ENGINEERING RESEARCH INSTITUTE THE UNIVERSITY OF MICHIGAN ANN ARBOR Quarterly Progress Report No. 8 AN INVESTIGATION OF INTERGRANULAR OXIDATION IN STAINLESS STEELS AND HIGH-NICKEL ALLOYS Clarence A. Sieb'ert Maurice,J.. Sinnott Lynn H. DeSmyter Project 2110 DEPARTMENT OF THE AIR FORCE CONTRACT NO. AF 33(616)-35553 PROJECT NO. 54-670-60 WRIGHT AIR DEVELOPMENT CENTER WRIGHT-PATTERSON AIR FORCE BASE March 1956

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN ABSTRACT One heat of commercial Type-310 stainless steel and one heat of Hastelloy B material were oxidized under stress at temperatures of 1700~, 1800~, 19000, and 2000~F, and 1200~, 14000, 1600~, and 1800~F, respectively, and the severity of intergranular penetration determined microscopically. In general, the deeper penetrations increase in depth and frequency as the stress is increased. The weight gained during oxidation (without stress) was determined for the Hastelloy B material. The oxidation of this material followed the parabolic law although the rate was quite high. ii

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN INTRODUCTION This research project has been undertaken under the sponsorship of the Wright Air Development Center of the U. S. Air Force. Its objectives are: 1. To investigate the effect of temperature and stress on the progressive intergranular oxidation and to determine the threshold stress of some stainless steels and high-nickel alloys at temperatures of 1200~F and above. 2. To measure the oxidation rates of these materials by means of the weight gained during oxidation. 3. To study the subsurface oxides. MATERIAL The material used during this portion of the investigation consisted of one heat of commercial Type-310 stainless steel and one heat of Hastelloy B. The Type-310 stainless was received in the form of cold-rolled and annealed strip 1.0 x.05 in. and the Hastelloy B in the form of hot-rolled strip 1.0 x.038 in., which had received a light cold-roll pass to improve the surface and a final anneal. The analysis of the Type-310 alloy is 19.26% Ni., 25.42% Cr, 0.62% Si, 1.58% Mn, 0.05% C, remainder Fe, and the Hastelloy B material is 5.19% Fe, 28.01% Mo, 0.26% Cr, 0.45% V, 0.74% Co, 0.24% Si, 0.60% Mn, 0.03% C, remainder Ni. PROCEDURE The oxidation procedure for the stress-oxidation tests was the same as that previously used, except that the samples were not quenched, but aircooled. Since x-ray analysis of the oxides is not now a regular part of the investigation, this simplification was possible. The weight-gain method is the same as that described in the WADC TR 55-470, Part I. The metallographic and penetration-counting procedure was the same as that used in the previous investigation. In addition to the above procedure, another method of evaluating and presenting the penetration data was used during this quarter. It is known -- _ —--------------- --------------------

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN that the most uncertain types of penetrations are the extremely small ones. These represent in the main true intergranular penetration as is shown by high-magnification examination. However, some are the result of polishing artifacts or are surface-roughness effects which usually occur during oxidation. In addition to being somewhat uncertain, these small penetrations do-i-.na'ce the evaluation with respect to total number and mean depth, since they are present to a greater extent than the deeper ones. To make the effect of stress and temperature on the deeper and more significant penetrations more noticeable, it was decided to make a modified number and meandepth calculation, excluding the smaller penetration. To do this the first group, and in a few instances the first and second groups, were excluded and a modified number and mean-depth calculation made. Each group is 0.0006 in. deep. A complete explanation of the penetration-counting and calculating procedure is given in WADC TR 55-470, Part I. The penetration frequency curves which have been a regular method of reporting data also give these data, since each curve represents the distribution of penetrations of various depths. However, this new method shows the changes in the penetration character in the critical regions to a more marked extent. RESULTS AND DISCUSSION The commercial Type-310 stainless steel and Hastelloy B material have been tested at varying stresses at temperatures of 1700~, 1800~, 1900~ and 2000~F, and 1200~, 14-00~, 1600~, and 1800~F, respectively. The stress levels were chosen so as to give a representative count of intergranular penetration in the low stress regions and were concentrated in the region where the stress caused an increase in the depth of these penetrations. This discussion will be divided into two classifications, commercial Type-310, Heat 25139, and Hastelloy B, Heat B-1400. COMMERCIAL TYPE-510 ALLOY, HEAT 25159 This material was exceedingly clean in the as-received condition, had a very good surface, and showed no significant surface defects when prepared metallographically in a manner similar to that used for the oxidized specimen. The variation in the number of intergranular penetrations with depth below the metal oxide interface at constant testing temperature is given in Figs. 1 through 4. These curves follow the general decay type as described in WADC 54-120. As the temperature of testing is increased, the number of deep penetrations increases. When the stress at a given temperature is increased, the number of deep penetrations also increases. -------------------------— 2 —------------

