US3357458A - Process and product for improved resistance to stress corrosion - Google Patents

Description

Dec. 12, 1967 Riled Aug. 5,

F. J. RADD ET AL 3,357,458 PROCESS AND PRODUCT FOR IMPROVED RESISTANCE TO STRESS CORROSION 1964 5 Sheets-Sheet 1 0 IO N o o o N o 0 z 2 n: o B 8 5 6 8 Q L o o It I 5 Lu c o 8- s a s O N o o O o o o o BELLBWVIG d0 NOIlOIIGEIH INVENTORS FREDERICK v.1v RADD LOU/5 H. WOLFE ATTORNEY Dec. 12, 1967 F. J. RADD ET AL 3,357,458

PROCESS AND PRODUCT FOR IMPROVED RESISTANCE TO STRESS CORROSION 5 Sheets-Sheet 2 Filed Aug. 5, 1964 lSd OOO'I) 88381.8

INVENTORS FREDERICK J. RADD BY LOU/S H. WOLFE TORNEY Dec. 12, I967 F. J. RADD ET AL 3,357,458

PROCESS AND PRODUCT FOR IMPROVED RESISTANCE TO STRESS CORROSION Filed Aug. 5, 1964 O 3 Sheets-Sheet IOO DESCALED DESCALED CYCLES (MILLION) (lSd OOO'I) $83818 INVENTORS FEEDER/CK J..RADD BY LOU/S H. WOLFE ZQ IMW United States Patent Ofiice 3,357,458 Patented Dec. 12, 1967 3,357,458 PROCESS AND PRODUCT FOR IMPROVED RESISTANCE TO STRESS CORROSION Frederick J. Radd and Louis H. Wolfe, Ponca City, Okla.,

assignors to Continental Oil Company, Ponca City,

Okla, a corporation of Delaware Filed Aug. 3, 1964, Ser. No. 386,884 6 Claims. (Cl. 138-177) This application is a continuation-in-part of copending application Ser. No. 321,027, filed Nov. 4, 1963, and now abandoned.

The present invention is directed to metal products having improved resistance to stress corrosion and a process for the preparation of said products.

As used herein, the term stress corrosion is defined in the broad sense to include corrosion which occurs during the application of externally applied stress and also situations wherein no external stress is applied. When applied, the external stress can be continuous or intermittent and can vary in magnitude.

Steel and other metals play an import-ant part in the production, refining and processing of petroleum. Many of the applications of metals in the petroleum industry involve corrosive environments. This is particularly true in the production of crude petroleum where environments are encountered which contain brines, that is, various metal halides, and in particular, sodium chloride; sulfurcontaining compounds, such as hydrogen sulfide; carbon dioxide; organic acids and other corrosive materials. Often these corrosive materials are present in the crude petroleum or, if not, are present in the formations which are adjacent the petroleum-producing zones. The problems of corrosion and how to combat it are particularly important to manufacturers of well casing and tubing which are employed in connecting the ground surface with the oil-bearing formations and to the manufacturers of sucker rods or oil well pump rods which are employed in conjunction with pumps which affect removal of crude petroleum through oil well casing. These sucker rods are steel rods which come in various diameters, usually 25 or 30 feet in length, having threads or other means at each end whereby the rods can be coupled together to form a string of sufiicient length to extend from the pump driving means which is stationed above ground down to the oil well pump which is located down near the oil-producing formation.

Sucker rods are ordinarily made from carbon or low alloy steels and are installed in service without any special treatment other than heat treatment such as normalizing or tempering. It is recognized that mill scale-bearing surfaces are undesirable, and some of the manufacturers treat the sucker rods to remove hard mill scale by flame treatment or, more commonly, by shot blasting or shot peening. These various processes for removing mill scale have been followed for up to fifty years or more. In general, these processes succeed in removing most but not all of the mill scale.

At the present time, the life expectancy of sucker rods is relatively short and in highly corrosive environments such as those containing a large amount of hydrogen sulfide, failure may occur in a very few months or even in a shorter time. Failures also occur at a very high rate in services wherein the stresses placed on the sucker rods in the pumping operation are very high. While not subject to externally applied stresses like the sucker rods, well casing and tubing also experience in corrosive environments a much shorter life than would be obtained in the absence of corrosive materials.

It is an object of this invention to provide an improved process for reducing the failure of metals due to stress corrosion.

