Because of unfavorable wetting conditions much residual oil is left when a porous material is flushed by water. Methods suggested to change reservoir wetting to improve oil displacement efficiency are generally expensive. The present laboratory study was undertaken to gain all understanding of the factors which determine reservoir wettability, and to find out if oil displacement efficiency might be improved by a wettability change accomplished at low cost in an oil reservoir.
Contact angle measurements were made on mineral surfaces using several sets of reservoir oil and water samples. Results of the contact angle studies suggest that reservoir wettability may be primarily determined by natural surface-active substances present in the reservoir fluids. The effect of changing salinity and pH of the water phase was studied. The results suggest that gross changes in preferential wettability might be accomplished by injection of water containing simple chemicals to alter pH or salinity in the reservoir. Such treatment could be much less expensive than injection of commercial surface-active agents.
Waterflood tests have also been made using synthetic cores and oil and water having wetting characteristics similar to those of reservoir fluids. Cores initially oil-wet were flooded in such a way that they were made preferentially water-wet by the advancing flood water. This reversal in preferential wettability achieved greater oil displacement efficiency than when either oil-wet or waterwet conditions were maintained throughout the flood. For the systems studied, the higher the oil viscosity the greater the percentage improvement obtained over conventional waterflood recovery. This suggests that a flooding process making use of wettability-reversal may extend the oil viscosity range over which water flooding is attractive.
Because a precise adjustment of reservoir wettability does not seem to be required, and because altering the pH or salinity in some reservoirs may be inexpensive, it appears that a waterflooding process employing wettability- reversal could find successful field application.
This paper presents the results obtained after calculating matches of the observed pressure performance of five fields completed in a common aquifer. A general description of the Central Basin Platform area in West Texas in which the five fields, Andector, Embar, Martin, TXL, and Wheeler, are located is contained in the paper. The method of utilizing the electric analyzer to calculate simultaneously matches of the observed pressure performance of the five fields is outlined. The determination of boundaries of pressure communication is discussed and the extent of pressure interference between fields consistent with the configuration of the area aquifer is shown graphically.
Anomalies in the pressure performance of the Andector Ellenburger Field resulted in an investigation of pressure interference between several Ellenburger fields located on the Central Basin Platform in West Texas. Four fields, Embar, Martin, TXL, and Wheeler, were found to contribute significantly to the pressure drawdown at Andector.
An initial investigation of the pressure performance of the Andector Ellenburger Field revealed that the reservoir pressure could not be matched by the calculation technique usually applied to water drive reservoirs. In the past, many have advanced the opinion that fracturing in the Ellenburger was a local condition restricted to the immediate uplifted area around the field, in which case each field would in all probability be surrounded by its own relatively small aquifer. Following several unsuccessful attempts to account for the pressure performance at Andector, a system assuming a common aquifer involving several fields in the immediate vicinity of Andector was placed on the electric analyzer and the effect of inter field interference was noted. A refinement of the initial analyzer calculations involved the determination of the approximate boundaries of the aquifer, the calculation of detailed matches of the pressure performance of the interfering fields, and an evaluation of the degree of communication between fields.
Geology and Development
The fields considered in this work are located in the Central Basin Platform, a north.south trending structural feature in West Texas lying between the Delaware Basin to the west and the Midland Basin to the east. The locations of various individual Ellenburger fields in this general are shown in Fig. 1 with the five major fields which appear to be definitely intercommunicating colored in black and located in the circled portion of the figure.
The various factors considered in recommending the initiation of a gas injection project in the southern portion of the Cedar Lake Field are discussed. Performance history under gas injection operations is reviewed and these data are analyzed, utilizing both the material balance method and the fractional flow and frontal advance expressions.
Results of the analysis of the performance data indicate that the injected gas has contacted and affected at least 60 per cent of the reservoir and a substantial increase in ultimate recovery can reasonably be expected. By holding the reservoir pressure appreciably above the bubble point, the well productive capacities have been maintained substantially above the level predicted for primary operations.
