US 3028912 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
April 10, 1962 v. J. BERRY, JR., ETAL 3,028,912
RECOVERY OF 01:. FROM AN UNDERGROUND FORMATION 2 Sheets-Sheet 1 Filed Sept. 28, 1956 T U 0 w D mi N m. F F 0 N D m mm m T w R a ME 1. m mm T m T M mT S ll.
D m .L F R H A W m m 5 o OEHK :0 1 mmPSS FlGURE l u m G 0 N T A m T V m N m 1 0 0 U 9. VI! u s S m m E D E G0. a u a m N J ."m A U E T. G RB m m m mm T T V m w o 0 3w 2 7 T A E R R. T U w w 06 F i 2% W C A u T A C S m m N 0 C W B 0 O O C 3 2 mmOa .rzmummm I :=O J 3Qmwm YZIM/Z M ATTORNEY P 1952 v. J. BERRY, JR., ETAL 3,028,912
RECOVERY OF OIL. FROM AN UNDERGROUND FORMATION 2 Sheets-Sheet 2 Filed Sept. 28, 1956 2 T v PM; do Pm? mmk 3 O O O m W m w 6 a 2 FIGURE 3 OIL- WATER- QUARTZ 0u BRINE- QUARTZ (25,000 p.p.m. N001) 0n.- BRlNE-QUARTZ (50,000 mm. NaCl) km; 50 km; mmk s O Q 0 Q o PH FIGURE 4 United States Patent 3,028,912 RECOVERY OF OIL FROM AN UNDER- GROUND FORMATION Virgil J. Berry, In, Robert 0. Leach, and Ovner Wagner, Jr., Tulsa, Okla, assignors to Pan American Petroleum Corporation, a corporation of Delaware Filed Sept. 28, 1956, Ser. No. 612,775 3 Claims. (Cl. 166-9) This invention relates to the art of recovering hydrocarbon oil from underground formations. More specifically, it relates to an improved waterfiood procedure.
In a waterfiood operation, water is injected through one or more input wells causing the oil to be displaced from the formation surrounding the input wells. This oil travels ahead of the waterflood front from the formation surrounding the input wells toward the output wells from which the oil is produced in the usual manner.
Various modifications of the waterflooding technique have been employed to further increase the ultimate recovery of oil from the formation. For example, chemical additives, such as sodium carbonate, surface active agents, and the like, have been added to the flooding waters. These additives have not been used extensively possibly due to the difficulty experienced both in the laboratory and in the field in selecting the proper additive for a specific formation.
Attempts have been made to use laboratory model flood tests, under conditions simulating formation conditions, to determine the proper flood water additives. Possibly the most desirable method of conducting flood tests which has been proposed in the past to collect and use a native state core. A native state core is a core cut from a formation and handled in such a manner that the physical and chemical properties of the formation are preserved within the core. The expense involved in collecting the cores has severely limited extensive use of this procedure.
Artificial cores, which are considerably more economical to use than native state cores, have been employed to determine the efiiciency of waterflooding using different additives. The workers have found numerous variables which influence the results of the artificial core flood tests. Flooding water compositions selected on the basis of such tests have been found to be applicable to some field operations but not to others. Even the laboratory results themselves have sometimes been erratic for unexplained reasons. In addition, the length of time required for conducting large numbers of flow tests is generally prohibitive if any considerable variety of compositions is to be tested whether the cores are artificial or native state.
An object of this invention is to provide an improved method for waterflooding an oil bearing formation. it is a further object of this invention to provide an improved rnethod'for determining the proper composition of flooding water for a given formation.
We have now discovered that many formations are preferentially oil wet due to the presence of naturally occurring surface active agents in the reservoir fluids, these agents being adsorbed on the solid surfaces of the formations. In such cases we have found that greatly increased oil recovery in a waterflooding operation can be obtained by causing the formation to become preferentially wet by the flooding water rather than by the formation of crude oil. It is essential that this reversal of preferential wettability not occur before arrival of the waterflood front. We have discovered that this can be accomplished by adjusting the pH in some cases. In other cases the preferential wettability can be reversed by changing the salinity of the flooding water. Frequently it is desirable to adjust both the pH and salinity. The required amount and direction of adjustment of the pH and salinity may be determined by measurements of the angle of contact of a solid surface with the interface between the crude oil and the flooding water. These tests will be described later in more detail. In general, the flooding water should change the contact angle from a value of more than degrees, as measured from the water side of the oil-water interface, to a value less than 90 degrees.
