ENGINEERING RESEARCH INSTITUTE UNIVERSITY OF MICHIGAN ANN ARBOR. MICH. PROGRESS REPORT TO THE NATIONAL ADVISORY COMMITTEE FOR AERONAUTICS COVERING RESEARCH ON HEAT-RESISTANT ALLOYS Submitted R. F. Decker John P.: Rowe J. W. Freeman June 30, 1956 Under Contract NAw 6457

ACKNOWLEDGE MENTS The examination of structures with the electron microscope has been a joint endeavor with Professor Wilbur Bigelow and his associate, Mr. J. Amy. Professor Bigelow's work is sponsored by the Metallurgy Research Branch of the Aeronautical Research Laboratory, WADC. Samples were prepared, etched and examined under guidance of Professor Bigelow. In addition, he will use the samples to improve techniques for phase identification and the results will become available to the NACA research. The excellent techniques developed by Professor Bigelow are now being standardized and adopted by metallurgists for more routine studies. Particular acknowledgements are due to Professor Bigelow for his excellent work in developing the techniques and phase identification procedures. The contribution of a number of people working on this project is acknowledged. Mr. Karl Kienholz contributed greatly to operating the vacuum furnace. Mr. Alex Dano did excellent optical and electron micrographic work. His development of etching procedures for macrostructures and optical micrography aided greatly in the study. Miss Christine Sadler has ably prepared metallographic samples and photographs. Mr. Jerry White assisted with much of the routine heat treating and testing.

PROGRESS REPORT OF RESEARCH ON HEAT-RESISTANT ALLOYS JUNE 30, 1956 This report presents the progress made since the last progress report dated September 15, 1955. The major experimental work has been concentrated on the influence of melting practice and hot-working conditions using the refractory Ti + Al hardened alloy known as Udimet 500. Therefore the rest of the report will be confined to this subject. In addition there were several other investigations in progress in September. S TATUS OF INVESTIGATIONS OF ABNORMAL GRAIN GROWTH, OVERHEA TING AND INFLUENCE OF HOT-WORKING CONDITIONS ON PROPERTIES OF WASPALOY Three other programs were in progress at the time the last progress report was submitted. The status of these programs is as follows: 1. Abnormal Grain Growth in Heat-Resistant Alloys This program has been completed and the following two final reports submitted covering the results: (1) "Abnormal Grain Growth in Nickel-Base Heat-Resistant Alloys" (2) "Abnormal Grain Growth in M252 and S816 Alloys" 2. Influence of Overheating on the Rupture Properties at 15000F of S816, HS-31 and M252 Alloys. Final reports have been submitted covering S816 and HS-31 alloys. The titles of the reports were:

(1) "Effect of Overheating on the Creep-Rupture Properties of S816 Alloy at 1500~F" (2) "Effect of Overheating on the Creep-Rupture Properties of HS-31 Alloy at 1500 F" The report covering M252 alloy is being prepared. The results for M252 showed that periodic exposures of Z minutes to 19000 and 2000'F prolonged rupture life at 1500~F. Thus, this type of exposure to these temperatures does not reduce strengths at 15000F as was the case for S816 and HS-31 alloys. Any reduction in life would require the presence of stress to accelerate creep. Some reduction in life, however, resulted from the periodic exposure to 16000 and 1800'F in a manner similar to that observed in S816 and HS-31 alloys. 3. Influence of Hot-Working Conditions on the Rupture Properties of Waspaloy. When the budget was reduced this investigation was temporarily stopped. As soon as the report on overheating is completed, the results to date will be prepared in report form. INFLUENCE OF MELTING PRACTICE, HOT-WORKING CONDITIONS AND HEAT TREATMENT ON THE HIGH-TEMPERATURE PROPERTIES AND HOT-WORKING CHARACTERISTICS OF A Ti + Al HARDENED ALLOY The study of the influence of melting conditions on the properties of heatresistant alloys using the refractory alloy known as Udimet 500 has been continued since September 1955 along the following lines: 1. Study of the influence of melting variables, particularly combinations of deoxidation and superheat practice, 2. Introduction of variations in hot-working conditions, 3. St!dy of the influence of heat-treating conditions on properties, 40 Development of a better understanding of the basic microstructures in the alloys with emphasis on the use of the electron microscope.

High-temperature properties were evaluated in all cases through survey rupture tests at 1600~F under a stress of 25,000 psi. Qualititive observations on 'iot-working characteristics have been made, although the restrictions imposed by the limited conditions of working have mainly limited this to cracking tendency. The general objectives of the investigation have been to isolate melting, hot working, and heat treatment variables having a significant effect on creep-rupture properties. With these established, the basic metallurgical causes of the observed variations were to be determined. Melting has been carried out in a vacuum chamber. Ingots were hot rolled to bar stock and heat treated. Detailed chemical analyses including determination of 02 and N2 contents were used to evaluate variables. The probability that any variations in melting or hot working would have their influence through alteration of the reactions involving precipitation of Ni3 (Al, Ti) has led to detailed analysis of microstructures. INFLUENCE OF MELTING PRACTICE The program on the effect of melting practice on the high-temperature properties of Udimet 500 was continued. Among the variables studied were deoxidation practice, superheat temperature, pouring temperature, ingot mold shape and refining temperature. Melting was restricted to vacuum with no introduced gases. Induction melting was carried out in the University of Michigan vacuum furnace in stabilized zirconia crucibles. Pressures before meltdown and after pouring were less than 5 microns as measured by both Stokes and thermocouple gages. Melt temperatures were measured with a Pt - Pt + Rh immersion thermocouple. Melting cycles for the heats melted are illustrated in figure 1. Aim analysis, in weight percent, on all heats was as follows:

