SPE 58711 An Optimal Foam Quality for Diversion in Matrix-Acidizing Projects

Jose M. Alvarez, Hercilio Rivas and Geidy Navarro, PDVSA-INTEVEP, Venezuela

Copyright 2000, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 2000 SPE International Symposium on Formation Damage Control held in Lafayette, Louisiana, 2324 February 2000. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

the foam quality condition for a particular matrix-acidizing project obtaining a more cost-effective job. Introduction Matrix acidizing is defined as a treatment used to remove damage near the wellbore 3 . This treatment intends to dissolve any damage and some particles of the porous medium increasing or reestablishing its permeability. The technique consists in injecting a reactive fluid (acid) into the porous medium at pressure bellow the fracturing pressure. The most common acids used in the petroleum industry are HCl and HF. Matrix acidizing has been used extensively in the oil industry. However, its success depends strongly in the proper placement of the acid into the more damage zones 4,5 . Unfortunately, much of the acid flows through the less-damage zones, leaving the damage area almost untreated. Therefore, diversion techniques are required in order to improve the efficiency of any matrix-acidizing job. In the last two decades, foams have been widely used as diverting agent in matrix-acidizing processes. Factors favoring the use of foam over solid diverters are the cost of the latter and the lack of confidence in the effectiveness by using solids for long intervals 6 . Bernard and Holm7 were the first to indicate that foam could selectively block high-permeability zones, or by analogy less-damage zones. Smith et al.8 proposed the use of foam as a diverter agent in acidizing jobs. They found, using two parallel coreholders packed with sand with a permeability ratio of 20, that 99% of the fluid was diverted into the less-permeable layer. Burman and Hall9 concluded that diversion efficiency increases with permeability and decreases as foam quality is increased. They experiments were performed in two cores with porous medium of permeabilities varying from 10 to 158 md. Thompson and Gdanski10 concluded that foam-acid diversion is limited by permeability differences. More specifically, they observed that good diversion is expected for permeability ratios less than 10:1. Their experiments were carried out with parallel limestone cores with permeabilities ranged from 1.5 to 9 md. They also pointed out that a minimum foam quality of 60% is required to block the acid flow and that foams of 70% or 80% were the most effective.

Abstract The proper placement of acid is a crucial factor in a matrixacidizing project. Unfortunately, much of the acid flows through the less-damage zones, leaving the damage area untreated. Foams have been used to divert the acid to the damage areas. However, the complexity of foam behavior in a porous media has reduced the efficiency of the diversion. Behenna1 claimed that if foam is generated or injected into the less-permeable layer (damage area), or if insufficient foam is injected into the more-permeable layer, diversion is not obtained. More recently, Alvarez et al.2 concluded based in a proposed model, the fg * model, that foam injection in the lowquality regime (low foam quality, fg ) is better for acid diversion. They claimed that in this regime, more foam is injected into the higher-permeability layer before switching from foam to acid. In this study experimental results confirming the implications of the fg * model are shown. Additionally, the experimental results suggest the existence of an optimum foam quality value for improving the efficiency of an acid-diversion work. Experiments were carried out in a composite core (parallel cores) having a permeability ratio of 15. The cores were packed with sand having different particle diameter to obtain the permeability contrast. An anionic surfactant was used in these experiments. Foam quality was evaluated from 95 percent to less than 40 percent. All the experiments were performed at 600 psi outlet pressure and room temperature. The gas used was nitrogen. The existence of an optimum foam quality for acid diversion implies that matrix-acidizing projects should be designed carefully. Experimental work, such as the one performed in this study, and the fg * model may help to choose in advance

