US3155162A - Propping fractures with glass balls - Google Patents

Description

United States Patent PROPTTNG FRAQTURES WETH GLAfsS BALLS Don H. Fliciiinger and George C. Howard, Tulsa, and

Frederick H. Rixe, near (intense, (Firia, assignors to Pan American Petroleum Corporation, Tulsa, @irla, a

corporation of Delaware No Drawing. Filed Nov. 20, 196i, Ser. No. 153,692

3 Claims. (Ql. 16642) This invention relates to hydraulic fracturing of formations penetrated by wells. More particularly, it relates to increasing the flow capacities of fractures in such formations. This application is a continuation-in-part of our copending US. patent application Serial Number 708,068, filed op January 10, 1958, now abandoned.

In US. Reissue Patent 23,733, Farris, it is suggested that hydraulically formed fractures in formations such as oil-bearing formations be propped open by use of spacer materials such as sand. In practical use of the fracturing process described in the Farris patent, sand is almost always used for this purpose. Other materials, such as metal particles, produce much greater fracture flow capacities than the sand. This is thought to be because the metals are malleable; While sand is brittle. That is, when the overburden load is sufficient to exceed the elastic limit of the metals, the metal particles simply deform until sufiicient area of contact develops between the particles and the formation to support the load. In the case of sand, however, when the elastic limit of the particle is exceeded, the material shatters rather than deforming non-elastically. Such crushing continues until enough small particles are produced to support the load.

Fractures propped by crushed sand have low flow capacities for two reasons. First, the small particles cannot prop the faces of the fracture very far apart. Second, the small particles have a wide particle size distribution. The small particles bridge the pores between larger particles, thus greatly reducing the rate of flow through the crushed sand.

In spite 'of the low flow capacities of fractures propped by sand, this is the material most Widely used commercially. The metals are not generally used. Apparently, three reasons explain this lack of use. First, most of the metals have high densities. It is sometimes difficult to transport a material as light as sand in the liquids used in fracturing. It is even more difficult to support the heavy metals. Second, most of the metals are susceptible to rapid corrosion by brines present in oil-bearing formations. Relatively corrosion-resistant alloys, such as stainless steels, may, of course, be selected, but these are subject not only to the objection of high density but particularly to the third objection. This third reason for not using metals is that metal shot, when available at all in the size ranges desired, is very expensive. Due to the high density of the metal, this is particularly true if the propping agent requirements are determined on a volume basis. Corrosion-resistant alloy shot is, of course, even more expensive. Those skilled in this art apparently prefer to use the inert, low-density and inexpensive sand rather than the metals in spite of the tendency of the brittle sand to crush.

An object of this invention is to provide a fracturing composition containing low-density, inert propping agents, said composition being capable of effecting formation ice fractures having high flow capacities. Another object of the invention is to provide a method for effecting formation fractures having high flow capacities. A more specific object is to provide a new low-density, inert propping agent which can be deposited in formation fractures to increase the flow capacities of such fractures to values greater than those now produced by propping agents such as quartz sand.

We have discovered that fractures having very high flow capacities can be formed by using as a propping agent balls of a certain glass within a particular size range and having at least a specific minimum strength.

Although quartz is stronger and harder than most ordinary giass, a glass head will usually support a greater load without crushing than a quartz sand grain of the same size. This is because the surface of the glass bead is substantially free from flaws, while the quartz sand grain has many scratches, cracks, crevices, cleavage planes, and specks of impurities, all of which are points of Weakness. Thus, quartz sand grains will almost always crush before glass beads of the same size when a load is gradually applied. If the load is such that the sand grains crush While the glass beads do not, the fracture propped by the glass beads has a much greater flow capacity than the fracture propped by the quartz sand.

If the load is such that both the sand grains and glass beads are crushed, the comparison is between crushed glass and crushed sand. Here, the stronger and harder nature of the quartz becomes important. The quartz is not crushed to as great a degree as the glass. Therefore, the quartz propping particles are larger, propping the fracture open farther, and providing larger openings between the particles than in the case of crushed glass. This seems to explain the erratic results noted when ordinary solid glass beads are used as fracture props or spacers. It also explains the apparent anomaly that the glass beads, which are individually stronger than most individual quartz sand grains, frequently provide propped fractures having slightly lower flow capacities than when sand is used as a propping agent.

