Wettability Alteration in Gas-Condensate Reservoirs To Mitigate Well Deliverability Loss by Water Blocking

Wettability Alteration in Gas-Condensate Reservoirs To Mitigate Well Deliverability Loss by Water Blocking Myeong Noh* and Abbas Firoozabadi, RERI Summary Liquid blocking in some gas-condensate reservoirs is a serious problem when the permeability is low (for example, of the order of 10 md or less). The current practice centers mainly on hydraulic fracturing to improve gas flow. In most cases, the frequency of application results in high costs. An alternative is the permanent alteration of wettability from liquid-wetting to preferentially gaswetting. In this work, we present an experimental study of wettability alteration to preferential gas-wetting using a multifunctional surfactant and polymer synthesized for this particular application. The treatments are performed with an alcohol-based-surfactant/ polymer solution. We treat Berea cores and reservoir-rock samples from two gas-condensate reservoirs. In one of the reservoirs, water blocking has resulted in a significant reduction of well deliverability. The treatment provides significant improvement on the phase mobility. In this study, our focus is the investigation of water/gas two-phase flow at high temperatures (80 and 140 C). Basic measurements such as contact angle, spontaneous imbibition, and the effect on the absolute permeability are discussed. The initial liquid saturation at the time of treatment may have an influence on the wettability alteration. The results of the treatment on oil-saturated and water-saturated cores are presented. The treatment by alcohol without using the polymer is compared and discussed. Two-phase-flow tests in single-phase and two-phase injections are performed before and after the treatment using brine and gas. Relative permeabilities of gas and water are measured, and the improvement after the treatment is presented. Various measurements in our work show that water and gas relative permeability increase significantly in a wide range, especially at high liquid saturation. Introduction The exploitation of low-permeability gas-condensate reservoirs has received increased emphasis in recent years. Liquid trapping (water or condensate blocking) around a wellbore is one of the major causes of reduced productivity in low-permeability gascondensate reservoirs. Formation damage by the loss of aqueous fluids by operations such as drilling, fracturing, or acidizing is a potential source of reduced productivity of gas reservoirs because it causes water accumulation (water blocking) near the wellbore. Water blocking is caused by capillary pressure, which tends to imbibe and hold the liquid phase, resulting in a reduction of gas mobility. The decrease in water saturation from hydraulic fracturing around the wellbore is a slow process when the permeability is low and the capillary forces are high. The investigation of the gas/water relative permeabilities can determine the gas-mobility loss associated with the change of water saturation. Early studies by McLeod and Coulter (1966), and McLeod et al. (1966) showed that the alcoholic acid mixture of isopropyl alcohol, or methanol, and hydrofluoric acid/hydrochloric acid (HF- HCl) helps the water-removal and cleanup rate in gas wells. They tested various compositions of the acid mixture, including alcohol; * Currently with Chevron Corporation. Copyright 2008 Society of Petroleum Engineers Original SPE manuscript received for review 1 June 2005. Revised manuscript received for review 17 March 2008. Paper (SPE 98375) peer approved 19 March 2008. the treatment improved the displacement of the fluid by decreasing the interfacial tension and increasing water vaporization in gas reservoirs, particularly those with high clay content. They also suggested that the alcohol treatment would be more effective in a high-pressure and -temperature reservoir because of lower interfacial tension. Poor gas production in low-permeability reservoirs after application of hydraulic fracturing is discussed in many papers (Tannich 1975; Liao and Lee 1993; Abrams and Vinegar 1985). Liquid invasion and leakoff during this operation can reduce well productivity. Tannich (1975) showed that the cleanup of water blocking is fast when the fracture conductivity is very high. Therefore, we can expect rapid cleanup when the reservoir is highly permeable because the capillary pressure tends to be small. Liao and Lee (1993) presented numerical studies of liquid cleanup and gasproduction performance for hydraulically fractured gas wells. They showed that the parameters for water blocking and cleanup are capillary pressure and relative permeability. When the reservoir rock is damaged by fracturing-fluid invasion, the pressure drop should be large enough to overcome the capillary end effect of the damaged zone at the fracture surface. Abrams and Vinegar (1985) used the capillary entry pressure to estimate the pressure drop required to initiate and maintain the gas flow into