Title:
Weldable, age hardenable, austenitic stainless steel
United States Patent 3895939
Abstract:
An age hardenable, austenitic stainless steel having superior weldability properties as well as resistance to degradation of properties in a hydrogen atmosphere has a composition of from about 24.0 to about 34.0 weight percent (w/o) nickel, from about 13.5 to about 16.0 w/o chromium, from about 1.9 to about 2.3 w/o titanium, from about 1.0 to about 1.5 w/o molybdenum, from about 0.01 to about 0.05 w/o carbon, from about 0 to about 0.25 w/o manganese, from about 0 to about 0.01 w/o phosphorous and preferably about 0.005 w/o maximum, from about 0 to about 0.010 w/o sulfur and preferably about 0.005 w/o maximum, from about 0 to about 0.25 w/o silicon, from about 0.1 to about 0.35 w/o aluminum, from about 0.10 to about 0.50 w/o vanadium, from about 0 to about 0.0015 w/o boron, and the balance essentially iron.
US Patent References:
Austenitic alloy
Aggen - November 1962 - 3065067
Austenitic stainless steel
Rundell - February 1967 - 3306736
Gas-shielded arc welding of 18% nickel steel
Corrigan - March 1967 - 3309243
Austenitic alloy
Aggen - November 1962 - 3065067
Austenitic stainless steel
Rundell - February 1967 - 3306736
Gas-shielded arc welding of 18% nickel steel
Corrigan - March 1967 - 3309243
Inventors:
Brooks, John A. (Livermore, CA)
Krenzer, Robert W. (Arvada, CO)
Krenzer, Robert W. (Arvada, CO)
Application Number:
05/411554
Publication Date:
07/22/1975
Filing Date:
10/31/1973
View Patent Images:
Export Citation:
Assignee:
The United States of America as represented by the United States Energy (Washington, DC)
Primary Class:
Other Classes:
420/586.100
International Classes:
C22C38/50; C22C38/46; C22C38/50; C22C38/44; C22C38/06
Field of Search:
75/128V,128T,124,128F,128W
Primary Examiner:
Rutledge, Dewayne L.
Assistant Examiner:
Steiner, Arthur J.
Attorney, Agent or Firm:
Horan, John King Dudley Resendez Ignacio A. W.
Claims:
What is claimed is
1. A stainless steel for use in hydrogen environment with good retention of properties having high weldability and minimum susceptibility to hot cracking during welding consisting of from about 0 to about 0.0015 weight percent boron, from about 0 to about 0.25 weight percent silicon, from about 0 to about 0.25 weight percent manganese, from about 0.01 to about 0.05 weight percent carbon, and the remainder consisting of from about 24.0 to about 34.0 weight percent nickel, from about 13.5 to about 16.0 weight percent chromium, from about 1.9 to about 2.3 weight percent titanium, from about 1.0 to about 1.5 weight percent molybdenum, from about 0 to about 0.01 weight percent phosphorous, from about 0 to about 0.010 weight percent sulfur, from about 0.1 to about 0.35 weight percent aluminum, from about 0.10 to about 0.50 weight percent vanadium, and the balance iron.
2. A stainless steel of claim 1 comprising about 0.0014 weight percent boron, about 0.16 weight percent silicon, about 0.11 weight percent manganese, about 0.027 weight percent carbon, about 24.7 weight percent nickel, about 14.7 weight percent chromium, about 2.18 weight percent titanium, about 1.25 weight percent molybdenum, about 0.004 weight percent phosphorous, about 0.005 weight percent sulfur, about 0.10 weight percent aluminum, about 0.25 weight percent vanadium, and the balance iron.
3. A stainless steel of claim 1, electron beam weldable in the solution treated condition without cracking at electron beam gun parameters of 75 inches per minute welding rate, 150 kilovolts, 17 milliamps, sharp focus, and 8 inch gun to workpiece distance.
4. A stainless steel of claim 1 wherein the maximum alloy weight percent for phosphorous is about 0.005 weight percent and for sulfur is about 0.005 weight percent.
5. A stainless steel of claim 1, wherein said hydrogen environment comprises a hydrogen atmosphere at about 3,500 pounds per square inch and at 200°C temperature.
6. A stainless steel of claim 1 with ductility properties not severely adversely affected when supersaturated with hydrogen.
1. A stainless steel for use in hydrogen environment with good retention of properties having high weldability and minimum susceptibility to hot cracking during welding consisting of from about 0 to about 0.0015 weight percent boron, from about 0 to about 0.25 weight percent silicon, from about 0 to about 0.25 weight percent manganese, from about 0.01 to about 0.05 weight percent carbon, and the remainder consisting of from about 24.0 to about 34.0 weight percent nickel, from about 13.5 to about 16.0 weight percent chromium, from about 1.9 to about 2.3 weight percent titanium, from about 1.0 to about 1.5 weight percent molybdenum, from about 0 to about 0.01 weight percent phosphorous, from about 0 to about 0.010 weight percent sulfur, from about 0.1 to about 0.35 weight percent aluminum, from about 0.10 to about 0.50 weight percent vanadium, and the balance iron.
2. A stainless steel of claim 1 comprising about 0.0014 weight percent boron, about 0.16 weight percent silicon, about 0.11 weight percent manganese, about 0.027 weight percent carbon, about 24.7 weight percent nickel, about 14.7 weight percent chromium, about 2.18 weight percent titanium, about 1.25 weight percent molybdenum, about 0.004 weight percent phosphorous, about 0.005 weight percent sulfur, about 0.10 weight percent aluminum, about 0.25 weight percent vanadium, and the balance iron.
3. A stainless steel of claim 1, electron beam weldable in the solution treated condition without cracking at electron beam gun parameters of 75 inches per minute welding rate, 150 kilovolts, 17 milliamps, sharp focus, and 8 inch gun to workpiece distance.
4. A stainless steel of claim 1 wherein the maximum alloy weight percent for phosphorous is about 0.005 weight percent and for sulfur is about 0.005 weight percent.
5. A stainless steel of claim 1, wherein said hydrogen environment comprises a hydrogen atmosphere at about 3,500 pounds per square inch and at 200°C temperature.
6. A stainless steel of claim 1 with ductility properties not severely adversely affected when supersaturated with hydrogen.