If slightly stronger glass beads are used, greater loads can be imposed before failure occurs. Unfortunately, however, it is never possible to predict exactly what load a bead will be called upon to support. Therefore, it is generally not possible to say whether a bead of a known strength will crush or not so long as it is only slightly stronger than an ordinary glass head or sand grain. Thus, as pointed out above, if the beads crush, use of the stronger beads may result in a flow capacity of the fracture less than if sand had been used.

Several attempts have been made to explain the erratic behavior of solid glass balls or beads. Additional data and information, however, have always shown the explanations to be incomplete or in error, or both. Therefore, it has not been possible to take advantage commercially of the very high flow capacities which are sometimes provided by the use of glass balls as fracture props. In all the Work with glass beads, there seems to have been only one type of glass balls which has always produced consistently good results even when used in loW concentrations in contact with very hard formations and under high overburden loads. This type of glass heads is the1 one used in tests 21 to 28 reported in the following tab e.

Cone. Over- Propping Particle Prop. burden Av. Free. Test No. Propping Agent Agent Size, Strength, Agent, Depth, Capacity, Type Book Remarks U.S. Sieve Lb./Part. Lb. /16 Feet Md-Ft.

Sq. Ft.

1 Ottawa Sand -16 -18 3 4, 500 1,000 Tensleep Sand grains broke.

d 20 -30 2 d D0. -16 -18 3 Do. -16 20 2 D0. -20 -40 2 Do. 10 12 D0. -16 -18 3 Do. -20 -30 3 Do. -12 D0.

10 2 Beads failed. 10 3 Beads failed in 3 tests. 10 4 -Bcads didnot fail. 7 2 Beads failed. 20 -25 3 D0. -14 -10 1 Some beads imbedded,

failed. -14 +16 3 5,900 23,000 Tensleep Lrcalized breakagc-28 day Q- -14 -16 2 6, 000 13, 200 Hard Sandstone... Little breakage in 1 test. -14 -16 2 d Beads crushed in 4 tests. do -14 -16 2 0 Some bead breakage. i do 14 -16 5% Short test. Glass Beads (Z) -12 -14 2 Very few beads failed. do 14 16 2 Do.

-12 -14 a Do. -12 -14 2 D0. -14 -16 2 Do. -12 -14 3 D0,." -12 --14 2 pa. 1 -12 -14 2 F%wtbeads failed-2t hour Reference is made above to solid glass balls or beads. 30 highest concentration of 14 to 16 mesh balls which could In general herein the terms balls and beads are used interchangeably. In some arts the term beads indicates balls with holes through them. As used herein, however, the meaning is used in the usual sense in the glass art where solid glass balls without holes are sold under the 'name beads for use in reflective paints and the like.

In the tests reported in the table, a 20-ton capacity hydraulic rain was assembled so that pressure could be applied to short cylindrical core sections. Some of the cores be packed in a single layer between the cores.

The crushing strengths of each i dividual bead were determined by placing a single bead between two tungsten carbide surfaces and slowly applying a load and observing the load at the time of failure of the bead. Other hard and strong materials can be used in place of tungsten carbide in this test. Malleable metals and alloys such as copper, brass, aluminum, and the like should be avoided as test surfaces since these surfaces become dented during were from the Tensleep formation and some were from a 40 the test, giving misleadingly high values for the crushing hard sandstone formation thought to be the Tensleep. These were obtained from wells cored through the formations inWyoming, Other cores were from the Cardium formation and 'were obtained from wells cored through this formation in the Pembina Field of Canada. The cores were sawed to expose smooth circular surfaces 3 /2 [inches in diameter. The propping material under test was uniformly distributed in the concentrations shown in ,the table over the surface between two of these core sections, the assembly mounted in the hydraulic ram, and the desired pressure applied. A hydraulic pump connected to the 20-to'n'rarn automatically maintained pressure on the system.