a brinesaturated core. On the basis of their work, Abrams and Vinegar concluded that the pressure drawdown has to be much higher than the capillary entry pressure to avoid significant water blocking. Cimolai et al. (1993) and Bennion et al. (1993) presented coreflood studies demonstrating that reservoirs with low initial water saturation readily accommodate the invaded water. They suggested that the improvement and reuse of drilling fluids can improve horizontal-well performance markedly by minimizing formation damage. Kamath and Laroche (2003) discussed the cleanup of water blocking in gas wells based on two regimes: (1) displacement of the fluids from the formation followed by (2) vaporization of water by the flowing gas. The second regime gives slow displacement, and the deliverability slowly increases during several months. Even if the flowing gas is fully saturated with water, water saturation can be reduced because the gas becomes undersaturated as the pressure decreases. They suggested that the conventional laboratory data exaggerate the deliverability loss caused by water blocking and that the alcohol treatment can improve the recovery significantly. Mahadevan and Sharma (2003) studied the effect of permeability, wettability, temperature, and surfactants on the cleanup of water blocking. The evaporation regime lasted for thousands of pore volumes (PV) of gas flow resulting in a slow cleanup and improvement in gas relative permeability. Mahadevan and Sharma (2003) suggested that the cleanup of water blocking can be improved by addition of solvent (methanol) to increase the vaporization rate increase of permeability change of wettability of the rock from water-wet to oil-wet by introducing surfactant increase in temperature increase in drawdown Wettability alteration to intermediate gas-wetting in porous media by treatment with a fluorochemical polymer has been established by researchers at RERI (Li and Firoozabadi 1979; Tang and Firoozabadi 2002, 2003). Tang and Firoozabadi (2002) measured the relative permeability before and after the treatment for Berea 676 August 2008 SPE Reservoir Evaluation & Engineering

sandstone using the fluorochemical polymer, FC-722. In most cases, the treatment increases the endpoint relative permeability of the liquid phase and decreases the irreducible liquid saturation and endpoint gas relative permeability. As a result, liquid mobility increases, while in some cases, there may be a reduction in gas relative permeability. But even when there is a reduction in gas relative permeability, because of the substantial increase in liquid relative permeability, the treatment results in an increase of gasflow rate for a given pressure drop, as will be shown later in this paper. The effect of the chemical treatment may depend on many factors, including initial fluid saturations, type of rock, reservoir temperature, and, most importantly, the type and composition of the chemical additives. The main purpose of this work is the study of chemical additives for two reservoir rocks to quantify the improvement of gas mobility after the treatment. Since the recent studies (Li and Firoozabadi 1979; Tang and Firoozabadi 2002, 2003), we have tested new chemicals in different types of core samples. In this paper, we present experimental results using the Fluorochemical Surfactant 11-12P and the Fluoroacrylate Copolymer L-19062. We use sandstone-core plugs from two different gas-condensate reservoirs and Berea cores. In the first set of experiments, we examine the effectiveness of the chemical additive on a particular rock type by measuring contact angle and spontaneous imbibition. Thereafter, we perform various flow tests with different chemical concentrations. In this paper, we show that the wettability of reservoir cores can be altered from oil-wetting to preferentially gas-wetting state. The results from the flow tests demonstrate the effectiveness of wettability alteration from improved gas flow in gas-water systems at 140 and 80 C. Fluids and Rocks Brine (1.0 wt% of NaCl) and nitrogen are used as aqueous and gaseous phases, respectively, in the experiments. Density and viscosity of brine are 1.0 g/cm 3 and 0.97 cp at 20 C. Gravity of pure nitrogen gas is 0.967. The interfacial tension between gas and water is 71 dyne/cm at the ambient condition. For the oleic phase, normal decane (nc 10 ) is used. Specific gravity of nc 10 is 0.73. In the past, we have compared imbibition- and flow-test results of nc 10,nC 14, and actual condensates. Results were comparable. We tested six Berea, and eight reservoir cores from two different gas-condensate reservoirs; the relevant properties are shown in Table 1. Some of the absolute-permeability data were measured by outside laboratories, and they are compared to our measurements. For some cases, a reduction in absolute permeabilities is detected after the treatment. A procedure to minimize the reduction of permeability has been devised. The results will be discussed in this work. High velocity coefficients are calculated by plotting gas-flow rates and pressure data. The absolute permeability and highvelocity coefficients are calculated using (Katz 1959; Firoozabadi and Katz 1979) p 2 1 p 2 2 M 2 ZRTL W A = 1 k + W A....