The analysis of the Cedar Lake project suggests that in certain limestone reservoirs, at least, the probable success of gas injection cannot be predicted simply from observation of permeability distribution throughout the pay section, as indicated by core analysis data, on either one or a number of wells. Further, the performance of this particular project fails to indicate any basis for classifying carbonate reservoirs in general as being inherently unsuited to a dispersed type gas injection program, thus indicating that each reservoir should be considered on its own merits, regardless of the composition of the reservoir rock.
Early in the life of the Cedar Lake Field, an extensive data gathering program was initiated to provide an accurate record of reservoir performance characteristics. From the study of these data it was apparent that there was a critical need for supplementing the natural reservoir energy in order to maintain well productivities and obtain the maximum ultimate oil recovery. Accordingly, detailed engineering studies were made of the various methods of secondary recovery which might be applicable. As a result of these investigations, the decision was made to initiate a gas injection program of sufficient intensity to maintain reservoir pressure at approximately 600 psia, or some 274 lb above the bubble point pressure of 326 psia.
Correlation of water with its reservoir zone or formation has been one ofthe applications of oil-field water analysis of greatest direct value to thepetroleum engineer. The water in each producing zone tends to have diagnosticcharacteristics by which it can be distinguished from every other water aboveor below that zone in that immediate vicinity. Representative analyses ofoil-field waters from producing oil and gas fields in the Rocky Mountain regionare included, and the diagnostic characteristics are discussed briefly. It isconcluded that the generally dilute nature of Rocky Mountain oilfield waters isa result of dilution by meteoric waters, and that there is no relationshipbetween presence or absence of commercial oil and the character of water in astructure.
The study of waters associated with oil and gas began more than 50 years agoand has been well recognized by operators, engineers, and geologists for about30 years. It appears unnecessary at this time to recite the history ofoil-field water analysis; suffice to say that it has proved its worth manytimes to the production engineer suddenly confronted with water problems inproducing oil and gas wells.
Chemists and geologists have studied the possible origin of these waters andas yet the subject is unsolved in many important phases. There are both localand regional problems connected with the explanation of concentrations andcharacteristics peculiar to each field, each subsurface zone and each province.Criteria useful in postulating the occurrence of oil or gas in one provincecompletely fail when applied to another, and data carefully prepared andanalyzed from one field may be actually misleading when applied to another.
Correlation of water with its reservoir zone or formation has been one ofthe applications of oil-field water analysis of greatest direct value to theengineer. The concentrations and characteristics of the waters are essential tothe engineers and geologists making interpretations of electric logs in thisregion. The behavior of water under conditions of reservoir temperature andpressure is dependent to a great degree upon its concentration andcharacteristics, and thus important to the reservoir engineer. With the adventof secondary recovery methods in the Rocky Mountain region the characteristicsand treatment of water will become increasingly important.
Premises Upon Which Correlation Is Based
A brief review of the geochemical history of oil-field waters will sufficeto present the premises upon which correlations of the waters are based.Sedimentary rocks which are now stratified were first sediments in seas, lakesand streams. These sediments were filled interstitially with connate water.With burial the sediments were compressed and consolidated, integrated andindurated into bed rock and much of the connate water was dispelled.
A mud-conditioning program found to be very effective for drilling andcompletion operations on routine field wells requiring relatively shortdrilling time involves a moderate alkaline-tannate-bentonite treatmentresulting in an ultimate filtration rate of 10.0 cc or less (API test). Mudweight schedules are planned from pressure information on completed wells inproducing reservoirs and drillstem test data obtained on other zones not beingproduced at present. In general, terminal mud viscosities average 45 sec(Marsh). This value has been found to be sufficient to remove cuttings from thewell bore on the average well, with the slush pumps in general use on therigs.
On field wells requiring drilling times in excess of approximately 30 days,an alkaline-tannate-lime-bentonite treatment has been effective in maintainingdesirable viscosities and filtration rates with a substantial reduction inchemical costs. This system has shown particular advantage on heavily weightedmuds and those with abnormal flow-line temperatures.
