CENTRAL RESEARY LISRARY | DOCHMENT GOLLECTION ., =~ 4 OAK RIDGE NATIONAL LABORATORY operated by UNION CARBIDE CORPORATION w | NUCLEAR DIVISION for the U.S. ATOMIC ENERGY COMMISSION LOCKHEED MARTIN ENERGY RESEARCH LIBRARIES ORNL- TMP ]906 I : 3 4456 05134L7 3 IN- AND EX-REACTOR STRESS-RUPTURE PROPERTIES | OF HASTELLOY N TUBING H. E. McCoy, Jr. | J. R. Weir, Jr. OAK RIDGE NATIONAL LABORATORY CENTRAL RESEARCH LIBRARY DOCUMENT COLLECTION LIBRARY LOAN COPY DO NOT TRANSFER TO ANOTHER PERSON If you wish someone else to see this document, send in nome with document and the library will arrange a loan NOTICE This document contains information of o preliminary nature ' and was prepared primarily for internal use at the Ook Ridge Mational Laboratery. It is subject to revision or correction and therefore does not represent a final report. — — LEGAL NOTICE —— = This report was prepored as an account of Government sponsored work., Meither the United States, nof the Commission, ner any person acting on behalf of the Commission: A. Mokes any worranty or representotion, expressed or implied, with respect to the accurocy, completeness, or usefulness of the information contained in this report, or thet the use of any informotion, apparotus, methed, or process disclosed in this report moy not infringe privately owned rights; or B. Assumes ony liabilities with respect to the use of, or for damages resulting from the use of any infermotion, apparatus, method, or process disclosed in this report. As used jn the above, "‘perscn octing on behalf of the Commission’ includes ony employee or contracter of the Commission, or employee of such contractor, to the extent thot such employee or cantractar of the Commission, or employse of such contractor prepares, disseminates, or provides occess to, any informotion pursuont to his employment or controct with the Commission, | or his -mplnymenr with such contractor., e ORNL-TM-1906 Contract No. W-7405-eng-26 METALS AND CERAMICS DIVISION IN- AND EX-REACTOR STRESS-RUPTURE PROPERTIES OF HASTELLOY N TUBING H. E. McCoy, Jr., and J. R. Weir, Jr. SEPTEMBER 1967 OAK RIDGE NATTONATL, LABORATORY Cak Ridge, Tennessee operated by UNION CARBIDE CORPORATION for the U.S. ATOMIC ENERGY COMMISSION LOCKHEED MARTIN ENERGY RESEARCH LIBRARIES 3 4456 051341L7 3 iii CONTENT'S Page Abstract . 1 Introduction . 1 Experimental Details . 3 Test Materials . 3 Test Specimens . 5 Testing Techniques . 5 Experimental Results . . . . « . « . « « v ¢ v v v v v v v v v o« .. 10 Discussion of Results . . . . . . . . . « . « « + o v v v v oo 21 Summary and Concluslons . . . . ¢ . . . i v v e e e e e e e e e 24 Acknowledgments . . « . . . .t it e e e e e e e e e e e e e e 24 IN- AND EX-REACTOR STRESS~-RUPTURE PROPERTIES OF HASTELLOY N TUBING H. E. McCoy, Jr. and J. R. Weir, Jr. ABSTRACT The stress-rupture properties of two heats of Hastelloy N tubing have been determined at 760°C in- and ex-reactor. ITrradiation reduced the rupture life and the rupture strain, but no effects on the creep rate were detectable. Small variations in behavior of tubular specimens tested during irradiation and small rod specimens tested after irradiation are explained on the basis of differences in stress states and sizes of test sections. The effects of irradiation are rationalized on the basis of the behavior of helium which occurs in the metal as a result of the thermal 10B(n,Od) transformation. INTRODUCTION Numerous cases have been reported where the high-temperature mechan- ical properties of nickel- and iron-base alloys deteriorated under neutron 1-9 irradiation. This deterioration manifests itself as both a reduction 1D. R. Harries, "Neutron Irradiation Embrittlement of Austenitic Stainless Steels and Nickel Base Alloys," J. Brit. Nucl. Energy Soc. 3, 74 (1966) . - °G. H. Broomfield, D. R. Harries, and A. C. Roberts, "Neutron Irra- diation Effects in Austenitic Stainless Steels and a Nimonic Alloy," J. Iron Steel Inst. (London) 203, 502 (1965). >N. A. Hughes and J. Caley, J. Nucl. Mater. 