Description:
BACKGROUND OF INVENTION
The present invention relates to a high strength, age hardenable austenitic stainless steel having superior weldability and good retention of properties in a hydrogen environment.
Nickel-chromium-iron alloys are known which may have a yield strength of 150,000 pounds per square inch (psi) at room temperature and have good resistance to corrosion and oxidation even at high temperatures, but are generally known to be highly susceptible to weld hot cracking, e.g., both heat affected zone (HAZ) microfissuring and fusion zone hot cracking. These alloys may also have a reduction of properties upon exposure to a hydrogen environment.
The susceptibility of these stainless steels to weld cracking has not only limited their applications but has also given rise to severe rework problems. Several hot cracking mechanisms have been suggested to explain weld HAZ cracking and fusion zone cracking in nickel-chromium-iron alloys of this type. For example, intergranular cracks associated with a secondary grain boundary phase (identified through x-ray diffraction as the Fe 2 Ti Laves phase) in specimens heated to about 1,340°C have been observed. This phase forms a eutectic with iron at about 1,300°C. It has been proposed that HAZ cracking may be associated with a liquation of the grain boundaries by the Fe--Fe 2 Ti eutectic which allows the boundaries to separate under the welding stresses.
Whatever theory may be proposed, weldability problems are encountered. Thus HAZ microfissuring and fusion zone cracking have been encountered not only in connection with the welding operation itself but also during age hardening treatments of these alloys. This invention determines the source of hot cracking in austenitic age hardenable stainless steels and identifies a eutectic phase and provides a novel alloy composition not susceptible to the objectionable weld hot cracking problem. Although it may be possible to minimize weld cracking on the varying compositions by reducing the welding rate, reducing heat input, welding thin sections, and the like, a large potential for cracking may still exist. This novel alloy provided in this invention enables the use of the high welding rates with minimum susceptibility to hot cracking in the weldment.
Although some nickel-chromium-iron alloys may have "good" weldability as compared to others, these may sustain significant loss of properties in a hydrogen atmosphere or environment. For example, hydrogen embrittlement tests on tensile specimens which have been exposed to a hydrogen environment at about 3,500 psi at a temperature of about 200°C for several days may result in such as a greater than about 34 percent loss, and in some instances, as much as about 50 percent loss, in ductility as well as a different fracture mode when tensile tested at ambient or room temperature (i.e. -- about 25°C).
SUMMARY OF INVENTION
In view of the above problems and limitations, it is an object of this invention to provide an improved nickel-chromium-iron base alloy or stainless steel which has minimal susceptibility to hot cracking during welding processes.
It is a further object of this invention to provide a stainless steel compatible with an environment of hydrogen such that tensile and yield strengths are not severely adversely affected when supersaturated with hydrogen.
It is a further object of this invention to provide a novel stainless steel having ductility properties determined during the cooling process by Gleeble testing similar to ductility properties determined during the heating process.
Various other objects and advantages will appear from the following description of the invention and the most novel features will be pointed out hereinafter in connection with the appended claims. It will be understood that various changes in the details and composition of the alloy components which are herein described in order to explain the nature of the invention may be made by those skilled in the art without departing from the principles and scope of the invention.
The invention comprises a stainless steel alloy which has reduced susceptibility to hot cracking during welding operations, which alloy contains not more than about 0.0015 weight percent (w/o) boron, not more than about 0.25 w/o silicon, from about 0.01 to about 0.05 w/o carbon, from about 0 to about 0.25 w/o manganese and further including from about 24.0 to about 34.0 w/o nickel, from about 13.5 to about 16.0 w/o chromium, from about 1.9 to about 2.3 w/o titanium, from about 1.0 to about 1.5 w/o molybdenum, from about 0 to about 0.01 w/o phosphorous, from about 0 to about 0.010 w/o sulfur, from about 0.1 to about 0.35 w/o aluminum, from about 0.10 to about 0.50 w/o vanadium, and the balance essentially iron.
DESCRIPTION OF DRAWING
FIG. 1 graphically plots spot varestraint test results for various alloy compositions.
FIG. 2 plots the Gleeble heating cycle employed.
FIG. 3 graphically plots percent reduction in area vs temperature for a specific alloy during the on-heating cycle and also during the on-cooling cycle.
FIGS. 4-11 graphically plot ductility curves and ultimate tensile strength curves for composition of Samples B through I.
DETAILED DESCRIPTION
A high strength, age hardenable austenitic stainless steel compatible with hydrogen (e.g., when supersaturated with hydrogen) and having a yield strength (at 0.2% strain or offset) of about 100,000 psi as well as high weldability, including minimal susceptibility to weld cracking, was desired. This invention description determines the cause of hot cracking in prior similar austenitic stainless steels and provides an alloy which has reduced susceptibility to hot cracking both in the HAZ and in the fusion zone as well as not having a degradation of properties upon exposure to hydrogen atmospheres at temperatures of about 200°C and pressures of about 3,500 psi.
Table I shows compositions of various samples of heats. A, B and J samples are of generally known compositions. A number of B samples were heat treated at about 1,350°C for times of from five minutes to about two hours. Electron microprobe analysis showed that, after heating, grain boundary phases of these samples were highly enriched in nickel, titanium, silicon and carbon and depleted in iron and chromium. Microprobe results indicated at least two different titanium phases in the boundaries, one heavily concentrated in titanium and appearing to be associated with carbon which was probably titanium carbide formed on cooling, and another phase enriched in titanium, nickel and silicon. X-ray diffraction on particles dissolved from the matrix showed two phases:
A tiC type (a o = 4.33 A) and an A 2 B type Laves phase (a o = 4.76 A, C o = 7.71 A).
contrary to previous theories, the high temperature phase forming in the boundaries is not the simple Fe--Fe 2 Ti eutectic but rather a complex A 2 B type phase highly enriched in nickel, titanium and silicon. Thus prior theories assumed the nature of the phase could not be affected since a titanium reduction would result in a loss of strength. This invention shows that the complex A 2 B type phase allows selective alteration of ingredients to yield a reduced susceptibility to cracking. This same phase is observed in Gleeble samples heated to a maximum temperature above 1,290°C (as shown in the graphs and as described hereinbelow) in about 6 seconds and rapidly cooled. Thus the end result of the high temperature segregation of alloying elements in materials having compositions similar to A, B and J samples, is a formation of nickel rich phases in the boundaries which play an important role in the hot cracking mechanism.