Contact angle tests will be erratic and unreliable, as experienced in much of the prior art work, unless two previously over-looked essential requirements are met. First, the reservoir fluids must be permitted to come to equilibrium with the mineral surface. That is, in the test, the amount of surface active agents on the mineral surface should be brought into substantial equilibrium with the surface active agents in the reservoir fluids as they exist in the formation. Second, the oil must be maintained in substantially air free condition to avoid destruction of some of the surface active agents by oxidation and the creation of others, since this frequently changes the nature of surfaces contacted by the oil. Preferably, the brine and the rest of the test system should also be maintained substantially air free although small amounts of air can be tolerated.
In the drawing:
FIGURE 1 is a curve showing the results of flooding a particular core, first with neutral water and then with acid water.
FIGURE 2 is a curve showing residual oil saturation of a core versus salt concentration of the flooding water.
FIGURE 3 is a curve showing the variation in contact angle with changes of pH in water in contact with a crude petroleum and quartz.
FIGURE 4 is a family of curves showing the variation of contact angle with changes in pH of various brines in contact with quartz and an oil solution of n-octylamine.
To determine if a formation is susceptible to our waterflooding method it must first be determined if it is preferentially oil wet. This determination can be made by measuring the angle of contact of the oil-water interface with the solid surface. This angle is usually termed the contact angle and is generally referred to as such herein. Unless otherwise specified, the angle stated will be that which may be measured by rotating 3 line from the solid surface through the water-phase to a line tangent to the oil-water interface at the solid surface. If the contact angle is about 90 the solid surface is wet substantially equally readily by both oil and water. That is, there is no preferential wetting. If the contact angle is greater than 90 degrees the formation is preferentially oil wet and our process is applicable where the preferential oil wettability is due to the presence of surface active agents. However, there may be some formations which are oil Wet because of a bituminous coating on the sand grains. In such cases preferential water wettability may not be easily attained by adjusting pH and salinity. The presence of bitumens can be detected by examination of a core. If the angle is less than 90 degrees the formation is preferentially water wet and our process is normally applicable to a much smaller degree.
The contact angle usually cannot be measured on the rough porous surface of the formation itself. A smooth, impermeable surface is generally preferred. Contact angles are normally measured on polished surfaces of quartz, dolomite, and calcite crystals, since these materials are'the ones predominantly exposed to oil and water in petroleum-bearing formations. If the composition of the formation surface is known to be something other than quartz or calcite, a solid surface as similar to the formation as possible should be used. The crystal surface used should be a natural face of the crystal, or if the crystal is cut to obtain a testing surface, the out should be parallel to a natural plane of the crystal. This is because surfaces cut at an angle to these natural planes sometimes give results other than those obtained on the natural crystal surface.
The natural preferential wettability of the formation is determined by allowing the crude petroleum and the brine from the formation to come into equilibrium with quartz and calcite crystal surfaces and measuring the contact angles with the two types of surfaces. The contact angles may differ only slightly. Even if the contact angles differ considerably against quartz and calcite they may both indicate the same type of preferential wettability. That is, they may differ only in the degree of preferential wettability rather than the kind. If one surface is preferentially water wet while the other is preferentially oil wet the operability of our method will depend upon which type of surface is predominantly present in the formation. Usually this is known. In some cases the predominant type of surface will not be known as in a quartz sandstone cemented with calcium carbonate. In such cases it should be presumed, for purposes of testing, that the surface is such that'our method is applicable. Then, after determining the proper flooding water composition as described later, this flooding water may be tested in a native state core from the formation to determine finally the applicability of the composition before it is used in the commercial flood. The native state core test can be eliminated if the cost of the salt, acid, or base, to obtain the desired flooding water composition is small. If the formation' is preferentially water wet, the adjustment in flooding water composition will generally be harmless and may produce some benefits. If the formation is preferentially oil Wet, however, the benefits will be great.