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN Figures 5 through 12 are the penetration-number and mean-penetrationdepth graphs. The number of graphs show the effect of stress at a constant temperature on the total number of penetrations and the number of pene'trations, excluding the minor groups. The mean-depth graphs show the effect of stress on the mean depth of penetration and a modified mean depth which is obtained by excluding minor groups. In both tyTes of graphs the elongation observed in 2 in. at the end of the 100-hour test is shown. The elongation is plotted for all samples tested, although it was impossible to obtain penetration data on some of the more highly stressed samples since they contained fissures throughout the matrix. In the 17000~18000~ and 20000F tests the number decreases as the stress is i1nc.reased when. all groaups are codnsidered. However, the number of deep penetrations is not decreased when the minor groups are excluded, as is shown by the nuimber curves. In Fig. 5 (1700~F) the curve excluding the first group showed an increase in penetration number with increasing stress, while the 1800~ and 2000~F graphs showed a decrease even when the first groups were excluded, as shown in Figs. 7 and 11. Excluding the first and secon.d g-roups showed. no appreciable effect of stress on number in the 1800~ and 2000~F gaphs. Figure 9 (1900~F) shows an increase in the number of penetrations for the total number as well as the condition when the minor groups are excluded. Figures 6 and 10 show that the mean depth, considering all penetrations, increases significantly in the 1700~ and 1900~F tests and that the deeper penetrations increase even more noticeably. The 1800~F graph shows a shift in the shape of the curve as the number of groups considered is decreased. Considering all groups, the mean depth decreases slightly, while excluding the first group shows no effect of stress, and excluding the first and second groups shows a significant increase in. the mean depth. HASTELLOY B HEAT B-1400 The Hastelloy B material was tested at temperatures of 1200~, 1400~, 1600~ and 1800~F in the stress-oxidatiorn units and weight-gain equipment as described in the procedure portion of this report. It is interesting to note that the oxide scale obtained during 14000, 16000, and 1800~F tests flaked off completely during cooling. The surface after this flaking was powdery, dark blue-grey in the 14000F samples, powdery, light blue-grey in the 1600~F samples, and shiny copper color in the 1800~F samples. The powdery surface could be rubbed off, exposing a steelblue surface in each case, while the copper-colored surface was very adherent. The 1200~F tests resulted in a fairly adherent dark blue-black oxide. These colors and conditions were quite reproducible and were independent of stress. ----- 35

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN Figure 13 shows the weight gained during oxidation at constant temperatures of 1200~, 1400~, 1600~, and 1800~F. The rate at 1600~ and 1800~F is far above any previously encountered in this investigation, as the constants at 1600~ and 1800~F are 0.047 and 0.36, the latter being more than 20-fold greater than any value obtained for the stainless steels at any temperatures investigated. This factor, combined with the flaking of the oxide, would limit the use of this material in this temperature range. Figures 14 through 17 are the penetration frequency graphs giving the relationship between penetration depth and number. These curves are of the decay-curve type previously discussed. The effect of stress in the critical region (that region where the material elongates considerably) could not be determined for the 1200~F tests since the units were not constructed to carry a load greater than 200 lb. A 1/4-in. cross section was the smallest practical specimen size, and therefore the maximum stress obtainable was about 20,000 psi, which was insufficient. The testing at 1200~F was therefore discontinued when 20,000 psi showed negligible elongation at the end of the test. The 1200~, 1400~, and 1600~F graphs are similar, although there are more deep penetrations present in the 1600~F graphs. The 1800~F graph differs from these three in that the distribution is shifted in the direction of the deeper penetrations. In general, increasing the stress causes a greater number of deeper penetrations. This effect is more pronounced at higher temperatures. Figures 18 through 25 show the effect of stress on the number and mean depth of penetrations. The 1200~F tests were conducted at a stress below the critical range, as previously discussed, and therefore do not show any appreciable effect. Since all of the penetrations encountered were extremely small, the slight increase in number is not considered significant. The large amount of scatter encountered in the 1400~F curve makes the effect of stress on number difficult to determine, while the 16000F shows an incresse and the 1800~F a slight decrease in number. Excluding the first group shows an increase in the number of deep penetrations in the 1800~F graph. This exclusion is of no value in the lower-temperature tests since the penetrations are for the most part in the first group. The mean depth of penetration is not appreciably affected by stress in the 1200~, 14000, and 1600~F tests as shown in Figs. 19, 21, and 253 In the 1800~F tests, however, the mean depth is increased significantly with increasing stress. This is especially noticeable in the deeper penetrations, as is shown by the curves excluding the minor groups. It was not possible to obtain a great deal of elongation in the 1800~F tests. Some specimens which fractured during the latter part of the test showed only 3.5% elongation. The proposed metallographic analysis of this material should shed some light on this problem. ----------------------- 4 —----------