The effective overburden pressure on the propping materials, as reported in the table, was calculated from the applied rampressure. The equivalent well depth was determined from the effective overburden pressure previously derived from field data. The fracture capacity was calculated from data obtained by flowing nitrogen under a measured differential pressure through the fracture from a central hole drilled in theupper half of each core assembly. 7 V

The concentration of propping agent has been reported :in the table as the number of pounds of propping material per 16 square feet of fracture. This rather arbitrary figure was chosen on the assumption that the average fracture has a clearance of about /1 inch, Such a fracture has a volume of 1 gallon if the area is 16 square feet.

Thus, there is at least a rough correlation between the concentration figure given in the table and the concen- "tration'in the fracturing fluid. Actually, however, there is little relationship between the concentration of propping agent in thefracturing fluid and that in the fracture itself. Forpurposes of comparison, the concentration of 5 /2 pounds per 16 square feet used in test 20 was the 75 streng-ths of beads.

in describing particle size and distribution of particle size, it is convenient to define the balls as being of a certain mesh range. For example, if balls are said to be in the 14 to 16 mesh range, all but'a percent or so will pass a number 14 US. Standard sieve while all but a percent or so will be retained on a number 16 US. Standard sieve.

The Ottawa sand used in the tests was a quartz sand '50 having unusually round particles. The strength of inddividual sand grains varies widely as noted in the table. All the samples tested had some weak and some strong particles. When compressed between the core faces, there was always considerable crushing of the sand par- 'ticles, producing a wide distribution of particle sizes a representative commercial type of lime glass beads used for such purposes as reflective signs and the like. It will be apparent that the fracture propping ability was about the same as when sand was used, or slightly less, except 1n one case. In this case, a fracture flow capacity was obtained about thirty times as great as the best results obtained using sand. These exceptional results were obtained by carefully screening the beads so they were all almost exactly the same size and then using a high concentration of them (4 pounds per 16 square feet). A 0 slightly lower concentration (3 pounds per 16 square feet) failed in three tests to produce similar results since in every case the beads crushed. These data are typical ofthe erratic nature of'resu'lts when using ordinary glass beads as propping agents.

The experimental composition (G) produced beads -12 to 16 m esh-size-- range.

which were somewhat stronger. These were prepared on small scale equipment resulting in some variation in strength of individual beads. As shown in the table, most of the beads failed under loads ranging from 100 to 150 pounds per bead. A few beads failed even outside of this range.

The results were somewhat better than when ordinary glass beads were used. When the type (G) beads were used with the soft Cardium cores, the results were quite encouraging. Several explanations of the failures of these beads, when used with the harder cores, have been proposed. No completely satisfactory explanation has been offered which could be used as a basis for predicting when such beads should be used.

Type (Z) glass beads were prepared from an entirely new composition having the following analysis:

Percent Calcium oxide 6.26 Boron oxide 5.9 Sodium oxide 2.9 Silica 28.6

Due-to the high melting point of the composition, manufactoring methods capable of operating at higher temperatures should be used for forming our heads in the Methods such as those described in US Patents 2,460,977, 2,461,011, or 2,524,613 should be considered forforming these beads. As usual, the properties of the beads depend to some degree on the method by which they are made. Therefore, representative beads having any given composition and produced by any given method should be checked for average individual strength by the test previously described.

The reasons for the excellent results produced by the type (Z) beads are not completely understood. However, the satisfactory nature of the beads even in contact with a very hard formation at a great depth is apparent from the data. Therefore, it would seem that these beads can be used with confidence as props in any formations at depths up to at least 10,000 feet. A full monolayer of these beads in a fracture requires about 5.5 pounds per 16 square feet for the 14 to 16 mesh beads and about 6.6 pounds per 16 square feet for the 12 to 14 mesh balls. Thus, the concentration of beads between the core faces in tests 21 to 28 reported in the table was never more than about /2 of a full monolayer and was sometimes less than /3 of a full monolayer. It will be apparent that by use of a higher concentration of beads, a larger load can be supported so the beads can be used at depths even greater than 10,000 feet.

