US3413166A - Fine grained steel and process for preparation thereof - Google Patents

Description

26, 1968 v. F. ZACKAY ETAL 3,413,165

FINE GRAINED STEEL AND PROCESS FOR PREPARATION THEREOF Filed Oct. 15, 1955 2 Sheets-Sheet l INVENTORS VICTOR E ZACKAY EARL R. PARKER KRAHAMADHATI v RAVI BY 24.2, ,1 Q.GLA4&W

ATTORNEY NOV: 26, 1968 v ZACKAY L 3,413,166

FTNE GRAINED STEEL AND PROCESS FOR PREPARATION THEREOF Filed Oct. 15, 1965 2 Sheets-Sheet z A-2 STEEL 8 LU 53 z o o I U 2 5| E 1 LU 3 x 0 o 49Cycle Tampering Temperature 47 El 625C I 1 l 1 I Il l 75 H5 I55 l95 QUENCHING BATH TEMPERATURE, C

A-lSTEEL m m i 47 0 Ct Ij 45 Tempering Temperature L; 0 450C :5 0 600C g [I 625C 43 l 1l l l 35 Il5 I55 QUENCHING BATH TEMPERATURE, C INVENTORS Q VICTOR E ZACKAY EARL R. PARKER KRAHAMADHATI V. RAVI BY 7Q, g, ,1 Cl CAM ATTQRNEY United States Patent 3,413,166 FINE GRAINED STEEL AND PROCESS FOR PREPARATION THEREOF Victor F. Zackay, Berkeley, Earl R. Parker, Orinda, and

Krahamadhati V. Ravi, Berkeley, Calif., assignors to the United States of America as represented by the US. Atomic Energy Commission Filed Oct. 15, 1965, Ser. No. 496,729 7 Claims. (Cl. 148-125) ABSTRACT OF THE DISCLOSURE A high strength, ductile steel is formed by mechanically deforming austenitic steel while below the tempering temperature, then cyclically heating and cooling the steel to progressively lower temperatures so that on each cooling cycle martensite is formed as small size plates in progressively larger numbers along dislocation networks, a high density network of fine martensite plates being formed in the austenite matrix. The cycling is stopped short of complete transformation of the austenitic form to the martensitic form.

This invention relates to alloy steels having high toughness, and more specifically to the production of a steel having a fine austenitic-martensitic micro-structure. The invention described herein was made in the course of, or under, contract W7405eng48 with the United States Atomic Energy Commission.

An object of this invention is to provide steels having high tensile strength and ductility.

Another object of the invention is to provide a process for producing a high strength steel having a very fine grained austenitic-martensitic micro-structure.

As used hereinafter the symbol M will refer to the temperature of the initial austenite to martensite transformation. Also M refers to the final austenite to martensite transformation temperature point at which all of the austenite is converted to martensite.

The process of this invention comprises the deforming of steel in an austenitic condition to form a heavy dislocation network, the deformation being followed by a cyclic heat treatment consisting of alternate heating at an elevated tempering temperature and cooling to progressively lower temperatures. At the lower temperatures a fine martensite micro-structure is formed. When the steel is cyclically tempered the carbides precipitate on the heavy dislocation network formed by the initial deformation. The martensite plates are fine in size because they are confined to the carbide dislocation grid matrix.

The first step in the process is the deformation of an austenitic steel at a temperature which is less than the tempering temperature. The preferred deformation temperature is around 500 C., which is the temperature at which the dislocations are mobile and tend to segregate in an ordered array rather than a random dispersion. Because the number of dislocations is approximately proportional to the increase in strength of the cyclically treated steel, the percentage of deformation will likewise affect the increase in tensile strength. However, a minimum deformation of about 25% is required to form the heavy slip bands on which the carbides will precipitate. Less than 25% deformation will not give the high strength desired. The maximum deformation of the steel is about 90%.

The next step in the process after the warm working is the rapid quenching of the steel to a temperature slightly below the M point. In the steel composition of this invention the M temperature was maintained well below room temperature, --30 C. to -40 C. in order to prevent inadvertent conversion to martensite. On quenching Patented Nov. 26, 1968 the steel, the martensite plates so formed are small because they are trapped in the alloy carbide-dislocation grid. This effect is shown in the accompanying optical photomicrographs and will be hereinafter discussed in more detail.

