ENGINEERING RESEARCH INSTITUTE THE UNIVERSITY OF MICHIGAN ANN ARBOR Final Report THE EFFECT OF PROCESSING VARIABLES AND THE INFLUENCE OF VACUUM MELTING UPON THE STRESS-RUPTURE PROPERTIES OF CAST NICKEL-BASE ALLOYS C. M. Hammond 5R. A. Fliin Project 2462 DEPARTMENT OF THE NAVY, BUREAU OF AERONAUTICS AIRBORNE EQUIPMENT DIVISION, MATERIALS BRANCH (AE-411) CONTRACT NO. NOas 56-359-d WASHINGTON, D.C. September 1956

The University of Michigan * Engineering Research Institute TABLE OF CONTENTS Page LIST OF TABLES iii LIST OF FIGURES iv SUMMARY v INTRODUCTION 1 OBJECTIVE 1 PROCEDURE 5 A, DETAILED PROCEDURE —PROCESSING VARIABLES 5 B. DETAILED PROCEDURE- VACUUM-MELTING EFFECTS 4 lo Nitrogen Additions 4 2. Vapor Collection and Analysis 6 35 Structural Studies 6 DATA AND DISCUSSION OF RESULTS 7 A. THE EFFECT OF PROCESSING VARIABLES UPON STRESS-RUPTURE PROPERTIES 7 1o 95-Percent Confidence Limits for Stress-Rupture Curves 7 2, 95-Percent Confidence Limits for Individual Rupture Tests 8 B. THE MECHANISM OF THE INFLUENCE OF VACUUM MELTING UPON STRESSRUPTURE PROPERTIES 20 1o Nitrogen Additions to Vacuum + Argon Melts 20 2. Vapor Collection and Analysis 20 3. Structural Studies 24 CONCLUSIONS 29 A, EFFECTS OF ADDITIONAL PROCESSING VARIABLES 29 B. VACUUM MELTING 29 SUGGESTIONS FOR FUTURE WORK 30 Ao EXAMINATION OF VACUUM-MELTED MICROSTRUCTURES 30 B. PHYSICAL CHEMISTRY OF VACUUM MELTING 31 C, INVESTIGATION OF THE EFFECT OF ORDERING UPON STRESS-RUPTURE PROPERTIES 31 BIBLIOGRAPHY 32 APPENDICES 33-44 ii

The University of Michigan * Engineering Research Institute LIST OF TABLES Table Page Io Effect of Melting and Casting AtmospheresUpon Stress-Rupture Properties at 1500~F 2 IIo Processing Variables Investigated 5 IIIo Nitrogen Additions to Vacuum-Melted Heats of the Guy-Type Alloy 6 IV. Effect of Charge Preheat Time Upon 100-Hour Rupture Strength of the Guy-Type Alloy 8 V, The Effect of Nitrogen Content of the Guy Alloy Upon 100-Hour Elongation 20 VIo Analyses of Vapor Collected from Vacuum-Melted Heats of the Guy-Type Alloy 22 VIIo Effect of Refining Time Upon 100-Hour Elongation of the GuyType Alloy 24 VIII. Inclusion Content of the Guy-Type Alloy as a Function of Melting Atmosphere 24 IXo Relative Concentrations of Iron and Boron in Minor-Phase Extracts of the Guy-Type Alloy 27 iii

The University of Michigan * Engineering Research Institute LIST OF FIGURES Figures Page 1o Effect of charge preheat time upon the stress-rupture properties of air= and argon-protected Guy-type alloys 9 2o Effect of pouring temperature upon the stress-rupture properties of argon-protected Guy-type alloys 11 3o Effect of pouring temperature upon the stress-rupture properties of vacuum + argon Guy-type alloys 12 4. Effect of pouring temperature upon the stress-rupture properties of argon-protected GMR-255-type alloys 13 5. Effect of pouring pressure upon the stress-rupture properties of argon-protected Guy-type alloys 14 6. Effect of pouring pressure upon the stress-rupture properties of argon-protected GMR-2355type alloys 15 7. Effect of holding time in the liquid state upon the stress-rupture properties of argon-protected Guy-type alloys 17 8o Effect of holding time in the liquid state upon the stress-rupture properties of argon-protected GMR-235-type alloys 18 9. Effect of mold preheat temperature upon the stress-rupture properties of air- and argon-protected Guy-type alloys 19 10o Effect of nitrogen content upon the stress-rupture properties of vacuum-melted Guy-type alloys 21 11o Effect of refining time upon the stress-rupture properties of vacuum-melted Guy-type alloys 23 12. Inclusion content typical of vacuum + argon heats of the Guytype alloy 25 13. Inclusion content typical of the Guy=type alloy melted under vacuum + argon (with nitrogen additions) and air atmospheres 25 14. Electron micrograph of rod-like precipitate found in air-melted specimens after testing (Guy-type alloy) 28 15. Electron micrograph of the general structure of the matrix and y? precipitate (Guy-type alloy) 28 ~~____________..... iv ---------------------------

The University of Michigan * Engineering Research Institute SUMMARY In the work of the previous year (September 1954 —September 1955)1 it was found that melting in vacuum improved the elevated-temperature properties of three prominent nickel-base alloys (Final Report, Contract No. NOas 55-11Oc). The objectives of this year's research were (L) to investigate additional processing variables such as melting time and pouring temperature and (2) to explore further the changes in properties associated with vacuum melting. A. EFFECTS OF ADDITTIONAL PROCESSTNG VARIABLESo The following variables were investigated: lo Effect of charge preheat before melting, 2o Pouring temperatureo 5. Pouring pressure, 4o Increased time of melt in liqaid state (nucleation effect)" 5. Mold preheat temperatureo The effect of charge preheat was pronounced, resulting in an increase of 100-hour rupture strength (1500F) from 40,000 to 50,000 psi for the Guy-type alloy. Variables 2-5 inclusive were without effect (95-percent confidence limits), Bo VACUUM-MELTING EFFECTS One of the principal differences in vacuum-melted material is low nitrogeno As nitrogen was increased in vacuum melts by late additions, the elongation of the Guy-type alloy was reduced from 7-10 percent to 2 percent. Strength was unaffectedAnalysis of metal vapors during vacuum melting exhibited surprisingly large quantities of undesirable elements such as lead and calcium, Refining time during vacuum melting reduced the elongation of the Guy-type alloy from 7-10 percent to 5-5 percento Electron microscopy, electron diffraction, x-ray diffraction, and spectroscopic investigations of vacuum- vs air-melted material were conductedo The phases Cr7C3, Fe2E, and Cb(CN) were identified along with the matrix lines in both meltso Electron microscopy disclosed the presence of a rod-like precipitate present only in air-melted samples, after testingo Also, a grainboundary precipitate has tentatively been identified only in air-melted microstructureso I- ----------------------— v ---------------

The University of Michigan * Engineering Research Institute INTRODUCTION In the past few years, many new heat-resistant alloys have been developed for high-temperature applications, such as turbine blades in jet engines. Nickel-base alloys containing major percentages of chromium, molybdenum and aluminum and varying percentages of boron have been prominent in these development programs. Prior to the initiation of this research, the majority of these alloy programs dealt only with variations of chemical composition which were correlated with stress-rupture properties at 15000 F. The effect of processing variables upon elevated-temperature properties of existing nickel-base alloys in order to demonstrate their ultimate potential had received relatively little attention. It appeared necessary, therefore, to parallel the investigations of compositional effects with research on the effects of processing variables, such as melting and casting atmospheres, pouring temperatures, etc., upon stress-rupture properties of cast nickel-base alloys. Previous research on this contract1 determined the effect of melting and casting atmospheres upon the stress-rupture properties of three commercial nickel-base alloys. Table I and the following discussion briefly review that research. As shown in Table I, vacuum + argon melting increases the elongation of all three alloys and also significantly raises the 100-hour rupture strength of the Guy-type alloy. Metallographic examination with the light microscope revealed no reasons for the above effects. The only positive difference found between vacuum+ argon and air or argon-protected heats was nitrogen content. OBJECTIVE Based upon the foregoing data, this year's research had two objectives. First, the remainder of the processing variables, such as melting practice, pouring temperature, pouring pressure, etc., were to be evaluated to determine their individual effects upon stress-rupture properties. Second, because of the pronounced influence of vacuum melting upon stress-rupture properties, further investigation to explain the mechanism for these effects was needed. 1 ___________