Several methods are available for measuring contact angle. In a preferred method, the test surface is first submerged in the oil. The solid should remain submerged within the crude oil a sufficient time for the surface active materials within the crude oil to reach the solid surface and come to equilibrium between this surface and the oil. The aging time allowed for equilibrium to be established may vary from a few hours to several days and must be determined by a series of tests for each separate system. After the oil and the surface come to equilibrium a drop of water, which has preferably been brought into equilibrium with the oil, is immersed within the oil above the submerged surface such that the water will move through the oil and contact the surface. A small amount of time is then allowed for the water drop to adjust itself to approximately the stable contact angle. The common practice is to measure the angle by visual means, preferably using a telescope fitted with a goniometer eyepiece. The contact angle is observed until it reaches an equilibrium value.
When using a crude which is not transparent, it is sometimes difficult to see the immersed drop of water through the oil, and therefore other means must be employed to determine the contact angle. uate procedure is to suspend the surface within the water, which is transparent, and then immerse a drop of oil within the water below the solid testing surface such that the oil will rise within the water to contact an area on the solid. The drop is then allowed to remain in contact with the solid until equilibrium is attained at the oil-solid surface as determined by a series of tests. When equilibrium has been attained the water is caused to advance across the oil-covered surface by means and for reasons explained later in more detail. The contact angle is then observed until it reaches an equilibrium value as previously described.
If a drop of water is added to the oil-covered surface, equilibrium is attained rather quickly since the water seems to displace the oil quickly from the surface. Even near the boundary of the oil and water phases, equilibrium is rather rapidly obtained, generally within about 3 or 4 hours. If a drop of oil is added to the water-covered surface, a long time-possibly several weeks or even A generally accepted alter- 4 longer-is required for the oil to displace water from the surface so that the interface between the liquids reaches its final equilibrium contact angle with the solid surface.
In general, on mineral surfaces equilibrium is attained quickly as water is advancing to replace oil while equilibrium is reached slowly if water is receding to be replaced by oil. Contact angles measured with Water advancing will be referred to as advancing contact angles, while those measured with water receding will be referred to as receding contact angles.
As previously noted, if an opaque crude oil is used as the oil phase, which is usually the case, a drop of the oil is added to the water-covered test surface. The water in this case is receding from the surface touched by the oil so a receding contact angle occurs. To avoid waiting the long period required for equilibrium to occur under such circumstances, we prefer to cause an oil-water interface to move toward the oil phase so an advancing contact angle can be measured. One method tested was to withdraw some oil from the oil drop by use of a pipette to decrease the volume of the oil drop and cause the water to advance across the oil-covered surface. This technique has produced erroneous results. A possible explanation for these erroneous results is that the decrease in surface meant the oil drop produces a greatly increased concentration of surface active agent at the interface. This results in contact angles which are erratic and usually much too high in value. In general, if the oil-water surface area of the drop is decreased significantly, unreliable contact angles are produced, while if this surface area is maintained constant or is increased, reliable values are obtained.
The oil-water interface may be moved across the test surface by immersing the tip of a pipette, a glass rod, or the like, in the drop and pulling the drop across the surface. This usually gives reliable advancing contact angles since the volume and surface area of the drop remain constant. In some cases a semi-solid interface forms between the oil and water. In such cases the oil drop when pulled across the test surface may leave a trail behind it or present a rough liquid interface in contact with the test surface. In either case, measurement of the exact contact angle becomes difilcult.
We prefer to produce an advancing contact angle by squeezing the oil drop between two parallel surfaces as soon as it is added to the test surface so the drop assumes a flattened shape. '1' hen, when the surfaces are moved apart, the water advances across the oil-covered surface producing an advancing contact angle. The area of the interface between the liquids is also increased. This avoids the difficulty observed when the area is decreased.