It is another object of this invention to provide an improved process for treating the surface of metals to increase their resistance to stress corrosion.

Still another object of this invention is to provide metal products having improved resistance to stress corrosion by a novel method of surface treatment.

Yet another object of this invention is to provide sucker rods and tubular goods having improved resistance to stress corrosion.

These and other objects of the invention will become more readily apparent from the following detailed description and discussion.

The foregoing objects are achieved broadly by provid ing metal products from which substantially all oxidedefective and decarburized surface layers have been skinned off the metal surface by a mechanical treatment and the surface thereafter subjected to surface coldworking to induce compressive stresses in the metal surface. The oxide-defective and decarburized surface layers are those which are formed on the metal usually either during manufacture or during subsequent heat treatment.

In the drawings:

FIGURE 1 shows in graphical form the relationship between reduction in diameter and depth of cold-working for A.I.S.I. 1036 steel.

FIGURE 2 shows the results of fatigue tests pertinent to the invention.

FIGURE 3 is another graphical illustration of fatigue tests pertinent to the invention.

In one aspect, the invention is directed to improved sucker rods and their method of preparation. In another aspect, the invention is directed to improved tubular goods and their method of preparation; such as, drill pipe, which is employed in drilling wells and in particular, wells for the production of hydrocarbons; tubular casing, which is employed in wells to retain the formation and through which petroleum products can be produced; production tubing, which can be disposed withinthe well casing for producing a plurality of formations; and line pipe which is employed for movement of petroleum above ground, for example, in gathering systems.

While it is realized that the individual steps of removing surface oxides and decarburized layers from metals by mechanical treatment and surface cold-working of metals are both very old in the art, it has been found unexpectedly that the combination of these two treatments provides a surface which has substantially greater resistance to stress corrosion than the ordinary metal surface. This finding is particularly unexpected when one considers that stress corrosion, particularly that which occurs in an environment of hydrogen sulfide, has been considered by many to be aggravated by stresses introduced in the surface of metals, for example, by surface cold-working. In addition, skinning-E metal oxides and decarburized layers to expose the bare metal surface per so would not be expected to provide a material having enhanced resistance to stress corrosion. Yet, as stated, the combination of these two operations has unexpectedly provided .a very substantial improvement in corrosion resistance.

In carrying out the process of this invention, the skinning off process by which the oxide-defective and decarburized layers are removed can be effected by any mechanical means, for example, by a cutting or machining operation, by milling, by grinding, or by any other mechanical operation which assures substantially complete removal of oxides and decarburized layers and preferably, in the case of sucker rods or well casing, resulting in a product of substantially uniform cross section. This metal removal process generally involves an operation to a depth of from 0.002 to about 0.010 inch and preferably, not greater than about 0.015 inch. Examination of the metal for effective oxide and decarburized layer removal can be made visually or with a suitable scanning device for a continuous commercial operation.

The second step of the process, the surface cold-working, is effected in such a manner so as to impart compressive stresses, preferably of a substantially uniform nature, in the surface of the metal. These stresses can be provided in any of the conventional manners by the employment.

of either constant or impact tooling, such as, shot peening, cold-rolling, cold-drawing, and swaging. Surface coldworking can provide tensile stresses as well as compressive stresses, therefore, it is important that the cold-working be carried out in such a manner as to assure the desired compressive stresses. The surface cold-working can be employed in varying degrees depending on the particular metal involved. Generally, it is carried out in such a manner as to provide cold-worked surface layers of a depth ranging from about 0.001 inch to as high as 0.02 inch or higher. The compressive stresses involved will range in magnitude from as low as 100 p.s.i. to as high as 50,000 p.s.i. or higher.

The process of producing cold-worked surface layers involves a reduction in thickness of the material being treated which differs in magnitude from the depth of coldworking. Usually, the reduction in thickness, expressed as reduction in diameter, is from about 0.25 to about 0.45 times the depth of cold-working. The relationship between reduction in diameter and depth of cold-working for A151. 1036 steel is shown in FIGURE 1 of the draw rngs.

While the process of treatment as described herein is referred to as cold-working, this treatment can be effected over a range of temperatures which can be subatmospheric, for example -20 F. or -30 F. or lower, extending upwardly, for steels, to as high as about 950 F. The temperature of cold-rolling as defined herein is the temperature at which the surface treatment is initiated, for example, cold-rolling at atmospheric temperature indicates no heating or cooling of the work piece prior to cold-rolling. During the cold-rolling operation, of course, heat is generated due to the work input to the metal which ordinarily results in an increase of the temperature ,of the work piece. In any case, the cold-working actions herein refer to temperatures which do not produce appreciable by-product annealing and/ or grain growth but which leave the grains in a strained condition.