(1) Here p 1 and p 2 are the inlet and outlet pressure, respectively; M and are molecular weight and viscosity of gas, respectively; W/A represents mass flux; k is the absolute permeability; R and Z are the gas constant and the gas deviation factor, respectively; T is temperature; and L is the length of the core. High-velocity coefficients may increase after the treatment; that is consistent with permeability reduction. For permeability measurements, we use outlet pressure higher than 50 psi to reduce the Klinkenberg effect. Apparatus Single-phase-water- and two-phase-injection tests in gas-saturated cores are performed to show the mobility improvement after the treatment. Fig. 1 shows a schematic of the experimental setup. A core holder is placed in an oven horizontally. Gas injection is controlled by a pressure regulator. Water is injected with a constant flow rate, and gas-flow rate is measured at the outlet. Fevang and Whitson (1996) suggested an experimental technique for measuring k rg as a function of k rg /k ro. Two-phase relative permeability experiments were proposed using the pseudosteadystate technique to measure relative permeabilities under conditions that are similar to the near-well region of a gas-condensate reservoir (Mott et al. 2000; Cable et al. 1999). They showed that the upper limit of k rg /k ro is approximately 10 for rich condensates and approximately 50 for lean condensates. An upper limit of 50 applies to practically all gas-condensate reservoirs because k rg is relatively high at k rg /k ro >50 and the extrapolation to higher k rg /k ro is straightforward. Therefore, condensate blocking depends only on relative permeabilities within the range of 1<k rg /k ro <50, and this range represents approximately 0.05<k rg or k ro <0.3. Here we measure the relative permeability using nitrogen/water two-phase flow focusing on 1<k rg /k rw <100. For two-phase-injection tests, we August 2008 SPE Reservoir Evaluation & Engineering 677

Fig. 1 Schematic of the apparatus for flow measurements. inject water for 20 PV to saturate the core and then inject water and gas simultaneously. Water- and gas-flow rates are recorded at steady state. Next, we decrease the water-flow rate and repeat the measurements to obtain drainage relative permeability data. Darcy s law is applied to calculate water relative permeability. For gas relative permeability, however, the high velocity coefficient from single-phase gas flow may not be valid in two-phase flow because it can be affected by the presence of a second phase. The reduction of effective porosity and the increase in effective tortuosity results in higher effective resistance when water is present. Some authors (Liu et al. 1995; Frederick and Graves 1994; Ali et al. 1997) include the relative resistance change in the presence of the second phase. Here we follow the correlation by Liu et al. (1995) and use the two-phase/single-phase ratio high-velocity coefficient given by rg k rg 1. The two-phase high velocity coefficient can be written as 2 = 1 rg = 1 k rg....(2) Combining Eqs. 1 and 2, we obtain the expression for gas-phase relative permeability: k rg = 1 k + W A 1 p 2 1 p 2 1 2 M....(3) 2 ZRTL W A Treatment We present the basic evidence of wettability alteration after the treatment. Fluorochemical Surfactant 11-12P is used to alter the wettability of core plugs to preferentially gas-wetting. We also use Fluoroacrylate Copolymer L-19062. We have tested two different batches of 11-12P. For convenience, the first batch and the second batch are named 11-12Pa and 11-12Pb, respectively. Contactangle measurements, spontaneous-imbibition tests, and the effect of the treatment on absolute permeability are discussed. Procedure and Adsorption. In this section, we first present the procedure for treatment for Core C411. Then we provide data for all the other cores. Core C411 is placed in an oven at 140 C. The alcohol-based solution with 11-12Pb is injected for 5 PV at 200 psig. The solution used in the treatment for this core consists of 83% ethanol, 10% water, 3% acetic acid, and 4% 11-12Pb. The surfactant solution has only 2% 11-12Pb; the rest of the 98% is the solvent (ethanol). The core is aged overnight (15 hours) at 200 psig and 140 C. Thereafter, we inject 100 PV of water to find out if the water injection washes out the surfactant adsorbed on the pore surface. The adsorption after washing with water is calculated using the weight difference before and after the treatment: weight increase after treatment Ĉ = weight after treatment = mg of chemical g of core....