The University of Michigan * Engineering Research Institute TABLE I EFFECT OF MELTING AND CASTING ATMOSPHERES UPON STRESS-RUPTURE PROPERTIES AT 15000 F. Research at The | University of Michigan Alloy Type Published __ Atmosphere Properties Air Argon- Vacuum ___. Protected + Argon 100-hour rupture strength (psi 49,000 42,000 41, 000 56,oo Guy Percent elongation 2-5 3 1-5 7-10 100-hour rupture strength (psi) 37-40,000 41,000 41,000 42,000 GMR 35 Percent elongation 6-10 6-9 3-7 14-19 100-hour rupture strength (psi) 43,000 37,00 4000 4,0 42,000 nco 00 Percent elongation 10 2-4 3-4 21 2

The University of Michigan * Engineering Research Institute PROCEDURE In order to accomplish the objectives of this research, the following experimental program was planned, involving A, processing variables in general, and B, the effects of vacuum melting: A. Processing Variables 1. Charge preheat time before melting in air and argon-protected atmospheres. 2. Pouring temperature. 3. Pouring pressure. 4. Holding time of the melt in the liquid state after all additions were made. 5. Mold preheat temperature. The above variables are considered to be the more important processing variables affecting elevated-temperature properties. B. Vacuum-Melting Effects 1. Nitrogen additions to vacuum + argon melts. 2. Collection and analysis of metal vapors evolved during vacuum + argon melting. 35 Structural studies to determine the effects of vacuum + argon melting upon metal structure. The nitrogen additions were made to increase the nitrogen of the vacuum + argon heats to the level of air and argon-protected heats. If nitrogen is an important factor in controlling stress-rupture properties, these additions should lower the strength and/or ductility of vacuum + argon melts to that of air and argon-protected heats. The vapor analyses were collected to determine whether any trace elements, which might be harmful to elevated-temperature properties, were evolved during vacuum melting. The objective of the structural studies was to compare the microconstituents (size, shape, number, etc.) of air and vacuum + argon heats. The detailed procedures for carrying out this program are presented below: A. DETAILED PROCEDURE —PROCESSING VARIABLES Two particular alloys of comparable chemical analysis, except for boron content, were selected for studying the effects of the processing variables. 5

The University of Michigan * Engineering Research Institute The alloys selected were the Guy type (.50 percent B) and the GMR 235 type (.06 percent B). The processing variables, such as pouring temperature, pouring pressure, etc., were varied individually, maintaining the remainder of the variables consistent with those used in determining the effects of melting and casting atmosphere upon stress-rupture properties.1 With this experimental design a statistical treatment of the data was possible in order to evaluate the individual effects of each processing variable upon stressrupture properties at 1500~ F. The processing variables studied and the conditions under which they were investigated are listed in Table II. The melting and casting techniques for the argon-protected GMR-235type alloy and the vacuum + argon melted Guy-type alloy were the same as those previously described,l while the melting of the Guy-type alloy, under air and argon, was slightly different. For the air and argon-protected Guy-type alloy, the charge was melted slowly at the beginning of the heat to avoid bridging and, consequently, severe overheating of the molten metal. The time required to melt down the heat increased from five minutes for the conventional practice to twenty minutes using this preheat practice. Since these alloys contained fairly high percentages of easily oxidized elements, such as Al, Ti, etc., a melting practice which occupied the least time from start-up to pour originally seemed to be the best melting practice. However, one heat of the Guy-type alloy with an air melting atmosphere, which was designed to evaluate the effect of the time of vacuum + argcn melting practice in air, displayed a 100-hour rupture strength comparable to vacuum + argon properties. This melting procedure was further investigated and it seemed advisable to use this practice for the remainder of the melts involving the effects of processing variables upon the rupture properties of the Guy-type alloy. The effect of the variables outlined above upon the properties of the GMR-235-type alloy had already been collected and up to the pres ent time these experiments have not been conducted using the slower melting practice. B. DETAILED PROCEDURE —VACUUM-MELTING EFFECTS Since vacuum + argon melting affected both the high-temperature strength and ductility of the Guy-type alloy, this alloy was selected in order to study the mechanisms for the vacuum effect. 1. Nitrogen Additions. —Nitrogen was added to vacuum + argon heats in the form of high-nitrogen (1.75 percent N) ferrochrome. This was added just prior to pouring, under an argon pressure of approximately 400 mm of mercury. The application of argon pressure insured reasonable nitrogen recovery in the melt. _______________________ 4 —---------------

The University of Michigan * Engineering Research Institute TABLE II PROCESSING VARIABLES INVESTIGATED Alloy Type Melting Heat Range of Variable Variable Used Atmosphere No.(l) Employed 1. Charge preheat time Guy air and argon R-285 before melting |R-218 Slow melting at beginning )R-300 of heat (20 minutes to melt R-213 charge vs 5 minutes of previous data) R-284 tR-301 2. Pouring temperature R-212 Guy argon R-213 2660 - 2950~ F. )R-214 1R-302 (R-223 Guy vacuum IR-224 2660 - 3000~ F. R-225 (R-150 GMR 235 argon |R-151 2660 - 2950~ F. (R-152 3. Pouring pressure (R-215 Guy argon tR-216 0 - 12 psig. (R-217 (~-153 GMR 235 argon R-154 0 - 15 psig. R-155 4. Holding time (R-219 Guy argon R-220 1/2 - 2 hours R-221 GMR 235 argon {R-222 1/2 - 2 hours 5. Mold preheat temperature (-278 Guy air and argon R-281 1800~ F. )R-279 R-280 (1) Full details given in Appendices I-V. I —-------------— 5 ----------------

The University of Michigan * Engineering Research Institute In order to evaluate the effect of a late chromium addition and argon pressure upon the vacuum + argon Guy-type alloy, a control heat was melted in which standard ferrochrome (essentially 0 percent nitrogen) was added prior to pouring and argon pressure of 400 mm applied to the heat. The nitrogen additions and appropriate heat numbers are listed in Table III. TABLE III NITROGEN ADDITIONS TO VACUUM-MELTED HEATS OF THE GUY-TYPE ALLOY Heat No. Percent N2 Added R-209 0 (control heat) R-211.01 R-210.05 R-287.10 2. Vapor Collection and Analysis. —Metal vapors evolved during vacuum melting were collected on a thin sheet of 99.90 percent nickel. The sheet was cleaned in acid and washed thoroughly in water before mounting it above the crucible. The vapors were collected throughout the entire heat and were easily flaked off after completion of the melt. The analyses of these vapors were conducted spectrographically by the J. H. Herron Company in Cleveland, Ohio. 5. Structural Studies. —The structure of the Guy-type alloy was examined using several different metallurgical techniques: a. Optical microscopy. b. X-ray diffraction. c. Spectroscopy. d. Electron diffraction. e. Electron microscopy. The metallographic specimens were again studied by optical microscopy. The inclusion content of alloys melted in vacuum + argon was compared to air and argon-protected heats. The ratings of the inclusion content of the vacuum + argon heats to which nitrogen was added were also included in this survey. X-ray diffraction patterns were obtained from solid and powder samples of air and vacuum + argon heats, before and after testing. Minor phases, extracted by a bromine-alcohol mixture2 from the metal samples of air and vacuum + argon heats, before and after testing, were analyzed by x-ray diffraction. I —---------------------— 6 —-------------

The University of Michigan * Engineering Research Institute The same extracts mentioned above were analyzed spectrographically to determine whether any compositional variations were present in the minor phases. Electron diffraction was used to attempt to identify the Chinesescript phase present in all Guy-type alloys. By directing a low-angle beam of electrons upon the Chinese-script phase, which remains in relief after polishing and etching, positive identification is possible from the diffraction pattern. Examination of the precipitates and other phases present in the structure and unresolved by optical microscopy was initiated with electron microscopy. The metal surface was prepared in the conventional manner and standard practices, developed by Bigelow and associates,2 for obtaining shadowed plastic replicas were used. The samples examined consisted of vacuum + argon and air-melted heats. DATA AND DISCUSSION OF RESULTS The presentation of the results of this research is most conveniently discussed under two major headings: A. The effect of processing variables upon stress-rupture properties. B. The mechanism of the influence of vacuum melting upon stressrupture properties. A. THE EFFECT OF PROCESSING VARIABLES UPON STRESS-RUPTURE PROPERTIES To evaluate the significance of the effect of the processing variables investigated upon stress-rupture properties, two methods of statistical analyses can be used: 1. 95-percent confidence limits for stress-rupture curves. 2. 95-percent confidence limits for individual rupture tests. 1. 95-Percent Confidence Limits for Stress-Rupture Curves. —If a sufficient number of rupture tests are conducted so that a representative stress-rupture curve can be drawn through the data, the first method of statistical analysis can be employed. Essentially, this analysis consists of determining whether the stress-rupture curve for the variable investigated belongs to the same universe as the data to which it is compared. If the rupture curve for the variable studied falls outside the 95-percent confidence 7