TABLE I ____________________________________________________________ ______________ HEAT CHEMISTRY, WT.%, BALANCE Fe ____________________________________________________________ ______________ Sample C Mn P S Si Cr Ni Al Mo V Ti B ____________________________________________________________ ______________ A 0.037 1.22 0.015 0.012 0.70 14.29 25.59 0.22 1.37 0.22 2.25 0.0046 B 0.063 1.50 0.004 0.005 0.58 14.45 25.40 0.22 1.27 0.26 2.25 0.0073 C 0.031* 1.51 0.004 0.006 0.63 14.50 25.70 0.24 1.27 0.26 2.22 0.0059 D 0.025 1.47 0.004 0.005 0.61 14.50 25.50 0.25 1.26 0.25 2.20 0.0009 E 0.028 0.13 0.006 0.005 0.25 14.77 26.01 0.23 1.25 0.26 2.18 0.0063 F 0.027 0.11 0.004 0.005 0.16 14.70 24.70 0.10 1.25 0.25 2.18 0.0014 G 0.028 1.54 0.004 0.005 0.63 14.25 34.70 0.25 1.24 0.25 2.15 0.0064 H 0.025 1.51 0.004 0.005 0.60 14.75 25.45 0.66 1.27 0.23 1.60 0.0012 I 0.025 1.39 0.010 0.008 0.58 15.20 23.50 0.24 1.25 0.25 2.63 0.006 J 0.068 1.32 0.018 0.003 0.63 14.90 24.93 0.19 1.25 0.21 2.15 0.004 ____________________________________________________________ ______________ 6 *Underlining denotes major changes from known compositions.
TABLE III ____________________________________________________________ ______________ Sample C Mn P S Si Cr Ni Al Mo V Ti B ____________________________________________________________ ______________ C 0.031 1.51 0.004 0.006 0.63 14.50 25.70 0.24 1.27 0.26 2.22 0.0059 K 0.027 0.09 <0.01* <0.01 0.1 14.34 25.50 0.23 4.98 0.25 2.55 0.0004 L 0.020 0.11 <0.01 <0.01 0.15 14.48 30.46 0.27 1.22 0.25 2.07 0.0010 ____________________________________________________________ ______________ *< = less than
It is shown that the relationship of silicon to the nature of the high temperature (1,290°-1350°C) grain boundary phase is an important factor in the mechanism of hot cracking since a reduction of silicon in the alloy has been found to reduce the hot cracking tendency.
Various alloy compositions are shown as heats or samples C through I in Table I, these being vacuum induction melted or other suitable furnace melted with charge additions of appropriate purity to obtain the alloy chemistry described herein and cast into ingots which were rolled and cut into appropriate desired sizes, such as a plate size of 3 × 3 × 0.25 inches. The purity and chemistry of the alloy may be optimized by remelting using a vacuum consumable arc remelt furnace after vacuum induction melting. These were then solution treated at about 995°C for about 45 minutes in an air atmosphere and water quenched to yield an ASTM grain size of about 6. Electron beam welding parameters which are known to be severe on ordinary compositions were employed for welding tests. Five 90 to 100 percent penetration circular welds with diameters of 1/2 to 21/2 inches were made on each plate. Electron beam welding parameters were 75 inch per minute welding rate, power of 150 kilovolts and 17 milliamps using an electron beam having maximum convergence on the workpiece surface, i.e., a sharp focus, the gun being about eight inches from the workpiece. These tests were designed to easily discern the relative hot cracking susceptibility of each composition. Table II illustrates the percent of weld sections exhibiting gross cracks. Major chemistry modifications from the standard samples A, B and J are underlined in Table I. The sample letter designated is the same for Table I and Table II.
TABLe II ______________________________________ ELECTRON BEAM WELD RESULTS ______________________________________ % Weld Sections Exhibiting Samples Gross Cracks ______________________________________ B 72 C 70 G 45 H 38 E 18 I 10 D 8 F 0 ______________________________________
As noted from Table I and Table II (which lists the samples in order of decreasing number of cracks) samples B and C show that the reduction of carbon from 0.06 w/o to 0.03 w/o has no beneficial effect on fusion zone cracking. Both the increase in titanium from 2.25 w/o to 2.63 w/o in sample I and the increase in nickel from 25 w/o to 35 w/o in sample G appear to have some limited beneficial effects in fusion zone cracking. In addition, samples D and E showed that both boron and silicon are detrimental to fusion zone cracking. Comparing the results from D and H, it was noted that the substitution of about half a percent of titanium with aluminum was not effective in reducing fusion zone cracking in welds. As Table II demonstrates, which confirms the discovery hereinbefore stated, no gross cracking was observed in sample No. F wherein the carbon, boron, manganese and silicon are all low.
From the welding results, it was noted that increases in both nickel and titanium and decreases in both silicon and boron were extremely effective in reducing fusion zone cracking and this indicated a novel alloy composition which has high yield strength, was age hardenable, and did not have a high susceptibility to hot cracking during welding as prior art compositions had. Tables I and II clearly verify that compositions that may be similar to this invention do not have the reduced susceptibility to weld cracking of this invention.
Scanning electron microscopy studies on the fusion zone cracks during the welding procedures verified that the cracks formed during the last stages of solidification. These fusion cracks exhibited segregation of a low melting point phase which, by using the nondispersive x-ray system of the scanning electron microscope, was found to be enriched in titanium-silicon-nickel and depleted in iron and chromium with respect to the matrix. Analysis of the HAZ verified that this phase existed in the HAZ as well as in the fusion zone.
Spot varestraint testing was conducted on compositions B through I to determine the relative HAZ hot cracking susceptibility. These spot varestraint tests show the effect of chemistry on HAZ cracking. Spot varestraint testing involves melting a spot on a specimen with a gas tungsten arc weld torch. After a predetermined time, the arc is extinguished and a ram of known radius applies pressure to the specimen. The augmented strain is the additional strain to the regular weld shrinkage stresses caused by the ram bending the specimen. By using rams with different radii, it is possible to plot curves relating augmented strain to total crack length in the HAZ surrounding the spot weld as shown in FIG. 1.