900 -- -",' 900 - m j —- 2500psi 000 a Io30psiZ Iii \ 700psi i —- 2350psi pi IiT 70 -- -2420 a 2245psi \r~ 500p 2710 a 2550psi 70 1580 a 1550psi 600- 600 ---- I mI 2710 Io 690psi oI oii.,, vC EL 400- 0040C o ii 0 ps 2000opsirn 200 - — 200 ---- - -1500,1550 a 1580psi I - 26160psi a 00: — ^ r"2385 -- 100 1065 a 1300psi ___ _ 0 \\ ~V^\ Y ^'~~~ looopsiT \03^\ \^ x I69Opsi 1'395psi 0 0.001 0.002 0.003 0.004 0.005 0.0060 0.001 0 0.00 0.003 0004 0.005 0.006 DEPTH OF PENETRATION, INCHES DEPTH OF PENETRATION, INCHES FIG. I PENETRATION VS. DEPTH BELOW SURFACE. FIG. 2 PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 STAINLESS, HT. 25139 17000F, 100 HOURS. TYPE 310 STAINLESS, HT. 25139. 18000F, 100 HOURS.Z

m z 9001 900 _______________ C'| -952ps90 r ----- 775 a l0 s 50si 650psm 800 lolopsi 800 8 Zuups2 I V......I'lhA - -~~~~~~~~~~~~ - 57 0 psi-. U.- -2000psii~?l 57 700 __000P___________________nr - 620 a 755psi m 700 700 620psi m 600 600____ 60 --- H 4 ----------— __ _ —-— 6 0_ _ _ —----- _ _ —- _ _______ _ C) i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 I1 I' zz O w _ _ _ _ - 4.500 a.-500 w Iw s ~ I _1_,_ I__ 3400 - 400 -- 200 \20C -j U.I 775ps1 It 00_______\\A200 a3 3~pi__ ________V't 300 D300 — 200 2000.001 0.002 0.003 0.00 0005 00060 0c00 -- ^- -003 0.004 0.005 0.006 DPHOF PENETRATION, INCHESpp DET F PNETATION, INCHES FIG. 3 PENETRATION VS. DEPTH BELOW SURFACE. FIG. 4 PENETRATION VS. DEPTH BELOW SURFACE.> TYPE 310 STAINLESS, HT. 25139. 1900"F. 100 HOURS.TYPE 310 STAINLESS, HT. 25139. 2000@F, 100 HOURS. Z

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN 2400 60 -1200 + 2000 - 50 -1000~ _ 0 o l l l number 0 5 z z a 1600 40 cJ 800Z w 5 ". z 3 U)\ 2 X w 0 w 4 1200 --— * 30 - 600' c:o 12C 300 30 w 0 w W T 80 —--------- 20 A 400 x eexcluding first gro up U. ~^ _/~L ]. 600 1200 1800 2400 3000: STRESS,PSI FIG. 5 SUMMARY PENETRATION-FREQUENCY CURVES. TYPE 310 ALLOY, HT. 25139, 1700~F, 100 HOURS. 1.5' 60 re 1.... 50 0o C, Xx~~~~~~~~~~~~~~~~~~~ w Xn excluding first group I z' 40 cj - Z w _ l- w 0.~5~4 -20 -1 W Z depth. ij. a 0~~~~~~~~~~~~~~~~~ii 600 1200 1800 2400 3000 STRESS,PSI FIG. 6 SUMMARY PENETRATION-DEPTH CURVES. TYPE 310 ALLOY, HT. 25139, 1700~F, 100 HOURS.