In our method, the wettability of a formation is determined by setting up a series of small test cells in which a drop of one of the liquids, usually formation oil, is squeezed between two test surface plates immersed in the other liquid, preferably brine from the formation. One cell is tested after a short time of an hour or so. Another after several hours, and so on. In each test the surfaces are moved apart. The advancing contact angle can then be observed until it reaches an equilibrium value. This can be compared to measurements made on samples aged for shorter periods of time to determine if the surface active agents in the crude petroleum hav reached equilibrium between the oil, water and solid surface. Such tests are repeated for any particular crude oil, solid surface, and brine until a true equilibrium advancing contact angle is obtained to indicate whether the formation is preferentially oil wet or preferentially water wet. Many variations of the above technique will be apparent to those skilled in the art. For example, several tests can be made in a single test cell moving the surfaces further apart on I successive tests.
If no brine is obtainable from the formation, a sample of the crude oil should be allowed to come into equilibrium with an artificial brine similar to that present in the formation.
The approximate natural brine composition can be obtained by grinding a sample of a core from the formation and extracting with fresh water the brine naturally present in the formation. The resulting brine can then be analyzed to determine the original natural brine composition. If no reliable analysis can be obtained, it is advisable to assume the formation is preferentially oil wet, determine the flooding water composition required to make the formation preferentially water wet, and check this composition in a native state core test. As previously stated, this final step may be eliminated in some cases, if desired.
The best flooding water composition for use in our method is preferably determined by measuring the iquilibrium advancing contact angle between a test surface and the interface between the formation crude oil and a series of flooding water compositions. The test surface may be quartz or calcite. Preferably both should be used. If the formation surface composition is known to be predominantly of a still different material, the contact angle should be measured against a solid impervious surface of this material. The test is carried out in the same way as the test to determine the original wettability of the formation. A few preliminary tests will generally indicate whether the natural salinity and pH of available flooding water should be increased or decreased. The degree of increase or-decrease can then be easily determined by setting up a series of contact angle test cells using flooding water compositions having various degrees of salinity and hydrogen ion concentration. The results of these tests will indicate how much adjustment of pH and'salinity should be made to insure water wettability of the formation when contacted by the flooding water.
We have found that the pH and salinity of the flooding water composition should be such as to reduce the contact angle to less than about 90 degrees.
It may be reasoned thatwater advancing with this contact angle should produce the best displacement of the oil. Thus, it might at first thought, appear that the water composition should be such as to produce an equilibrium contact angle of about 90 degrees. This has not been found to be true in practice. For best results, the water composition should produce an equilibrium contact angle somewhat less than 90 degrees, preferably no greater than about 70 degrees. The explanation for this observation i not certain. It may be that in actual flooding operations dilution of the flooding water by water naturally occurring in the formation, or adsorption of the acid, base, or salts by the formation, or even reaction with the formation or naturally occurring brine-s may occur to decrease the effects of the flooding water. operation sufficient time is not available for static equilibrium contact angles to be reached at the flooding front. That is, the flooding velocity is such that the dynamic contact angle is somewhat higher than the equilibrium value. Thus, a flooding water which will produce a static equilibrium contact angle of about 70 degrees could produce a dynamic contact angle during the flood of about 90 degrees, under'the conditions of the flood. Whatever the explanation, the flooding water composition should prefcrably be such as to produce an equilibrium contact angle of no greater than about 70 degrees.
Compositions producing contact angles smaller than about 70 degrees generally are also desirable. It should be noted, however, that higher concentrations of base, acid, or salt are required to achieve such smaller contact angles. Economic considerations limit the amounts of additives to be used. Technical considerations such as adverse efiects on clays and the like also may limit the concentrations which can be employed.
Many types of salts, acids, and bases may be used. Each one will produce slightly different results. The differences, however, usually are not great. Therefore, a salt, base, or acid should generally be selected which is not expensive and which will not adversely react with the in the figures.
It is also possible that in the floodingnr.
brine, oil, or formation. For these reasons, hydrochloric acid, sodium hydroxide and sodium chloride are generally used as the acid, base, and salt, respectively. It may be advisable in some cases, to use a buffer which will aid in holding the pH at the desired value. For example, if a pH of about 4 to 5 is found to be desirable, sodium acetate may be used as a buffer. If a pH of 7 or 8 is required, the buffer may be sodium boratc.