Alloy steels which have been treated according to this process had the composition shown in Table I:

TABLE I Alloying Element, Weight Percent C Cr Ni Mo Si An important requirement which had to be met in the processing of the alloy is that the M temperature be well below room temperature, typically about 30 C. to -40 C. With an M below room temperature, it is convenient to work the steel and insure against inadvertent martensite transformation. A second requirement is that a balance is maintained between the alloying carbide formers and the carbon in the steel.

In balancing the carbon content against carbide formers sufiicient chromium and molybdenum was added to combine with all the carbon in the alloy. Three chromium carbides are known to be present; Cr C CI'7C3 and Cr C The carbide in greatest abundance in alloy steels with Cr is Cr C In the case of molybdenum, MoC and Mo C are known. However it has been shown that the precipitate in ausformed alloys is MoC. Accordingly, in the preparation of the alloys here it was assumed that Cr forms Cr C and Mo forms MoC. The nickel was added to lower the M of the alloy. Silicon was added to enhance temper resistance. It has been found in this invention that the Si addition is beneficial in giving somewhat higher ultimate tensile strengths with ductility being improved or maintained.

- The next step in the process is the tempering of the martensite formed in the first quench. Tempering of the martensite precipitates the carbides out of solution and forms carbides on the grain boundaries which limit the size of martensite formed in subsequent quench operations. Precipitation of the carbides lowers the initial M temperature and the steel is quenched again to a lower incremental value. The ideal incremental value was found to be 20 to 25 C. lower at each quench. However, this ideal differential varies with the composition of the steel. After each cyclic quench the steel is tempered. The steels were cyclically quenched down to liquid nitrogen temperature (196 C.) in less than ten cycles, depending on the increment used. Thus the process may be described as the cyclic quenching with intermediate tempering of an austenitic steel.

Previous experiments in the development of this invention were primarily concerned with producing a steel of martensite by cyclic cooling to below the M temperature. The martensitic steel produced in this way had the expected high strength because of the very fine microstructure, but was very brittle, having a ductility of less than 4%. In this invention it has been found that by retaining some austenite and forming a predominantly martensitic micro-structure in the austenitic matrix, a steel having an unusually high combination of strength and ductility is produced. Specifically, an A-l steel was produced with a tensile strength of 276,000 p.-s.i. at an elongation of 27%. This steel had a micro-structure of approximately 60% martensite-40% austensite as determined from the photomicrograph shown as FIGURE 1.

FIGURE 1 is an optical photomicrograph of 800x magnification for one type of steel. The steel depicted in the picture is classified as Al, and has the composition shown in Table I. Referring specifically to FIGURE 1 there is shown an A-l steel which has been processed portion was used. Liquid nitrogen was added to the alcohol to reduce the temperature to successively lower incremental values. An iron-constantan thermocouple was employed to measure the temperature of the cryogenic according to the following steps: 5 bath which was stirred frequently to minimize tempera- Ingots of induction melted Al steel approximately ture gradients. Close temperature control was assured by 2 /2" diam. and length, were cast in an inert atmosemploying a large amount of liquid and the use of fairly phere. These were then forged to 7 by A bars. Followsmall specimens. ing forging the ingots were ground to A by /4" bar For the initial cycle, the specimens were heated to the stock. 10 constant tempering temperature of 600 C. for 10 minutes Bars of 4" lengths were austenitized at a temperature followed by quenching to the lower temperature of of 1100 C. (1200 C. for A2 steels). No protective 35 C. for cryogenic treatment for a period of 5 minatmosphere was employed as the high chromium and utes. The cycle was repeated nine times using a constant nickel contents have the effect of conferring suflicient tempering temperature of 600 C. throughout and at sucoxidation resistance to the steels at the temperatures used. cessively lower quenching temperatures at C. incre- An austenitizing time of one hour was employed for all ments. After the final quench at 196 C. (liquid nitroof the steels tested. The bar stock was air cooled from gen) the specimen was subjected to tempering at a temthe austenitizing temperature to room temperature and perature of 600 C. for a period of minutes. The examined by metallograph and hardness measurements to specimen was electropolished and then etched electrolytiinsure that the samples were essentially 100% austenitic. 20 cally. Tensile tests were conducted on an Instron machine The bar stock was cleaned of any scale formed during using a crosshead travel speed of 0.1 cm./minute.

TABLE II Processing Strength, K p.s.i. Alloy Elongation, Reduction Steel Deformation Cycle Tempercent in area, (percent) at pering Yield Tensile percent Temp. C C.) Temp., C.