Spot varestraint testing equipment incorporates a delay timer which may be set such that the ram strikes the sample from 0 to 500 milliseconds after the gas tungsten arc torch is extinguished. In subsequent tests, a delay of 140 microseconds was employed. Welding parameters of 130 amps, 8 volts, using argon shielding gas for a time of 8 seconds on the 0.25 inch thick specimens were used. Surface cracks in the HAZ only were measured under 30 power and 60 power magnification to determine the total crack length for each sample.
FIG. 1 illustrates the results obtained from the spot varestraint testing and compares the relative hot cracking susceptibility of all of the experimental alloy compositions. The effect of chemistry modifications for HAZ microfissuring should be compared to sample C since all samples except sample B have low carbon content. As further shown in FIG. 1, the increase in titanium from 2.22 w/o to 2.63 w/o (sample I) is extremely detrimental to HAZ hot cracking, even though it is beneficial to reducing fusion zone hot cracking. Also, the decrease in carbon from 0.063 to 0.031 w/o has some detrimental effects. Note that compositions F and H which lie on the abscissa of FIG. 1 did not have any HAZ microfissuring using the above test parameters. The decrease in manganese and silicon (sample E), the decrease in boron from 0.0059 to 0.0009 w/o (sample D) and the direct substitution of 0.4 w/o aluminum for titanium all had beneficial results on the reduction of HAZ microfissuring. The increase of 10 w/o nickel (sample G) had very little, if any, significant effect. No cracking was observed in samples F and H even after 8% augmented strain.
Hot ductility (Gleeble) testing was conducted on the various alloy compositions to predict the susceptibility of each material to HAZ hot cracking during welding. Both on-heating to the zero strength temperature (ZST) and on-cooling testing from the ZST were conducted. The hot ductility testing results provided information as to the role different elements play in the hot cracking mechanism as well as in predicting the relative susceptibility to HAZ hot cracking. Both on-heating and on-cooling tests were conducted on 1/4 inch diameter samples using the temperature cycle illustrated in FIG. 2. On-cooling experiments were conducted from three temperature ranges on J sample, these being 2,125°F, 2,200°F, and 2,375°F, while on-cooling was conducted from slightly below the ZST on the other experimental composition alloys.
In FIGS. 3 through 11, % R.A. refers to % reduction in area of the fracture surface; T.T. -- °F refers to test temperature, degrees Fahrenheit; U.T.S.KSI refers to ultimate tensile strength, thousand pounds per square inch; continuous curves are on-heating curves and the dash line curves are on-cooling curves. Curves with blocks denoting points are for ultimate tensile strength and curves with dots or circles denoting points are for percent reduction in area. The 3 dash-line curves in FIG. 3 are on-cooling curves.
The on-heating ductility for sample J is shown in FIG. 3. There is a large drop in ductility at 2,100°F and an ill-defined ZDT at approximately 2,175°F. The large number of data points around 2,100°F shows the drop in ductility occurs over an extremely small temperature range. Sample B exhibits the same large drop in on-heating ductility at 2,100°F as sample J as shown in FIG. 4. The strength does not exhibit a sharp drop at 2,100°F but decreases continuously from 1,800°F to the ZST of 2,275° to 2,300°F.
The reduction in carbon from 0.063 to 0.031 in sample C has no significant effect on the on-heating behavior, FIG. 5. There is a large drop in ductility at about 2,100°F which may be shifted about 25°F higher in temperature, but there is no significant difference between the two samples. The on-heating strength of sample C, FIG. 5, is also very much the same as that of sample B. Both the ZDT, which is difficult to define because of the tailing out of the ductility above 2,200°F and the ZST about 2,330°F, are very close to the same for both the standard and low carbon samples.
The hot ductility data for samples D through I should be compared to sample C, low carbon, since all the modified samples also have low carbon. The increase in 9 w/o nickel in sample G had no significant effect on either the on-heating strength or ductility, FIG. 6. The increase in titanium by 0.4 w/o in sample I does have a significant effect on the on-heating behavior. Ductility is reduced almost 100°F as shown in FIG. 7. Also, the ZDT is reduced 50 to 75°F and the ZST reduced close to 100°F to about 2,250°F. The reduction in boron in sample D had a large effect on the on-heating behavior. The steep drop in ductility at 2,100°F is no longer present as shown in FIG. 8. The ZST is increased about 50°F to 2,350°F. Also, relatively good strength and ductility are maintained to within 50° to 75°F of the ZST.
The on-heating behavior of heat H, FIG. 9, and sample F, FIG. 10 are very similar to the low boron sample D. Again the drop in ductility around 2,100°F is absent and both samples experience relatively good strength to the ZST and ZDT of about 2,350°F.
An on-heating anomaly is observed in both the strength and ductility of sample E, (FIG. 11). The large drop in ductility again occurs at 2,100°F, the same as in the low carbon sample C, but at 2,150°F the ductility starts to recover from 5 to 30 percent at 2,200°F and then drops to the ZDT of 2,300° to 2,350°F. A corresponding minimum was also observed at 2,150°F in the strength.
The on-cooling ductility from three different peak temperatures of sample J is shown in FIG. 3. When testing on-cooling from 2,125°F, about 25°F above the large on-heating ductility drop, the recovered ductility is very similar to the on-heating ductility except shifted about 50°F lower in temperature. When tested from 2,200°F, the recovery in ductility is again similar to the on-heating except now is shifted to about 100°F below the on-heating data. When testing on-cooling, from 2,375°F, about 25°F above the ZST, little ductility was recovered even as low as 1,875°F.
All the on-cooling testing of the samples was conducted slightly (10°-15°F) below the ZST except for samples B and G for which no on-cooling tests were conducted.
Sample C (low carbon), FIG. 5, exhibits little recovery in strength or ductility as much as 300°F below the ZST. Sample I, (low carbon, high titanium) also does not exhibit any significant recovery and ductility until cooled 300°F below the ZST, FIG. 7. A significant amount of strength is recovered 150°F below the ZST.