Cr Co Mo Ti Al Mn Si C Ni 20.0 15. 0 4.00 3. 10 3. 10 0. 12 0. 12 0. 08-0. 15 Bal. Electrolytic Ni, Cr, Co, and Mn melting stocks were used. The Mo was arc -melted low carbon stock; the Ti was Ti 55 stock and the Al was 99. 99 percent purity ingot stock. Ten-pound heats were poured into massive copper molds. Hot-working practice was kept constant for all the heats used in the study of melting practice. This practice included: 1. Homogenizing ingot 1 hour at 23000F, air cooling, 2. Surface grinding ingot, 3. Rolling at 21500F to 7/8 inch bar stock using 22 passes with 21 reheats of 10 minutes between passes. The last pass was a 5 percent reduction followed by air cooling. Chemical analyses were obtained from samples cut from the rolled bar stock. This metal originally was in the center of the as-ca:t ij-.ots. The analyses are included in Table I. Evaluation of high temperature properties of the heats was made after the as-ro!led stock was solution treated at 2150~F and air cooled. Rupture samples were tested at 25,000 psi and 1600~F. The results are tabulated in Table II. The results will be reviewed in the following sections. Effect of Deoxidation Practice Deoxidation was carried out using varying amounts of C in the charge and with Si and Al. One objective was to melt down with excess oxygen and then hold the melt under high vacuum for possible reaction of the oxygen with impurities to refine the melt. Final deoxidation with carbon required a prolonged time period due to the necessity of slowly adding the carbon to avoid excessive gas evolution.

Effect of a High 02 Refining Period The effect of a refining period in the melting cycle was studied by varying the method of carbon addition as follows: Heat Method of Carbon Addition 1129 100%o of carbon mixed with crucible charge to deoxidize melt during meltdown. 1 133 50%o of carbon mixed with crucible charge to partially deoxidize melt during meltdown. 50%o of carbon added to complete deoxidation after a 20 minute refining period at 2700'F. 1136 0%o carbon mixed with crucible charge. 100% carbon added to deoxide melt after a 20 minute refining period at 27000F. Holding times for the three heats were not constant because of the difficulty of excessive gas evolution during carbon additions after meltdown. While deoxidation of I-at 1129 occurred during meltdown, deoxidation of Heat 1136 required 1. 5 hours after meltdown. After deoxidation, Si, Al, Ti, and Mn additions were made. Then the melts were superheated to and poured at approximately 29000 to 3000 F. The rupture properties obtained are listed in Table II. 'They were: Heat Rupture Time Elongation Reduction in Area (hours) (percent) (percent) 1129 82.3 4.7 1133 90.7 4.0 4.0 1136 99.5 6.0 4.0

Effect of Deoxidizing Material Al, Si, and chunk C were used to deoxidize the melt after meltdown and a 20 minute refining period at 2700~F. Then, after additions of the remaining alloying elements, the melts were superheated to and poured at approximately 3000'F, with the exception of the Al deoxidized heat which was superheated to 3100~F. The heats and variables were: Heat Deoxidizing Material 1136 chunk C 1.138 Al 1139 Si followed by chunk C (Si did not completely deoxidize the melt) The rupture re sults were: Rupture Time Elongation Reduction in Area Heat (hours) (percent) (percent) 11T3 99.5 6.0 4.0 1138 147.4 5.0 4.6 1139 75.5 4.0 3.1 Effect of Ingot Mold Shape Heat 1137 was melted with identical procedures as those used in Heat 1,136 but was cast in an ingot mold with 50 taper whereas Heat 1136 was cast in a straight mold. The rupture results obtained were: Rupture Time Elongation Reduction in Area Heat (hours) (percent) (percent) 1136 99.5 6.0 4.0 1137 87.1 6.0 6.4

Effect of Superheat and Pouring Temperatures Several conditions of superheating were studied to determine if purification of the melt by vaporization was a factor in properties. In addition, evaluation of the effect of pouring temperatures upon ingot structures, cracking during working and rupture properties was desired. In order to establish the effects of superheat and pouring temperatures, Heats 1141 and 1142 were melted with identical practice as 1138 except for variations in superheat and pouring temperatures as listed below. Heat Superheat Temperature Pouring Temperature 1-138 3100*F 3000~F 1141 2975'F 2 750*F 1142 2750'F 2750~F It should be noted that the carbon contents obtained in Heats 1141 and 1142 were high compared to Heat 1138. The rupture results are summarized as follows: Rupture Time Elongation Reduction in Area Heat (hours) (percent) (percent) 1138 147.4 5.0 4.6 1141 81.2 8.0 11.5 1142 50.8 8.0 12.3 As-cast macrographs and micrographs illustrating the effect of pouring temperature on structures are shown in figure 2. Effect of Refining Temperature It seemed possible that the temperature of the melt during the refining period might affect the purification reactions.