The University of Michigan * Engineering Research Institute limits of previously established data, the variable has a significant effect upon-stress-rupture properties. 2. 95-Percent Confidence Limits for Individual Rupture Tests. —If too few rupture tests are conducted to calculate a representative stress-rupture curve for the particular variable, the second method of statistical analysis must be used. In this case, an individual rupture test is compared to the 95-percent confidence limits for individuals of previously established data. If an individual rupture test for a variable falls outside the confiderce limits of the previously established data, the variable has a significant effect upon stress-rupture properties. Otherwise, the effect is not significant. In this section the results of each processing variable will be discussed separately. a. Charge Preheat Timeo The effect of charge preheat time upon stress-rupture properties at 15000 F. of the Guy-type alloy, melted under air and argon, is shown in Fig. 1I In order to determine the significance of charge preheat time with regard to stress-rupture properties, both types of statistical analyses were used. Since the rupture curve for the 20-minute charge preheat time falls outside the 95-percent confidence limits for a preheat time of five minutes, charge preheat time is a significant processing variable. Also, 50 percent of the individual tests fall outside the 95-percent confidence limits for individual rupture tests when only one would show a significant effect. Therefore, it is evident that preheat time is a significant processing variable and considerably affects 100-hour rupture strength (Table IV). TABLE IV EFFECT OF CHARGE PREHEAT TIME UPON 100-HOUR RUPTURE STRENGTH OF THE GUY-TYPE ALLOY 100-Hour Rupture Percent Time Atm Strength (psi) Elongation 5 air, argon 40,000 1 - 3 20 air, argon 50,000 2 - 4 20 vacuum 56,000 7 - 10 _____l_____________________ 8

C Cr Mo Al Fe Cb B Compositions of R218.11 11.78 5.80 6.22 4.65 1.78.43 (chrg. for all other heats the same) Analysis range for previous.08-.26 12.83-15.35 4.42-5.87 4.33-7.31 4.36-6.94 1.52-2.30.28-.48 Gu-alloy data Heat No. Atm Symbol Heat No. Atm Symbol Pouring temp. for all heats - 2800~F Pouring press. for all heats - 5 psig. R285 air Q R213 argon 4 R218 6 R284 " Elongation given near point -~ R300 " 0- R301 " representing rupture test. R ~|up~\~~ Il |Equation for improved melting practice lines: ptuure c, log Xr = 4.921 -.115 log X2 630000 Prev Confide ls l f 55,~000 a rOidividwl rupture urvtests fro universl line from previous airprevious air and argon data air and argon data 5_00 (5 Min. preheot).. - 4.55 000 PO 40,000. tC,) 357000 1.5 5 I0 20 50 I00 200 500 I000 TIME, HRS Fig. 1. Effect of charge preheat time upon the stress-rupture properties of air- and argon-protected Guy-type alloys.

The University of Michigan * Engineering Research Institute It is also of interest to note that the 100-hour rupture strength is nearly comparable to the vacuum + argon melted Guy alloy. The charge preheat time, however, has no significant effect upon elongation. The influence of this variable upon rupture strength is undoubtedly due to the elimination of severe overheating of the heat during melting. b. Pouring Temperature. The Guy-type heats, melted under argon and vacuum + argon atmospheres were poured at temperatures of 2660~, 27700, and 2950~ F. to evaluate the effect of pouring temperature upon stress-rupture properties. Similarly, the GMR-235-type heats, melted under argon, were poured at 2660~, 2760~, and 2950~ F. The stress-rupture results from these experiments are shown in Figs. 2, 3, and 4. To evaluate the influence of pouring temperature upon rupture properties of the two alloys, the second method of statistical analysis was used. In Figs. 2, 3, and 4 all the rupture tests fall within the appropriate confidence limits for individual tests. Therefore, it can be concluded that pouring temperatures, over the range investigated, do not significantly affect the 100-hour rupture strength of argon-protected and vacuum + argon melted Guy-type alloys or the GMR-235-type alloy melted under argon. Also, pouring temperature gave no significant change in 100-hour elongation for the Guy alloy. The 100-hour elongation of the GMR 235 alloy increased, compared with the values previously reportedl (6-12 percent vs 3-7 percent). The effect seemed to be general for all GMR 235 heats regardless of the processing variable investigated. This increase in elongation, however, is possibly due to the lower molybdenum content of the present heats (3.0 percent) compared to higher contents (approx. 5.0 percent) published previously.l c. Pouring Pressure. The Guy-type and GMR-235-type alloys melted under argon were poured at pressures of 0, 8, and 12 psig and 1.25, 10, and 15 psig, respectively. The stress-rupture results are shown in Figs. 5 and 6. The second statistical method was used to evaluate the effect of pouring pressure. Since the individual rupture tests for all pressures investigated were within the appropriate 95-percent confidence limits for individual tests, pouring pressures, over the range investigated, did not significantly affect 100-hour rupture strength of the Guy- or GMR-235-type alloys melted under argon. The 100-hour elongation of either alloy was not affected by pouring pressure...........l_________________ 10.....

C Cr Mo Al Fe Cb B Composition of heat R212.10 12.86 5.74 6.24 4.94 1.54.58 - (chrg. of heat R213, R214, = and R302 same as R212) Typical composition of data.11 11.78 5.80 6.22 4.65 1.78.43 used to calc. 95% confidence' limits w Rupture curve for pouring temperature data as a whole: OQ 95% Confidence limits W 5,___00____ /1'individual tests from 20 minute 65,000 - - - / charge preheat data _ 60,000 U) - \ 55,000 -- 5_, / _, ) o50o000 --- w ~ ~ ~_ 3.0 Cl 45,000 40,000 - - Equation of line: log X1r = 4.9529 -.1222 log X2 35,000.5 5 10 20 50 100 200 500 1000 TIME, HRS'A Heat No. Atm Symbol Pouring Temp.(~F) Superheat (~F) Pouring pressure - 5 psig for all heats R212 argon * 2950 440 20-minute charge preheat practice R213 " 2770 250 used in these heats. R214 "- 2660 150 R302 " 2660 150 Fig. 2. Effect of pouring temperature upon the stress-rupture properties of argon-protected Guy-type alloys.

-I C Heat No. Symbol Pouring Temp. (0F) Superheat (CF) R225 5000 480o R225 2800 280 R224 2660 140 0 3\ O Rupture curve for pouring S temperature data as a whole go* 65,000 C') 60,000 cn 55,000 (n Ir 50,000 95% Confidonce limits - --- ^~~~~n~~~~ ~individual tests from previous 45,000 vacuum and argon data 40,0000 Equation of line: log X1, = 4.9876 -.1205 log X2 Elongation given near point ( representing rupture test. 35,000C i.5 5 10 20 50 100 200 500 1000 TIME -HRS 0 Fig. 5. Effect of pouring temperature upon the stress-rupture properties of vacuum + argon Guy-type alloys. r^ r^ r+

C Cr Mo Al Fe Ti B Composition of heat R150.14 14.24 2.85 3.03 9.13 1.56.06 - (chrg. of R151 and R152 same as R150) Analysis range for previous.15-.27 15.03-15.99 4.31-5.39 2.37-4.66 7.65-11.10 1.90-2.30.06-.12 GMR-235-Type heats.' 0 Heat No. Atm Symbol Pouring Temp (~F) Superheat (~F) Pouring Press. - R150 argon 2950 440 5 psig R151 " 2760 250 " R152 " 2660 150 Equation of line: log Xlr = 4.8070 -.1021 log X2 Elongation given near point representing rupture test. 65~00 -650,000 /- Rupture curve for pouring 60,000 temperature as a whole 55,000^~ /_^ A^-~95 % Confidence limits for individual m 55,000 C~1 -. - | ——. /\ tests from previous argon data Q 50,000- O O m 45,000 TIME, HRS. 3 LU " 35,000 30,001.5 5 I0 20 50 100 200 500 1000 TIME, HRS.: Fig. 4. Effect of pouring temperature upon the stress-rupture properties of argon-protected GMR-2355-type alloys.