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN + 2400 60 1200 ~~~~~~~~~~~~~~z ~~~~A Z,, 80 0 I 0 400 4 1600 800 excluding first grou z I z \I Weo.~~~~~~~~~ ~o T,, 1, 120 600.0 z 0 1 5 —---------------------— d ---------— 0 00I - 0 S 2xldnw 00 1000 1500 2000 2500 4a~~00 I000 15002 2 000 0 TYPE 310 ALLOY, HT. 25139, 1800F, 100 HOURS. 3.0'....' 3...60 r 2.5 50 x uw W A Z 0 X Z 2.0 ----- 40 C "' A z ~. excluding lot and 2nd groups w I,, _o_- i __1.5 30 z I z - o 0 S!a ~~~~excluding first group elongation - w 0 z 1.0 1 200 2500 w a. z, e ~~~~~~~~~~~~~~Z~~w 0 z w~r t4 AW w a. 0.5 10 depth A 500 1000 1500 2000 2500 STRESS, PSI FIG. 8 SUMMARY PENETRATION-DEPTH CURVES. TYPE 310 ALLOY, HT. 25139, 1800~F, 100 HOURS. 8

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN 2400 60 t200M. U, a. 42000 r 150 KX)O z numb er z a 1600 10 N o800 w a) z w 0 d U) z 0 2-r, excluding I ~~0 elongation ~~~W 500 1000 1500 2000 2500 STRESS, PSI FIG. 9 SUMMARY PENETRATION-FREQUENCY CURVES. TYPE 310 ALLOY, HT. 25139, 1900~F, 100 HOURS. no F- I- - --— o 50 w, z / xcludinqfirst group z 1.0w - -- -- 40 C1 z ~ 400 I 0 -200"U I ^ ^-^' ~ / I 30I 50 z 1.0 40 N z z SUMMARY PENETRATION- CURVES. TYPE 310 ALLOY, HT. 25139, 1900~F, 00 HRS. 1-.5. 0 w1 a. FIG. 10 SUMMARY PENETRATION -DEPTH CURVES. g

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGANL X + 2400 60 r1200' 2000 50 -1000 I 0 o number | z z 160 0 N- 800 z 0. Q (L z ~ _1 excluding first group z x L~. W0-+; 600z 2 w LU. 0 ac a: 800 20 ^400 FIG.mII SUMMRY PENTRATIO FREUENCY URVES w ~E4001 Y 1 AO0 200 000 excluding irst aond 2nd grouu ^ -------------- ------- -------------- 50 3n iecluding first group z o 0 | 0. 0 Lw 0 z z 0: 2W~~~~~~~~ e~ ~~ ~~W elongation 500 1000 1500 2000 2500 STRESS,PSI FIG. 12 SUMMARY PENETRATION-DEPTH CURVES. TYPE 310 ALLOY, HT. 25139, 2000~F, 100 HOURS. ~~~10~ 0 a~~~~~~~~~~TES S ~~~I G 12 SUM R P N TAIO - E T C U R VES. ~~~~~~TP 10ALOH.2513,20~,I0HUS w 20 ~~~1

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN 1900 II I 1600 E l'l' z g0 200 1800 F/ 200 ----- --- -----!* i i i i i i liiii --— iiiiiiii -F W. 0 1000 2000 3000 4000 5000 6000 TIME- MIN FIG. 13 APPLICATION OF PARABOLIC RELATIONSHIP TO WEIGHT-GAIN DATA; HASTALLOY B, HT. B-1400. 11

900 ——. T -'- 0 —-------- |C PF z O9 |i~I -o- 1570 p=i 800 SOO ^ f ^3~O p8180 —-- I lj 1 5704960psi:! -10250 psi i I 1 0psi I~ l l U; - 7775psi m i 1000, 8900, 19250 0 psiI 10000, 118003 I11 --- -- 19250 psi i| I aB 16900psi 600 — 600CC — I i iii~~~~~~~~~~~~~~~~~i"f - I 7470 a 8650psi o I 0 Ij [|_, z "I z! S! ic- 2050 psi #! w w 5 _ a. 00 5UC I i o I, 0' w w!Il-*< 2"400 S,.C 400 o0 0~~~~0 o!D m HASTALLOY B, HT. B-1400, 1200F, 100 HOURS. HASTALLOY B. HT. B-1400, 1400~F 100 HOURS.Z

m z 900 - -" 70 900 —--- <a' 80900\ psi 3900psi Z 734 6800 psi p 1500 a 3950 psi I l —— 7340 B 7530psi 800 - -80 - -3500 a 180opo i 3000 a 3800psi Z i%;~~~~~, l!/ioo 500psi 700!'70I11 700 -:= 7800ps - 700 p m I ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~m -— 385 Opsi 600 I. 600 — ~ i —- 30o0 -,ooo \:::: o5OC - L =__ U, z Qr. 50 = t ~ w C, (/ i, i'|~^ H~~~~~~I:', IzJ a 0. -a - 400m'I I II% mc: w ~~~~~~~~~~~~~~~~w --- - 7800psi - - C "400 --- E400 = co a Z 1 200 - _ _ _ _ _ _ _ _ _:1 200 3BOOpsi'l ooI 3800psi 00 100 A\ 402Opsi'IV ~~~~~~/89~ ~~100* ~i'l/. 3500 8 4020psi 8900psi 1 8230ps 3850psi ~,-.-_ _.____- _< 950psi 0 0.001 0.002 0.003 0.004 0.005 0.006 0 0.001 0.002 0.003 0.004 0.005 0.006' DEPTH OF PENETRATION, INCHES DEPTH OF PENETRATION,INCHES FIG. Il6 PENETRATION VS. DEPTH BELOW SURFACE. FIG. 17 PENETRATION VS. DEPTH BELOW SURFACE. HASTALLOY B, HT. B-1400, 1600F, 100 HOURS. HASTALLOY B, HT. B -1400, 1800F, 100 HOURS.