(4) Unlike Core C411, which is treated with 4-wt%-surfactant solution, Core H122 is treated with 2 wt% of 11-12Pb. The adsorption is 1.6 and 0.53 mg/g for Cores C411 and H122, respectively. Core O319 is treated with 4% 11-12Pa, and the adsorption is 0.24 mg/g. Chemical L-19062 is used in the treatment of core plugs H131, H113, and C422. The adsorption results are summarized in Table 2. Generally, the treatment with high concentration results in a higher adsorption; also, the tighter the core, the lower the adsorption. A small permeability reduction is expected because of the adsorption of the chemical onto the pore surface. However, in some cases, there is a significant reduction in permeability because of extensive chemical adsorption in the inlet part before the solution spreads throughout the core. The result is a significant reduction of permeability. The pretreatment procedure is applied before the chemical injection. The pretreatment solution consists of water, acid, and ethanol, and it is injected into a core for 5 PV. Next, we inject 5 PV of various concentrations of L-19062 and ethanol so that the reaction occurs only in the core. We have recently carried 678 August 2008 SPE Reservoir Evaluation & Engineering

out a comprehensive study on the nature and extent of chemical reaction. Results will be published at a later date. The two-step treatment procedure is applied in a Berea core with 12% L-19062; there is only 5% reduction of permeability. The concentrations of 2 and 4% L-19062 are also applied in tight reservoir cores. The results are presented in Table 1. Note that with the pretreatment, there is little change in permeability. Contact Angle. We perform a simple test to determine the wettability of samples. When a drop of nc 10 is placed on the untreated Core H122, immediate imbibition occurs and the drop becomes invisible rapidly (Fig. 2a). However, the water drop stays on the untreated sample and forms a contact angle of approximately 90. Therefore, the untreated core sample is strongly oil-wetting. After the treatment with 2% 11-12Pb solution, the contact angles of water and nc 10 change, as shown in Fig. 2b. The water drop forms a contact angle at approximately 150, and nc 10 drop shows a contact angle of approximately 60. The results in Core O319 show a smaller contact angle (20 ) for water before the treatment; after the treatment with 4% 11-12Pb, the results are similar to those of Core H122 (Fig. 2d). Core C411 behaves similarly. 4% treatment with L-19062 in Core C422 shows similar water contact angle, but nc 10 imbibes slowly, as shown in Fig. 2f. Visual observation shows that the wettability of the core sample is altered after the treatment, but there is little difference in contact angle with respect to the chemical concentration (see Figs. 2b and d). Spontaneous Imbibition. To quantify the wettability change of a core sample, we use the spontaneous-imbibition tests with water and nc 10 at ambient conditions. The air-saturated core is immersed in the liquid, and the weight is recorded throughout the test. Fig. 3 shows that water and oil saturations in the untreated Core C411 are approximately 0.15 and 0.37 at the test termination, respectively. As mentioned before, the core is strongly oil-wetting and the nc 10 imbibition shows twice as much saturation as water. A comparison of water and nc 10 imbibition before and after the treatment is shown in Fig. 3. Altering wettability from oil-wetting to preferentially gas-wetting reduces liquid imbibition. Water saturation decreases from 0.17 to 0.1, and nc 10 imbibition decreases from 0.36 to 0.21. Therefore, the treatment reduces liquid saturation by approximately 40% at termination. The result of spontaneous imbibition of Core H122 with 2% 11-12Pb treatment is shown in Fig. 4; 2% treatment results in approximately 50% reduction of water saturation. S w decreases from 0.12 to 0.06. The 2% treatment with 11-12Pb is as effective as 4% treatment in the spontaneous imbibition of water. Fig. 2 Contact angle before and after treatment; H122 before treatment (a); H122 after treatment (b); O319 before treatment (c); O319 after treatment (d); C422 before treatment (e); and C422 after treatment (f). August 2008 SPE Reservoir Evaluation & Engineering 679

Fig. 3 Spontaneous imbibition of water and nc 10 for Core C411 before and after treatment. The treatment with 11-12Pa in Core O319 results in less imbibition than that with 11-12Pb, as shown in Fig. 5. The water saturation decreases from 0.37 to 0.04, and oil saturation decreases from 0.47 to 0.24. The treatment reduces water imbibition by 90% and nc 10 imbibition by 50% at 500 minutes. Because the water contact angle in Core O319 changes significantly (from 20 to 150 ) from the treatment, compared to Cores C411 and H122, the reduction in water imbibition is also more pronounced. Absolute Permeability. The absolute permeability is measured using nitrogen. Before the treatment, the permeability of Cores C411 and H122 is 1.37 and 27 md, respectively. After the treatment with 11-12Pb, the absolute permeabilities decrease to 0.65 md for Core C411 and 21 md for Core H122. There is approximately 50% reduction of permeability by 4% treatment and approximately 22% reduction by 2% treatment. We