For the most part, the chemical treatment utilized in wildcat drillingfollows closely the program used on field wells, depending on depth andduration of drilling operations. Wildcat mud programs are planned frominformation available from various geologic and operational sources, takinginto consideration the possibility of encountering mud problems of a specialnature in the wildcat area. Careful planning of wildcat mud programs has provedto have definite value in avoiding most serious mud problems.
An analysis and recommended treatment are presented on several special mudproblems which have been encountered in the area in the past. Those problemsdiscussed are lost circulation, blowouts, sloughing shale, excessive chloridecontamination, sulphate contamination, and prevention and correction of cementcontamination.
Drilling-mud control in the Southwest Texas area, in general, does notinvolve the multiplicity of problems encountered in other coastal areas. Forinstance, it is rare that a mixture of problems such as lost circulation,abnormal pressure, severe sloughing shale, and others, occurs on one wellwithin close enough limits in vertical depth not to be taken care of by asensible casing program, together with a relatively clear-cut mud program.There are conditions such as the fairly widespread occurrence of abnormalpressure in the lower Frio and Vicksburg zones, and the tendency for sloughingshale in the Jackson section which make for rather expensive mud control.However, the availability of fairly accurate geologic and operationalinformation makes possible a reasonable degree of standardization on mudtreatment over a relatively large area. It may be said that, over most of thisarea where drilling is currently in progress, the natural mud made is very poorin quality, but responds readily to moderate chemical treatment to yield verygood wall-building characteristics.
Laboratory apparatus has been devised which permits study of the displacementof oil from cores by water and by gas. The cores used contained interstitialbrine as well as oil.
Experiments were run to determine the comparative effect of varying theproperties of the fluids used. No great effect was noted on the maximumdisplacement achieved. This observation made it unnecessary in initial work touse fluids in their exact reservoir conditions. Consequently, the displacementswere run at near-atmospheric pressure in Pyrex glass equipment, using strippedcrude oils.
Introduction and Theory
The chief object of this work has been to determine the efficiency of gasand water as primary agents for displacing oil from reservoir rock underlaboratory conditions in which capillary phenomena were predominant. To thisend the maximum displacement of oil from cores has been ascertained. Thismaximum displacement may not be equal to the maximum displacement from areservoir; but it will he a close approximation to it sometimes, and othertimes the laboratory information will be useful in reservoir engineeringpredictions. It is believed that the laboratory experimental maximum representsthe upper limit for the reservoir recovery.
The experiments were carried out by obtaining cores of interest from thereservoir, and filling the pores with interstitial brine and oil with therestored state technique. Then the oil was displaced from the core as describedlater, either by brine from below, or by gas from above. The former type ofdisplacement suggests analogy to production by water drive, but not to waterflooding, for reasons discussed below. The latter type of displacement isbelieved to simulate production by gas cap displacement.
The displacements were performed by what may be termed thecapillary-pressure method. The cores are placed in capillary contact with anoil-wetted membrane which has very small pores (about I micron in diameter).Pores of this size will transmit oil hut prevent the passage of gas or water,unless the pressures used are higher than the capillary pressures employed inthis work. Accordingly, use of the membrane makes it possible to apply acapillary pressure differential between the displacing phase and the oil in thecore.
The effects of various antifreeze agents on the formation of hydrogensulphide hydrate have been studied. On a molar basis the relative lowering ofTM (the maximum temperature at which solid hydrogen sulphide hydrate can existin equilibrium with the given solution) for the various agents is: sodiumchloride 1.00, calcium chloride 1.71, methyl alcohol 0.57, ethyl alcohol 0.68,ethylene glycol 0.73, diethylene glycol 0.73, sucrose 0.87, dextrose 0.71. On aweight basis the relative lowering of TM is: sodium chloride 1.00, calciumchloride 0.91, methyl alcohol 1.08, ethyl alcohol 0.89, ethylene glycol 0.69,diethylene glycol 0.41, dextrose 0.23, and sucrose 0.15.