C Cr Mo Al Fe Cb B Composition of Heat R216.10 12.24 6.0 6.19 4.75 1.76.41 (chrg. for R215 and R217 - same as R216) Typical composition of data used.11 11.78 5.80 6.22 4.65 1.78.43 to calculate 95% confidence limits Heat No. Atm Symbol Pouring Press.(psig) Pouring Temp.(~F) O R215 argon * 8 2800 R216 if 0 t R217 " 12 ". m Equation of line: log Xl = R /-Rupture curve for pouring Elongation given near point 4.963 -.148 log X2 temperature data as a whole representing rupture test. 65,6nnnn______________________0^^^^ l_______000~l | ~ - 95% Confidance limits 56,000 " — /~ individual tests from 20 minute _ _ _ _ _ _ ^^. / \ charge preheat data m o 5500 —cn' 45,0 0 0-, —-- |: 4 20-minute charge preheat C 40,000 time used in melting -- these heats., | 35,000' 1.55 5 10 20 50 100 200 500 1000 TIME, HRS Fig. 5. Effect of pouring pressure upon the stress-rupture properties of argon-protectedC Guy-type alloys.

-4 Composition: Charge for heats R155, R154, Elongation given near point and R155 same as R150 shown in Fig. 4. representing rupture test. Equation of line: log Xir = 4.8053 -.1004 log X2 Heat No. Atm Symbol Pouring Press, (psig) Pouring Temp. (~F) R155 argon * 10 2850 R154 15 2850 R155 0 2850 65,000 _ /-Rupture curve for pouring 60-000 pressure data as a whole 6c____0C Rupture curve for pouring _______-95% Confidence limitsfor individual_______ F-55,0002^-5 tests from previous argon protected 50,000 d t m 45,000 10 type 9a a- 40,000 I w cc~~~~~~~~~~~~~~~~~~~~~~~~~~ t~35,000 1I5 5 1 0 20 50 100 200 500 000 TIME) HRS Fig. 6. Effect of pouring pressures upon the stress-rupture properties of argon-protected Gv1-255 type alloys.

The University of Michigan * Engineering Research Institute d. Holding Time. The Guy-type and GMR-255-type alloys, melted under argon, were maintained in the liquid state for one-half and two hours after all additions had been made. The stress-rupture data are shown in Figs. 7 and 8. The second statistical method was used to evaluate the data. All the individual rupture tests, for the holding times investigated, were within the appropriate 95-percent confidence limits for individual tests. Therefore, holding time does not significantly affect 100-hour rupture strength. Also, no effect upon 100-hour elongation was observed. e. Mold Preheat Temperature The Guy-type alloy, melted under air and argon, was poured into molds preheated at 18000 F. instead of the usual 1600~ F. preheat temperature. The stress-rupture data are included in Fig. 9. The second method of statistical analysis was employed to evaluate the data. All the individual rupture tests fell within the appropriate 95percent confidence limits for individual tests. Therefore, mold preheat temperature, varied from 1600~ to 18000 F., does not significantly affect 100hour rupture strength of the Guy-type alloy. Also, no effect upon 100-hour elongation is evident from the data. To summarize the effects of processing variables upon stress-rupture properties of the two alloys investigated, the following statements can be made. 1. The charge preheat time significantly affected the 100-hour rupture strength of the Guy-type alloy. 2. The remainder of the variables did not significantly affect the rupture strength of either alloy. 3. None of the processing variables affected the 100-hour elongation of the Guy- or GMR-235-type alloys. 4. In general, the 100-hour elongation of the GMR-235-type alloy increased, compared with the values previously reported (6-12 percent vs 3-7 percent). The experimental data for all the variables investigated are included in Appendices I-V. _______________________ 16

.I -4 Heat No. Atm Symbol Holding time in liquid state (hrs) R219 argon * 2 R220 " 1/2 o Elongation given near point 1 20-minute charge preheat practice representing rupture test. used for these heats. -Q 65,0000 60,000 4-... —....- - -- ------... 55,000 ___ —-- - ^ ~50,000a> 45,000 —u) 40,000. Cn V 95 % Confidence limits __ r 38,000 individual tests from 20 minute 35,000- t I Icharge preheat data, 35,000' 1.5 5 10 20 50 100 200 500 1000 TIME, HRS "I 0 Fig. 7. Effect of holding time in the liquid state upon the stress-rupture properties of argon-protected Guy-type alloys. l l~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~f

C Cr MO Al Fe Ti B Composition of Heat R221.16 14.86 5.67 5.50 10.38 1.86.06 (Same chrg. as R222) Analysis Range for previous.15-.27 15.03-15.99 4.31-5.39 2.57-4.66 7.65-11.10 1.90-2.30.06-.12 GMR-235 data - Pouring temp. 2850~F-Pouring press. 5 psig 0 Heat No. Atm Symbol Holding time in. liquid state (hrs) R221 argon 0 2 R222 4 1/2 5I Equation of line: log Xjr = Elongation given near point 4.8562 -.1218 log X2 representing rupture test. 0o 65,000 ar ~r(, ________________ ____________________ ____________ ____ ____ ______ m^ 600C 00 Rupture curve for holding time 55,000 data as a whole ---------— _ ^ 5 0,0 0 0C CL —n ------ ^ 45,000ur 40,000,~-~ _____ ~~~95% Confidence limits for C ^'* ** >_____^ " ^ Lf ___ 35,000 individual tests from previous 10~~~~~~ argon-protected data ^" ^^.0 30,000 -----------------------' 1.5 5 10 20 50 100 200 500 1000 TIME, HRS Fig. 8. Effect of holding time in the liquid state upon the stress-rupture properties of argonprotected GMR-255-type alloys.

Heat No. Atm Symbol Mold Temp.'I R278 air O 1800 R281 " 6 r R279 argon 0 " R280 4 o ^*^^ |^~~ ~Elongation given near point representing rupture test. 65,000 60,000 rH (j 55,000 Equation of line: 4 3 m ^ 50,000 log X = 4.7054 -.012 log X2 1 / _ 45,000 ---- 0 40,000 -3.0 o 4.0 -95 ~/0 Confidonce limits. individual tests from 20 minute 3 Pouring temperature - 2000~F charge preheat data oo 35,000 Pouring pressure - 5 psig 20-minute charge preheat time employed for both air- and argon-protected heats 30,000 \ 1 1.5 5 10 20 50 100 200 500 1000' TIME, HRS 0 Figo 90 Effect of mold preheat temperature upon the stress-rupture properties of air- and' argon-protected Guy-type alloyso l l~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~f

The University of Michigan * Engineering Research Institute B, THE MECHANISM OF THE INFLUENCE OF VACUUM MELTING UPON STRESS-RUPTURE PROPERTIES The results of this segment of the research are conveniently discussed under the following headings: 1. Nitrogen additions to vacuum + argon melts. 2. Vapor collection and analysis. 53 Structural studies. The Guy-type alloy was selected for the above analysis because of the effect of vacuum + argon melting upon 100-hour strength and ductility. The stress-rupture properties were obtained at 15000 F. 1. Nitrogen Additions to Vacuum + Argon Melts. —Nitrogen reduces ductility and does not significantly affect the high-temperature strength of the Guy-type alloy (Table V and Fig. 10). TABLE V THE EFFECT OF NITROGEN CONTENT OF THE GUY ALLOY UPON 100-HOUR -ELONGATION Percent Percent Percent Nitrogen Added Nitrogen Analyzed Elongation (100 Hour) 0 6Ol -7 lO 2Oa.01.02 5 - 6.03.03 4 - 4o5.10 --- 2 The second method of statistical analysis was used in determining the insignificant effect of nitrogen content upon stress-rupture properties of the Guy-type alloy. All individual rupture tests are within the appropriate 95-percent confidence limits for individual tests. 2. Vapor Collection and Analysis., —The analyses of the metal vapors collected during vacuum + argon melting are shown in Table VI. Two different refining times were used to determine the importance of this procedure. More metal vapor was collected during the 20'minute refining period. It is evident from the data that considerable refining of trace elements occurs during vacuum melting. ------------------------— 20 —-------------