10, 60 (1963). “F. C. Robertshaw et al., Am. Soc. Testing Mater. Spec. Tech. Publ. 341, 372 (1963). 5N. E. Hinkle, Am. Soc. Testing Mater. Spec. Tech. Publ. 341, 344 (1963) . = ®P.C.L. Pfeil and D. R. Harries, "Effects of Irradiation in Austenitic Steels and Other High-Temperature Alloys,"” Am. Soc. Testing Mater. Spec. Tech. Publ. 380, 202 (1965). 7W. R. Martin and J. R. Weir, "The Effect of Irradiation Temperature in the creep-rupture life and in the rupture ductility. This effect has been correlated with the thermal neutron dose and has been attributed to the helium produced by the transmutation of 1°B to 7Ii and %He.®,9,10 However, the sensitivity of different materials to helium produced by this process varies greatlyl and our present state of understanding of this problem necessitates that we study each material individually. We have studied Hastelloy N, an alloy developed at Oak Ridge National Laboratory specifically for use with molten fluoride salts.?l Tt is nickel base, solid solution strengthened with about 16% Mo, and contains 7% Cr for moderate oxldation resistance. However, this alloy has become a candidate material for use in several other reactors and we are attempting to learn as much as possible about this material in nuclear environments. Recent studies have shown that the mechanical properties of this material are indeed altered by irradiation. 12,13 In this study we compared the in- and ex-reactor properties of two vacuum-melted heats of Hastelloy N tubing at 760°C. The testing techniques used in this program will be described and the resulting test data pre- sented. These data will be compared with those obtained for the same material in uniaxial postirradiation stress-rupture tests. on the Post-Irradiation Stress-Strain Behavior of Stainless Steel," Am. Soc. Testing Mater. Spec. Tech. Publ. 380, 251 (1965). 8J. T. Venard and J. R. Weir, "In-Reactor Stress-Rupture Properties of a 20 Cr—25 Ni Columbium-Stabilized Stainless Steel,” Am. Soc. Testing Mater. Spec. Tech. Publ. 380, 269 (1965). °P.C.L. Pfeil, P. J. Barton, D. R. Arkell, Trans. Am. Nucl. Soc. 8, 120 (1965). = 10p,R.B. Higgins and A. C. Roberts, Nature 206, 1249 (1965). W, D. Manly et al., "Metallurgical Problems in Molten Fluoride Systems," Progress in Nuclear Energy, 2 (IV), Technology, Engineering, and Safety, pp. 164~79, Pergamon Press (1960). 12y, R. Martin and J. R. Weir, "Effect of Elevated Temperature Irra-~ diation on the Strength and Ductility of the Nickel-Base Alloy, Hastelloy N," Nucl. Appl. 1(2), 160-67 (1965). 13y. R. Martin and J. R. Weir, "Postirradiation Creep and Stress Fupture of Hastelloy N," Nucl. Appl. 3, 167 (1967). EXPERIMENTAL. DETAILS Test Materials The two lots of material used in this study were 10,000 1b vacuum- melted heats obtained from Allvac Metals Company. This material was con- verted to tubing by Superior Tube and had a nominal inner diameter of 0.540 in. Heat 5911 was initially fabricated with a 0.015 in. wall, but the same material was redrawn (designated 5911R) to obtain a 0.010 in. wall., Heat 281-4-0143 was fabricated into tubing with a 0.010 in. wall. The working schedule used to manufacture the tubing is proprietary and details are not available. The chemical analysis of the fabricated mate- rial is given in Table 1. A typlcal cross section of the as-recelved tubing (Heat 281-4-0143) is shown in Fig. 1. This tubing was coated for a specific application. Neither the coating composition nor the technique for i1ts application to the metal can be made available, but temperatures as high as 1150°C were encountered in the processing. The microstructure Table 1. Chemical Analysis of Test Material Content, wt % Element Heat Number Heat Number 281-4-0143 5911 Pe O . 28 O ¢ 03 Cr 7.00 6.14 Mo 16.88 17.01 Ni Ral Bal C 0.05 0.056 Mn 0.50 0.21 B 0.0006 0.0010 5 0.009 0.002 o 0. 002 0.002 gi 0.28 0.05 cu 0.03 <0.01 Co 0.01 0.04 Al 0.20 0.15 T 0. 04 0.067 W 0.01 0.01 0 0.0006 0.0014 N 0.0025 <0.0005 H 0. 