The solid hydrate of hydrogen sulphide, H2S?6H2O1, is in equilibrium withwater and liquid hydrogen sulphide at 85?F at a partial pressure of hydrogensulphide approximately equal to the vapor pressure of hydrogen sulphide at thattemperature. At temperatures below 85?F, in the presence of excess hydrogensulphide (greater than the amount required to react with the water present) thehydrate is in equilibrium with liquid hydrogen sulphide at approximately thevapor pressure of hydrogen sulphide.
In the presence of excess water, the solid hydrate is in equilibrium withliquid water, containing dissolved hydrogen sulphide, at a pressure lower thanthe vapor pressure of liquid hydrogen sulphide at the same temperature. Thelower the temperature, the greater the difference between the decompositionpressure of the hydrate and the vapor pressure of liquid hydrogen sulphide atthat temperature.
Conditions conducive to the formation of solid hydrogen sulphide hydrate mayexist in wells producing a gas containing a high percentage of hydrogensulphide under high pressure, or in the lead lines from such wells. This papergives the results of tests made with various antifreeze agents which might beuseful in decomposing the hydrate or in preventing the formation of the hydrateof hydrogen sulphide.
Fig I shows a diagram of the apparatus used. The sample under observationwas contained in a glass tube 34 cm long, having a 6 mm id and a wall thicknessof 2.5 mm. The lower end of this tube was closed with a plug of butyl rubber.The upper end of the tube was connected to a diaphragm gauge, with a stainlesssteel diaphragm, a manometer, a vacuum pump, and a cylinder of hydrogensulphide.
The tube was held nearly horizontal, with the liquid resting along the lowerside of the tube. It was found that when the tube was held in a verticalposition, it broke with explosive violence as soon as it became filled withsolid hydrate. Presumably this was caused by the expansion which occurred onsolidification of the hydrate. The difficulty disappeared when the tube washeld in an inclined position during the tests.
The S.P. log is shown to be a measurement of the potential drop along thedrill hole, caused by ohmic effect in the mud. The notion of static S.P. isbrought forward, and its relation to the S.P. log is discussed. Other factorsinfluencing the shape and amplitude of the log are considered; attention isgiven to conditions encountered in practice. Numerous figures are givenillustrating graphically the results; these figures are of particular interestfor comparison with field examples.
The S.P. log, although indicating permeability, is not an absolutemeasurement of permeability, nor of porosity, of the formations traversed by adrill hole. It is affected by several parameters, such as resistivity offormations and mud, thickness of formations, and others, which should beappraised carefully. Simple rules have been established for a betterdistinction of the boundaries of permeable sections, particularly in difficultcases, such as those encountered in highly resistive formations. A systematicapplication of the established principles will assist in obtaining moreinformation from the S.P. log than was possible thus far; for instance, underfavorable conditions, presence of oil may be detected, or amount of shale insands may be estimated.
The S.P. log, or spontaneous potential log, has been known and widely usedduring the last 15 years for the location of permeable beds traversed by drillholes.
In electrical logging practice, the S.P. log is shown on the left hand sidetrack of the record (as may be seen in later examples) where it can be easilycorrelated and interpreted with the resistivity curves located to the right.Usually, the S.P. log consists of a base line, more or less straight, havingexcursions or "peaks" to the left. The base line frequently has beenfound to correspond to impervious beds, while the peaks are usually foundopposite permeable strata.
Measurements which will indicate positively the presence of permeability inthe formations, and which will give accurately the boundaries of the permeablezones, are of great importance in oil-field practice. Thus far, the S.P. log isthe best approach to such determinations; unfortunately, its interpretation isnot always evident.
With respect to the base line of the S.P. log, it may be noticed that thisline is not always at a definite location on the chart. Sometimes it may shiftabruptly, while other times a gradual drift is apparent.
As far as the peaks are concerned, their shape is not uniform; some arerounded while others are sharp. Also, from other data, it may be found thatoccasionally the peaks extend appreciably beyond the boundaries of permeablezones into zones which are not everywhere permeable.