The on-cooling behavior of samples D, F and H, FIGS. 8, 10 and 9 are all similar and exhibit much better on-cooling properties than the other experimental heats. All recover significant strength and ductility within 100° to 150°F below the ZST. Of the three samples, sample F recovers both strength and ductility most rapidly when compared, for example, at 2250°F. These results confirm both the welding and spot varestraint results that boron is extremely detrimental to weld hot cracking.
Sample E (low carbon, manganese, silicon) recovered little ductility until about 1,900°F, 450°F below the ZST. This recovery is similar to that of sample C. The recovery of strength in sample E is also extremely poor, but somewhat better than sample C.
Tensile specimens of alloys having the compositions listed in Table III, wherein sample L has the composition of this invention, were placed in a furnace having a hydrogen atmosphere at 3,500 pounds per square inch, heated to 200°C and held at temperature for about 96 hours. This hydrogen "charging" effected a substantial room temperature supersaturation. Immediately after charging, these specimens were tested in ambient or air environment together with specimens of the same alloys in the "uncharged" or unsaturated condition. Results of these tests conducted at ambient temperature are given in Table IV.
TABLE IV ____________________________________________________________ ______________ % Reduction % Loss in Yield Strength in Area of Reduction Sample (at 0.2% Offset) Fracture Surface in Area ____________________________________________________________ ______________ C-Unsaturated 115,000 51 35 Saturated 102,000 33 K-Unsaturated 153,000 23 Saturated 127,000 5 78 L-Unsaturated 127,000 55 Saturated 103,000 49 11 ____________________________________________________________ ______________
Specimens of C and K exhibited a change in fracture mode in the hydrogen charged condition from ductile to partially intergranular brittle behavior. Sample L did not have a change in the fracture mode when tested in the hydrogen saturated condition.
Alloys of this invention having the properties shown in Table IV as well as an ultimate tensile strength of about 150,000 pounds per square inch and a 23 percent elongation in 1 inch may be obtained by heat treating the alloy in a furnace for one hour at about 1,800°F, air cooling and subsequently heating to 1,325°F and holding at temperature for about 16 hours followed by air cooling to ambient temperature.
Welding tests confirm the spot varestraint results and the ductility data obtained by the Gleeble test. Further, these tests verify that the novel alloy composition herein described and claimed having novel concentration ranges of from about 0.01 to about 0.05 w/o carbon, from about 0 to about 0.25 w/o manganese, from about 0 to about 0.25 w/o silicon, from about 0 to about 0.0015 w/o boron, which provide excellent weldability properties together with from about 24.0 to about 34.0 w/o nickel, from about 13.5 to about 16.0 w/o chromium, from about 1.9 to about 2.3 w/o titanium, from about 1.0 to about 1.5 w/o molybdenum, from about 0 to about 0.01 w/o phosphorous, from about 0 to about 0.010 w/o sulfur, from about 0.1 to about 0.35 w/o aluminum, from about 0.10 to about 0.50 w/o vanadium and the balance being essentially iron, provides a high weldability material which is not susceptible to the objectionable hot cracking either in the HAZ or in the fusion zone. Although the cited ranges may be used, susceptibility to cracking may be further minimized by using a maximum of 0.005 phosphorous and 0.005 sulfur together with the other recited ranges herein described.
The present invention relates to a high strength, age hardenable austenitic stainless steel having superior weldability and good retention of properties in a hydrogen environment.
Nickel-chromium-iron alloys are known which may have a yield strength of 150,000 pounds per square inch (psi) at room temperature and have good resistance to corrosion and oxidation even at high temperatures, but are generally known to be highly susceptible to weld hot cracking, e.g., both heat affected zone (HAZ) microfissuring and fusion zone hot cracking. These alloys may also have a reduction of properties upon exposure to a hydrogen environment.
The susceptibility of these stainless steels to weld cracking has not only limited their applications but has also given rise to severe rework problems. Several hot cracking mechanisms have been suggested to explain weld HAZ cracking and fusion zone cracking in nickel-chromium-iron alloys of this type. For example, intergranular cracks associated with a secondary grain boundary phase (identified through x-ray diffraction as the Fe 2 Ti Laves phase) in specimens heated to about 1,340°C have been observed. This phase forms a eutectic with iron at about 1,300°C. It has been proposed that HAZ cracking may be associated with a liquation of the grain boundaries by the Fe--Fe 2 Ti eutectic which allows the boundaries to separate under the welding stresses.
Whatever theory may be proposed, weldability problems are encountered. Thus HAZ microfissuring and fusion zone cracking have been encountered not only in connection with the welding operation itself but also during age hardening treatments of these alloys. This invention determines the source of hot cracking in austenitic age hardenable stainless steels and identifies a eutectic phase and provides a novel alloy composition not susceptible to the objectionable weld hot cracking problem. Although it may be possible to minimize weld cracking on the varying compositions by reducing the welding rate, reducing heat input, welding thin sections, and the like, a large potential for cracking may still exist. This novel alloy provided in this invention enables the use of the high welding rates with minimum susceptibility to hot cracking in the weldment.
Although some nickel-chromium-iron alloys may have "good" weldability as compared to others, these may sustain significant loss of properties in a hydrogen atmosphere or environment. For example, hydrogen embrittlement tests on tensile specimens which have been exposed to a hydrogen environment at about 3,500 psi at a temperature of about 200°C for several days may result in such as a greater than about 34 percent loss, and in some instances, as much as about 50 percent loss, in ductility as well as a different fracture mode when tensile tested at ambient or room temperature (i.e. -- about 25°C).
SUMMARY OF INVENTION
In view of the above problems and limitations, it is an object of this invention to provide an improved nickel-chromium-iron base alloy or stainless steel which has minimal susceptibility to hot cracking during welding processes.
It is a further object of this invention to provide a stainless steel compatible with an environment of hydrogen such that tensile and yield strengths are not severely adversely affected when supersaturated with hydrogen.
It is a further object of this invention to provide a novel stainless steel having ductility properties determined during the cooling process by Gleeble testing similar to ductility properties determined during the heating process.
Various other objects and advantages will appear from the following description of the invention and the most novel features will be pointed out hereinafter in connection with the appended claims. It will be understood that various changes in the details and composition of the alloy components which are herein described in order to explain the nature of the invention may be made by those skilled in the art without departing from the principles and scope of the invention.