Since Heat 1138 had relatively good rupture properties and hot workability, a similar melting practice was used in making Heat 1143. While the melt temperature was kept at 2700'F during the 20 minute refining period with Heat 1138, in Heat 1.143 this temperature was maintained at 2900~F. In addition, a polished piece of pure Ni was suspended above the melt during this refining period in order to collect metal vapors volatilized from the melt. Chemical analysis of the condensed metal vapors yielded the following quantities of elements in the condensate: Elements Elements Elements Elements Elements over 10% 1% to 10%o 0. 1% to 1%o 0. 01%0 to 0. 1o 0. 001% to 0. 01%O Cr, C o, Mn Ni P Al, Cu, Pb B, Ca, Mg, Mo, Si Rupture testing revealed that Heat 1143 did not reproduce the properties of Heat 1.138o Rupture Time Elongation Reduction in Area Heat (hours) (percent) (percent) 1.138 147.4 5.0 4.6 1 1 -3 85.6 15.0 9.3 1143 (duplicate) 86.5 8.5 10.0 Effect of Melting Practice on Cracking During Rolling Since a constant hot-working practice was employed in rolling the ingots of the heats used to study melting practice, it was possible to compare the tendency to crack during working. Figure 3 shows photographs of the as-rolled bars, Several differences in tendency for cracking during rolling can be observed. 1. Heat 1129, deoxidized during meltdown without a refining period, had intermediate cracking.

2. Heats 1133, 1136, and 1137, deoxidized by carbon after a refining period, underwent relatively severe cracking. 3. Heats 1138, 1141, and 1142, deoxidized by Al after refining at 2700~F cracked very little. However, it is possible that in Heats 1141 and i142 the high carbon content contributed to the reduced tendency to crack. 4. Heat 1139, which was deoxidized incompletely with Si and then completely with C, had an intermediate tendency for crack formation. Likewise, Heat 1143, which was deoxidized by Al after refining at 2900~F, was intermediate. INFLUENCE OF HOT-WORKING VARIABLES Prior experience has indicated that in order to produce bar stock from the cast ingots with a minimum of cracking, the working conditions employed must remain within a rather narrow range of variables. The best procedure found to hot rll the ingots has been to reduce about 4 to 6 percent in one pass and then reheat for 10 minutes. Metal temperatures had to be kept between 2100~ and 2200~F. The cracking problem thus limited the range of variables which could be considered in the study of hot working. Effect of Homogenization of Ingots The effect of heating to 23000F on the workability of Heat 1129 was studied. A portion of the ingot was rolled without any exposure to 23000F, and then the effect of exposure for one hour prior to rolling and for one hour prior to rolling followed by another hour part way through the rolling was also included. The bars so produced were rupture tested at 1600'F and 25, 000 psi after solution treatment both at 1975~ and 2150~F. The influence of this homogenization may be summarized as follows:

10 Effect on Workability The bars from this study are shown in figure 4. Some corner cracking was evident on all three of the bars but was not severe. The cracks were not deep and apparently had no effect on the metal below them in the bars. The homogenization apparently had no effect on the workability of this heat. Effect on Rupture Time The rupture time for Heat 1129 (fig. 5) does not appear to have been influenced significantly by the 2300~F homogenization. When the final heat treatment was 1975 F, homogenization at 2300~F may have been slightly detrimental, while it may Aid<:; been somewhat benificial when the final heat treatment was at 2150OF, Effect on Microstructure The microstructures of two of the bars following rolling are shown in figure 6. All three exhibited a mixed grain size which did not appear to vary appreciably from bar to bar. After solution treatment at either of the two temperatures used for this study, recrystallization resulted in a more uniform grain size (fig. 7), The same structures were retained after rupture testing at 16000F (figs, 8 and 9) with some additional precipitation appearing in the grain boundaries. The microstructures after rupture testing also show at X1000D the presence of a fine dispersion of the Ni3 (Al, Ti) precipitate. Effect of Amount of Finishing Reduction Two heats, 1141 and 1142, were used in the study of the effect on rupture properties of the amount of reduction given in the last passes of the rolling The

ingots were homogenized one hour at 2300'F and surface ground prior to the start of rolling. They were then rolled from the 2-1/2 inch ingot size down to 1 inch broken-cornered squares using the usual procedure of single pass reductions of about 4-6 percent with reheats for ten minutes between passes. The bars were then cut in half and one half finished to 7/8 inch squares using the usual practice of three passes with two reheats. The remaining two halves were then reduced to 7/8 inch squares in two passes with no reheats. The samples obtained from these bars were then heat treated and tested at 1600'F and 25, 000 psi. Although there was some scatter in the data obtained for the two heats and the two working conditions (table II and fig. 10), there do not appear to be any differences in rupture properties as a result of the heavier reduction in the final passes on these two heats. EFFECT OF HEAT TREATMENT AFTER ROLLING In all of the prior studies discussed in this report the material has been rupture tested as solution treated at 2150~F and air cooled. Several other heat treatments also were studied in order to determine the effect on properties of the various steps of normal heat treating. The following conditions were evaluated on many of the heats produced in this investigation: 1. Solution treat 4 hours at 19750F, air cooled, 2. Solution treat I or 2 hours at 2150~F, air cooled, 3. Solution treat 2 hours at 21500F, air cooled, plus solution treat 4 hours at 19750F, air cooled, followed by aging 24 hours at 15500F plus 16 hours at 14000F, 4. Solution treat 2 hours at 21500F, air cooled plus the same double age as in (3). The results of this study on most of the heats prepared indicate the following general trends as a result of the heat treating conditions (fig. 11)