No.61--- so at 450- 500 176 192.5 15 No. 62... so at 450 500 218.2 270.9 9 No. 63 so at 450- 500 142.2 186.8 8 11-1.. 80 at 400- 600 197 242 26 Al. so at 400- 600 201.2 261 26 11-1 so at 400 600 206.6 276.1 27 36 11-1 so at 400 600 157.1 247.1 28 3s A2 so at: 600 450 229.1 286.6 16 14 A2 80 at 600----.- 460 220.1 290.0 16 15 A2 so at 600.---.. 450 235.7 281.1 15 15 21-2 so at 600 450 182.4 340.5 14 13 A2 so at 600 450 172. 7 331.0 19

1 Shown in Figure 1.

2 With final temper at 600 C. for 30 minutes. 8 With final temper at 550 C. for 15 minutes. 4 With final temper at 550 C. for 30 minutes.

austenitizing by grinding. The bar was cut into short lengths and sealed in /z" diam. 109 mm. wall stainless steel tubing. The function of the tubing was to insure uniformity of temperature during the rolling operation.

The specimens were heated in a tube furnace prior to and during the rolling operation. A preheat time of 15 minutes was used prior to the rolling operation in order to bring the stainless steel casing and the specimen up to the rolling temperature of 400 C. The deformation process was carried out in steps of 25 mils per pass during the initial stages of deformation, and 15 mils per pass at the final stages of deformation. After every two passes through the rolling mills, the specimen was returned to the furnace for a period of approximately two minutes for reheating. An infrared pyrometer sighted on the specimen during rolling was employed to insure that the specimen temperature remained relatively constant throughout the rolling operation. The total time for 80% reduction in thickness of the specimen was less than one hour. After the last pass through the rolling mill, the specimen was water quenched to retain the 'austenite phase at room temperature. The specimen was removed from the stainless steel jacket by cutting and samples were prepared for metallography.

The specimen was heated from room temperature to a pre-selected tempering temperature (cycle tempering temperature) followed by alternate quenching to decreasing cryogenic temperatures and intermediate tempering.

The heating medium for cycle tempering was a molten salt bath of nitrates in a resistance heated container. As shown in Table II tempering temperatures of 450 C., 500 C. and 600 C. were employed depending upon the steel composition. For the cryogenic quenching treatment a mixture of ethanol and liquid nitrogen in suitable pro- The tensile data for the Al steel process as described previously is shown in Table II, together with corresponding data for the other steels. Also shown in Table II is the tensile data for the Al steel which has not been given the 30 minute final tempering treatment. Without the final temper treatment the yield strength is 206.6K p.s.i., ultimate is 276.1K p.s.i. at 27% elongation. With the final tempering treatment it is seen that a slight increase in ductility (to 28%) is obtained, but with a concomitant loss in yield strength (to 157.1K p.s.i.). The same effect of increasing the ductility by the final tempering treatment is shown for the A2 steel, although the loss in yield strength is less.

The Al steel processed as described above without the final 30 minute temper is shown in FIGURE 1. The steel was electrolytically polished and etched with 2% nital. Upon close inspection of FIGURE 1, it can be seen that the carbides which have been precipitated out by the tempering treatment are uniformly distributed along the dislocation boundaries. The low carbon content of the Al steel results in a microstructure having a large percentage of untransformed austenite as shown by the white areas. The austenite is estimated to be 60% and the martensite is about 40% as shown by the dark areas. The appearance of the tempered martensite in FIGURE 1 indicates a plate size of very fine proportions; less than 1.0 micron. It is evident from studying the microstructure that the cycling process has the effect of forming a small amount of martensite with each cycle and the plates are confined to the heavy dislocation network. Transformation to martensite further increases the dislocation density due to the plastic deformation of the austenite by the volume expansion of the transformation reaction. This process, which could be termed internal cold-working,

increases the dislocation density and hence prevents the formation of agglomerated carbides by providing additional nucleation sites for carbide precipitation. These dislocations act as nucleating sites for carbide formation which is also facilitated by the tempering treatment between quenches.

The cycle tempering temperature has a significant effect on the properties of the product. Lower tempering temperatures are found to be suitable for optimum results with high carbon steel such as A-2, while higher tempering temperatures yield good results with low carbon alloys such as A-l. The lower temperatures in the case of low carbon steels are not high enough to induce carbide precipitation while the high temperature above the optimum overages the carbides, leading to a fall in hardness and hence strength. The high carbon alloy A-2 showed a drop in hardness at higher cycle tempering temperature of 625 C., this being due to excessive carbide agglomeration. Referring to FIGURE 2 there is shown the effect of the cycle tempering temperature on the final hardness for the A2 steel. As can be seen for the 625 C. cycle temper, there is a catastrophic drop in the Rockwell C hardness after several cycle treatments. This indicates the optimum cycle tempering temperature is in the range of 400 C. However, for the low carbon A1 steel a higher cycle tempering temperature produces the optimum hardness as shown in FIGURE 3. In this case, a cycle tempering temperature of 625 C. gives the highest Rockwell C hardness. Thus it is seen that the cycle tempering affects the hardness of the steel due to carbide agglomeration.