JOSE M. ALVAREZ, HERCILIO RIVAS AND GEIDY NAVARRO

SPE 58711

Zerhboub et al.4 reported that diversion efficiency increases as foam volumes is increased. They also found that diversion is more efficient in high permeabilities cores. This conclusion was obtained by comparing the results performed in parallel cores. The diversion efficient was higher in the experiment using 18.3 darcy and 6 darcy cores, as high and low-permeability layers respectively, than using 920 md and 300 md cores (note that the permeability ratio is 3.1 in both experiments). An important finding reported by Zerhboub et al. is that a shut-in period after foam injection increases diversion efficiency. Parlar et al.11 , using parallel Berea cores with different permeabilities, found that surfactant type, surfactant-preflush and foam slug sizes are critical parameters for liquid diversion into the less-permeable layer. They observed, using a foam quality of 65%, that a huge surfactant preflush (e.g. 30 PV) and a large foam slug (e.g. 0.5 PV or more) lead to post-liquid anti-diversion (liquid injected after foam diverted into the high-permeable layer), a bad result for matrix-acidizing jobs. Behena1 , performing parallel-core tests with a fixed foam quality of 75%, claimed that post-foam liquid diversion cannot be obtained unless specific permeability conditions are encountered. He pointed out that if foam is generated or injected into the less-permeable layer, or if insufficient foam is present in the more permeable layer, diversion is not obtained. Therefore, permeability in the low-permeability layer should be low enough to completely prevent foam from entering there. Modeling foam in porous medium has vital importance for improving understanding and maximizing efficiency. Zhou and Rossen12 developed a model for foam diversion in matrix acidization of sandstones based on fractional-flow theory and assuming that foam obeys the Pc* model. They found that the key to acid diversion is trapping as much of the gas in the foam as possible. Additionally, they concluded that surfactant preflush helps placing a large foam slug into the morepermeable layers improving diversion. Kibodedeaux et al.13 , using fractional flow theory, concluded that preflush size is a secondary factor in acid diversion. However, results from Osterloh and Jante14 , suggesting the existence of two flow regimes (Figure1), may imply that all the matrix-acidizing jobs are located in the second flow 15 regime where P c* model does not apply . This conjecture may suggest that previous results should be reviewed. More recently, Alvarez et al.2 proposed a model, the fg * model, that unites foam behavior in acidizing projects and gasdiversion IOR projects and reconciles apparently contradictory data. The model can be explained as follows: In the high quality regime (vertical contours in Fig. 2), Pc* controls bubble size and therefore gas mobility. The key in this regime is the ability of the surfactant to stabilize bubbles. In the low-quality regime (horizontal contours in Fig. 2), bubble size is fixed and pressure gradient depends on porous medium, but not on ability of surfactant to stabilize foam.

The transition between the two regimes occurs roughly where the contour lines from the high-quality regime intersect the contour lines from the low-quality regime. Therefore, the transition zone is sensitive to both the nature of the porous medium and ability of surfactant to stabilize foam. Based in the fg * model, Alvarez et al. suggested that injection in the low quality-regime (i.e., low foam qualities) is better for acid diversion, since more foam will be injected into the more-permeable layers (Fig. 3). A very striking result reported in their work is the fact that permeability has a big impact on fg * (foam quality at the transition zone, Fig. 2), reducing fg * as permeability decreases. This result implies that for a fixed foam quality, the permeability will set in which flow-regime the experiment or the field project is carried out, affecting the diversion efficiency of the process. This also may explain the contradictory results found in the literature, especially if the experimental conditions are not the same. Foam-acid diversion research has increased the knowledge in this area. However, there are contradictory conclusions and some paradoxes that need to be solved in order to improve the efficiency of this process. The objective of this work is to present experimental data supporting the implications of the fg * model. Experimental Approach Materials All the experiments were carried out with silica sand. An anionic surfactant was used in all the experiments. The surfactant solution was 1 wt.% in deionized water. Nitrogen was used as the gas phase during the experiments. Surfactant solution was injected after foam to emulate the acid. Apparatus Figure 4 shows a schematic representation of the equipment used to perform the experiments. The system consisted of two identical coreholders placed in parallel with the coreholders in vertical position and with the injection from the top. The coreholders are standard stainless steel cells. The diameter of the cells and the length are 1.37 inches and 6.29 inches, respectively. Water or surfactant solution were injected with a Waters reciprocating-piston pump (model 590). The injected liquid mass was measured with a Mettler balance providing a check on pump injection rates. A Brooks massflow controller was utilized to regulate the nitrogen flow rate. A back pressure regulator (BPR), Tescom, is placed at the outlet of each core holder to control the outlet pressure. Individual liquid flow rates from each core were monitored by collecting the effluent at the exit of each BPR. All experiments were carried out at 600 psi outlet pressure and room temperature. Two Validyne pressure transducers with a 500psi diaphragm were used to measure the total pressure drop across both cores. A Validyne pressure transducer, 5-psi diaphragm, was placed between the two effluent lines in order to ensure zero pressure drops between the BPRs. This is crucial in parallel core experiments using two BPRs since small differences may distort the final results.