1l With only one exception, solution treating at 21500F resulted in higher strengths than solution treating at 1975~F. Although the strength levels were different from heat to heat, the magnitude of this increase appeared to be about the same for all heats. 2. The effect of aging on the strength of the material after solution treating at 2150~F was only studied on two heats. The results are not conclusive and show an improvement in life for one the heats and a decrease in life for the other one when compared to the strength as solution treated at 2150'F with no aging. 3. The effect of a double solution treatment followed by a double age (condition 3 above) was also difficult to evaluate, In general, the strength appears to be -7 out the same as it was after solution treating at 2150~F. Since cooling rate after solution treatment was found to influence the precipitation reaction, a limited investigation was carried out to find the effect of cooling rate B:efore aging and rupture testing. Specimens of Heats 1138 and 1139 were heat treated by 2 hours at 2150~F, ice-brine quenched; plus 4 hours at 1975~F, ice-brine quenched; plus 24 hours at 1550~F; plus 16 hours at 1400~F. The specimen from Heat 1139 fractured intergranularly during machining. Other cracks were visible on the machined surface. The specimen from Heat 1138 had no visible cracks after machining, but ruptured upon loading at 16000F. Microstructural investigations showed the presence of a discontinuous precipitate at the grain boundaries after aging in both samples. However, specimens which had been air cooled after solution treatment did not exhibit discontinuous precipitation and were not embrittled.

13 Effect of Testing Procedures Testing of all samples prepared was conducted at 1600'F and 25, 000 psi. Loading the specimens into the furnace was done by a standard practice of placing the test in a hot furnace, adjusting temperature and loading in four hours. In order to know the effect of variations from this standard procedure,deliberate differences in the preheat time were evaluated. The results of these tests were as follows as conducted on Heat 1130: Time at 1600~F before applying Rupture Time stress(hoUr) (hours) 2 80.0 4 102.2 24 23.4 These results show that there is a definite influence on rupture time of the time at temperature prior to loading. These rupture times, plotted together with hardnesses after aging at 16000F (fig. 12) indicate a similar trend for both effects. STRUCTURAL STUDIES OF UDIMET 500 The variations in properties may be controlled by the distribution of the Ni3 (Al, Ti) precipitates and other secondary phases. Electron micrographic studies offer a means of studying this possibility. Accordingly, a program is in progress to provide familiarity with the fine structures revealed by the electron microscope, The initial phase of this study involves determination of the size and distribution of the secondary phases in one heat of the alloy as influenced by exposure to several temperatures for varying times, Ultimately it is expected

14 that sufficient background information will be developed so that it will be possible to determine if melting practice and processing variables have their influence on properties at high temperature through the secondary phase structures. The structural studies included work on solution of the secondary phases and precipitation of the secondary phases as described in the following sections. Effect of Solution Treatment Specimens from rolled bar stock of Heat 1138 were homogenized 2 hours at 2150~F and ice-brine quenched. These specimens were treated 4 hours at 1700~, 1800~, 1900~, and 2000~F and ice-brine quenched. Vickers penetration hardnesses obtained on these samples are presented in figure 13. In addition, hardness of a sample aged at 16000F is included. While treating at 1600' and 1700~F increased the hardness above the original value obtained by homogenization at 21500F followed by ice-brine quenching, treating at 1800~ and 1900~F decreased the hardness significantly. Treating at 2000~F gave a slightly lower hardness than the homogenized and quenched value. Figures 14 and 15 include micrographs of the samples. A uniform precipitate was present in the matrix after treatments at 17000, 1800~ and 1900'F. This precipitate increased in size and spacing as the treatment temperature increased to 1900~F. The precipitate is assumed to be undissolved Ni3 (Al, Ti). Treating at 20000F maintained the precipitate in solution so that none is visible in the micrographs (figs. 14d, 15d). In addition, it is apparent that considerable amounts of secondary phases agglomerated at the grain boundaries at temperatures up to 1900~F. These were not present after treatment at 2000~F. It is evident from this study that the solution temperature for Ni3 (Al, Ti) in Udimet 500 lies between 19000 and 20000F for a 4 hour heat treatment. The

15 massiveness of the precipitate particles after treatment at 19000F suggests that the solution temperature could be as high as 1950'F in Heat 1138. Microstructures of several of the heats after solution treatment at 19750F have shown presence of undissolved Ni3 (Al, Ti). In some cases the precipitate was uniform throughout the structure, while in other cases the precipitate appeared in bands. It is believed that this condition arises because the solution temperature of Ni3 (Al, Ti) is close to 1975~F. The presence of the precipitate after the 1975~F treatment seems to depend somewhat upon prior history. It is probable that some of the erratic rupture test results obtained after the 19750F treatment can be attributed to failure to completely dissolve the Ni3 (Al, Ti). In cases where the precipitate was visible after the 19750F treatment, increasing the solution treating temperature to 21500F increased rupture life. The hardness data in figure 13,showing low hardness after the 1800~ and 1900'F treatments,indicate that these treatments can lead to severe overaging of the Ni3 (Al, Ti) and matrix depletion. In fact, the hardnesses after these treatments were lower than those obtained after complete solutioning and ice-brine q uenc hing. Precipitation Effects Samples of rolled bar stock of Heat 1138 were solution treated 2 hours at 2150~F. One set of samples was ice-brine quenched, one set was air cooled and one sample was water quenched after the treatment. Effect of Cooling Rate after Solution Treatment Comparison of hardnesses of samples ice-brine quenched, water quenched, and air cooled after the treatment at 21500F is included in figure 16.