The invention comprises a stainless steel alloy which has reduced susceptibility to hot cracking during welding operations, which alloy contains not more than about 0.0015 weight percent (w/o) boron, not more than about 0.25 w/o silicon, from about 0.01 to about 0.05 w/o carbon, from about 0 to about 0.25 w/o manganese and further including from about 24.0 to about 34.0 w/o nickel, from about 13.5 to about 16.0 w/o chromium, from about 1.9 to about 2.3 w/o titanium, from about 1.0 to about 1.5 w/o molybdenum, from about 0 to about 0.01 w/o phosphorous, from about 0 to about 0.010 w/o sulfur, from about 0.1 to about 0.35 w/o aluminum, from about 0.10 to about 0.50 w/o vanadium, and the balance essentially iron.
DESCRIPTION OF DRAWING
FIG. 1 graphically plots spot varestraint test results for various alloy compositions.
FIG. 2 plots the Gleeble heating cycle employed.
FIG. 3 graphically plots percent reduction in area vs temperature for a specific alloy during the on-heating cycle and also during the on-cooling cycle.
FIGS. 4-11 graphically plot ductility curves and ultimate tensile strength curves for composition of Samples B through I.
DETAILED DESCRIPTION
A high strength, age hardenable austenitic stainless steel compatible with hydrogen (e.g., when supersaturated with hydrogen) and having a yield strength (at 0.2% strain or offset) of about 100,000 psi as well as high weldability, including minimal susceptibility to weld cracking, was desired. This invention description determines the cause of hot cracking in prior similar austenitic stainless steels and provides an alloy which has reduced susceptibility to hot cracking both in the HAZ and in the fusion zone as well as not having a degradation of properties upon exposure to hydrogen atmospheres at temperatures of about 200°C and pressures of about 3,500 psi.
Table I shows compositions of various samples of heats. A, B and J samples are of generally known compositions. A number of B samples were heat treated at about 1,350°C for times of from five minutes to about two hours. Electron microprobe analysis showed that, after heating, grain boundary phases of these samples were highly enriched in nickel, titanium, silicon and carbon and depleted in iron and chromium. Microprobe results indicated at least two different titanium phases in the boundaries, one heavily concentrated in titanium and appearing to be associated with carbon which was probably titanium carbide formed on cooling, and another phase enriched in titanium, nickel and silicon. X-ray diffraction on particles dissolved from the matrix showed two phases:
A tiC type (a o = 4.33 A) and an A 2 B type Laves phase (a o = 4.76 A, C o = 7.71 A).
contrary to previous theories, the high temperature phase forming in the boundaries is not the simple Fe--Fe 2 Ti eutectic but rather a complex A 2 B type phase highly enriched in nickel, titanium and silicon. Thus prior theories assumed the nature of the phase could not be affected since a titanium reduction would result in a loss of strength. This invention shows that the complex A 2 B type phase allows selective alteration of ingredients to yield a reduced susceptibility to cracking. This same phase is observed in Gleeble samples heated to a maximum temperature above 1,290°C (as shown in the graphs and as described hereinbelow) in about 6 seconds and rapidly cooled. Thus the end result of the high temperature segregation of alloying elements in materials having compositions similar to A, B and J samples, is a formation of nickel rich phases in the boundaries which play an important role in the hot cracking mechanism.
TABLE I ____________________________________________________________ ______________ HEAT CHEMISTRY, WT.%, BALANCE Fe ____________________________________________________________ ______________ Sample C Mn P S Si Cr Ni Al Mo V Ti B ____________________________________________________________ ______________ A 0.037 1.22 0.015 0.012 0.70 14.29 25.59 0.22 1.37 0.22 2.25 0.0046 B 0.063 1.50 0.004 0.005 0.58 14.45 25.40 0.22 1.27 0.26 2.25 0.0073 C 0.031* 1.51 0.004 0.006 0.63 14.50 25.70 0.24 1.27 0.26 2.22 0.0059 D 0.025 1.47 0.004 0.005 0.61 14.50 25.50 0.25 1.26 0.25 2.20 0.0009 E 0.028 0.13 0.006 0.005 0.25 14.77 26.01 0.23 1.25 0.26 2.18 0.0063 F 0.027 0.11 0.004 0.005 0.16 14.70 24.70 0.10 1.25 0.25 2.18 0.0014 G 0.028 1.54 0.004 0.005 0.63 14.25 34.70 0.25 1.24 0.25 2.15 0.0064 H 0.025 1.51 0.004 0.005 0.60 14.75 25.45 0.66 1.27 0.23 1.60 0.0012 I 0.025 1.39 0.010 0.008 0.58 15.20 23.50 0.24 1.25 0.25 2.63 0.006 J 0.068 1.32 0.018 0.003 0.63 14.90 24.93 0.19 1.25 0.21 2.15 0.004 ____________________________________________________________ ______________ 6 *Underlining denotes major changes from known compositions.
TABLE III ____________________________________________________________ ______________ Sample C Mn P S Si Cr Ni Al Mo V Ti B ____________________________________________________________ ______________ C 0.031 1.51 0.004 0.006 0.63 14.50 25.70 0.24 1.27 0.26 2.22 0.0059 K 0.027 0.09 <0.01* <0.01 0.1 14.34 25.50 0.23 4.98 0.25 2.55 0.0004 L 0.020 0.11 <0.01 <0.01 0.15 14.48 30.46 0.27 1.22 0.25 2.07 0.0010 ____________________________________________________________ ______________ *< = less than
It is shown that the relationship of silicon to the nature of the high temperature (1,290°-1350°C) grain boundary phase is an important factor in the mechanism of hot cracking since a reduction of silicon in the alloy has been found to reduce the hot cracking tendency.