In field operations, the entire volume of flooding water may be adjusted to the desired composition. However, it may be satisfactory to adjust the composition of only the first portion of the injected water. This bank of water may then be displaced through the formation by any water composition available which does not seriously react with the ingredients of the flooding water. Preferably, the volume of the desired flooding water composition should make up at least about 25 percent of the estimated total volume of water to be used in the flooding operations.
An example of a crude petroleum oil containing surface active agents as described herein is the oil obtained from the Cardium sand of the Pembina Field in Canada. This crude oil, when left in contact with quartz or calcite for a period of a few days, causes these surfaces to be preferentially oil wet. To determine a possible explanation for this observation a sample of the crude oil was extracted to remove the surface active agents. The extracting solution was an aqueous solution of hydrochloric acid. After separation of the extract from the oil this aqueous solution was neutralized with sodium hydroxide and was extracted with a narrow boiling petroleum fraction consisting of hydrocarbons having predominantly from 10 to 12 carbon atoms per molecule. This petroleum fraction is referred to hereinafter as a C -C fraction. The C -C fraction does not normally cause surfaces to be preferentially oil wet. The extract of the neutralized aqueous solution, however, caused a quartz surface to become definitely preferentially oil wet. Acidified water in contact with this oil-wet surface converted the surface to a preferentially water wet condition.
A solution of n-octylarnine in C -C fraction behaves in much the same manner as the C -C extract obtained from the Pembina crude oil. The n-octylamine solution avoids many of the complications present when using crude oils. Therefore, this system was adopted for some of the fundamental tests, the results of which are shown In several cases, however, the results were checked using crude oils from several sources. Some of these results are also presented in the figures. Considering these figures in more detail:
7 FIGURE 1 presents the results of a test giving direct comparison of an ordinary waterflooding process and our flooding method. In this test a synthetic core 24-inches long was prepared by packing a Lucite tube 1% inches ID. with No. 16 Ottawa quartz sand. Equal volumes of mineral oil having a viscosity of about 35 cpse. and water were equilibratcd with about A percent by volume of n-octylamine. The core was then saturated with the oil phase and allowed to stand overnight, after which another pore volume of oil was put through the core. Experience has shown that this treatment is necessary in order to assure surface equilibrium. We have found that no flow test is valid unless such equilibrium is attained. The core was then flooded at a constant flooding rate of one foot per day. The flooding water was in equilibrium with oil and amine. As soon as the water to oil ratio reached a value of about 12 to l, the composition of the flooding Water was changed to a 0.025 N hydrochloric acid solution and the flooding was continued. As shown in FIGURE 1, a large additional amount of oil was recovered. FIGURE 1 also shows that when, in
ordinary flooding water. The quartz was water wet, however, when the acid water was present. This test demonstrates two very important points. First, flooding with water which changed the formation from preferentially oil wet to preferentially water wet condition produced a considerably increased quantity of recovered oil over that normally expected. Second, our method is applicable to reflooding of formations that have already been flooded by conventional water drive. In general, the refiooding will be more efficient if the oil phaseremaining is continuous over the sand surfaces.
The test from which data shown in FIGURE 2 were derived was run in much the same manner as that explained in connection with FIGURE 1 except that the oil phase was the C -C fraction previously described, and instead of changing the pH of the flooding water the salinity was changed. The salt used was sodium chloride. In FIGURE 2 the numbers above the curve indicate the equilibrium contact angles of the brine-oil interface on a quartz surface. It will be noted that in the flood test, increased recovery was obtained by use of a flooding water composition producing an equilibrium contact angle as low as 68 degrees. Part of the increased recovery using-50,000 p.p.m. salt rather than 25,000 p.p.m. may have been due to the further decrease in contact angle, but was probably due more to the ability of the stronger salt solution to reverse wettability more quickly as the flood front of water moved through the core; Thus, while the equilibrium angles are nearly the same for the two salinities, the dynamic angles actually existing during the flood may have been considerably different.