16 Figure 16 also contair3optical and electron micrographs of the three conditions. Review of these data reveals that considerable precipitation of Ni3 (Al, Ti) occurs during air cooling after solution treating. Although the optical micrograph contains no evidence of Ni3 (Al, Ti) in the air cooled sample, the high hardness level and the electron micrograph show its presence. While the micrographs of the ice-brine quenched and water cooled samples do not show a precipitate, the higher hardness level of the latter sample indicates that some precipitation occurred even during water quenching. Effect of Aging Time and Temperature Two sets of samples, one set ice-brine quenched after solution treatment and the other air cooled, were aged 1, 10 and 100 hours at 10000, 1200~, 1400~, and 1600~F. Vickers penetration hardnesses for these samples are summarized in figures 17 and 18, Observations of these figures allows several general conclusions to be -drawn. 1. Aging at 1000~F increases hardness little. 2, Hardness increases progressively up to 100 hours at 12000 and 1400~F. No overaging was evident at these temperatures in 100 hours. 3, Ov/eraging occurred at 1600~F during the first 100 hours. In addition several comparisons can be made between figure 17 and figure 18 to show the effect of initial treatment before aging. 1. The difference in the initial precipitate present in the air cooled and ice-brine quenched conditions affects the aging curves at 1000~ and 1200~F. Longer aging times are needed to reach a given hardness level when the initial condition was ice-brine quenched.

17 2. At 1400~F, although the hardness after 1 hour is lower in the ice-brine quenched sample, longer aging times give approximately the same hardness level with both startinlg conditions. 3. During aging at 16000F, the samples initially ice-brine quenched resisted overa:ging to a greater extent. Representative optical micrographs of the aged samples are included in figures 19 and 20. Electron microscopic studies in the aging program have not been completed as yet. However, two representative micrographs are shown in figure 21. It was observed from the micrographs that: 1. The Ni3 (Al, Ti) precipitate was not clearly resolvable by optical microscopy after any of the aging treatments of the study. Aging did develop the precipitate so that evidence of it was seen in optical micrographs as a general greying or spotting of the background. 2. Preliminary electron micrograph studies have shown that the precipitate is resolvable after aging at or above 1000~F. 3. Significant changes in the grain boundaries occurred during aging. In general, the grain boundaries broadened with increasing temperature and time. 4. A particularly striking phenomenon occurred at the grain boundaries during aging after ice-brine quenching. This has been tentatively identified as a discontinous precipitation; it appeared to be most pronounced during aging at 1200~ and 1400'F. It should be noted that microstructural evidence of this was obtained in the weak and brittle ice-brine quenched rupture specimens mentioned in the section on heat treating variables 5. Electron micrographs of specimens aged at 1600~F showed two agglomerated phases at the grain boundaries. One of these phases eteched like Ni3 (Al, Ti).

18 Future Structural Studies Electron micrography will be completed for the aged specimens and for specimens after rupture testing. This will serve as a basis for relating properties at high temperatures to possible structural differences resulting from variations in melting practice and prior history. The latter studies are the main objective of this part of the investigation. At present it seems as if the size and distribution of the Ni3 (Al, Ti) particles may be the main cause of variation. The grain boundary precip - \ 1d during aging suggest that attention should also be directed to this factor. The available information also suggests that stability of precipitates may be a factor and should be evident in both microstructures and hardness values.

TABLE I Results of Chemical Analyses of Heats (weight percent) Vacuum Fusion Kjeldahl Heat C W0 N 2 Ti Al Mo Cr Co Si Mn S P 2 2 2 1129.05.0006.0003.013 2. 82 3. 30 4. 00 21 2 14. 9.12.13.021.003 1130.04.0007.0004.010 3. 17 3. 45 4. 00 21. 7 14. 6.14 <. 10 025.004 1133.06.0004.0008.005 3. 22 3. 35 4. 00 20. 7 14. 8.11.16.025.005 1136.05.0003.0005.007 2. 98 3.10 4. 10 18. 1 15. 2.10.10.021.006 1137.09.0003.0006.005 3. 18 3. 15 4. 20 18.2 15.1.10.10.025.008 1138.08.0012.0004.008 3. 14 3. 14 4. 15 18. 8 15. 1.10 <.10 026.008.008 (1) 1139.08.006 3. 30 3. 35 4.10 19.4 15. 0.14.15.026.006 1141.20.006 2. 98 3. 00 4. 20 19. 2 14. 5.22.15.020.007 1142. 19. 006 2.93 2. 85 4.10 19. 8 14. 5.23.13.018.004 1143. 13. 005 3. 05 3. 15 4.10 19. 8 15. 2.12.10.008.005 (1) Check analysis at second laboratory.

TABLE II Stress-Rupture Data at 25, 000 psi and l6000F Solution Treating Aged Reduction-_ Carbon Time Temperature 24 hrs at 15500F+ Rupture Time Elongation of Area Hot Working Heat (Weight Percent) (hr) (OF) 16 hrs at 14000F (hr) (percent) (percent) Procedure (1) 1129 0,05 4 1975 none 116.7 -- 6.3 (2) 1 2150 none 80.7 5.0 6.3 (2) 4 1975 none 63.9 5. 0 8.0 (3) 1 2150 none 82.3 -- 4.7 (3) 4 1975 none 73.6 8.0 5.6 (4).1 2150 none 124,1 4. 0 8.5 (4) 1130 0.04 4 1975 none 2 hr (5) 80.0 5.9 6.9 Standard 4 hr (5) 102. 2 2. 0 5. 6 Standard 24 hr (5) 23,4 13,0.13.0 Standard 2 2150 none 117, 7 8.0 3.2 Standar, 1133 0.06 4 1975 none 62,1 6,0 7.9 Standard 2 2150 none 90.7 4,0 4.0 Standard 1136 0.05 4.1975 none 79.0.12, 0 16.7 Standard 1 2150 none 99.5 6,0 4,0 Standard.1137 0.09 4 1975 none 29,3 13.0 18, Standard.1 2 150 none 87. 1 6.0 6.4 Standard