SPE 58711

AN OPTIMAL FOAM QUALITY FOR DIVERSION IN MATRIX-ACIDIZING PROJECTS

Experimental Procedure Each cell was packed with silica sand having different particle-diameter distribution, ensuring cores with different permeabilities. The cells were then fully saturated with water and water permeability determined. A cell of 8000 md. (highpermeability cell) and a 550 md. cell (low-permeability cell) were obtained and used during all the experiments. Before starting each experiment, water was injected with the two cores open to flow in order to adjust the BPRs, insuring that the flow rate ratio was the same as permeability ratio. Then 4 PV of surfactant solution, 1 wt.%, were injected before nitrogen foam injection. Nitrogen and surfactant solution flow rates were set to obtain a specific foam quality, fg , at the outlet conditions of 600 psi. Total flow rate was kept constant and equal to 1 cc/min in all experiments. Foam was injected until steady state was obtained (i.e. constant pressure drop through the cores). After steady state was achieved, surfactant solution was injected at 1 cc/min to emulate the acid. Individual liquid flow from each core was collected to measure diversion efficiency as a function of foam quality, fg . After concluding each experiment, 50 PV of water were injected. Then the core was placed in a oven at 140 F for 24 hr. After that, the core was water saturated and permeability measured to ensure no petrophysical changes. Strategy Five experiments were carried out in order to evaluate the influence of foam quality on liquid diversion. Table 1 shows the working conditions for each experiment. Foam quality was varied from 95% to 30%. Results Figure 5 shows the pressure drop during foam injection for each experiment. The trend observed in these experiments is clear, an increment in pressure drop is observed as foam quality increases. The relative flow (i.e., flow collected from each core divided by total flow injected) as a function of foam quality is shown in Figure 6. Three different regions can be observed in Figure 6. Region I (the improving region) no good diversion is observed, region II (the optimizing region) excellent diversion is obtained and finally region III (the declining zone) where the diversion efficiency is declining. Figure 7 presents the flow ratio (i.e., flow through the highpermeability core divided by the flow through the lowpermeability core) as a function of fg . Note that a fg equal to zero (i.e., only liquid injection) the flow ratio is equal to permeability ratio. Figure 8 is similar to Figure 7 but reducing the y-axis scale. Discussion Although is not clear how to relate some findings of the fg * model developed in single-coreflood experiments to parallelcoreflood experiments. Alvarez15 pointed out that the higherpermeability layer dominated the foam behavior in a heterogeneous core, affecting the fg * average (i.e., the fg * observed during the experiment, no the particular fg * for each particular layer, Figure 9). He explained that much of the flow

