate Formations. 8th International Symposium on Reservoir Wettability,Houston, 16–18 May.

Zhou, X., Morrow, N.R., and Ma, S. 2000. Interrelationship of Wettability,Initial Water Saturation, Aging Time and Oil Recovery by SpontaneousImbibition and Waterflooding. SPEJ 5 (2): 199–207. SPE-62507-PA.DOI: 10.2118/62507-PA.

Zou, L. et al. 1997. Enthalpy of Adsorption and Isotherms for Adsorptionof Naphthenic Acid Onto Clays. J. Colloid & Interface Science 190 (2):472–475. DOI: 10.1006/jcis.1997.4898.

SI Metric Conversion Factors°F (°F–32)/1.8 � °Cin. × 2.54* E+00 � cm

*Conversion factor is exact.

Yongfu Wu now works with the Dow Chemical Company as achemistry specialist, and previously was a staff research scien-tist at the California Institute of Technology. His research inter-ests are in the area of surfactant and polymer science, with anemphasis on chemical EOR and reservoir conformace controltechnology. Wu holds BS and MS degrees from the University ofScience and Technology of China, and MS and PhD degreefroms the City University of New York, all in chemistry. Patrick J.Shuler is on the research staff at the California Institute of Tech-nology. His research interests include use of oilfield chemicalsassociated with production operations, especially for EOR.Shuler holds a BS degree from the University of Notre Dame,and MS and PhD degrees from the University of Colorado, all inchemical engineering. Mario Blanco is a senior researcher atthe California Institute of Technology. His research interests in-clude application of chemicals in several industries, includingoil and gas production. He holds a BS degree from the Univer-sidad Del Valle de Guatemala and a PhD degree from theUniversity of California at Los Angeles, both in chemistry. Yong-chun Tang is director of the Power Environmental Energy Re-search Center at the California Institute of Technology. He isalso serving as professor and director for the Green Chemicaland Environmental Energy Research Institute of Shanghai Uni-versity. His major research interests are applying molecularmodeling and experimental simulation techniques for manyenergy-related areas. Tang holds a BS degree in chemistryfrom Shanxi University and a PhD degree from Ohio University.William A. Goddard, III is the Charles and Mary Ferkel Professorof Chemistry, Material Science and Applied Physics and direc-tor of Materials and Process Simulation Center at Cal Tech. Hisresearch interest is modeling of complex chemical systems,including those applied in oil and gas production. He holds aBS degree from the University of California at Los Angeles anda PhD degree from the California Institute of Technology, bothin engineering science.

34 March 2008 SPE Journal