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN 2400 60 2000 50 ~I I Z z z C.)z a 1600 Ci 40 mo w a. z Uw z hcr~~~~~~~~~~~~ 0 Un 120030' LLE R~number -0 0 US~~~~~~~~~~ Sw o w 400 ----------- - ------------------------------------------ 10 4 800 20 w z CD w z w a. 400 I 0 _a I - I 4000 8000 12000 16000 20000 STRESS, PSI FIG. 18 SUMMARY PENETRATION-FREQUENCY CURVES HASTALLOY B, HT. B-1400, 1200eF, 100 HOURS. 1.5, 60 50 x w w z 3t~~~~~~~~~~~~ yz I a.a~~~~~~~~~~~~~~' z w 0 __.'' _ 30 z z a. wI-~~~~~~~~ O_~~w.5 20 4 ao w HASTALLOY B, HT. B8-1400, 12000F,,00 HOURS. 4000 8000 12000 16000a2000

ENGINEERING RESEARCH INSTITUTE ~ UNIVERSITY OF MICHIGAN 2400 60 2000 50 /) w I I o 0 z z - 1600 40 CN aa w ~ U) 1200 30 ~ — Z u. 0'~' number * - IU0 ~ o 400 ------ -----—? —- - 0 _- 800__ STRESSPSI FIG. 20 SUMMARY PENETRATION -FREQUENCY CURVES. HASTALLOY B, HT. B-1400, 1400~F, 100 HOURS. 1.5 - -- -- - - - -- -- - -- i1 -- --- 60 flo -------- ------- ------- --------------— 50 0 x w I z 40- I 40 N0 ~~~~~~~~~~~~~~~~~~~~~~~i iz 1.5 60 z z o 0 IZ C)CD 4000 8000 12000 600 2000'aII0. z 1 - kW ~~~~~(A Z4000 8000 12000 16000 20000 I.B- 40 F 10 HO

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN 2400 60 2000 50 o 0 z z c 1600..... 40 N w x 1200 30Q U. 0 | 800 20 R w z STRESS,, PSI I ~ I I I 0 Ld z w 16 1.0~~~a FIG. 22 SUMMARY PENETRATION-FREQUENCY CURVES. 40~~~~~~~~~~~~~~~0 z 0n w o 30, w z _._z Is -_ I0 lm ~ tilt I I ~ i I IIII IIII IAI 2000 4000 6000 8000 10000 STRESS, PSI FIG. 25 SUMMARY PENETRATION —DEPTH CURVES. HASTALLOY B, HT. B-1400, 16000F, 100 HOURS. 16 - z~~~~~1

ENGINEERING RESEARCH INSTITUTE * UNIVERSITY OF MICHIGAN 2400 60 -600 0 2000I 0 5000 U) w 1 o oX 0 z z 0 1600 40 | 400N U z a. z 0 -J _ number Cz w 3.01 ----- - ------------- I 60 0 u )1200 t 30 300 x ~ 40. 80 0 _ —- -- -2000 z wo a. 400 I0 I00. 0 elongltion, ~f, II II- I I I 9:.3 _._:._ i K) 0 0 2000 3000 4000 5000 STRESS, PSI FIG.24 SUMMARY PENETRATION-FREQUENCY CURVES. HASTALLOY B, HT. B-1400, 1800~F, 100 HOURS 3.0' 160'o) 2.5., _ _50 0 oc x. UJ l~ Ill) Z I Z 2.0 40 cN -r I I I ~~excluding Ist and 2nd I — x' w _X 1.5 30 z excluding first roup w w 204 z 1.0 0.. - LU Z L w J 0.5 elongatiOn -000 2000 3000 4000 5000 STRESS, PSI (I) fractured during tet~ FIG. 25 SUMMARY PENETRATION-DEPTH CURVES. HASTALLOY B, HT. B-1400, 1800~F, 100 HOURS. 17