A comparison with permeability measurements made on cores has oftenconfirmed that there was no definite correspondence between the magnitudes ofthe peaks and the permeability values.
This paper presents a theory for estimating the rate of gravity drainage ofa liquid out of a sand column. Account is taken of the variation inpermeability to the liquid as the saturation in the upper part of the sandbecomes less than 100 pct.
The theory is confirmed by previously published experimental data.
Petroleum engineers have expressed the need for a theory of gravitydrainage. Brunner, in particular, has pointed out that some type ofmathematical theory is necessary to begin the application of laboratory data tofield problems.
Muskat and his associates have recently made contributions to the theory ofgas-drive behavior and have indicated an intention to apply their methods towaterdrive systems. No theory of gravity drainage rates has been developed,however, and it seems desirable to formulate one at this time.
Differential Equations of Capillary Flow
The flow of liquids in partially saturated porous media has been studied bymany investigators. Richards presented derivations of fundamental differentialequations governing two-phase capillary flow; and used simplified forms ofthose equations in solving a steady-state problem. Muskat and Meres presentedand used equations different from those of Richards. Their equations did notexplicitly involve capillary pressure gradients; but included, on the otherhand, terms expressing the effects of the evolution of gas from the liquidphase during flow.
Leverett stated in 1940 that "previous work on the flow of fluidmixtures in porous solids [had] failed adequately to account for all of thethree influences that cause motion of the fluids: capillarity, gravity, andimpressed external pressure differentials." Leverett's basic equations,however, were specialized forms of the general equations of Richards, which hadactually taken account of the three influences mentioned by Leverett, but hadnot been used in a problem involving all three.
The fact is that our knowledge of capillary flow and our ability to expressthis knowledge in differential equations exceed our ability to solve theequations except in a few cases. General differential equations have usuallybeen of little more than formal value. In solving practical problems, it hasbeen necessary to develop specific equations, preserving terms that involvedthe factors important in those problems, and purposefully neglecting otherterms that were not of predominating influence. This is the method followedhere. It is believed that the solution of the particular problem and the schemeof the solution itself are new.
The term oil shale is defined. Foreign oil shale developments are outlinedbriefly. The richest and most extensive oil-shale deposits in the United Stateslie in western Colorado. The Bureau of Mines is opening two adjacent areas forunderground mining in western Colorado. From the first area the oil shale ismined selectively as required by the retorting plant. The second area is beingdeveloped as a unit of a full scale oil-shale mine to ascertain the mostpractical methods of mining oil shale and to determine costs. In both of thesemining areas, research will be carried on to develop new and improved practicesfor mining oil shale.
Oil shale is a sedimentary rock containing a solid mineraloid of indefinitecomposition known as kerogen. On heating oil shale, the kerogen undergoesthermal decomposition and some of the vapor products of the decomposition canbe condensed as shale oil. Oil from shale can be a major source of liquid fuelssince extensive deposits of oil shale are known to exist throughout the world.The United States considers the establishment of an oil-shale industry onlywhen doubt exists of the adequacy of the domestic supply of crudepetroleum.
On the other hand, oil shale is not strange to foreign oil men. The industryis an old one in other parts of the world. The French industry has been inexistence since 1838, the Scottish since 1859, the Australian since 1860, theEstonian and Swedish industries since World War I, and the Manchurian industrysince 1929.
The most highly exploited oil-shale deposits are the Carboniferous shales ofScotland. About 20 seams ranging from 4 to 12 ft in thickness are worked. Theyield during recent years in Scotland has been 18 to 25 gal of shale oil perton, from 3,000,000 tons per year. The French and Swedish oil-shale depositsare of lower grade than the Scottish. The Estonian oil shale deposits are 7 ftthick, and the oil yield averages 50 gal per ton. Over 1,000,000 tons are minedeach year, half of which is used directly as fuel to substitute for coal andthe other half is used for the production of oil.