Analysis Heat No. C Cr Mo Al Fe Cb B N2 S R209.16 12.43 6.00 5.40 5.36 2.11.43.01 R211.15 12.43 5.96 5.45 5.36 1.92.43.02 R210.14 12.27 6.09 5.93 5.06 2.0.43.05 1 Elongation given near point - Equation of line: log Xlr = 4.941-.1082 log X2 representing rupture test. ~ j^ |^" -^K-95% Confidence limits for:_ _._"'~ _ /~\~ individual tests from previous vac. W 65p,000 ----— _ —____-_-_- d I t+argon daoa 3 55,000 * —-. Rupture curve for nitrogen - 7 _ r 4 I 50000 data as a whole ____|____ _____ CL Pouring temperature for all heats - 2900~F.5 6 Q 45,000 Nitrogen added as high-nitrogen ferrochrome. j A n Nitrogen addition was made under 400 mm of argon pressure. | | cr 0 r In control heat, R209, nitrogen-free ferrochrome was added uC under 400 mm of argon. 35,000_ I I_\ I _Q 1.55 510 20 50 100 200 500 1000 TIME, HRS. Heat No. Atm Symbol % N2 added % N2 (analysis) % Elong., 100 hr n R209 vac 0 0.01 6-7 R211 " d.01.02 5-6' R210 " O.03.03 4-4.5 - R287 ".10 2 = Fig. 10, Effect of nitrogen content upon the stress-rupture properties of vacuum-melted Guy-type alloys.

The University of Michigan * Engineering Research Institute TABLE VI ANALYSES OF VAPOR COLLECTED FROM VACUUM-MELTED HEATS OF THE GUY-TYPE ALLOY Estimated Percentage Range Elements Present Zero Refining Time 20-Minute Refining Time Cr.10 - 1.0.001 -.01 Mn.10 - 1.0.01 -.10 Ni over 10 percent over 10 percent (major constituent) (major constituent) Al.10 - 1.0.01 -.10 Cu.10 - 1.0.01 -.10 Pb.01 -.10.01 -.10 Ca.01 -.10.01 -.10 Mg.01 -.10.01 -.10 Mo 01 - 10.01 -.10 Si.10 - 1.0.10 - 1.0 B.001 -.01.001 -.01 The effect of refining time upon stress-rupture properties at 15000 F. is shown in Fig. 11. The second statistical method was used to evaluate the effect of refining time. All the rupture tests fall within the appropriate 95-percent confidence limits for individual tests and, therefore, refining time does not affect the 100-hour rupture strength of the Guy-type alloy. However, zero refining time results in lower ductility at 100 hours as shown in Table VII. 22

-4 -I -, ct~ 0 Rupture line for refining time data as a whole Elongation given near point representing rupture test. 65,000 rK0 60,000 - 55,000 -------- m W, 50,000 O 50,000 Heat No. Atm Symbol Refining Time Elong Cf 4n0n (min.) (100 hr) S 45)000 R2 vac M 0 5 -\ "ft-Om W R227 vcO3( 40,000 R297 0 0 5 ---- vw~~~~~~~ ^ — ~~~~~~~~~~~~~95% Confidence limits for individual tests 3 from previous vac. + argon data using a 20 35^QQ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ min. refining time 1.5 5 10 20 50 100 200 500 1000 TIME, HRS. Fp 0 3F Figo 11o Effect of re-fining time upon the stress-rupture properties of vacuum-melted Guy-type alloy. 3 rt C ro

The University of Michigan * Engineering Research Institute TABLE VII EFFECT OF REFINING TIME UPON 100-HOUR ELONGATION OF THE GUY-TYPE ALLOY Refining Time (minutes) Percent Elongation 20 7-10 0 3 -5 Therefore, removal during vacuum melting of the trace elements listed in Table VI affects elongation in the same manner as nitrogen. 5. Structural Studies. —In order to determine whether any structural differences existed as a result of air vs vacuum + argon melting of the Guy-type alloy, these structures, before and after rupture testing, were examined by optical microscopy, x-ray diffraction, spectroscopy, electron diffraction, and electron microscopy. The results are conveniently discussed under the above headings. a. Optical Microscopy. The microstructures of air, vacuum + argon, and vacuum + argon heats to which nitrogen was added were examined for inclusion content. The inclusion ratings are listed in Table VIII. TABLE VIII INCLUSION CONTENT OF THE GUY-TYPE ALLOY AS A FUNCTION OF MELTING ATMOSPHERE Inclusion Type (ASTM Inclusion AtmosphereDesignation for Steels, Globular-Type Oxides) vacuum + argon D - 1 vacuum + argon D - 4 (nitrogen additions) air D - 4 Figures 12 and 15 exemplify the ASTM designations given in Table VIII. It is evident from the table and figures that a smaller number of inclusions result from vacuum + argon melting compared with air or nitrogen additions to vacuum + argon melts. 24

The University of Michigan * Engineering Research Institute r I- ~; / ^ / - p.. ^'''.' -- -~~~~~~~~~~~~~~ Fig. 12. Inclusion /content typical of vacuum + argon heats of the Guy-type alloy (250X, unetched). Remarks: Nao inclusions present in entire cross section of test bar. 9/ * * ~. * * -. -. *-... *~ ~~~~~~~~~~~~~~~~~~~~~~~~~ ***>t... w' ~ I' \ ~". **: / "',^' Fig. 15. Inclusion content typical of the Guytype alloy melted under vacuum + argon (with nitrogen additions) and air atmospheres. (250X, unetched) Remarks: Considerable amount of inclusions present, micrograph representative of entire cross section of test bar. 25

The University of Michigan * Engineering Research Institute b. X-ray Diffraction. X-ray diffraction patterns have been obtained from minor-phase extracts, powder samples, and solid samples of the Guy-type alloy for both air and vacuum + argon melting atmospheres. Since the phases present in the ascast condition may disappear and new phases appear during rupture testing at 15000 F., as-cast and after-rupture-testing patterns were obtained for each of the above atmospheres. The diffraction results are shown in Appendices VII-X. The phases Cr7C3, Fe2B, and Cb(CN) have been identified along with the F. C. C. matrix lines. No appearance or disappearance of any of these phases occurs with melting atmosphere or with the as-cast or after-rupture-testing condition of the specimen. Other diffraction data indicate the presence of either a precipitate based on the Ni3A-(7') phase or the existence of a su1ri lattice formation withinl the alloy matrix. c. Spectroscopy. To insure the presence of the above elements (Cr, B, Fe) and to measure their relative concentrations, the minor-phase extracts were analyzed spectrographically. The lines for the above three elements were definitely present in the spectrum of the vacuum + argon, as-cast, and after-rupturetesting extracts and the air, as-cast, and after-rupture-testing extracts. To measure the relative concentrations of Fe and B in the four samples, the intensities of a sensitive iron line, one of the lines of the boron doublet, and a chromium line, whose intensity remained relatively cons stant for all four samples, were measured. (Chromium was the major element present in the extracts.) The intensity ratios Fe/Cr and B/Cr were calculated from the above measurements and are shown in Table IX. These data show that air heats contain a greater percentage of boron and a smaller percentage of iron than vacuum + argon melted heats. As a result of the data in Table IX, it is apparent that Fe2B is not a pure stoichiometric compound but actually a solid solution with small boundary limits. d. Electron Diffraction. Electron diffraction patterns have been obtained from etched surfaces of the four different specimens (air melted, before and after testing; vacuum melted, before and after testing), using reflection diffraction techniques.2 The patterns obtained to date have been generally of rather poor quality, consisting of weak, diffuse spots and having rather heavy backgrounds. Because