would occur through the more-permeable layer in the absence of diversion. Therefore, the high-permeable layer should contribute the most in the fg * average. In these experiments, in the absence of diversion, 94% of the total flow would occur through the more-permeable layer. The trend observed in Figure 5 indicates that at least foam qualities below 90% below the fg * average. This inference can be explained using Figure 2. At fixed total flow rate (e.g., 1cc/min in all the experiments), P increases with decreasing foam qualities until fg * is reached; then P decreases with further decreases in foam quality in the low-quality regime (as observed in Figure 5). Therefore, the transition zone between the two regimes (i.e., fg * average), in our experiments, should be elsewhere between 90% and 100% foam-quality values. The high foam-quality value at fg * observed in this work is in agreement with the values reported by Osterloh and Jante14 , and Alvarez et al.2 for sandpack cores with high permeability. Region I (the improving zone) in Figure 6 is for foam qualities above 90%. The improving zone coincides with foam qualities higher than or close to the value of fg * (i.e., foam quality at the transition zone in Figure 2), in this region more foam is injected to the less-permeable layer reducing the efficiency of the acid diversion, as shown schematically in Figure 3. However, even at the quality of 95% some diversion is obtained, flow ratio is reduced from 15 to less than 2 (Figure 7). The optimizing zone (region II) is for foam qualities from 90% to 40% (Figure 6). In this zone an excellent diversion is observed. The flow ratio is less than 1 indicating that more acid is diverted into the less-permeable layer. The main attribute of this zone is that the slope of the line in Figure 8 is small in comparison with the slopes in the other two regions. However, the diversion efficiency still increases as foam quality decreases (e.g., the flow ratio is equal to 0.33 for 90% foam quality and 0.2 for 60% foam quality). This region coincides with values of foam quality below the fg * average. Therefore, less foam is being injected into the lowpermeability layer improving the diversion efficiency as shown schematically in Figure 3. Considering that the fg * average is higher than the fg * value for the less-permeable layer, as shown schematically in Figure 9, an increment on the diversion efficiency should be expected for injected foam qualities below the fg * average. But the increment should stop once the foam quality injected approaches the fg * value of the low-permeable layer. Foam qualities injected below the fg * of the low-permeable layer should not improve the diversion efficiency, because there is not more reduction in foam flow through this layer. This conjecture was observed in these experiments, flow ratio is equal to 0.2 for 60% and 40% foam qualities. Therefore, it may be inferred that the fg * value for the less-permeable layer should be between 60% to 80% foam qualities, in agreement with values reported by Alvarez15 for cores with permeabilities around 550 md. The last region, the declining region, is observed for foam qualities below 40%. In this zone the diversion efficiency decreases as foam quality is reduced (e.g., flow ratio equal to

JOSE M. ALVAREZ, HERCILIO RIVAS AND GEIDY NAVARRO

SPE 58711

0.2 and 2.4 for foam qualities of 40% and 30%, respectively). This striking result may be not explained by the fg * model. The fg * model suggests not changes in the acid-diversion efficiency for foam qualities being injected below the fg * value. However, the low-pressure drop obtained at foam qualities below 40%, may be not sufficient for acid diversion. The existence of three regions implies the presence of an optimal foam quality value for acid-diversion jobs. This optimal value clearly is located in the second region, the optimizing region. However, it should be noted that more surfactant is used as foam quality is reduced. Therefore, a trade-off between diversion efficiency and costs will set the final foam quality used in a specific process. As pointed out by Alvarez et al.2 , the fg * value is set by permeability and other parameters. Thereupon, the optimal value for a specific acid-foam treatment will rely on some reservoir and operational conditions (e.g., permeability and injection flow rate). Experimental work as the one presented in this paper but at the conditions relevant to the treatment, should be accomplished before performing the field job. Conclusions 1. Three different regions were found by changing foam quality. Foam qualities in the optimizing region should be selected to obtain the best results in a foam-acid diversion job. 2. Insights gained from this work can guide the selection of foam quality for field applications in foam-acid diversion. 3. The results presented in this study supports some suggestions introduced by the fg * model, that acid-foam diversion is better at low foam qualities (i.e., foam qualities in the low-quality regime). 4. The results suggest the existence of an optimal foamquality value for diversion, However, the final decision will depend on diversion efficiency and costs. Since, more surfactant is used at low foam qualities. 5. No fluid anti-diversion was observed during the realization of the experiments in this study. 6. Contradictory results found in the literature can be reconciled with the results shown in this study. Nomenclature fg = gas fractional flow fg * = gas fractional flow at transition between flow regimes k = permeability, md PV = pore volume Pc = capillary pressure, psi Pc* = limiting capillary pressure, psi P = pressure gradient, psi/ft Uw,Ug = superficial velocities of water or gas, ft/d Ut = total superficial velocity, ft/d