have used a different batch 11-12Pa with various concentrations in various other core samples and have not observed a considerable reduction in permeability. For example, the permeability of Core O319 stays the same after the treatment. The pretreatment process solves the problem of permeability reduction to a large degree (see Table 1). Initial Liquid Saturation. We have examined the effect of initial liquid saturations on the treatment. Five treatments with different Fig. 5 Spontaneous imbibition of water and nc 10 for Core O319 before and after treatment. Fig. 4 Spontaneous water imbibition for Core H122 before and after treatment. initial liquid saturations have been performed using Berea cores. The experimental conditions are described in Table 3. For 8% 11-12P treatment, 79% ethanol, 10% water, and 3% acetic acid are used. For the alcohol treatment (BrC), we use 87% ethanol. After each treatment, we performed water- and nc 10 -imbibition tests. The results are shown in Figs. 6a and 6b, respectively. The alcohol treatment without fluorochemical in Berea Core BrC shows less water imbibition than the untreated imbibition test. Water saturation decreases by 18%, and nc 10 saturation is approximately the same. Compared to 11-12P treatment, however, alcohol treatment (BrC) shows considerably more liquid imbibition. 8% treatment (LBm5) with 11-12Pa in dry core reduces water saturation by 95% and oil saturation by 80% at termination. For the test in Berea cores, treatment with 11-12P is more effective for water imbibition than for nc 10 imbibition. The treatment with L-19062 in BrF shows slightly more water imbibition compared with that with 11-12P. However, the nc 10 imbibition is almost the same as alcohol-treated Core BrC and untreated Core BrD. The imbibition results are consistent with contact-angle tests. When the cores are initially saturated with liquid either water or nc 10 the effect of treatment is still significant in imbibition. The treatment with liquid-saturated cores may be slightly less effective than that in the dry cores, as shown in Fig. 6. However, these differences are small. The treatment effectiveness is, therefore, independent of the initial-saturation state. Flow Test This section includes experimental results of two-phase-flow tests using single-phase and two-phase injections. We inject the aqueous phase in a gas-saturated core. Next, water and gas are injected simultaneously and flow rate and pressure drop are measured at steady state. Two-phase relative permeabilities are calculated and compared before and after the treatment. Single-phase nc 10 injection in a gas-saturated core is also tested at a high temperature. Single-Phase Injection. The gas-saturated core, C411, is placed in an oven at 80 C with the overburden pressure of 1,000 psig. An aqueous phase is injected at 1 cm 3 /min, and the pressure drop is recorded with time. The results are shown in Fig. 7 for the untreated and treated core. At steady state, at approximately 30 PV, the untreated core has 107 psi and the treated core has 69 psi of pressure drop. As mentioned before, the absolute permeability decreases from 1.37 to 0.65 md without pretreatment. Considering this reduction, the relative mobility of the aqueous phase increases significantly. The relative permeability of water at steady state increases from 0.39 to unity after the treatment. Note that because of permeability reduction, the core has become heterogeneous and the reduction is mainly toward the inlet. As a result, the relative permeabilities at steady state may not have physical meaning. 680 August 2008 SPE Reservoir Evaluation & Engineering

The pressure-drop history is shown in Fig. 8 for 2% treatment in Core H122. Because we perform this experiment with two different injection rates, the data are presented in terms of the dimensionless pressure drop defined by p D =ka p/q w w L. The dimensionless pressure drop at 30 PV decreases from 4.9 for the untreated core to 3.3 for the treated core at 2-cm 3 /min of water injection rate. At 4-cm 3 /min water-injection rate, the dimensionless pressure drop decreases from 4.7 to 3.0. Therefore, the water relative permeability at steady state increases from 0.2 to 0.31 for 2-cm 3 /min of water injection rate and from 0.21 to 0.33 for 4-cm 3 / min water-injection rate. Water shows 60 to 70% of mobility increase after the treatment. Unlike the spontaneous imbibition results, the flow test shows that the 11-12Pb treatment with higher concentration (4%) in Core C411 is more effective than 2% treatment in Core H122. We also perform a series of water-injection tests in gassaturated Berea cores at ambient conditions (20 C). As noted before, these cores are treated at different initial liquid saturations (see Tables 1 and 3). Fig. 9 shows that all three treatments with 11-12P behave similarly and dimensionless pressure drop stabilizes at approximately 1.2. The water relative permeabilities of three treated cores at steady state are approximately 0.83. The relative