The University of Michigan * Engineering Research Institute of their spotty character, accurate measurements of interplanar spacings have not been possible. It appears, however, that the patterns are essentially the same for all four specimens. TABLE IX RELATIVE CONCENTRATIONS OF IRON AND BORON IN MINOR-PHASE EXTRACTS OF THE GUY-TYPE ALLOY Melting Atm Condition Fe/Cr B/Cr vacuum + argon as-cast.89 1.58 vacuum + argon after testing.89 1.57 air as-cast.76 1.98 air after testing.64 2.21 e. Electron Microscopy. To date, microstructure examination of electron microscopy has yielded one definite difference between air- and vacuum-melted structures. In all air-melted heats, a rod-like precipitate appears in the microstructure of after-tested specimens. This precipitate is shown in Fig. 14. The rod-like phase appears to precipitate along definite crystallographic planes. The general structure of the y' (based on Ni3Al) precipitate within the matrix is shown in Fig. 15. No difference in size, shape, etc., of the y' precipitate is evident for air- or vacuum-melted samples. The etchant used in the above work is designed to delineate the over-all microstructure. Several other etchants, which bring out grain boundaries, grain-boundary precipitates, and selectively etch the y' phase, have only been tried on a preliminary basis. These preliminary tests indicate the possibility of a grain-boundary precipitate present in air-melted but not vacuum-melted samples. The research on the mechanism of the influence of vacuum melting upon stress-rupture properties is summarized below:.1. Nitrogen and zero refining time during vacuum melting reduces ductility of vacuum + argon melts to air-melted values. 27

The University of Michigan * Engineering Research Institute Fig. 14o Electron micrograph of rod-like precipitate found in air-melted specimens after testing (Guytype alloy). (10,000 X) Remarks: Etched electrolytically in HF, glycerine, and alcohol mixture Fig. 15o Electron micrograph of the general structure of the matrix and y? precipitate (Guy-type alloy). (10,000 X) Remarks: Etched electrolytically in HF, glycerine, and alcohol mixture. -------------- 8 —------------

The University of Michigan * Engineering Research Institute 2. Vacuum melting removes many elements which might be harmful to stress rupture properties. 5. The phases Cr(C3, Fe2B, and Cb(CN) have been identified in both airand vacuum-melted heats. 4. A rod-like precipitate is present only in the microstructure of air samples, after testing. CONCLUSIONS Following the arrangement of the report, the conclusions are best given under two divisions: A, Effects of Additional Processing Variables, and B, Vacuum-Melting -ffe-ts. A. EFFECTS OF ADDITIONAL PROCESSING VARIABLES 1. Charge preheat time, that is, the time during which the charge is visibly heated before melting, has a major effect. An increase from a preheat period of 5 to a period of 20 minutes improves the 100-hour, 1500~ F. rupture strength of the Guy-type alloy from 40,000 to 50,000 psi. 2. Pouring temperatures of 26600, 27700, and 29500 F. for the Guytype alloy and of 26600, 2760, and 29500 F. for the GMR-255-type alloy resulted in no change in stress-rupture properties at 15000 F. (95-percent confidence limits), 5. Variations in pressure from 0 - 15 psig during pouring were without effect upon the stress-rupture properties (15000 F.) of Guy- and GMR255-type alloys. 4. Variations in the time during which the melt was in the liquid state (under argon) up to two hours were without effect upon the stress-rupture properties (15000 F.) of Guy- and GMR-255-type alloy's 5. Preheat temperature of the mold was raised to l800~ F. in place of the usual l600~ F. without effect upon the stress-rupture properties (15000 F.) of Guy-type alloy. B. VACUUM MELTING 1. Late nitrogen additions from 0 to.10 percent N to vacuum melts of Guy-type alloy decreased the elongation regularly from 7-10 percent to 2 ______________________________ 29 —------

The University of Michigan * Engineerrg Research Institute percent elongation (at 15000 F., 100-hour). No effect upon strength was noted. 2. Zero refining time during vacuum melting decreased the elongatio from 7-10 percent to 3-5 percent in Guy-type alloys. No effect upon strength was noted. 5. Vapor samples collected during vacuum melting showed appreciable amounts of deleterious elements, such as lead. 4. Structural studies employing electron diffraction, spectroscopic, and x-ray diffraction techniques identified the phases Cr7C3, Fe2B, and Cb(CN). These three phases were present in air- and vacuum-melted samples. 5. Electron microscopy disclosed a rod-like precipitate which appeared only in air-melted samples, after testing. SUGGESTIONS FOR FUTURE WORK The work of the past two years has demonstrated that vacuum melting produces very definite improvement in the 100-hour elongation of nickel-base alloys. It is also evident that the 100-hour elongation of vacuum heats is decreased to the elongation of air-melted heats by nitrogen additions and zero refining time. In view of the above discussion, the following future research would greatly add to the understanding of the effects of vacuum melting upon stress-rupture properties: A. Further examination of vacuum-melted microstructures by electron microscopy. B. An investigation of the physical chemistry involved during vacuum melting. C. The investigation of the effect of ordering upon stress-rupture properties. A. EXAMINATION OF VACUUM-MELTED MICROSTRUCTURES The preliminary investigations with different selected etchants revealed the possibility of a grain-boundary precipitate present only in airmelted heats. Since this type of precipitation is known to decrease ductility, the presence of such a grain-boundary phase could explain the difference in ductility of air- and vacuum-melted heats. Therefore, it is suggested that research along these lines be continued to explain the elongation differences in vacuum- vs air-melted heats of the Guy-type alloy. 50

The University of Michigan * Engineering Research Institute B. PHYSICAL CHEMISTRY OF VACLJM MELTING Preliminary analysis of vapors collected during vacuum melting disclosed important amounts of deleterious elements such as lead. This work indicates that further quantitative analyses of the purification during vacuum melting by evolution of undesirable elements might also explain the improvements in properties which have been obtained.' The entire field of vacuum melting is of such widespread interest and importance that the development, in this way, of basic data might constitute major progress. It is proposed, therefore, first, that vapor samples be collected over melts prepared under various conditions and that the vapor analyvsis be correlated with elevated-temperature proper ties.Secoi..d, ie elements Jhich seem to have major effects could be reintroduced separately into vacuum melts7 and individual effects upon elevatedtemperature properties determined. At the same time^ the evolution of deleterious elements from the melts should be followed quantitatively along the classical lines of' physical chemistry, developirng activity data for the important harmful trace elements. C. II'WVESTIGATION OF 7iHE E FFECT` OF ORDERING:UPON STRESSR7PTR7?E PROPER7 IES It is suggested that a small amount of' research be devoted to the exploration of the effect of ordering upon the elevated-temperature properties of these nickel-chromium-aluminum alloys. A hign-temperature x ray camera operating with a bent calcium fluoride monochromat is now available and could be used to determine ordering t+emperatures for setveral alloys. The changes in elevated-temrperature properties accompanying ordering could be developed. In addition to the foregoing, there is a portion of work to be completed involving other processing variables, A number of unsuccessful attempts were made to evaluate the effect of grain size. A proceduire has now been developed at Michigan to produce very fine grained material using graphite molds and it is recommended that representative specimens be tested. 31

The University of Michigan * Engineering Research Institute BIBLIOGRAPHY 1. Hammond, C. M. and Flinn, R. A. "The Effect of Melting and Casting Atmospheres on the Stress-Rupture Properties of Cast Nickel-Base Alloys," Department of the Navy, Bureau of Aeronautics, Airborne Equipment Division, Instruments Branch (AE-43), Contract No. NOas 55-110-c, Washington, D.C. 2. Brockway, L. 0., and Bigelow, W. C., "The Investigation of the Minor Phases of Heat Resistant Alloys by Electron Diffraction and Electron Microscopy," WADC Technical Report 54-589, Wright Air Development Center, May, 1955............................ 52.................

The University of Michigan * Engineering Research Institute APPENDICES 55

APPENDIX I EFFECT OF CHARGE PREHEAT TIME UPON THE STRESS-RUPTURE PROPERTIES OF AIR AND ARGON-PROTECTED GUY-TYPE ALLOY Pouring Temperature, 2800~F. Testing Temperature, 15000F. Pouring Pressure, 5 psig. Heat Atm Chemical Composition (1)Stress Rupture Percent (2) C Cr Mo Al Fe Cb B Life (Hours) Elongation Ih R-218 air.11 11.78 5.8o 6.22 4.65 1.78.45 60,000 59.5 5 3 55,000 67.2 5 50,000 111 5 45,000 116 P R-285 air charge same as R-218 60,000 7.5 1.5 50,000 175 2.5 m R-500 air charge same as R-218 60,000 36.9 2 45,000 117 - R-215 argon charge same as R-218 60,000 14.2 50,000 69.3 2 45,000 128 2.5 40,000 282 5 R-284 argon charge same as R-218 50,000 150 4 0 R-501 argon charge same as R-218 60,000 57.7 5 - 50,000 85.7 2 (1) Balance Ni (2) Percent elongation in one inch measured after fracture.