FIGURE 3 shows the results of measuring the contact "angles of systems in which the solid surface was quartz,
the water had various degrees of acidity and the oil phase was crude oil from the First Wall Creek Sand of the Salt Creek Field in Wyoming. Use of water having a pH in the range of about 2.4 to about 5.2 resulted in the quartz having a surface which was preferentially oil wet. Use of water solutions either more strongly acidic or more strongly basic resulted in the quartz having a surface which was preferentially water wet. Since the natural pH of water in this sand is about 8.4 the formations should be naturally water wet and this is known to be the same. Thus, our method of waterfiooding probably is not particularly applicable to this formation. if the pH of the naturally occurring water had been in the range from 3 to 4, it is apparent that the sand would have been preferentially oil wet and our method would have been applicable.
In FIGURE 4 the data were obtained using a solution of n-octylamine in the C -C fraction as the oil phase, quartz as the solid surface and water samples having various degrees of acidity as the water phase. In the upper curve the water contained substantially no salt. in the middle and bottom curves the water contained 25,000 p.p.m. and 50,000 p.p.m. respectively, of sodium chloride. The similarity of the top curve to that of FIGURE 3 is apparent except that due to the strongly basionature of the amine the peak of the curve is displaced into a higher pH range. A comparison of the upper curve in FIGURE 4 with the curve in FIGURE .3 shows that quite diiferent flooding water would be required for these two systems. The proper flooding water in each case could be determined by the method we have proposed. The difference between the three curves in 8 FIGURE 4 shows again the effects of salinity on the preferential wettability of surfaces. When 50,000 p.p.m. sodium chloride were present in the brine, the solid surface was preferentially water wet in all pH ranges.
It will be apparent from the above description that our discovery of the reason for preferential oil wettability of certain formations, with the discovery that changes of pH and salinity may reverse this wettability, together with our method of determining when the formation is preferen tially oil wet and the degree of changes in pH and salinity required to cause reversal of preferential wettability makes possible a method of recovering larger percentages of the oil from such formations than are recovered by ordinary waterflooding methods.
1. In a process for water flooding an oil-bearing formation which has been determined to he preferentially oilwet wherein flooding water is injected through an input well into said formation and oil is recovered through a producing well, and wherein an additive is used in said flooding water to reverse the wettability of said formation, the improvement comprising obtaining an air-free sample of crude oil from said formation, placing said crude oil in contact with a smooth, solid surface having a composition substantially the same as that of the surface exposed to the oil and water within said formation, placing water in contact with said surface and said oil, determining the equilibrium contact angle of the oil-water interface with said surface, repeating the determination of the contact angle using water compositions in which at least one of the pH and salinity is changed from preceding water compositions until ranges of pH and salinity of water have been determined which will make said surface preferentiallg, water-wettable in the presence of said crude oil, and adding to said flooding water at least one material selective from the group consisting of acids, bases, and salts to ajust the pH and salinity of said flooding water to values falling within the ranges which were determined to cause said surface to be preferentially water-wettable.
2. The method of claim 1 in which said contact angle is measured by contacting said solid surface in one of the liquids, thereafter applying a drop of the other liquid to said surface, and observing the contact angle of the oilwater interface with said solid surface until it reaches a substantially constant value.
3. The method of claim 2 in which said oil-water interface is moved in the direction of the oil wet surface to produce an advancing contact angle while maintaining the surface area of said drop at least as great as the original surface area of said drop, and then observing the contact angle until it reaches a substantially constant value.
References Cited in the file of this patent UNITED STATES PATENTS A. a... are.
UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent NO. 3,028,912 April 1O 1962 Virgil J. Berry, Jr. et al. It is hereby certified that error appears in the abo' ent requiring correction and that the said Letters Patent corrected below.
ve numbered patshould read as Column 1, line 32, after "past" insert is -3 column 5 lines 14 and 15, for "iquilibrium" read equilibrium column 7, line 44 for "same" read case column 8, line 37,, for "ajust" read adjust Signed and sealed this 28th day of August 1962.,
STON G. JOHNSON Commissioner of Patents