Various alloy compositions are shown as heats or samples C through I in Table I, these being vacuum induction melted or other suitable furnace melted with charge additions of appropriate purity to obtain the alloy chemistry described herein and cast into ingots which were rolled and cut into appropriate desired sizes, such as a plate size of 3 × 3 × 0.25 inches. The purity and chemistry of the alloy may be optimized by remelting using a vacuum consumable arc remelt furnace after vacuum induction melting. These were then solution treated at about 995°C for about 45 minutes in an air atmosphere and water quenched to yield an ASTM grain size of about 6. Electron beam welding parameters which are known to be severe on ordinary compositions were employed for welding tests. Five 90 to 100 percent penetration circular welds with diameters of 1/2 to 21/2 inches were made on each plate. Electron beam welding parameters were 75 inch per minute welding rate, power of 150 kilovolts and 17 milliamps using an electron beam having maximum convergence on the workpiece surface, i.e., a sharp focus, the gun being about eight inches from the workpiece. These tests were designed to easily discern the relative hot cracking susceptibility of each composition. Table II illustrates the percent of weld sections exhibiting gross cracks. Major chemistry modifications from the standard samples A, B and J are underlined in Table I. The sample letter designated is the same for Table I and Table II.
TABLe II ______________________________________ ELECTRON BEAM WELD RESULTS ______________________________________ % Weld Sections Exhibiting Samples Gross Cracks ______________________________________ B 72 C 70 G 45 H 38 E 18 I 10 D 8 F 0 ______________________________________
As noted from Table I and Table II (which lists the samples in order of decreasing number of cracks) samples B and C show that the reduction of carbon from 0.06 w/o to 0.03 w/o has no beneficial effect on fusion zone cracking. Both the increase in titanium from 2.25 w/o to 2.63 w/o in sample I and the increase in nickel from 25 w/o to 35 w/o in sample G appear to have some limited beneficial effects in fusion zone cracking. In addition, samples D and E showed that both boron and silicon are detrimental to fusion zone cracking. Comparing the results from D and H, it was noted that the substitution of about half a percent of titanium with aluminum was not effective in reducing fusion zone cracking in welds. As Table II demonstrates, which confirms the discovery hereinbefore stated, no gross cracking was observed in sample No. F wherein the carbon, boron, manganese and silicon are all low.
From the welding results, it was noted that increases in both nickel and titanium and decreases in both silicon and boron were extremely effective in reducing fusion zone cracking and this indicated a novel alloy composition which has high yield strength, was age hardenable, and did not have a high susceptibility to hot cracking during welding as prior art compositions had. Tables I and II clearly verify that compositions that may be similar to this invention do not have the reduced susceptibility to weld cracking of this invention.
Scanning electron microscopy studies on the fusion zone cracks during the welding procedures verified that the cracks formed during the last stages of solidification. These fusion cracks exhibited segregation of a low melting point phase which, by using the nondispersive x-ray system of the scanning electron microscope, was found to be enriched in titanium-silicon-nickel and depleted in iron and chromium with respect to the matrix. Analysis of the HAZ verified that this phase existed in the HAZ as well as in the fusion zone.
Spot varestraint testing was conducted on compositions B through I to determine the relative HAZ hot cracking susceptibility. These spot varestraint tests show the effect of chemistry on HAZ cracking. Spot varestraint testing involves melting a spot on a specimen with a gas tungsten arc weld torch. After a predetermined time, the arc is extinguished and a ram of known radius applies pressure to the specimen. The augmented strain is the additional strain to the regular weld shrinkage stresses caused by the ram bending the specimen. By using rams with different radii, it is possible to plot curves relating augmented strain to total crack length in the HAZ surrounding the spot weld as shown in FIG. 1.
Spot varestraint testing equipment incorporates a delay timer which may be set such that the ram strikes the sample from 0 to 500 milliseconds after the gas tungsten arc torch is extinguished. In subsequent tests, a delay of 140 microseconds was employed. Welding parameters of 130 amps, 8 volts, using argon shielding gas for a time of 8 seconds on the 0.25 inch thick specimens were used. Surface cracks in the HAZ only were measured under 30 power and 60 power magnification to determine the total crack length for each sample.
FIG. 1 illustrates the results obtained from the spot varestraint testing and compares the relative hot cracking susceptibility of all of the experimental alloy compositions. The effect of chemistry modifications for HAZ microfissuring should be compared to sample C since all samples except sample B have low carbon content. As further shown in FIG. 1, the increase in titanium from 2.22 w/o to 2.63 w/o (sample I) is extremely detrimental to HAZ hot cracking, even though it is beneficial to reducing fusion zone hot cracking. Also, the decrease in carbon from 0.063 to 0.031 w/o has some detrimental effects. Note that compositions F and H which lie on the abscissa of FIG. 1 did not have any HAZ microfissuring using the above test parameters. The decrease in manganese and silicon (sample E), the decrease in boron from 0.0059 to 0.0009 w/o (sample D) and the direct substitution of 0.4 w/o aluminum for titanium all had beneficial results on the reduction of HAZ microfissuring. The increase of 10 w/o nickel (sample G) had very little, if any, significant effect. No cracking was observed in samples F and H even after 8% augmented strain.
Hot ductility (Gleeble) testing was conducted on the various alloy compositions to predict the susceptibility of each material to HAZ hot cracking during welding. Both on-heating to the zero strength temperature (ZST) and on-cooling testing from the ZST were conducted. The hot ductility testing results provided information as to the role different elements play in the hot cracking mechanism as well as in predicting the relative susceptibility to HAZ hot cracking. Both on-heating and on-cooling tests were conducted on 1/4 inch diameter samples using the temperature cycle illustrated in FIG. 2. On-cooling experiments were conducted from three temperature ranges on J sample, these being 2,125°F, 2,200°F, and 2,375°F, while on-cooling was conducted from slightly below the ZST on the other experimental composition alloys.
In FIGS. 3 through 11, % R.A. refers to % reduction in area of the fracture surface; T.T. -- °F refers to test temperature, degrees Fahrenheit; U.T.S.KSI refers to ultimate tensile strength, thousand pounds per square inch; continuous curves are on-heating curves and the dash line curves are on-cooling curves. Curves with blocks denoting points are for ultimate tensile strength and curves with dots or circles denoting points are for percent reduction in area. The 3 dash-line curves in FIG. 3 are on-cooling curves.
The on-heating ductility for sample J is shown in FIG. 3. There is a large drop in ductility at 2,100°F and an ill-defined ZDT at approximately 2,175°F. The large number of data points around 2,100°F shows the drop in ductility occurs over an extremely small temperature range. Sample B exhibits the same large drop in on-heating ductility at 2,100°F as sample J as shown in FIG. 4. The strength does not exhibit a sharp drop at 2,100°F but decreases continuously from 1,800°F to the ZST of 2,275° to 2,300°F.