The pressure employed during cold-working will depend on the type of means used for effecting this treatment and the depth and rapidity of the cold-working operation. Pressure is not critical and any amount required to effect the cold-working can be employed.

Steels are the principal metals of construction employed in the oil industry, therefore, the invention finds particular application in the ferrous metals and particularly steels. More preferably, the invention is directed to non-austenitic steels, that is, those of a pearlitic or ferritie strucd ture. In addition to steels, the invention also finds applications to other metals and metal combinations, that is, alloys, which in their preparation or processing have obtained an oxide-defective and decarburized surface layer. Some examples of other metals and alloys which come within the scope of the invention are those involving K- Monel, nickel, copper, aluminum, cobalt, titanium, zirconium, vanadium, iron, rhodium, platinum and the like.

Sucker rods are normally prepared from solid bar stock, therefore, the improvements contemplated by this invention are carried out on the outer surface of the sucker rods. It is possible, however, that for some services a hollow sucker rod might be desirable, for example, when it is desired to introduce a corrosion inhibitor or some other material to a pumping well. In this event, it is contemplated that the process of the invention can also be Example I Thirty fatigue samples were cut from one piece of inch Axelson 60 sucker rod. Axelson 60 is a fully normalized rod that is made of A.I.S.I. 1036 steel. (This rod is labeled rod A in Table I.) The specimens were turned down with a lathe to a diameter of approximately 0.50 inch. A gage section was provided in the middle of each specimen having a radius of curvature of 0.5 inch and a minimum thickness of 0.30 inch.

The thirty fatigue specimens were sanded circumferentially with #1 paper until all of the tool marks were removed. Then they were normalized in the laboratory furnace. They were heated to 1650 F. for one hour and then air cooled. This heat treatment left a black oxide (Fe O scale on the samples.

After the samples were normalized, six of them were tested in R. R. Moore Rotating Beam Fatigue Machines at approximately 1725 r.p.m. without disturbing the black oxide coating. The results of this test are shown in FIG- URE 2 on the curve labeled Scale Present.

The remaining twenty-four samples were sanded circumferentially with #1 grit paper until the scale decarburized material was removed. (Each specimen was examined with a low power microscope to ascertain that the scale and decarburized materialwas removed.) Eight of these specimens were then tested in the fatigue machines, and the results are shown in FIGURE 2 on the curve designated Descaled."

Six of the remaining specimens were cold-worked with Az-inch balls. The balls were rotated around the center 'of the gage section of the sample. By using a tapered cone, the balls were forced into the sample as they rotated. The samples were cold-worked until their diameter was reduced by 0.010 inch. This produced a coldworked band that was about 7 of an inch wide. When these specimens were run in the fatigue machines, an increased performance was observed, but all specimens broke outside of the cold-worked area in a larger diameter section. (The stress is inversely proportional to the diameter.) The results of these tests are shown in Table I,

The depth of the cold-working was measured microscopically to be about 0.020 inch.

The remaining ten specimens were cold-worked with one-inch balls to produce a cold-worked band throughout the entire gage section. The reduction of the diameter was measured to be about 0.010 inch. The specimens were then tested in the fatigue machines with the results being shown on the curve Descaled and Cold- Worked in FIGURE 2.

The fatigue tests were conducted in air. Brine (3 percent NaCl in distilled water) was dropped on the center of the sample at a rate that was suflicient to keep it wet. The samples were tested at pressures varying from 20,- 000 to 50,000 p.s.i., i.e., approximately the normal working stress levels for oil Well pumping rods. Usually three samples of each set were tested at each stress level.

Subsequently, additional samples were prepared from another A.I.S.I. 1036 steel sucker rod. This set of specimens was from one, /s-inch diameter, Oilwell, Type N rod. (The Oilwell rod is labeled rod B in Table I.) This grade rod was also normalized at the factory as a standard manufacturing procedure.