Acknowledgments We thank PDVSA-INTEVEP for supporting the realization of this experimental work and for authorizing the publication of this study. References
1. Behenna, F.R.: "Acid Diversion from an Undamaged to a Damaged Core Using Multiple Foam Slugs," SPE 30121 presented at the 1995 SPE European Formation Damage Symposium, The Hague, Netherlands, May 15-16. 2. Alvarez, J.M., Rivas, H.J., and Rossen, W.R.: "Unified Model for Steady-State Foam Behavior at High and Low Foam Qualities," SPE 56825 presented at the 1999 SPE Annual Technical Conference and Exhibition, Houston, Texas, October 03-06. 3. Chambers, D.J.: "Foams for Well Stimulation," in Foams : Fundamentals and Applications in the Petroleum Industry, Schramm, L.L. (ed.) ACS Advances in Chemistry Series, 3, No.242, Am. Chemical Soc., Washington, D.C., (1994). 4. Zerhboub, M., Ben-Naceur, K., Touboul, E., and Thomas, R.: "Matrix Acidizing: A Novel Approach to Foam Diversion," SPE Production & Facilities, May 1994. 5. Hill, A.D., and Rossen, W.R.: "Fluid Placement and Diversion in Matrix Acidizing," SPE 27982 presented at the 1994 SPE Centennial Petroleum Engineering Symposium, Tulsa, Oklahoma, August 29-31. 6. Gdanski, R.D.: "Experience and Research Show Best Designs for Foam-Diverted Acidizing," Oil & Gas Journal, Sept. 1993. 7. Bernard, G.G., Holm, L.W., and Jacobs, W.L.: "Effects of Foam on Trapped Gas Saturation and on Permeability of Porous Media to Water," SPE J. (Dec. 1965) 5, No. 4, 295-300. 8. Smith, C.L., Anderson, J.L., and Roberts, P.G.: "New Diverting Techniques for Acidizing and Fracturing," SPE 2751 presented at the 1969 SPE California Regional Meeting, San Francisco, Nov. 6-7. 9. Burman, J.W., and Hall, B.E.: "Foam Diverting Technique Improved Sandstone Acid Jobs," Proc., World Pet. Cong., New Orleans, 1987. 10. Thompson, K.E., Gdanski, R.D.: "Laboratory Study Provides Guidelines for Diverting Acid With Foam," SPE Production & Facilities, Nov. 1993. 11. Parlar, M., Parris, M.D., Jasinski, R.J., and Robert, J.A.: "An Experimental Study of Foam Flow Through Berea Sandstone With Applications to Foam Diversion in Matrix Acodizing," SPE 29678 presented at the 1995 Western Regional Meeting, Bakersfield, Mar. 8-10. 12. Zhou, Z.H., and Rossen, W.R.: "Applying Fractional-Flow Theory to Foams for Diversion in Matrix Acidization," SPE 24660 presented at the 67th Annual Technical Conference and Exhibition of the SPE, Washington, DC, Oct. 4-7, 1992. 13. Kibodeaux, K.R., Zeilinger, S.C., and Rossen, W.R.: "Sensitivity Study of Foam Diversion Processes for Matrix Acidization," SPE 28550 presented at the 69th Annual Technical Conference and Exhibition of the SPE, New Orleans, LA, Sept. 26-28, 1994. 14. Osterloh, W.T., and Jante Jr, M.J.: "Effects of Gas and Liquid Velocity on Steady-State Foam Flow at High Temperature," SPE/DOE 24179 presented at the 1992 SPE/DOE Eighth Symposium on Enhanced oil Recovery, Tulsa, Oklahoma, Apr. 22-24. 15. Alvarez, J.M.: "Foam-Flow Behavior in Porous Media: Effects of Flow Regime and Porous-Medium Heterogeneity," dissertation, the University of Texas at Austin (1998).