Regardless of the concentration of the glass balls in the fracturing liquid, the final concentration in the fracture tends to be about the same. The liquid leaks away from the balls, leaving them closely packed together. Thus, a closely packed layer of beads in the fracture generally is obtained automatically. If it is desired that the balls be spaced apart, this can be most conveniently done by use of a temporary diluting solid with the balls. The diluting solid is then deposited in the fracture with the balls and spaces them apart. When the well is produced, the diluting solid disappears, preferably by dissolving, leaving the balls spaced apart. If the zone to be fractured is oil bearing, the temporary material should, of course, be oil soluble. Examples of suitable materials include naphthalene, paradichlorobenzene, hexachlorocyclohexane, or some of the synthetic resins manufactured from coumarone, indene, dicyclopentadiene, or the like. If a waterproducing well is to be fractured, the temporary diluting solid should be a water-soluble material such as sodium chloride. If a gas zone is fractured, the temporary diluting solid should be a slowly volatile material such as naphthalene,

The particles of the temporary diluting solid should be within about the same size range as that of the glass beads; that is, they should be no more than about 1.2 times as large and no smaller than about 0.5 times as large as the average size of the balls. The fracturing liquid should preferably be saturated with the temporary diluting solid in order that the size of the particles of this material will not become so small as they are pumped down the well that there is danger of their forming a seal over a bridge of the balls at the face of the fracture. The use of temporary diluting solids with permanent fracture props is described in more detail and claimed in U.S. patent application l00,404, filed by C. R. Fast and G. C. Howard on April 3, 1961.

The portion of the fracture near the well is the most important, of course, so it may be desirable in some cases to inject other propping agents, such as quartz sand, first. The glass balls can then be used to prop the fracture near the well where the highest velocity flow occurs.

The carrying liquid for the balls should preferably. be a low-penetrating liquid as defined in US. Reissue Patent 23,733, Farris. Almost any mineral oil, Water, or mixtures thereof can be used as a carrying liquid if desired. The term mineral oil is intended in its broad sense to include crude petroleum and fractions thereof. The mineral oil, water, or mixtures of the two, may be used without additives or may contain materials in solution or suspension. For example, water may be fresh or salty and may contain colloidal solids, such as starch, to decrease the loss of the water to the formation. The mineral oil may be, for example, kerosene or diesel fuel without additives, or it may contain materials such as naphthalene in solution or colloidal solids such as heavy metal soap. When a fracturing fluid issaid to consist essentially of a carrying liquid and the beads, it will be understood that small amounts of othermaterials, such as those named above or other special purpose additives not substantially detrimental to the properties of the fracturing fluid, may also be present.

The type of carrying liquid and the average size and concentration of glass balls will depend somewhat upon the geographical location of the formation to be fractured. In some areas the formations are easily fractured and propping agents are easily placed in the fractures. In such cases the fracturing liquid may have a low viscosity, the balls may be large in diameter, and the concentration of balls in the liquid may be high. In other areas it is difiicult to form fractures, in such areas the carrying liquid should be viscous, the balls should be small in diameter, and the concentration of the balls in the liquid should be low.

Many other variations of the procedures described above will occur to those skilled in the art. In general, the glass balls can be used in most of the fracturing fluids or methods previously proposed, the balls being substituted for the spacer materials, such as sand, previously used or suggested for use in any fracturing operations. The balls used in the tests were not perfect spheres al though they were nearly so. The balls need not be perfect spheres, although the more nearly-spherical, the better. For our purposes, the average ball should have a smallest dimension at least about percent of the greatest dimension. If the formation fracture is horizontal and a few of the nonspherical balls have their long dimension vertical, the formation must crush down around these balls until those balls deposited with their short dimension vertical bear a part of the weight. It is even possible that, when the formation crushes to produce a greater area of support on the first balls contacted, the total loads on some'of these balls will exceed their individual crushing strengths, resulting in their shattering. The more nearly spherical the balls and the more nearly all are the same size, the less chance of such shattering. To avoid excessive shattering, the average smallest dimension of the balls should be at least about 80 percent of the average largest dimension.

We claim:

1. A composition for forming and propping a fracture in an earth formation comprising a carrying liquid and solid balls made from glass having an analysis consisting essentially of about 62 percent calcium oxide, about 6 percent boron oxide, about 3 percent sodium oxide, and about 29 percent silica, said balls being within the size range passing a number 12 US. Standard sieve and being retained on a number 16 U.S. Standard sieve, said balls having an average crushing strength of at least about 200 pounds, and most of said balls having a minimum diameter which is at least about 80 percent of the maximum diameter.

