GB2058889A - Process for recovery of oil from subterranean reservoirs - Google Patents

Abstract

A process for recovering petroleum from a subterranean oil reservoir containing a high percentage of swellable clays wherein production from the reservoir formation is obtained by driving a thickened water solution from an injection well to a production well in which the water is thickened with uniformly substituted cellulose sulfate polymer having a degree of substitution of at least 1. The thickened water can be used in conjunction with other injectables such as surfactants to enhance the recovery of oil and control mobility of the injected water.

Description

SPECIFICATION Process for recovery of oil from subterranean reservoirs This invention relates to the recovery of oil from subterranean reservoirs and, more particularly, to new and improved secondary recovery operations utilizing a thickened flood water.

In the recovery of oil from oil-bearing reservoirs, it usually is possible to recover only a minor portion of the original oil in place by the so-called primary recovery methods which utilize only the natural forces present in the reservoir. Thus, a variety of supplemental recovery techniques have been employed in order to increase the recovery of oil from subterranean reservoirs. In the supplemental techniques, which are commonly referred to as secondary recovery operations, although in fact they may be primary or tertiary in sequence of employment, fluid is introduced into the reservoir in order to displace the oil therein to a suitable production system through which the oil may be withdrawn to the surface of the earth.The displacing medium may be a gas, an aqueous liquid, such as fresh water or brine, an aqueous liquid thickened with a polymer, an oil-miscible liquid such as butane, or thickened water combined with a surfactant. Generally, the most promising of the secondary recovery techniques involves the injection into the reservoir of an aqueous flooding medium, either alone or in combination with other fluids.

One difficulty which often is encountered in secondary recovery operations is the relatively poor sweep efficiency of the injected displacing liquid. That is, the displacing liquid exhibits a tendency to channel through certain portions of the reservoir and to bypass other portions. Such poor sweep efficiency is occasioned by differences between the viscosity of the injected displacing medium and the in situ reservoir oil and also by permeability variations within the reservoir. The reservoir ni'ay comprise a plurality of fairly well-defined zones of widely diverse permeabilities. The injected displacing fluid preferentially fiows through the more permeably zones of the reservoir thus leading to premature breakthrough of the displacing fluid at the production well or wells.Permeability is also adversely affected in reservoirs containing a high percentage of clays containing minerais which swell when the connate (or produced) water is displaced with fresh water or brine.

Even when the reservoir exhibits a relatively uniform permeability throughout, a situation referred to as instability fingering may develop in those instances where the viscosity of the injected displacing fluid is significantly less than the viscosity of the in situ reservoir oil. In this situation, the less viscous displacing fluid tends to develop fingers or bulges which may be caused by points of minute heterogeneities in the reservoir. These fingers of displacing fluid tend to become extended in the direction of flow and travel at a faster rate than the remainder of the injected fluid, thus again resulting in premature breakthrough at the production system.

Various techniques have been proposed in order to improve the sweep efficiency of the injected displacing fluid and thus avoid premature breakthrough. For example, it has been proposed in waterflooding operations to add thickening agents to at least a portion of the aqueous flooding medium in order to increase the viscosity thereof. The viscosity of the flooding medium may be increased prior to its injection into the reservoir or alternatively the viscosity may be increased in situ in order to avoid a reduction in injectivity at the injection well. For example, in U.S.Patent No. 3,208,518 to Patton, there is disciosed a waterflooding process wherein the viscosity of the flooding medium is increased in situ through the use of high molecular weight polymers such as ionic polvsaccharides produced by the fermentation of carbonhydrates by bacteria of the genus Xanthomonas, under controlledpH conditions.

It has now been found that a sulfated cellulose prepared by the methods disclosed in U.S. Patent No. 4,143,226, the disclosure of which is incorporated herein by reference is useful in mobility control in enhanced oil recovery. This polymer when substituted to a degree of substitution of at least 1 and preferably from 1 to 1.3 shows superior and unexpected results when tested under conditions encountered in rock and sand strate containing Berae and Bradford rock and sandstone rock core samples such as those encountered in ares such as the Claymore field which contains a high percentage of swellable clays.