TABLE II (continued) Solution Treating Aged Reduction Carbon Timne Tempe rature 24 hrs at 1550dF+ Rupture Time Elongation of Area Hot Working Hea' (Weight Percent) (hr) (~F) 16 hrs at 1400~F (hr) (percent) (percent) Procedure (1) 1138 0.08 4 1975 none 71.7 7.0 7.9 Standard 2 2.150 none 147.4 5.0 4.6 Standard 2 2.150+ 4 1975 aged 152.9 3.0 5.6 Standard 2 2150 (5) + 4 1975 (5) aged Broke on Loading Standard 1139 0.08 4 1975 none 126.9 5,3 Standard 2 2150 none 4.0 3 1 Standard 2 2.150+ 4 1975 aged 91.9 3.0 4.0 Standard 2 2150 (6) + 4 1975 aged Broke during Machining Standard 1141 0.20 4 1975 none 28.2 7.0 9.0 Standard 2 2150 none 81.2 8.0 11.5 Standard 2 2150 + 4 1975 aged 42. 1 9.7 11.7 Standard 2 2150 aged 66.9 8.5 10.0 Standard 4 1975 none 55.9 10.0 11.0 (.) 2 2150 none 53.4 8.0 9.5 (U)

TA:~r' E II (continued) Solution Treating Aged Reduction Carbon Time Temperature 24'hrs at 15500F+ Rupture Time Elongation of Area Hot Working Heat (Weight Percent) (hr) (~F) 16 hrs at 1400~F (hr) (percent) (percent) Procedure (1' 1141 0.20 2 2150 + 4 1975 aged 36.5 8.8 11.7 (7) 1142 0.19 4 1975 none 37.7 9.0 10.1 Standard 2 2150 none 50.8 8.0 12.3 Standard 2 2150 + 4 1975 aged 52.5 10. 0 10. I Standard 2 2150 aged 85.4 8.0 8.8 Standard 4 1975 none 37.3 10.0 13.8 (7) 2 2150 none 56,2 9.0 11.7 (7) 2 2150 + 4 1975 aged 57.6 12.0 14.8 (7) 1143 0.13 4 1975 none 82.5 5.7 11.5 Standard 2 2150 none 85.5 15.0 9.3 Standard 86.5 8.5 10.0 2 2150 + 4 1975 aged 80.6 11.5 13.0 Standard Notes I through 7 on following page.

TABLE II (continued).(1) "Standard" procedure as follows: 1 hour at 23000F, surface ground, rolled at 21500F to 7/8 inch bar in 22 passes with 21 reheats with 5%o in last pass. (2) Same as standard procedure with omission of surface grinding and homogenization at 23000F. (3) Same as standard procedure with omission of surface grinding, (4) Same as standard procedure with omission of surface grinding, plus 1 hour at 23000F after completion of first 10 passes (bar size 1-3/16 inches). (5) Preheat time before applying load in rupture testing. (6) Ice-bine quenched after solution treating. (7) Same as standard procedure through first 19 passes (bar size 1 inch). Then rolled to 7/8 inch at 21500F in 2 passes with no reheats (15 percent reduction).

Meltdown Addition of Si, Al, Ti, Refining Period Superheating Deoxidation 0 Pour 3200 3100 1129, 1130 1143 ~ 1139 3000 1133 1138 2900........ 1137 k ~~~~~~~~~~~~~114 114 1136 0 2800 H 2700 1142 2600 2500 80 0 1 2 3 4 Time, hr Figure 1. - Melting cycles used in study of effect of melting variables on high temperature properties of UDIMET 500.

Actual Size Actual Size (a) Macrograph of ingot section (c) Macrograph of ingot section of Heat 1138 poured at 30000F of Heat 1141 poured at 2750~F '~:~-?'~'~' Y: a,.. ' ' ~ ~ 36_am -Ag,e - '_0. to ' '.,... i. ) a.",/ / ~,.. ~. (0. 08C) and Heat 1141(0. ZOC).

;: - working practice. ' ' I.......:.. I fr. {j *X..~:;..~...... -. Figure 3.- Surfaces of 7/8 inch bars rolled from heats melted in study of effect of melting practice. Bars were rolled with constant hotworking practice.

(a) No homogenization (b) Homogenized 1 hour at 23000F before rolling (c) Homogenized 1 hour at 23000F before rolling plus 1 hour during the rolling Figure 4. - Photographs of as-rolled 7/8 inch bars from Heat 1129 showing the influence of homogeniztion at 2300~F on the workability.

Heat Treatment after Rolling 0 Solution treated 4 hours 1975~F, air-cooled a Solution treated 1 hour 2150~F, air-cooled 200 - - 180 160 140 _.. 1 20 *100 3 80 '-, '. 60 40 20 None Before Before plus During Point of exposure to 23000F treatment Figure 5.- Influence of exposure to 2300~F for 1 hour before and during the rolling of Heat 1129 on its rupture time at 1600~F and 25,000 psi.