permeability of water is 0.98 in Core BrF for the 12% L-19062 treatment at steady state. Dimensionless pressure drop for the untreated core and the alcohol-treated core are 2.67 and 2.13, respectively. These values correspond to water relative permeabilities of 0.38 and 0.47, respectively, at steady state. Similar to the spontaneous-imbibition results in Fig. 6, the alcohol treatment increases water mobility but treatments with 11-12P and L-19062 at any initial liquid saturations improve the phase mobility beyond the untreated core or the alcohol-treated core. In subsequent measurements, we find that the treatment with alcohol results in a temporary change of wettability. Water and nc 10 injection at 140 C were also performed in Core O319. There was no permeability reduction from treatment. Water and nc 10 are injected at 0.2 cm 3 /min, and the pressure drop is recorded with time. The results are shown in Fig. 10 before and after the treatment. For aqueous-phase injection in Core O319, the pressure drop at 20 PV decreases from 44 psi (P D 2.76) to 21 psi (P D 1.38) at 0.2 cm 3 /min of water-injection rate. The water relative permeability at steady state increases from 0.35 to 0.72. In the case of nc 10 injection, the untreated core has 40.4 psi (P D 2.57) of pressure drop and the treated core has 36.3 psi (P D 2.31) at 5.5 PV. Therefore, the mobility of the oleic phase at steady state increases by 10%. The test shows that the selected chemical is effective without the reduction of permeability. The relative mo- Fig. 6 Water and nc 10 imbibition with various initial saturations in Berea cores; water imbibition (a) and nc 10 imbibition (b). Fig. 7 Pressure drop at water-injection rate of 1cm 3 /min in gassaturated core: C411, 80 C. August 2008 SPE Reservoir Evaluation & Engineering 681

Fig. 8 Dimensionless pressure drop for water injection at two rates: H122, 80 C. Fig. 9 Dimensionless pressure drop with different treatment conditions for water injection: Berea cores, 20 C. bility of the aqueous phase doubles from the treatment. However, 4% 11-12Pa treatment increases oleic-phase mobility by only 10% at steady state. The treatment with 4% L-19062 improves water relative permeability at steady state by 54 and 120% in Cores H113 and C422, respectively (see Fig. 11). Because the water relative permeability at steady state for untreated Core H113 is relatively high (0.53) compared with Core C422 (0.32), Core C422 has more room for improvement. The single-phase water-injection test in Core O334 shows more improvement compared with Cores H113 and C422. The water relative permeability at steady state for Untreated Core O334 is 0.22, and it increases to 0.58 after treatment with 4% of L-19062. From single-phase-injection tests, we find a considerable increase in liquid mobility even with the absolute permeability decrease. The treatment with higher concentration provides further improvement. Two-Phase Injection (Relative Permeability Measurement). Relative permeability measurement is performed in Berea Cores BrD, LBm5, and BrF. Core BrD is untreated, and Cores LBm5 and BrF are treated with 8% 11-12Pa and 12% L-19062, respectively (see Table 3). Fig. 12 presents the relative permeability measurements for treated and untreated cores. We use the constant pressure drop of 30 psi throughout the experiments in Berea cores. The data from Tang and Firoozabadi (2002) are also plotted in Fig. 12. The treatment with 2% FC-722 shows only slight improvement compared to the untreated condition; treatments by 11-12Pa and L-19062 clearly show further improvement in Berea cores. Fig. 13 shows relative permeability data in which constant pressure drops of 120 and 25 psi are used for Cores C411 and H122, respectively. Fig. 13 shows the increase of gas relative mobility when k rw is large. When k rw is small, it seems that the gas mobility decreases after the treatment. Note that the region with small k rw represents large k rg /k rw ( 100 or larger) and relatively low water saturation, in which the water-blocking problem may not be significant. When the core is treated with the 2% 11-12Pb, there is no significant increase in gas relative permeability; on the other hand, there is a significant increase with the 4% treatment. Fig. 14 shows relative permeability data of untreated core and treated core with L-19062. Unlike the previous results, we observe that the gas relative permeability is improved throughout the whole range of k rw. Fig. 14 also shows that the treatment with higher concentration improves mobility further. We apply different pressure drops for relative permeability measurements in Cores H121 and H122, which are from the same formation and close to each other. Whitson et al. (1999) showed that the plot of k rg vs. k rg /k ro shows a dependency on the capillary number and that gas relative permeability increases with capillary number. The experimental study by Tang and Firoozabadi (2001) suggested that the effect of pressure gradient is more significant in intermediate-wet