The fatigue performance of this rod (descaled, 3 percent NaCl) was checked by running one sample at each of the following stress levels: 20,000, 30,000 and 40,- 000. It was run as nearly as possible under the same conditions as the descaled Axelson 60 rod, and its aerobic corrosion fatigue behavior was close to the performance of the Axelson rod. The results of these tests are also shown in Table I.

Example 11 Fatigue samples from the Oilwell rod of Example I were treated in the following manner.

(1) All specimens were sanded circumferentially with #1 grit paper.

(2) Twelve specimens were cold-worked with oneinch diameter balls, The one-inch balls cold-Worked the entire gage section. The diameter was reduced by 0.010 inch.

enclosure where the stress was substantially lower, but where air and a small amount of H 8 and brine were present. The third sample ran 22,367,000 cycles and broke in the gage area.

The results of these tests are shown in FIGURE 3. Hardness tests. were carried out on one specimen on two Rockwell ranges with the following results. (The specimen was from rod B.)

Cold-worked area:

This shows a consistent increase in hardness in the cold-worked area.

All of the data obtained in Examples I and II are pre- (3) Three of the non-cold-worked specimens were sented in numerical forminTable TABLE I Cycles Scale Present Descaled Descaled and Cold-Worked Stress 3% NaCl 3% N 2.01 3% NaCl 3% NaOl+H2S 3% N aCl 1 3% NaCl 3% NaOl+H S (Rod A) (Red B) (Rod A) (Rod B) (Rod A) (16" (Red A) (1" (Rod B) Balls) Balls) 20, 000 2, 497, 000 2, 340, 000 3, 020, 000 5, 018, 000 6, 092, 000 37, 863, 000 2 50, 549, 000 1,089, 000 4, 045, 000 3 550, 000 6, 000, 000 31,430, 000 F 52, 090, 000 1, 771, 000 2, 055, 000 6, 618, 000 11, 255, 000 34, 224, 000 Average 1, 786, 000 3, 040, 000 5, 818, 000 7, 782, 000 34, 506, 000

30, 000 761, 000 1, 158, 000 1, 072, 000 5, 684, 000 3, 141, 000 23, 225, 000 4 I4, 249, 000 799, 000 1, 614, 000 3, 223, 000 4, 681, 000 10, 194, 000 4 12, 368, 000 596, 000 831, 000 2, 599, 000 6, 599, 000 14, 127,000 22, 367, 000 Average 719, 000 1, 172, 000 3, 835, 000 4, 807, 000 15, 849, 000 16, 328, 000

11, 987, 000 Average 13, 027, 000 50, 000 87, 000 7, 660, 000 5, 278, 000 Average 6, 469, 000

1 The balls cold-worked a short length 2 Sample did not break, but it was cracked 3 This sample is not included in the average. 4 Sample broke outside of the gage section.

tested in the fatigue machine at 20,000 p.s.i. and three were tested at 30,000 p.s.i. These specimens were tested in 3 percent NaCl saturated with H 8. During the test, the sample was enclosed in a sealed container which made the system fairly air free.

(4) Two of the cold-worked samples were tested in the H 8 anaerobic system at 20,000 p.s.i., and they ran 50,000,000-I- cycles without breaking. Three of them were tested at 30,000 p.s.i. and two of them ran 12,000,-

of the gage section and all samples broke outside of the cold-worked area.

It is noted from the data in Table I that the removal of scale from the specimens provided an increase in fatigue life (cycles before failure) ranging from about 25 to 60 percent in the presence of sodium chloride. Further treatment (cold-working with l-inch balls) of the descaled specimens provided an increase in life over the untreated specimens from about 1800 to 2100 percent in the sodium chloride atmosphere.

It is particularly surprising that substantial improve- 000-| cycles. Both samples broke outside of the sealed ment in fatigue life is obtained in the corrosive environment containing hydrogen sulfide which is generally held to be much more damaging than an environment of sodium chloride alone. Normally, when steels increase in hardness, there is an increased hydrogen sulfide cracking susceptibility.

Example III Three additional specimens of the Oilwell rod (rodB) were prepared following the procedure previously described. The specimens were tested at 30,000 psi. in

Example IV Twelve additional specimens of the Oilwell rod (rod B) were prepared following the procedure previously described. The specimens were cold treated to different depths and then were tested at 20,000 and 30,000 p.s.i. in aerobic 3 percent NaCl. The results of the test are shown in Table III.