The present invention provides a process for recovering oil from subterranean oil reservoir formations by injection of water thickened by a thickening agent wherein there is used as thickening agent a uniformly substituted cellulose sulfate polymer having a degree of substitution of at least about 1.

The preparation of cellulose sulfate esters by a homogeneous method as described in the above U.S Patent employs a cellulose nitrite ester as a soluble chemical intermediate, which undergoes facile transesterification reactions with sulfur trioxide to result in mixed cellulose nitrite sulfur ester. Sodium cellulose sulfate is obtained from the mixed ester by removing the residual nitrite groups through a reaction with a protic solvent followed by neutralization with sodium carbonate.Thus, soluble products can be prepared within a wide range of degree of substitution (D.S.) of about 0.3-2.0 To perform as a mobility control agent in tertiary oil recovery, a polymer should exhibit such properties as complete solubility to eliminate plugging, high viscosity at low concentrations, compatibility with salt, particularly sodium, calcium, and magnesium chlorides, minimal depolymerization under high shear, and good solution stability at reservoir temperature over long periods to time. If the chacteristics of the polymer are positive in most, or preferably, all of these points, the product can be considered a good candidate having a reasonable chance of success as a mobility control agent.

The characteristics of the cellulose sulfates indicating applicability in tertiary oil recovery are as follows: 1. Compability of the various cellulose sulfates with mono-, di-, and trivalent metal salts is excellent as shown in Table 1. A solution of cellulose sulfate remains clear and viscous without the formation of the precipitate or gel when salts of the ions indicated are added. Only if the product exceeds a D.S. of about 1.3 is incompatibility observed with Fe3+, Ce3+, and Ba2+. In general, solutions can be even saturated with the salts indicated without showing any sign of precipitation, unless the D.S.

is relatively close to 0.3 which is the minimum D.S. required to render the cellulose water soluble. At such a low D.S., signs of incompatability are observed at salt concentrations of as low as 510%.

TABLE 1 Compatibility of cellulose sulfate with metal ions Sodium Cellulose Sulfate Metal lon D.S. < 1.3 D.S. > 1.3 Na+, K+, NH4+ + + Mg2+, Ca2+, Sr2+ + + Zn2+, Cu2+, Co2+, Ni2+ + + Fe2+, Cd2+, Hg2+, Pb2+ + + At3+, C + + Fe3+, Ce3+, Ba2+ + + = compatible - = incompatible 2. Dilute aqueous solutions of cellulose sulfate provide a relatively high viscosity at low concentrations. Figure 1 shows polymer concentrations required to produce a viscosity of 10 centipoises (cps) in solutions containing various amounts of sodium or calcium chlorides.The viscosity was measured with a Brookfield Viscometer, LVT Model at 6 rpm and 20 C. The following observations can be made from Figure 1: a. Low D.S. products provide more viscosity than the higher D.S. products; b. The addition of salt has a viscosity reducing effect up to about 0.51% salt, and no further significant viscosity reduction occurs above this level; c. The extent of the viscosity reducing effect decreases as the D.S. of the cellulose sulfate decreases.

Thus, depending on the concentration of salt, a viscosity of 10 cps is obtained with about 700-1300 parts per million (ppm) of a product with a D.S. of 0.6--0.7 while about 900-2200 ppm are required if the D.S. is 1.4-1.5. This indicates that a product with a D.S. of less than 0.6 probably should provide a 10 cps viscosity even at lower concentrations.

3. The cellulose sulfates are completely soluble and solutions do not contain highly swollen gel particles or aggregates. This is shown by the fact that aqueous solutions pass through fine filters without plugging and without any significant flow rate reduction. Table 2 shows filtration factors of aqueous solutions containing various amounts of sodium or calcium chlorides. The concentration of the cellulose sulfate was chosen such that the viscosity was about 9-10 cps. The solutions were prepared by dissolving the polymer with mechanical stirring for 1-2 hours, adding the required amount of salt, shearing the solution two minutes in a Waring Blendor at high speed, and filtering through Watman No.