The reduction in carbon from 0.063 to 0.031 in sample C has no significant effect on the on-heating behavior, FIG. 5. There is a large drop in ductility at about 2,100°F which may be shifted about 25°F higher in temperature, but there is no significant difference between the two samples. The on-heating strength of sample C, FIG. 5, is also very much the same as that of sample B. Both the ZDT, which is difficult to define because of the tailing out of the ductility above 2,200°F and the ZST about 2,330°F, are very close to the same for both the standard and low carbon samples.
The hot ductility data for samples D through I should be compared to sample C, low carbon, since all the modified samples also have low carbon. The increase in 9 w/o nickel in sample G had no significant effect on either the on-heating strength or ductility, FIG. 6. The increase in titanium by 0.4 w/o in sample I does have a significant effect on the on-heating behavior. Ductility is reduced almost 100°F as shown in FIG. 7. Also, the ZDT is reduced 50 to 75°F and the ZST reduced close to 100°F to about 2,250°F. The reduction in boron in sample D had a large effect on the on-heating behavior. The steep drop in ductility at 2,100°F is no longer present as shown in FIG. 8. The ZST is increased about 50°F to 2,350°F. Also, relatively good strength and ductility are maintained to within 50° to 75°F of the ZST.
The on-heating behavior of heat H, FIG. 9, and sample F, FIG. 10 are very similar to the low boron sample D. Again the drop in ductility around 2,100°F is absent and both samples experience relatively good strength to the ZST and ZDT of about 2,350°F.
An on-heating anomaly is observed in both the strength and ductility of sample E, (FIG. 11). The large drop in ductility again occurs at 2,100°F, the same as in the low carbon sample C, but at 2,150°F the ductility starts to recover from 5 to 30 percent at 2,200°F and then drops to the ZDT of 2,300° to 2,350°F. A corresponding minimum was also observed at 2,150°F in the strength.
The on-cooling ductility from three different peak temperatures of sample J is shown in FIG. 3. When testing on-cooling from 2,125°F, about 25°F above the large on-heating ductility drop, the recovered ductility is very similar to the on-heating ductility except shifted about 50°F lower in temperature. When tested from 2,200°F, the recovery in ductility is again similar to the on-heating except now is shifted to about 100°F below the on-heating data. When testing on-cooling, from 2,375°F, about 25°F above the ZST, little ductility was recovered even as low as 1,875°F.
All the on-cooling testing of the samples was conducted slightly (10°-15°F) below the ZST except for samples B and G for which no on-cooling tests were conducted.
Sample C (low carbon), FIG. 5, exhibits little recovery in strength or ductility as much as 300°F below the ZST. Sample I, (low carbon, high titanium) also does not exhibit any significant recovery and ductility until cooled 300°F below the ZST, FIG. 7. A significant amount of strength is recovered 150°F below the ZST.
The on-cooling behavior of samples D, F and H, FIGS. 8, 10 and 9 are all similar and exhibit much better on-cooling properties than the other experimental heats. All recover significant strength and ductility within 100° to 150°F below the ZST. Of the three samples, sample F recovers both strength and ductility most rapidly when compared, for example, at 2250°F. These results confirm both the welding and spot varestraint results that boron is extremely detrimental to weld hot cracking.
Sample E (low carbon, manganese, silicon) recovered little ductility until about 1,900°F, 450°F below the ZST. This recovery is similar to that of sample C. The recovery of strength in sample E is also extremely poor, but somewhat better than sample C.
Tensile specimens of alloys having the compositions listed in Table III, wherein sample L has the composition of this invention, were placed in a furnace having a hydrogen atmosphere at 3,500 pounds per square inch, heated to 200°C and held at temperature for about 96 hours. This hydrogen "charging" effected a substantial room temperature supersaturation. Immediately after charging, these specimens were tested in ambient or air environment together with specimens of the same alloys in the "uncharged" or unsaturated condition. Results of these tests conducted at ambient temperature are given in Table IV.
TABLE IV ____________________________________________________________ ______________ % Reduction % Loss in Yield Strength in Area of Reduction Sample (at 0.2% Offset) Fracture Surface in Area ____________________________________________________________ ______________ C-Unsaturated 115,000 51 35 Saturated 102,000 33 K-Unsaturated 153,000 23 Saturated 127,000 5 78 L-Unsaturated 127,000 55 Saturated 103,000 49 11 ____________________________________________________________ ______________
Specimens of C and K exhibited a change in fracture mode in the hydrogen charged condition from ductile to partially intergranular brittle behavior. Sample L did not have a change in the fracture mode when tested in the hydrogen saturated condition.
Alloys of this invention having the properties shown in Table IV as well as an ultimate tensile strength of about 150,000 pounds per square inch and a 23 percent elongation in 1 inch may be obtained by heat treating the alloy in a furnace for one hour at about 1,800°F, air cooling and subsequently heating to 1,325°F and holding at temperature for about 16 hours followed by air cooling to ambient temperature.
Welding tests confirm the spot varestraint results and the ductility data obtained by the Gleeble test. Further, these tests verify that the novel alloy composition herein described and claimed having novel concentration ranges of from about 0.01 to about 0.05 w/o carbon, from about 0 to about 0.25 w/o manganese, from about 0 to about 0.25 w/o silicon, from about 0 to about 0.0015 w/o boron, which provide excellent weldability properties together with from about 24.0 to about 34.0 w/o nickel, from about 13.5 to about 16.0 w/o chromium, from about 1.9 to about 2.3 w/o titanium, from about 1.0 to about 1.5 w/o molybdenum, from about 0 to about 0.01 w/o phosphorous, from about 0 to about 0.010 w/o sulfur, from about 0.1 to about 0.35 w/o aluminum, from about 0.10 to about 0.50 w/o vanadium and the balance being essentially iron, provides a high weldability material which is not susceptible to the objectionable hot cracking either in the HAZ or in the fusion zone. Although the cited ranges may be used, susceptibility to cracking may be further minimized by using a maximum of 0.005 phosphorous and 0.005 sulfur together with the other recited ranges herein described.