2. A process for increasing the productivity of fluids from a formation into a Well penetrating said formation comprising injecting a fracturing fluid through said well and into said formation, said fracturing fluid consisting essentially of a carrying liquid and solid balls made from glass having an analysis consisting essentially of about 62 percent calcium oxide, about 6 percent boron oxide, about 3 percent sodium oxide, and about 29 percent silica, said balls being within. the size range passing a number 12'U.S. Standard sieve and being retained on a number 16 US. Standard sieve, 'said balls having an average crushing strength of at least about 200 pounds, and most of said bails having a minimum diameter which is at least about 80 percent of the maximum diameter, and said fracturing fluid being injected into said formation at a rate sufiicient to fracture said formation and place said balls in the fracture.

3. A process for increasing the productivity of fluids from a formation into a well pentrating said formation comprising injecting a fracturing fluid through said Well and into said formation, said fracturing fluid including spaced individual strong a carrying liquid, strong, brittle balls, and particles of a temporary diluting solid, said fracturing fluid being injected at a rate sufiicient to fracture the formation and deposit said balls and said particles of a temporary diluting solid in said fracture, said balls being made from glass having an analysis consisting essentially of about 62 percent calcium oxide, about 6 percent boron oxide, about 3 percent sodium oxide, and about 29 percent silica, said balls being within the size range passing a number 12 US. Standard sieve and being retained on a number 16 US. Standard sieve, said balls having an average crushing strength of at least about 200 pounds, and most of said balls having a minimum diameter which is at least about 80 percent of the maximum diameter, and said particles, of a temporary diluting solid having substantially the same average size as said strong, brittle balls, and said temporary diluting solid having a melting point greaterthan the temperature of the formation to be fractured, whereby when formation fluids are produced from said well, said temporary diluting solid is" removed leaving the fracture propped open by Widely brittle balls to form high-capacity channels for flow of fluids from the formation to the well. j p

References Cited in the file of this patent UNITED STATES PATENTS 2,846,011 Miller Aug. 5, 1958 2,859,822 Wright NOV. 11, 1958 2,886,476 Dumesnil May 12, 1959 2,950,247 McGuire Aug. 23, 1960 UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent No. 3 155 162 November 3, 1964 Don Hy Flickinger et a1.

It is hereby certified that error appears in the above numbered patent requiring correction and that the said Letters Patent should read as corrected below.

Column 5, line 18, for "6.26" read 6206 Signed and sealed this 13th day of April 1965.

(SEAL) Attest:

ERNEST W. SWIDER EDWARD J BRENNER Attesting Officer Commissioner of Patents UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent No. 3,155,162 November 3, 1964 Don H, Flickinger et a1.

It is hereby certified that error appears in the above numbered patent requiring correction and that the said Letters Patent should read as corrected below.

Column 5, line 18. for "6.26" read 62.6

Signed and sealed this 13th day of April 1965.

(SEAL) Attest:

EDWARD J. BRENNER ERNEST W. SWIDER' Commissioner of Patents Attesting Officer

Claims (1)

1. A COMPOSITION FOR FORMING AND PROPPING A FRACTURE IN AN EARTH FORMATION COMPRISING A CARRYING LIQUID AND SOLID BALLS MADE FROM GLASS HAVING AN ANALYSIS CONSISTING ESSENTIALLY OF ABOUT 62 PERCENT CALCIUM OXIDE, ABOUT 6 PERCENT BORON OXIDE, ABOUT 3 PERCENT SODIUM OXIDE, AND ABOUT 29 PERCENT SILICA, SAID BALLS BEING WITHIN THE SIZE RANGE PASSING A NUMBER 12 U.S. STANDARD SIEVE AND BEING RETAINED ON A NUMBER 16 U.S. STANDARD SIEVE, SAID BALLS HAVING AN AVERAGE CRUSHING STRENGTH OF AT LEAST ABOUT 200 POUNDS, AND MOST OF SAID BALLS HAVING A MINIMUM DIAMETER WHICH IS AT LEAST ABOUT 80 PERCENT OF THE MAXIMUM DIAMETER.