~i-;'~\~- -c.','> - y"L\. Z~~-.'?.'..li~i,::".2 '"'... - jQ....': '.g:, ":&I. _)\.~Z, -....q ~.,~ x~...~. '" ~/ ~II (b) NoHomogenize1horati Z0~bfr ollng.4 ~ ~ ~ ~ ( (a) Noo e a 1 hour n rolling of H a 1a ~'-I1/~i~'rLCjl{j - \\ ~x/ B Y.7: t r3 ' V't ~~~;~c ~ i~ ~~ci V~;"L",~~r \6f >" JQ- ~~~4- jI X100D (b) H omognied i htoura 30Fbfr oln pls1or duigroln Figure~~~~~~~~~~ 6. -~ Inlec f ooeiato t200Fo h icotutr of eat112 i th asroledcondition

~.w,.,,,.,I.. K...., -~ 'i., —." '': —.-:....:,.. t' \,:'~,,'...,..>....,... ' pI r r!~. ~~ ' ~I '.'" _.. ' '1 -~ ~1..' A Y,:~~~~~,.:?-' /...X. OO X100D.,.-~.'' r~ ~ ~ ~ ~ pu 1 hordrinrolg,' '..>~i'.; ~..'...':..:" '.; ~~~~~~~~.,....,.'4~~~~~~~~ 2. ', —"'...'-' '-. *. f-,,,.,'.~..... ~ i' ~ ~~~. -i.,.,,.4 ~ ~' t No homogenization Homogenized 1 hour before rolling plus i hour during rolling (b) Bars solution treated41 hours 21950F, air-cooled \i~~~~~~~ ~~~ ~~~~~~~~~~~~~~~~ ~.. ~,, Fiur 7. -. Iflenc of-d' hooeizto at 23.0 on th mcosrctr of Heat 11r at.fter sol'i- '(. '-' ' '-~ ", r. f ',~,:.. ~/~~~~~~~~~... I I~~~~.;..x' —'''.'~".', ~~~~,,,-;.. ~,. ~,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ X100D X100D No homogenization Homogenized 1 hour before roiling plus 1 hour during rolling (b) Bars solution treated 1 hours 1950~F, air-cooled Figur 7.- Infuneo hooeizto at 23 0~ on~ th tutr ofeat112 afe souiontetet

'-uuwaLU;91 uoTnloos P,94VOTUTj rD1;la~j BTsd 000 'gZ pu-e Jo009T 4e p~sa4 aan~dn~i pule uoT4.ezu a2o tuo qT 4fnoq!TM PalTTo 6ZTI VeaH jo saan4ln.so-.rTy -'8 a;.2n' T;q L. '0n9 a;ncdnU 'paIoo0-.T.e '~.o0sZ.;noq T paev;[ uot.nioS (q) cIoooTX CIOOTX au; i I *t:, ~~~~~~~~I. ~..-.' \.;!' ' " ' ~ ~.;-~~.....' ~Tq: L. "9 TT a-zundnkI 'patoo:) —.jT- '..I j/.9.6T s.aoq V, pa~,va~x4 uo~,inoS (e) C1000TX CIO0TX es:,C:,~ ';if" ~a; -~~lif:~~.:; ' $ ~ \ ~ ~~\:A~~~~~~jjjjv~~~~~~~~;~.g...;*';~i. MO VI~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I,*,IC. qfle ~. *.,t.I31,~,.: I~t, ~.... ~..,-".:,::',~i, 't;: i~: ~~~~~~~~~~~~~~~~~~~ r V~~~~~~~~~~~' ".2' '~7 ~ft ~~ '*~ ~ I,~ "'~",,t:-::.~ ~"~~:.;r t':":'":s, ~: ~* '. '~ *'".....,r:'" r~..~....". i' ~.: ~*;. I:,'.;. i-1~ ~'?. ~" _ i,..:t~i; r-t,14:r ~,..... *4* \.. 4l. I~~~f-lii: ~ ~ ~.'Y 4~~~~~~~~~~~~. 4 ~~.4. "~rI:+ ~~~~~~~~CI P' -::r~~ ariq -"~? *b- ~ ~

~i~~~~~VI O f 441,!r~ C's~~~~~~~~~~~~~~~~~" XIOOD ~~~~~X1000D (a)Soltio trate 4hours 1975~F, air-cooled, Rupture 73.46h A tw ~~;i N XNI f / 4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~o 77 44 X100D XIO00D (a) Solution treated 4 hours 21950F, air-cooled, Rupture 734. 6 hr

Heat Mode of Last 15-percentReduction 2Passes -No rehats 3 Passes- reeats 1141 -- -- 1142 0 -o 100 -- - 90 80 e 70 60 5Q 4 0 3 0 2 0 1 0 0 A B CD Heat treatment after rolling Figure 10. - Influence of mode of last 15-percent reduction by rolling on the response of two heats to the following subsequent heat treatments: A- 4 hours 19750F, air-cooled B- 2 hours 2150 0F, air-cooled D- B+ ag as in C. R i i pI I I tH-L-4-t-Y~~~,, I~ I I I T f

Heat Number Heat Number 1129 * 1139 1133 A 1141 1136 0 1142 A 1.137 V 1143 1138 200 180__ _ _ _ _ _ _ _ 160__ _ _ _ _ _ _ _ ~~~~~~140~ ~~~~ ~~ ~ ~., _l~~~~~~~~~~~~~~~~~~~ I ~ 120 ________________ ________________ S~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~/100 1 I 80 z~~~~~~~~~~~~~~~~~~~~~~~~~ 60 A B C D Heat treatment after rolling Figure 11.- Effect of heat treatment after rolling on the rupture time at 1600'F and 25, 000 psi for several heats. All heats processed by "tstandard" rolling procedure. Heat treatments as follows: A- 4 hours 1975 0F, air-cooled B-2hus25yarcoe C-B++ g 2 hur 550F lu 1 hus,40,

O VPN for Heat 1138 l I T I T T I 1 ~A~~~~ Rupture Time for Heat 1130 160,,, 380 140 II_ 120 Cl) 100 -0 0 40, 80 -- — 340 ' = 0 0 1 10 100 0 20__ _ _ __ _ _ _ 0 ____,_ ___ __ -300 0 1 ~~~~~~~~10 100 1000 Time at temperature, hr Figure 12. - Influence of time at temperature before app;1lying stress on the rupture time at 1600~F and 25, 000 psi for Heat 1130 compared with the time-hardness curve of Heat 1138 at 1600~F.