Additional test results for the other steels are shown in Table IV at varying deformation and cycle tempering temperatures.Steel #51, 62 and 63 were subjected to the complete thermo-mechanical treatment. All three steels had high strength and ductility, although in relation to the A-l and A-2 steels the toughness is much less.Steels #51, 61 and 63 contained no silicon or molybdenum as noted in Table I, and thus are less temperature resistant to the cycle tempering treatment. Inspection of the micrographs of #51 and #62 steels indicates non-uniform carbide distribution as compared to FIGURE 1 for A-1 steel containing Si and Mo. As indicated previously, FIGURE 1 shows homogeneous carbide distribution along the dislocation bands.

As can be seen from Table IV, the best combination of strength and ductility is obtained by the steels containing Si and Mo, with the proper cycle tempering temperature.

Initially, it was determined what efiect the amount of deformation and the temperature had on the martensite plate size formed on quneching of an austenitic steel. An alloy steel having a composition consisting of C, 0.86%; Cr, 9.78%; Mn, 0.74%; S, 0.10%; Si, 0.59%; Ni, 0.33%; was austenitized and deformed at three deformation rates (50%, 70%, 85%) and five differing temperatures. Upon quenching this steel the largest martensite plate lengths obtained were determined by optical metallography and are listed in Table III.

As can be seen from Table III, the percentage of deformation directly alfects the martensite plate size in an inverse relationship. A greater amount of deformation increases the dislocation density and thus limits the size of the martensite plates. Likewise, an increasing temperature limits the martensite plate size. This temperature effeet is presumably due to an increasing amount of carbide precipitation at the higher temperature.

Had no deformation of the austenite steel been performed the martensite plate size would be much higher as has been observed by examination of photomicrographs of an A-1 steel which has been austenitized at 1100 C. for 1 hour and, without deformation, direct quenched to liquid nitrogen temperature. The steel is essentially 100 martensite and has a martensite plate length of greater than 3.0 microns, more than twice the plate lengths shown in Table III. It can be seen that a substantial amount of deformation is required in order to achieve the fine martensite plate size in the final product.

In determining the exact nature of the strengthening mechanism in forming the A-1 steel shown in FIGURE 1, the process was altered by deleting certain of the steps given in the above example to determine the effect each had on the microstructure and hardness of the steel.

Sample of three A1 steels, after being austenitized at 1100 C. for a period of one hour, were given treatments corresponding to theProcesses 2, 3 and 4 listed in Table IV, which shows hardness data for the steel of FIGURE 1 (corresponding to Process 1) and Processes 2, 3 and 4. Photomicrographs with 800xmagnification following Processes 2, 3 and 4 were analyzed.

InProcess 2 the steel has been deformed at 400 C. and direct quenched in liquid nitrogen (-196" C.). The martensite plates were seen to be broken up by the dislocation bands. The familiar lightning bolt morphology of the martensite is evident. This can be understood in terms of the break-up of continuity of the austenite lattice by multiple slip resulting from deformation of the austenite. If a moving martensite plate is to continue to propagate along the same orientation it starts with, whenever it crosses a dislocation band in austenite, it will have to be displaced. In some cases the discontinuity may be too much for the plate to continue so a new plate will be propagated but displaced from the original. Also any precipitation on slip bands may stop martensite plates from propagating. The above described effect is clearly seen in photomicrographs of steel treated according toProcess 2 where the jagged martensite plates are blocked at the dislocation network and are prevented from propagating. The hardness of the steel afterProcess 2 is high, but is less than the cycled steel shown in FIGURE 1.

The other parameter, cycling, was isolated from deformation by cycling an austenitized sample of A-1 steel without prior austenite deformation as in Process 3. The hardness obtained was less than the direct quenched alloy (Process 2) and the deformed and cyclic quenched alloy (FIGURE 1). The effect of deformation on hardness is seen to be considerable. Following Process 3 the martensite formation is random and not confined by the slip bands, since no heavy dislocation network is formed. However, the cyclic quenching forms martensite plates of small size as compared with steel ofProcess 2 which is direct quenched. The finer martensite plates formed on cyclic quenching and tempering is probably due to the lowering of the M temperature during intermediate tempering. When the steel is intermediately tempered, carbides will precipitate out of solution, thus lowering the M point. Subsequent cyclic quenching at a temperature below the new M temperature forms additional martensite, but of a smaller plate size.