1 filter paper to remove extraneous material. A portion of 1,000 ml was then filtered through a Millipore filter at 40 pounds per square inch (psi). The filtration factor was obtained by dividing the sum of the time periods required for passage of the 16th, 18th, and 20th fifty milliliter (ml) portions by the sum of the time periods required for passage of the second, fourth, and sixty 50 ml portions. Therefore, a factor of 1 indicates a constant flow rate, i.e., perfect injectivity, and a factor of 2 a flow rate reduction of 50%.

All factors are significantly below 2, and a typical plot of flow rate vs. cummulative throughput in Figure 2 shows the flow rate becomes constant indicating that there is no plugging. The addition of salt has no significant influence on flow rate although, in some cases, there appears to be a very slight flow rate reduction as the concentration of NaCI increases or when NaCI is replaced by CaCI2. Similarly there may be a slight increase in flow rate reduction as the D.S. of the product decreases, but the extent of such influence, if any, appears to be small.

TABLE 2 Millipore filtration of sodium cellulose sulfate solutions Filtration Factor D.S.of NaCI CaCI2 NaCell. SO4 0.03 0.5 3.0% 0.03 0.5% 0.6--0.7 1.7 1.7 1.9 1,7 1.5 0.9 1.4 1.6 1.8 1.7 1.9 1.1-1.2 1.4 1.3 1.4 1.5 1.6 1.5 1.3 1.4 1.4 1.5 1.7 Table 3 shows the influence of polymer concentration on filtration factor indicating that substantially increasing the polymer concentration causes only a slight increase of the filtration factor.

The influence of prior shearing is shown in Table 4. It indicates that the polymer is not completely dissolved after 1-2 hours of moderate stirring, and that, for complete solubilization, addition shearing or mixing is required.

TABLE 3 Effect of concentration on filtration factor Na Cell. SO4, ppm Filtration Factor 700 1.5 1000 1.5 1300 1.6 1600 1.8 1900 1.6 TABLE 4 Effect of time of shearing in a waring blendor on filtration factor Shearing, min. Filtration Factor 0 4.7 1 1.7 2 1.6 3 1.4 5 1.4 4. Depolymerization under high shear was determined by mixing a 0.5% solution containing 0.5% NaCI in a Waring Blender for various periods of time and measuring the viscosity reduction. Table 5 is a summary of the results. The results show that, after 15 minutes of mixing, only about 3035% viscosity loss is encountered irrespective of the D.S. of the cellulose sulfate. Other cellulosics, such as hydroxyethyl cellulose (HEC) and carboxymethy cellulose (CMC), lose about 6070% of their viscosity under the same conditions.Xanthan gum appears to be similar to the cellulose sulfates while polyacrylamide (Pusher 700) is by far the most sensitive. Other workers have shown recently that, after an initial viscosity reduction of about 35%, the cellulose sulfate is more stable than the other polymers tested under extreme shear.

TABLE 5 Viscosity loss of polymer solutions after high shear (waring blender, 1 5 min) Product Viscosity Reduction, % NaCell.SO4, D.S.0.3 34.0 NaCell.SO4, D.S. 0.5 23.4 NaCell.SO4, D. S. 0.8 34.0 NaCell.SO4, D.S. 1.0 33.0 NaCelí.SO4, D.S .1.4 37.7 HEC (Natrosol 250H) 58.6 HEC (Cellosize lOOM) 70.0 CMF (7H4F) 62.8 Xanthan (Keltrol) 28.0 Polyacrylamide (Pusher 700) 83.5 In view of the favorable results discussed above, initial injectivity tests were carried out employing cores of (a) fired Berea sandstone, (b) Pennzoil's Bradford sandstone, and (c) sandstone from the Claymore field in the North Sea for which no polymer has been found suitable yet.The sodium cellulose sulfate used in these tests has a D.S. of about 0.6 unless indicated otherwise and was used "as is" without any pretreatment, such as prefiltration through a Millipore filter or diatomaceous earth.