state compared to strongly water-wet state. Fig. 15 presents the gas and water relative permeabilities at three different pressure drops. The relative permeabilities in Fig. 15 do not represent the same capillary number. Generally, a high pressure drop is associated with high capillary number. For the untreated core, Fig. 15a shows that the relative permeability at the pressure drops of 100 and 200 psi is the same. However, for the treated core in Fig. 15b, gas relative permeability increases with increase in pressure drop. This result is consistent with the studies of Whitson et al. (1999) and Tang and Firoozabadi (2001). We mentioned earlier that the spontaneous imbibition in some cores treated with 2% and 4% concentrations is the same. However, flow tests show that the treatment with the higher concentration provides further improvement in relative permeability. Conclusions We have altered the wettability of cores from two reservoirs, with permeabilities in the range of 0.3 to 27 md, and Berea sandstone cores. For one of the reservoirs, water blocking around the wellbore is a major issue. In the other reservoir, hydrocarbon blocking is significant. Results presented in the paper show that for both reservoir rocks, wettability alteration can be effective for reducing water blocking. These conclusions are based on the treatment at reservoir temperature of 140 C, as well as flow testing at 140 C. From the experimental results of this work, the following conclusions can be drawn: 1. The wettability of reservoir rock can be altered from oil-wetting to preferentially gas-wetting by treatment with 11-12P and L-19062. 2. The spontaneous-imbibition tests show 40 to 90% reduction of water saturation for the reservoir rocks after the treatment. Increasing surfactant concentration has negligible effect on water imbibition for some concentration range. 3. In some cases, the treatment decreases absolute permeability. When there is permeability reduction, the problem can be overcome by using the pretreatment process. 4. The treatment is effective with or without initial liquid saturation (oil or water). 5. The single-phase water-injection tests in the gas-saturated core show considerable increase in water mobility. The treatment with higher concentration results in higher water mobility. The effect of the treatment is the same at various temperatures (20 to 140 C). 682 August 2008 SPE Reservoir Evaluation & Engineering

Fig. 11 Dimensionless pressure at water-injection rate of 1 cm 3 /min (O334), 1 cm 3 /min (H113), and 0.5 cm 3 /min (C422) in gas-saturated cores: 140 C. Z gas deviation factor high-velocity coefficient viscosity phase Subscript D dimensionless value g gas o oil r relative value w water Fig. 10 Dimensionless pressure drop for oil and water injection at 0.2 cm 3 /min in gas-saturated core: O319, 140 C; water injection (a) and oil injection (b). Acknowledgments This work was supported by the member companies of the Reservoir Engineering Research Institute (RERI). Their support is appreciated. We also thank the 3M Corporation for providing the chemicals. The valuable suggestions by and discussion with Manny Arco of 3M Corporation are highly appreciated. 6. The gas relative permeability may increase in a wide range from the treatment. The treatment with higher chemical concentration may be more effective in the concentration range we studied. We also observe that there is an increase in mobility in the gas/oil system. However, the improved mobility is much less pronounced than the gas/water system. Our belief is that we have now developed a process for water blocking for field testing in gascondensate reservoirs. Nomenclature A area k absolute permeability L length M molecular weight p pressure q flow rate R universal gas constant S saturation T temperature W mass rate Fig. 12 Effect of treatment on relative permeability: Berea cores, 20 C. August 2008 SPE Reservoir Evaluation & Engineering 683

Fig. 13 Effect of treatment on relative permeability: C411 and H122, 80 C. Fig. 14 Effect of treatment on relative permeability: H131 and C422, 20 C. References Abrams, A. and Vinegar, H.J. 1985. Impairment Mechanisms in Vicksburg Tight Gas Sands. Paper SPE 13883 presented at the SPE/DOE Low Permeability Gas Reservoirs Symposium, Denver, 19 22 March. DOI: 10.2118/13883-MS. Ali, J., McGauley, P.J., and Wilson, C.J. 1997. The Effects of High- Velocity Flow and PVT Changes Near the Wellbore on Condensate Well Performance. Paper SPE 38923 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, 5 8 October. DOI: 10.2118/38923-MS. Bennion, D.B., Cimolai, M.P., Bietz, R.F., and Thomas, F.B. 1993. Reductions in the Productivity of Oil and Gas Reservoirs due to Aqueous Phase Trapping. Paper CIM 93-24 presented at the Annual General Meeting of the Petroleum Society of CIM, Calgary, 9 12 May. Cable, A., Mott, R., and Spearing, M. 1999. Experimental Techniques for the Measurement of Relative Permeability and In-Situ Saturation in Gas-Condensate Near-Wellbore and Drainage Studies. Paper 