In Process 4 the standard microstructure of an undeformed steel direct quenched forms essentially martensite. The microstructure reveals the elongated martensite plates Which are typical when austenitic steel is direct quenched to below the M; temperature. The steel is very brittle with low ductility. The martensite plate size of the A-l steels fromProcesses 2, 3 and 4 is substantially greater than the steel in FIGURE 1.

For A-l steel, the first process shown corresponds to the optical photomicrograph of FIGURE 1, Numbers and 6 correspond to the steel with no quenching and are essentially 100% austenite. Step 5 indicates the effect on hardness by the process of 80% deformation, which essentially is a normal coldworking.

While the invention has been disclosed with reference to certain specific examples, it will be apparent that numerous variations and modifications are possible Within the scope of the invention and it is not intended to limit the invention except as defined by the following claims.

What is claimed is:

1. The method of forming and heat treating a steel, comprising:

(a) heating a mass of steel at a temperature sufficient to render the structure austenitic, said steel having a chemical composition such that the austenite is stable at room temperature,

(b) quenching said steel rapidly to avoid any transformation to pearlite or bainite, whereby said uastenitic structure is retained,

(c) deforming said austenitic structure at a temperature wherein a high density dislocation network with heavy slip bands is formed, said deforming being sufiicient to effect a deformation of at least 25% locally,

(d) quenching the austenitic steel mass to a first cyclic quench temperature slightly below the martensitic transformation temperature, whereby elongated martensitic plates are formed in the austenite matrix,

(e) temperating the martensite microstructure so obtained at a temperature from about 150 to about 625 degrees centigrade,

(f) performing a plurality of additional cyclic quenching and tempering operations on said steel at successively lower quenching temperatures than said first cyclic quench temperature, said tempering being at a temperature in the range from about 150 to about 625 degrees centigrade, whereby each successive cyclic quench forms additional martensite of smaller grain size than the initial martensite,

(g) terminating the cyclic quenching prior to the complete transformation of the austenite to martensite, and

(h) final tempering the martensitic-austenitic microstructure at a temperature in the range from about 150 to about 625 degrees centigrade, whereby the ductility is increased.

2. The method of forming and heat treating a steel described in claim 1 wherein said cyclic quenching is terminated when a microstructure comprising at least 10% austenite is obtained, the remainder being essentially martensite.

3. The method of forming and heat treating a steel described in claim 1 wherein said steel has a chemical composition containing sufficient carbide forming elements selected from the group consisting of chromium and molybdenum in such proportion that the austenite is stable at room temperature.

4. The method of forming and heat treating a steel as described in claim 1 wherein said mass of steel has a chemical composition consisting essentially of carbon 0.10 percent to 0.80 percent, chromium 2.0 percent to 8.0 percent, nickel 8.0 percent to 22.0 percent and the balance iron.

5. The method described in claim 4, wherein the cyclic tempering temperature of steps (e) and (f) is selected from the temperature range of about 400 degrees centigrade to about 625 degrees centigrade.

6. The method of forming and heat treating a steel described in claim 4 wherein said initial tempering of the martensite microstructure at step (e) is a temperature in the range from about 400 to about 625 degrees centigrade and wherein the cyclic quenching of step (f) is terminated when said steel has a composition comprising about austenite and about 40% martensite.

7. A high strength alloy steel produced by the process of claim 1 comprised of fine grained tempered martensite in an austenitic matrix which has a metallurgical structure in which carbides are uniformly distributed along dislocation boundaries, the martensite having a plate size less than one micron across.

References Cited UNITED STATES PATENTS 2,934,463 4/1960 Schmatz et al. 148143 3,028,270 3/1962 Morita et al. 148143 3,178,324 4/1965 Grange et al 148143 X 3,189,493 6/1965 Chen 148-143 X 3,201,288 8/1965 Grange 148-124 3,250,648 5/1966 Grange et a1. 148--12.4

OTHER REFERENCES Bullens, Steel and Its Heat Treatment, vol. III, 1949, John Wiley & Sons Inc., New York, pp. 573582.

CHARLES N. LOVELL, Primary Examiner.

Images