Experimental Injectivity Results Results obtained by employing fired Berea sandstone are summarized in Table 6. The polymer concentration was 1000 ppm in all cases and the concentration of salt 200 ppm NaCI plus 500 ppm, 1000 ppm, and 5000 ppm CaCl2. The resistance factor varied within a range of about 6-9 depending on the concentration of CaCl2. The rezidual resistance was relatively low at 1.7-2.0 indicating minimal adsorption. This compared rather well with xanthan gum solution containing 1000 ppm CaCl2 that was prefiltered through a 1.2ju Millipore filter. Without such prefiltration, xanthan gum and other microbial polymers would plug severely, particularly in relatively low permeability oil sands.This is mostly due to the presence of cell debris which, in the case of cellulose sulfate, a chemically modified cellulose, is not a factor that has to be dealt with.

TABLE 6 LAB INJECTIVITY TESTS EMPLOYING FIRED BEREA SANDSTONE Residual Injection Fluid Permeability Permeability Porosity Viscosity cps Resistance Resistance Composition to Air, md to water, md Percent Initial Final Effluent Factor Factor NaCell.SO4, 1000ppm 104 76 18.3 13.4 14.0 12.5 9.2 1.7 CaCl2, 50ppm NaCl, 200ppm NaCell.SO4, 1000ppm 109 87 18.3 6.2 6.0 6.1 7.0 2.0 CaCl2, 1000ppm NaCl, 200ppm Xanthan, 1000ppm 113 99 18.4 19.0 16.4 18.6 8.3 2.5 CaCl2, 1000ppm NaCl, 200ppm NaCell.SO4, 1000ppm 112 85 18.3 5.5 5.7 5.2 5.7 1.7 CaCl2, 5000ppm NaCl, 200ppm Significant is the fact that high Brookfield viscosities are not necessarily indicative of high resistance factors.For example, the xanthan gum solution with a viscosity of 1 8-1 9 cps had a resistance factor of only 8, while the cellulose sulfate solutions had significantly lower viscosities of 5-13 cps but relatively high resistance factors of 6-9. Figures 3-6 are plots of the resistance factor vs. fluid injected. The resistance factor becomes constant after injection of 2 to 4 pore volumes indicating that there is no plugging. In this respect, the three tests employing cellulose sulfate compare favorably with xanthan gum and indicate good performance even at CaC12 concentrations at which xanthan gum is known to develop severe problems.

Table 7 is a summary of the results of similar tests but employing cores of low permeability Bradford sandstone. Solutions of prefiltered xanthan gum and unfiltered sodium cellulose sulfate in Bradford injection water were compared at concentrations of 545 ppm and 830 ppm, respectively. At these concentrations, both solutions had similar Brookfield viscosities of about 10 cps. Bradford injection water essentially is fresh water, and its composition is shown in Table 7 also. Figures 7 and 8 are plots of resistance factor vs. fluid injected showing that neither solution was plugging. It is remarkable that, in spite of the similarity of the viscosities, the resistance factor of the cellulose sulfate solution was about twice as high as that of the xanthan gum solution and that, to obtain equal resistance, the concentration of cellulose sulfate probably can be significantly reduced. The same trend was observed with the Berea rock, which casts some doubt on the value of Brookfield viscosities for the evaluation of polymers in this particular application. It is conceivable that a capillary viscometer would be more suitable than a rotational viscometer since it will be closer to simulating the flow through porous media. The best viscometer, however, is a core because it measures viscosity by determining relative flow rates directly in a porous medium.

TABLE 7 LAB INJECTIVITY TESTS EMPLOYING PERMEABILITY BRADFORD SANDSTONE Residual Injection Fluid Permeability Permeability Porosity Viscosity, cps Resistance Resistance Composition to Air, md to Water, md Percent Initial Final Effluent Factor Factor NaCell.SO4, 830ppm 16 15 16.1 9.5 9.4 8.5 9.9 4.3 Simulated Bradford Injection Water Xanthan, 545ppm 19 15 16.0 10.5 8.8 5.1 2.1 Simulated Bradford Injection Watera (a) Bradford injection water contains: Hardness as CaCO3, 164 ppm Chloride, 31 ppm Sodium, 15 ppm Sulfate, < 0.1 ppm Calcium, 43 ppm Magnesium, 14 ppm The final test was carried out using a core of North Sea sandstone from the Claymore field by injecting a 0.1% sodium cellulose sulfate solution prepared in simulated sea water as shown in Table 8.