9928 presented at the International Symposium of the Society of Core Analysts, Golden, Colorado, 2 4 August. Cimolai, M.P., Gies, R.M., Bennion, D.B., and Myers, D.L. 1993. Mitigating Horizontal Well Formation Damage in a Low-Permeability Conglomerate Gas Reservoir. Paper SPE 26166 presented at the SPE Gas Technology Symposium, Calgary, 28 30 June. DOI: 10.2118/26166-MS. Fevang, Ø. and Whitson, C.H. 1996. Modeling Gas-Condensate Well Deliverability. SPERE 11 (4): 221 230. SPE-30714-PA. DOI: 10.2118/ 30714-PA. Firoozabadi, A. and Katz, D.L. 1979. An Analysis of High-Velocity Gas Flow Through Porous Media. JPT 31 (2): 211 216. SPE-6827-PA. DOI: 10.2118/6827-PA. Frederick, D.C. and Graves, R.M. 1994. New Correlations to Predict Non-Darcy Flow Coefficients at Immobile and Mobile Water Saturation. Paper SPE 28451 presented at the SPE Annual Technical Conference and Exhibition, New Orleans, 25 28 September. DOI: 10.2118/ 28451-MS. Kamath, J. and Laroche, C. 2003. Laboratory-Based Evaluation of Gas Well Deliverability Loss Caused by Water Blocking. SPEJ 8 (1): 71 80. SPE-83659-PA. DOI: 10.2118/83659-PA. Katz, D.L. 1959. Handbook of Natural Gas Engineering. New York City: McGraw-Hill Higher Education. Li, K. and Firoozabadi, A. 2000. Experimental Study of Wettability Alteration to Preferentially Gas-Wetting in Porous Media and Its Effects. SPEREE 3 (2): 139 149. SPE 62515-PA. DOI: 10.2118/62515-PA. Liao, Y. and Lee, W.J. 1993. Production Performance of Hydraulically Fractured Gas Wells. Paper SPE 25465 presented at the SPE Production Operations Symposium, Oklahoma City, Oklahoma, 21 23 March. DOI: 10.2118/25465-MS. Liu, X., Civan, F., and Evans, R.D. 1995. Correlation of the Non-Darcy Flow Coefficient. J. Cdn. Pet. Tech. 34 (10): 50 53. Fig. 15 Effect of pressure drop on relative permeability for untreated and treated cores; untreated core is H121, 20 C (a) and treated core is H122, 20 C (b). 684 August 2008 SPE Reservoir Evaluation & Engineering

Mahadevan, J. and Sharma, M. 2003. Clean-Up of Water Blocks in Low Permeability Formations. Paper SPE 84216 presented at the SPE Annual Technical Conference and Exhibition, Denver, 5 8 October. DOI: 10.2118/84216-MS. McLeod, H.O. and Coulter, A.W. 1966. The Use of Alcohol in Gas Well Stimulation. Paper SPE-AIME 1663 presented at the SPE Eastern Regional Meeting, Columbus, Ohio, 10 11 November. DOI: 10.2118/ 1663-MS. McLeod, H.O., McGinty, J.E. and Smith, C.F. 1966. Deep Well Stimulation with Alcoholic Acid. Paper SPE 1558 presented at the Annual Meeting of the Society of Petroleum Engineers of AIME, Dallas, 2 5 October. DOI: 10.2118/1558-MS. Mott. R.E., Cable, A.S., and Spearing, M.C. 2000. Measurements of Relative Permeabilities for Calculating Gas-Condensate Well Deliverability. SPEREE 3 (6): 473 479. SPE-68050-PA. DOI: 10.2118/68050-PA. Tang, G.-Q. and Firoozabadi, A. 2001. Effect of Pressure Gradient and Initial Water Saturation on Water Injection in Water-Wet and Mixed- Wet Fractured Porous Media. SPEREE 4 (6): 516 524. SPE-74711-PA. DOI: 10.2118/74711-PA. Tang, G.-Q. and Firoozabadi, A. 2002. Relative Permeability Modification in Gas/Liquid Systems Through Wettability to Intermediate Gas Wetting. SPEREE 5 (6): 427 436. SPE-81195-PA. DOI: 10.2118/81195-PA. Tang, G.-Q. and Firoozabadi, A. 2003. Wettability Alteration to Intermediate Gas-Wetting in Porous Media at Elevated Temperatures. J. of Transport in Porous Media 52 (2): 185 211. Tannich, J.D. 1975. Liquid Removal from Hydraulically Fractured Gas Wells. JPT 27 (11): 1309 1317. SPE-5113-PA. DOI: 10.2118/ 5113-PA. Whitson, C.H., Fevang, Ø., and Sævareid, A. 1999. Gas Condensate Relative Permeability for Well Calculations. Paper SPE 56476 presented at the SPE Annual Technical Conference and Exhibition, Houston, 3 6 October. DOI: 10.2118/56476-MS. SI Metric Conversion Factors API 141.5/(131.5+ API) g/cm 3 cp 1.0* E 03 Pa s dyne 1.0* E 02 mn F ( F 32)/1.8 C psi 6.894 757 E + 00 kpa *Conversion factor is exact. Myeong Noh is a research engineer at Chevron Corporation in Houston. His research interests include multiphase flow in porous media, carbon dioxide sequestration, and enhanced oil recovery for heavy-oil reservoirs. Noh holds a BS degree from Hanyang University, Seoul, South Korea, and MS and PhD degrees from the University of Texas at Austin, all in petroleum engineering. Abbas Firoozabdi is senior scientist and director at the Reservoir Engineering Research Institute (RERI) in Palo Alto, California and a professor of chemical engineering at Yale University, New Haven, Connecticut. His main area of focus is hydrocarbon energy production and environmental stewardship on the basis of thermodynamics and mathematical analysis. Firoozabadi holds a BS degree from the Abadan Institute of Technology, Abadan, Iran and MS and PhD degrees from the Illinois Institute of Technology, Chicago, all in gas engineering. He is the recipient of the 2002 SPE Anthony Lucas Gold Medal, and the 2004 SPE John Franklin Carll Award. August 2008 SPE Reservoir Evaluation & Engineering 685