APPENDIX II EFFECT OF POURING TEMPERATURE UPON THE STRESS-RUPTURE PROPERTIES OF ARGON-PROTECTED AND VACUUM + ARGON MELTED GUY-TYPE AND ARGON-PROTECTED GMR-255-TYPE ALLOYS Pouring Pressure Under Argon Protection, 5 psig. Improved Melting Practice Used Only For Argon-protected Guy Alloy Standard Vacuum Procedure Employed Type Heat Pouring Atm. Chemical Composition (1)Stress Rupture Percent (2) Alloy Temp. C Cr Mo Al Fe Cb B Life (Hours) Elongation - Guy R-212 2950 argon.10 12.86 5.74 6.24 4.94 1.54.38 50,000 190 6 protect- 45,000 286 5 ed 0 Guy R-215 2770 argon charge same as R-212 60,000 14.2 5 protect- 50,000 69.3 2 ed 45,000 128 2.5 - 4o,000 282 5 Guy R-214 2670 argon charge same as R-212 55,000 86.3 5 protect- 50,000 95 5 ed Guy R-02 2670 argon charge same as R-212 50,000 92.8 5 protect- 40,000 251 2 aed Guy R-225 2800 vacuum charge same as R-212 65,000 27.5 5.5 + argon 60,000 50.7 5 56,000 100 6 m 50,000 219 8 Guy R-224 2660 vacuum charge same as R-212 65,000 28.4 4 + argon 60,000 61.5 5 56,ooo 105 8 50,000 249 7 Guy R-225 5000 vacuum charge same as R-212 50,000 242 10 + argon GMR255 R-15O 2950 argon.14 14.24 5.0 5.05 9.15 1.56.06 45,ooo 58.9 10 protect- 4oOOO 126 10 ed 55,000 278 6 W GMR255 R-151 2760 argon same charge as R-150 45,ooo 19 7 protect- 55,000 529 7 ed GMR255 R-152 2660 argon same charge as R-150 45,000 35.6 15.5 protect- 40,000 118 9 ed 55,000 346 12 (1) Balance Ni (2) Percent elongation in one inch measured after fracture.

-I APPENDIX III EFFECT OF POURING PRESSURE UPON THE STRESS-RUPTURE PROPERTIES OF ARGON-PROTECTED GUY- AND CGMIR-235-TYPE ALLOYS Pouring Temperature of Guy Alloy 28000 F. Pouring Temperature of GMR 235 28500 F. Improved Melting Practice Employed Only For Guy Alloy 0 Type Pouring Chemical Composition Rupture Percent Alloy Heat Pressure C Cr Mo Al Fe Cb B Stress Life (Hours) Elongation Guy R-216 0.10 12.24 6.0 6.19 4.25 1.76.41 50,000 39.3 2.5 45,000 128 2.5 3 Guy R-215 8 charge same as R-216 55,000 68.1 3 o\ 50,000 61.4 3 45,000 141 4 m Guy R-217 12 charge same as R-216 45,o000oo 105 -- 40,000 219 3 C Cr Mo Al Fe Ti B GMR 235 R-154 15.13 14.09 2.82 3.10 9.05 1.50.08 45,oo000 45 8 40,000 113 10 35,000 304 9 GMR 235 R-153 10 charge same as R-154 45,o000oo 23 6.5 40,000 128 9 GMR 235 R-155 0 charge same as R-154 40,000 130 6 35,000 348 13 ______________________________________________________________________________________________ w~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-,

APPENDIX IV EFFECT OF HOLDING TIME IN THE LIQUID STATE UPON THE STRESS-RUPTURE PROPERTIES OF ARGON-PROTECTED GUY- AND GMR-235-TYPE ALLOYS Pouring Temperature for Guy 2800~ F. Pouring Temperature for GMR 235 2850~ F. 0 Improved Melting Practice Employed For Both Alloys Pouring Pressure for both alloys 5 psig Type Holding Chemical Composition Rupture Percent' Alloy Heat Time (Hours) C Cr Mo Al Fe Cb B Stress Life (Hours) Elongation Guy R-219 2.10 12.40 5.67 6.19 4.95 1.48.40 50,000 149 6 isw^~~~~~~~~~~~~~~~~~~~ 45,000 366 6 Guy R-220 1/2 charge same as R-219 50,000 114 4 45,000 400 6 C Cr Mo Al Fe Ti B GMR 235 R-221 2.16 14.86 5.67 3.50 10.38 1.86.06 50,000 17.4 14 40,000 107 12 35,000 309 10 GMR 235 R-222 1/2 same charge as R-221 50,000 28.9 10 40,000 102 6 - 35,000 459 8 r+ _ _~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~f

The University of Michigan * Engineering Research Institute APPENDIX V EFFECT OF MOLD PREHEAT TEMPERATURE UPON THE STRESS-RUPTURE PROPERTIES OF AIR ANID ARGON-PROTECTED GUY ALLOY Improved Melting Practice Used For Both Air and Argon-Protected Heats Pouring Temperature 28000 F. Pouring Pressure 5 psig Mold Rupture Percent Heat Atm Temp. Chemical Composition Stress Life Elong_ _____ (OF) C Cr Mo Al Fe Cb B (Hours) ation R-278 air 1800 50, 000 119 4 45,000 517 1 R-281 air 1800 50,000 164 5 R-279 argon 1800 50,000 526 3 45,000 88.7 5 R-280 argon 1800 50,000ooo 91.6 5 45,000 220 4 ---------------------- 8 —-----------

APPENDIX VI EFFECT OF NITROGEN CONTENT UPON THE STRESS-RUPTURE PROPERTIES OF VACUUM + ARGON MELTED GUY-TYPE ALLOY C Pouring Temperature 29000 F. Nitrogen Added As High-Nitrogen Ferrochrome Nitrogen Addition Was Made Under 400 mm. Of Argon Pressure In Control Heat, R-209, Nitrogen-Free Ferrochrome Was Added Under 400 mm. Of Argon Percent Chemical Composition Rupture Percent Heat No. N2 Added C Cr Mo Al Fe Cb B IN2 Stress Life (Hours) Elongation ) R-209 0.16 12.45 6.o 5.40 5.56 2.11.4.01 65,000 22.1 7 w 56, 000 75.23 w^~~~~~~ ~~~~ ~~~~~~ ~~56,000 67.2 5 \o^~~~~~~~~~~. ~~~~ ~~50,000 199 8 50,000 107 6 ri, R-211.01.15 12.45 5.96 5.5 5.56 1.92.4.02 65,000ooo 29.56 56,000O 83.5 6 55,000 75.1 5 50,000 511 5 50,000 154 5 R-210.05.14 12.27 6.09 5.95 5.06 2.0.43.05 65,000ooo 15.2 5 56,000 68.1 4 50,000 148 4 50,000 120 5 R-287.10,07 56,000 27.7 1.0 50,000 95.2 2.0

The University of Michigan ~ Engineering Research Institute APPENDIX VII EFFECT OF REFINING TIME UPON THE STRESS-RUPTURE PROPERTIES OF VACUUM + ARGON MELTED GUY-TYPE ALLOY Pouring Temperature 2900~ F. Refining Chemical Composition Rupture Percent Heat Time C Cr Mo Al Fe Cb B Stress Life Elong_____ (Min. ) (Hours) ation R-227 0 60,000 61.1 4 56,000 88.1 5 56,000 25 2o5 50,000 191.2 3 R-297 0 65,000 50 4.5 56,000 83.8 5 R-298 20 56,000 78.4 4 Control ~~~Heat~~,40