A Millipore filtration of the saline polymer solution (Figure 9) produced excellent results indicating no adverse effect on filtration rate by the metal ions or anions in solution. However, the graph showing resistance factor vs. fluid injected of the core test in Figure 10 where a low D.S. cellulose sulfate is employed indicates a certain degree of plugging. In combination with the Millipore results, it indicates that the reason for plugging apparently is adsorption on the rock rather than incompatibility with the soluble ions in the aqueous phase. Results indicate that certain clays appear to associate to some extent with the low D.S. cellulose sulfate products but that such association is reduced or eliminated as the D.S. of the product, i.e., its negative charge, increases.Translated to the present novel results, it indicates that a cellulose sulfate with a higher D.S. shows highly reduced or negligible adsorption, so that the resistance factor will level off to assume a constant value. This is confirmed in Figure 11 where a cellulose sulfate with a D.S. of about 1 is employed.The resistance factor reaches a constant value of 5.2 indicating no plugging while the residual resistance of 1.6 is low indicating negligible adsorption. TABLE 8 LAB INJECTIVITY TESTS EMPLOYING SANDSTONE FROM THE CLAYMORE FIELD IN THE NORTH SEA Residual Injection Fluid Permeability Permeability Porosity Resistance Resistance Composition to Air, md to Water, md Percent Viscosity, cps Factor Factor NaCell.SO4, 1000 ppm 135 107 20.0 5.1 10.0 4.04 Sea Water Na Cell.SO4 152 126 20.4 4.5 5.2 1.6 (D.S. 1.0) 100ppm Sea Watera (a) Composition of Sea Water:Chloride, 19,900 ppm Sodium, 11,100 ppm Bicarbonate, 153 ppm Calcium, 430 ppm Sulfate, 2,770 ppm Magnesium, 1,340 ppm The principai advantages of these new cellulosics over polymers presently used are: (a) they are completely soluble without the formation of microgel particles and, therefore, can be dissolved easily and injected without any further pretreatment, such as prefiltration; (b) their exceptional compatibility with mono- and polyvalent metal ions permitting the use of highly saline water and produced water without adversely affecting injectivity; (c) their stability to high shear, and provided that oxidative and enzymatic degradations are exluded by suitable additives, to elevated temperature; and (d) the possibility of tailoring these products within wide ranges of degree of substitution and molecular weight to reduce or eliminate such problems as excessive adsorption and to optimize performance.

The improved and unexpected results shown by the cellulose sulfate polymer having a D.S. of at least about 1.0 appears to be caused by the high negative charge on the cellulose sulfate polymer. As a result, absorbtion on the rock and the tendency to plug the formation is highly reduced or eliminated.

It appears that the high negative charge on the polymer, which is comparable to that of a sulfonate surfactant, reduces the tendency towards fingering into the surfactant slug because of repulsive forces.

Adsorption of the polymer should be further reduced when a surfactant slug precedes the polymer solution for the same reason.

Claims (8)

1. A process for recovering oil from subterranean oil reservoir formations by injection of water thickened by a thickening agent wherein there is used as thickening agent a uniformly substituted cellulose sulfate polymer having a degree of substitution of at least about 1.

2. A process as claimed in claim 1 wherein the substituted cellulose sulfate polymer has a degree of substitution of from 1 to 1.3.

3. A process as claimed in claim 1 or claim 2 wherein the thickened water is injected following a surfactant injection slug.

4. A process as claimed in any of claims 1 to 3 wherein the water mixed with the thickening agent is produced water.

5. A process as claimed in any of claims 1 to 3 wherein the water mixed with the thickening agent is sea water.

6. A process as claimed in any of claims 1 to 5 wherein the subterranean oil reservoir formation is of the type encountered in the Claymore field in the North Sea reservoir.

7. A process as claimed in claim 1 substantially as herein described with reference to the drawings.

8. Oil when recovered by a process as claimed in any of claims 1 to 7.