SPE 58711

AN OPTIMAL FOAM QUALITY FOR DIVERSION IN MATRIX-ACIDIZING PROJECTS

SI Metric Conversion Factors cm3/min x 1.67 ft x 3.048 ft/day x 3.528 md x 1.0132 psi x 6.895 psi/ft x 2.262 E-08 = m3/s E-01 = m E-06 = m/s E-03 = m2 E+03 = Pa E+04 = Pa/m

Table 1. Experimental conditions

Core Low K : 550 mD High K: 8000 mD Surfactant Anionic 1 wt% Experimental Conditions P: 600 Psi T: Room temperature Flow rate: 1 cc/min Exp. Div-1 Div-2 Div-4 Div-7 Div-9 Div-10 fg 90 60 40 95 30 100

JOSE M. ALVAREZ, HERCILIO RIVAS AND GEIDY NAVARRO

SPE 58711

High-Quality

Regime:

Effect on P uncertain More difference in foam rate Less foam flows through lowperm. Layer as fg* decreases

Gas flow rate

fg2

fg1
Low-quality

Fig

Regime:
Higher K, bigger pores, less capillary resistance

Fig 1. Pressure drop as a function of flow rates (from Ref. 14)

Liquid flow rate

3. Effect of fg injected; more foam is placed in the highpermeability layer, reservoir represented as two layers with different K.

P
Gas flow rate

Transition foam quality, fg *

C fg Pump

varying Ug; Uw

Fixed foam quality, fg

N2
K: Pack K: Empaques Tp : Pressure-Drop Tp : Transductor transducers diferencial de BPR: Back presin Pressure regulators V c :Vlvulas de Sp :Phase contrapresin Spseparator : Separador de Cfg : Mass-Flow fases Cfg :Controller Controlador (g a s ) de flujo

K Tp2

Tp

fixed

S p2

Tp

varying Uw; Ug

Fixed UT, varying fg

Sp BPR 1 BPR
2

fixed
Liquid flow rate

Fig 2. Osterloh and Jantes type of plot (schematic); identification of important parameters (from Ref. 2).

Figure 4. Schematic of experimental apparatus.

SPE 58711

AN OPTIMAL FOAM QUALITY FOR DIVERSION IN MATRIX-ACIDIZING PROJECTS

250

fg
200 95% 90% 60% 40% 30%

Pressure drop psi) (

150

100

50

0 0 50 100 150 200 250 300 350 400 450

Time (minutes) Fig. 5 Pressure drop as function of time; effect of foam quality on pressure drop.

100

high K low K

80

Relative flow (%)

60

Region III
40

Region II

Region I

20

0 0 20 40 60 80 100

Foam quality, fg

Fig. 6 Relative flow as function of foam quality; three regions are observed.

16 12 Flow ratio 8 4 0 0 20 40 60 Foam quality, fg 80 100

Fig. 7 Flow ratio as a function of foam quality.

JOSE M. ALVAREZ, HERCILIO RIVAS AND GEIDY NAVARRO

SPE 58711

3
2,5 2 1,5 1 0,5

Flow ratio

0
20 40 60 80 100

Foam quality, fg

Figure 8. Flow ratio as function of foam quality.

fg (high K) fg average (observed in the experiment)

Gas flow rate

fg (low K)

Liquid flow rate

Fig. 9 Schematic representation of the f g average.