0 4 Hour Heat Treatment 360 ___ ___ rn0 2 Hour Heat Treatment 340..C ~0.~ 320...... U) "0.2 320 U)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 300......I-I 280....,... 260. '........... 1500 1600 1700 1800 1900 2000 2100 2200 2300 Heat treating temperature, ~F Figure 13. - Effect of heat treating temperature on hardness of Heat 1138. Rolled bar stock was annealed 2 hours at 2150F and ice-brine quenched prior to heat treatment. Samples were ice-brine quenched after heat treatment.

alidueS *'auou4eoal leaq ol ao.;ad joOSIZ are sanoq Z p~aUTX~uoSs deq Pq llod *8STT TeaH Jo olnlonlsoa:)WuL uo saan4-eaadUXu pawueTsap he svusauea4 4eaTq InoI{ V J~O 4 J -11:Ta.l aIooooz (p) 3oOO6T() o000Tz (p) io0 61O <TOOOTX <GOOOIX t->r At~%~: 4:... 7 -:A >./OST iN. 0Z o: CIOOOTX' '<;~:: -. j

-a:)E aam sid~u-eS 'luoauiljzo;[l 4lo o4:[o-.:d,_,o06TZ ~z snoq2 Palauu-e s'eIA >ioos aieq PlITOU '%TT 1VeH JO ~:[n4Dnacso~z:rlTI uO saaj-e~aT~ulol po4-euOiT.sop 4-e s~wauou. a t 4o:[~ aoInoq V J to3~ ~ t"4 r-'T ga2T 00ooooz (P).o006O O) aooo1'ux aoX000'Tx ~~i~~~j., w~~R,~ *: mi *W~4W~ 24 V 4 4 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 4,~~~~~~~~~~~~~~~~~~~~~ ~~1 4/4~~~~~~~~ ~~~ 44'L ) 9 4) 4> 44,T'q ")00;. (e 4444) )C100ET 4) '4 '44 ~ ',,~' All~~~~~~~~~~,' vp~

Optical Micrographs I IElectron X1000Dx13, OQOD 350/ B~~~~~~~~~~~~~ 340 - 330 320 U 310 300 Air Water Ic e-B rine Cooled Quenched Quenched Cooling rate Figure 16. - Effect of cooling rate after 2 hours at 2150'F on hardness and microstructueofHa 113 8

Aging Temperature 400 (OF)____ 0 1000 1200 1400 380 -"~360 - _ _ _ _ _ _ __ _ _ _ 0.74 340 - 320/ 300 A_____ 0 110 100 Aging time, hr Figure 17. - Effect of aging temperature and time on the hardness of Heat 1138. Treatmen rirt aging was 2 hours at 21500F, ice-brine quenched'.

Aging Temperature (oF) - 0 1000 A 1200 0 1400 1o 1600 400 M~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 380 -_ _"<_ 0 360.... (U u 340 _,, 320 ____ 0 1 10 100 Aging time, hr Figure 18.- Effect of aging temperature and time on hardness of Heat 1138. Treatment prior to aging was 2 hours at 2150~F, air-cooled.

s.lnoq Z S-ee UXU@ le;T T'eTIUI %TT H P lnfloSnaso.zI aq4 uo.Tt4eaadtui p'eU2?TSap V 2UI2V anoi- 001T JO z; JJ1 - anS aoO09 (P) JoOOW, ( ) aooo rx rrnTOO ~oOOZI (q,~,_oOOOI (~e):-000 TX (IPOOTY ~;ov - I, i;;' 4 4 I..4 =.it A.1...A.' P 3! kbso,{s 9' - j{O00ZT (q' adooi ( 00 aOOOTX rOOOTX JK i I-:?,s,.:*:?~s.

sanoi Z seM 1uaw!tUe -a4 uI '8ETT BUSH JO afTn~znj4SO.ITU DT4 UG to n4vaadtu p@4au2-sEp q4 a 4 Ue i2ue In2 O 001 JO 43oaJq -*Oz ajn3 El 0oO09T (P) 3oOO T (O) CIOOOTX aOOOTX!~ ' '.:, i t v Goooix cioooix! / / VII~ ~~7 1:;> IP- ~*. 8~~~Ao0T (q) foooT (.,).IOO, OTX CIOOO i*'t~~:0 [<) 'C^e"~ C~t~p t. a~;"'B ~:: p i

".:-':': 3 f: Figr 21,, - E ffecti of:: a on:!1 ti -:'I... t * vs;?.ii f:: S ':"-i, f: 0;,.:....".-..: -,: S; 0 X i'. y... o:i~I:::i:'::: <:: i;.. X 13, OOOD X13, QOOD (b) Aged 100 hours at 1600~F Figure 21. - Effect of aging on the microstructure of Heat 1138. Initial treatment was 2 hours at 2150~F, air-cooled. Electron micrographs.

UNIVERSITY OF MICHIGAN 3 9015 02519 72061111 11111 3II 9015 02519 7206