TABLE III Reduction Reduction Stress Of Cycles of Cycles (p.s.i Diameter, Diameter,

inch inch Average. 0. 0063 5, 723, 000 0. 0183 14, 570, 000

The results of this example show that cold-working to a depth of about 0.021 inch (reduction in diameter from 0.0171 to 0.0192 inch) provides substantially improved performance over a lower amount of cold-working, namely, to a depth of about 0.017 inch (reduction in diameter from 0.0059 to 0.0091 inch).

Example V Additional specimens were prepared from K-Monel in dimensions corresponding to those set forth in Example 1. These specimens were sanded and annealed at 1300 F. followed by a water quench. Half of them were coldworked to a depth of 0.015 inch (reduction in diameter 0.010 inch) and were then tested at 30,000 psi. in aerobic 3 percent NaCl. The results of these tests are presented in TablelV.

It is apparent from the data in Table IV that greatly improved life under corrosive conditions is also obtained by treating a metal other than steel in accordance with the process of this invention.

The reasons for the great improvement in resistance to stress corrosion obtained by the process and products of this invention is not readily apparent. For example, as pointed out previously, sucker rodshave been treated by shot peening which would induce compressive stresses in the surface of the rods. Rods treated in this manner, however, have been shown by tests to be not substantially better than untreated rods. Thus, it appears that the descaled step of the present invention is of major importance even though this step per se does not provide a significant improvement in resistance to stress corrosion. Since removal of scale and decarburized layers is so important, it is possible that previous methods of treatment such as shot peening' have been ineffective in their failure to remove substantially all of the oxide scale and decarburized layers from the metal surface. In any event, notwithstanding any speculations as to the reasons for the improvement obtained, it is apparent that the combined oxide and decarburized layers removal and cold-working steps of the present invention result in major improvements in resistance of metals to stress corrosion.

Although this invention has been described with respect to particular embodiments thereof, it is not to be so limited as changes and modifications may be made therein which are within the spirit andscope of the invention as defined by the appended claims.

We claim:

1. A process for the treatment of a steel sucker rod for use in corrosive environments containing hydrogen sulfide which comprises skinning-off by mechanical treatment substantially all oxide-defective and decarburized surface layers and thereafter inducing compressive stresses in the surface of said sucker rod by surface cold-working.

2. Sucker rods prepared in accordance with the process of claim 1.

3. The process of claim 1 in which the cold-working is efiected to a depth not exceeding about 0.020 inch.

4. A process for the treatment of steel oil well piping for use in corrosive environments containing hydrogen sulfide which comprises skinning-off by mechanical treatment substantially all oxide-defective and decarburized surface layers and thereafter inducing compressive stresses in the surface of said oil well piping by surface cold-work- 5. Oil well piping process of claim 4.

6. The process of claim 4 in which the cold-working is effected to a depth not exceeding about 0.020 inch.

References Cited prepared in accordance with the UNITED STATES PATENTS 1,178,813 4/1916 Lloyd 138-177 1,779,478 10/1930 Leech 72-40 2,680,938 6/1954 Peterson 134-16 X 2,701,408 2/1955 Borger 7253 OTHER REFERENCEs Romanov, V. V.: Stress Corrosion Cracking of Metals, Jerusalem, Israel, Program for Scientific Translations, 1961, page. 142.

Waber, I. T., and McDonald, H. 1.: Stress, Corrosion Cracking of Mild Steel, Pittsburgh, Corrosion Publishing Company, 1947, pp. 66-68, and 84.

SAMUEL ROTHBERG, Primary Examiner. B, E. KILE, T. L. MOORHEAD, Assistant Examiners.

UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent No. 3,357 ,458 December 12, 1967 Frederick J. Radd et a1.

fied that error appears in the above numbered pat- It is hereby certi on and that the said Letters Patent should read as ent requiring correcti corrected below.

Column 7, line 54, for "1300 F." read 1600 F.

Signed and sealed this 14th day of January 1969.

(SEAL) Attest:

EDWARD J BRENNER Edward M. Fletcher, 11'.

Commissioner of Patents Attesting Officer

Claims (1)

1. A PROCESS FOR THE TREATMENT OF A STEEL SUCKER ROD FOR USE IN CORROSIVE ENVIRONMENTS CONTAINING HYDROGEN SULFIDE WHICH COMPRISES SKINNING-OFF BY MECHANICAL TREATMENT SUBSTANTIALLY ALL OXIDE-DEFECTIVE AND DECARBURIZED

Images