The University of Michigan * Engineering Research Institute APPENDIX VIII X-RAY DIFFRACTION RESULTS FOR THE GUY-TYPE ALLOY MELTED IN VACUUM + ARGON, AS-CAST Cu Rad. Cr Rad. Cu Rad. Y' Precipitate Extract Powder Solid Sample Cr7C3 Fe2B CbC Cr3C2 Or Ordered F.C.C. Pattern Pattern Pattern CbN F.C.C. Matrix Matrix d I d I d I d I d I d I d I d I d I 3.62 vw 3.56 3.22 vw 3.22 mw 3.14.4 3.12 w 3.09 m 3.01 mw 2.87 w 2.74.7 2.70 w 2.65 w 2.59 m 2.56 s 2.55 wm 2.55 vs 2.55.8 2.52 2.48 mw 2.48 mw 2.48.5 2.46 m 2.46.5 2.38 m 2.30 1.0 2.26 w 2.26.6 2.23 1.0 2.22 w 2.20 w 2.20 s 2.16 w 2.15 w 2.13 m 2.12 s 2.10 s 2.10 mw 2.10.7 2.07 m 2.07 vs 2.07 2.07 s 2.04 s 2.03 wm 2.02 vs 2.01 s 2.00 mw 1.99.8 1.98 s 1.96 m 1.86 s 1.86 1.0 1.85 w 1.84 s 1.84 w 1.82 w 1.82 mw w 1.82 mw 1.0 1.79 w 1.79 1.79 s 1.78 w 1.78.7 1.76 m 1.76.7 1.71 m 1.66.4 1.64 mw 1.62 mw 1.62. 1.61 mw 1.60 1.58.5 1.56 m 1.56 vw 1.56 ms 1.54 vw 1.53.8 1.52 w 1.51 m 1.50.8 1.46 w 1.46 1.44.5 1.43 s 1.43 s 1.42.4 1.41 1.0 1.40 vw 1.40 vw 1.58 w 1.58.4 1.36 vw 1.55 s 1.34 ms 1.34 vw 1.34 ms 1.5 w 1.55 ms 1.33.4 1.28 wm 1.27 ms 1.27 s 1.27 ms 1.27 mw 1.27 1.27 m 1.26 m 1.26.8 1.24 vw 1.24 vw 1.19 1.15 1.09 m 1.08 1.04 mw 104 1.04.892 w.892.85 m.83 41

The University of Michigan * Engineering Research Institute APPENDIX IX X-RAY DIFFRACTION RESULTS FOR THE GUY-TYPE ALLOY MELTED IN VACUUM + ARGON, AFTER RUPTURE TESTING* y' Precipitate Powder Solid Sample Cr7C3 Fe2B CbC Cr3C2 Or Ordered F.C.C. Pattern Pattern CbN F.C.C. Matrix Matrix d I d I d I d I d I d I d I d I 3.62 vw 3.56 3.22 mw 3.14 vw 5.14.4 3.01 mw 2.74.7 2.70 w 2.65 w 2.61 m 2.58 w 2.57 vw.2.55 wm 2.55 vs 2.55.8 2.52 2.50 mw 2.49.5 2.48 mw 2.46.5 2.38 m 2.30 1.0 2.26 w 2.26.6 2.23 1.0 2.22 w 2.20 s 2.17 w 2.15 w 2.13 ms 2.13 m 2.12 s 2.11 wm 2.10.7 2.07 ms 2.07 2.07 s 2.06 vs 2.04 s 2.02 vs 2.01 vs 2.00 ms 1.99.8 1.96 m 1.90 w 1.90 1.0 1.87 w 1.86 s 1.86 1.0 1.85 w 1.84 s 1.84 w 1.82 mw 1.82 w 1.82 1.0 1.81 w 1.79 vs 1.79 1.79 s 1.78 s 1.78 w 1.78.7 1.76 m 1.76~.7 1.71 m 1.66.4 1.65 vw 1.64 mw 1.62 mw 1.62.5 1.61 vw 1.61 mw 1.60 1.59.5 1.56 vw 1.56 ms 1.53.8 1.51 m 1.50.8 1.46 w 1.46 1.44.5 1.43 s 1.45 s 1.41 1.0 1.40 vw 1.58 w 1.58 w 1.58.5 1.55 s 1.34 vw 1.34 m 1.33 w 1.33 ms 1.33.4 1.28 wm 1.27 vs 1.27 vs 1.27 mw 1.27 1.27 m 1.26 s 1.26.8 1.19 1.15 1.09 m1.09 1.08 1.04 1.04 1.04.892.892.853.83 * Same extract pattern as shown in Appendix VII ---------------------- -42

The University of Michigan * Engineering Research Institute APPENDIX X X-RAY DIFFRACTION RESULTS FOR THE GUY-TYPE ALLOY MELTED IN Alh, AS-CAST* 7' Precipitate Powder Solid Sample Cr7C3 Fe2B CbC Cr3C2 Or Ordered F.C.C. Pattern Pattern CbN F.C.C. Matrix Matrix d I d I d I d I d I d I d I d I 3.62 mw 3.62 vw 3.56 3.22 mw.3.14.4 3.13 w 3.01 mw 2.88 w 2.74.7 2.70 w 2.65 w 2.59 m 2.56 vs 2.56.8 2.55 wm 2.54 mw 2.52 2.49 mw 2.48 mw 2.48.5 2.38 m 2.30 1.0 2.26 w 2.26.6 2.23 1.0 2.22 w 2.22 w 2.20 s 2.13 m 2.12 mw 2.12 s 2.10.7 2.07 vs 2.07 2.07 s 2.06 vs 2.04 s 2.02 vs 2.01 vs 1.99 m 1.99.8 1.96 m 1.90 w 1.90 1.0 1.86 s 1.86 1.0 1.84 s 1.84 w 1.82 mw 1.82 w 1.82 1.0 1.79 m 1.79 1.79 s 1.78 w 1.78.7 1.76 m 1.76 m 1.76.7 1.75.7 1.71 m 1.66.4 1.64 mw 1.62 mw 1.62.5 1.61 mw 1.59.5 1.56 ms -.53.8 1.51 m 1.50.8 1.46 w 1.44.5 1.43 s 1.42.4 1.41 1.0 1.39.4 1.38 w 1.37.5 1.35 s 1.35 s 1.34 w 1.33 w 1.33 ms 1.30.3 1.28 wm 1.28.5 1.27 vs 1.27 m 1.27 m 1.27 1.27 i 1.26 s 1.26.8 1.19 1.09 1.09 m 1.08 1.04 1.04 1.04.892.892.85 *Same extract pattern as shown in Appendix VII. ------- --------- 443

The University of Michigan * Engineering Research Institute APPENDIX XI X-RAY DIFFRACTION RESULTS FOR THE GUY-TYPE ALLOY MELTED IN AIR, AFTER RUPTURE TESTING* y' Precipitate Powder Solid Sample Cr7Cs Fe2B CbC Cr3C2 Or Ordered F.C.C. Pattern Pattern CbN F.C.C. Matrix Matrix d I d I d I d I d I d I d I d I 5.68 mw 3.62 vw 3.56 3.44 vw 3.22 3.14.4 3.01 mw 2.74.7 2.70 w 2.65 w 2.61 m 2.56 mw 2.56.8 2.55 wm 2.55 vs 2.52 2.48 mw 2.48 mw 2.48.5 2.38 m 2.32 w 2.30 1.0 2.26 w 2.26.6 2.23 1.0 2.20 s 2.13 wm 2.12 m 2.12 s 2.10.7 2.07 vs 2.07 2.07 s 2.06 s 2.04 s 2.02 wm 2.01 vs 1.99 m 1.99.8 1.96 m 1.90 w 1.90 1.0 1.86 s 1.86 1.0 1.84 s 1.82 mw 1.82 mw 1.82 1.0 1.80 m 1.79 s 1.79 1.79 s 1.78 w 1.78.7 1.76 m 1.76.7 1.71 m 1.66.4 1.64 mw 1.62 mw 1.62 mw 1.62.5 1.60 w 1.59.5 1.56 w 1.56 ms 1.55.8 1.51 m 1.50.8 1.46 w 1.44.5 1.43 m 1.43 s 1.42.4 1.41 1.o 1.39.4 1.38 w 1.38 ms 1.57.5 1.55 s 1.535 s 1.55 vw 1.50 3. 1.28 wm 1.28.5 1.27 vv 1.27 m 1.27 m 1.27 1.27 m 1.26 s 1.26.8 1.19 1.13 1.09 1.09 1.08 1.04 1.04 1.04.892.892.85.83 ~Extract pattern same as Appendix VII. 44