0C04 I Y-731 14 Y |'>r. | X | i - | e | - | | ! I . i 5 ‘ - § - & . - 4 5 .l wJ v L= - - ; - | 3 Ll ¥ -+ - v ) - - oy E = - 5 * = e = - - L ' & . o . - 4 * . " | - " ¢ '\ . ¥ . - & - o - ” o C - & o, . . o T . s ik & - . - s " ~ d o ' » -2 - & Q 0 < » 2 - o . 5 - ¥ - L oo t. - * ® - o™ < - ' O. o s : . & i - o e , - - - L 1 - - - 2 8o - o = - & Ra . » 4 o4 - e - - ¢ 3 . 0e - - ’ ~ o - e - ew Q Ll Fig. 1. Photomicrographs of As-Received Hastelloy N Tubing-Heat 281-4-0143. Etchant: glyceria regia. of the tubing before it was tested is shown in Fig. 2. Because of the brittle nature of the coating, it probably exerted little influence on the properties of the tubing. Test Specimens A drawing of the test specimens used in this study is shown in Fig. 3. The assemblies were prepared by Atomics International(AT) and shipped to ORNL for testing. Testing Techniques The apparatus used in ex-reactor tube burst tests is shown schemat- ically in Fig. 4. Because of the relatively weak welds at both ends of the tubes, only the center 3 in. section was heated. The furnace con- sisted of three 1 in. zones with a thermocouple located on the specimen at the center of each zone for monitoring the temperature. A single pro- portioning controller was used with three variable power supplies for controlling the temperature. The controller received its signal from the thermocouple on the center zone and the variable power supplies were adjusted to obtain a uniform temperature over nearly the entire center 3 in. of the specimen. The specimens were pressured with argon and the external environment was air. After the pressure was adjusted manually, the specimen was isolated from the gas supply. TFailure was detected by a reduction in the system pressure. This pressure change actuated a switch that cut off a timer. Thus, failure was detected when the first crack penetrated the tube wall. The in-reactor experiments were run in a similar fashion although the equipment was somewhat more complicated. TFigure 5 shows a schematic diagram of the test equipment. The purge gas in these experiments was He—1 vol % O2. An experiment in two stages of completion is shown in Figs. 6 and 7. The entire experiment is built on the framework shown in Fig. 6. The sides of the can are welded on to obtain an integral unit that can be immersed in the poolside of the ORR. The furnaces on these specimens are similar to those used ex-reactor, the primary difference Y-73126 W' Y-73128 | y Ve { _-""1-\.—._._ o — T I,r‘ —— o . — . 5 S - . . - - . - - o / - e = i . - i . . I / rd = & Fig. 2. Photomicrographs of Processed Hastelloy N Tubing-Heat 281-4-0143, BEtchant: glyceria r CRNL-DWG 67-4935 /HASTELLOY N Sy CERAMIC BARRIER 0.5614-in. DIAM A 304 STAINLESS STEEL TUBE Yg-in. OD X0.034-in. WALL/ - 4.500 in. Fig. 3. Schematic Drawing of Test Specimen. ORNL-DWG 67-4934 PRESSURE RELEASE PRESSURE GAGE X PRESSURE SWITCH PRESSURE P SOURCE ° o TIMER o 0 o o=—— FURNACE 0 0 Q o SPECIMEN Fig. 4. ©Schematic Diagram of Apparatus Used for Ex-Reactor Tube Burst Tests. UMCLASSIFIED ORNL-DWG 64-1433 OPERATIONS BALCONY ALARM CIRCUIT OFF -GAS _______ POOLSIDE T[] JUNGTION | | BOX "¢" LA PRESSURIZING ATMOSPHERE PATCH PANEL TUBE (TYPICAL 10 EACH ) ' PRESSURIZING ~{_JCOMPUTRAN TUBE— 10 EACH /MANIFOLD MIXED GAS SUPPLY | HIGH PRESSURE HIGH PURITY GAS EXPERIMENT — PRESSURE CAN CELL VENT Fig. 5. BSchematic Diagram of Apparatus for Running In-Reactor Tube Burst Tests. Phato 85309 Fig. 6. Partially Assembled In-Reactor Experiment with the Tube Burst Specimens Mounted in Place. Photo 85585 e A= —— T ‘E:Ebr IM —— =S i L R Y m“rfl“m ST e ““H"“m|1|4n|l|.|||q|,|- 'u“ L TR e J?h " . .MJmhhijflJm; Fig. 7. In-Reactor Tube Burst Experiment Assembled Except for Cutside Container. being that each zone has a separate contreller. The heaters are only sufficient to maintain the test temperature (760°C) while the reactor is in operation. The three in-reactor experiments which were run in this series of tests utilized positions P-5 and P-6 in the ORR poolside b nave very similar flux levels. Cobalt-doped type 302 stainless steel monitors were used. The thermal flux was obtained from the 5gflo(n,y)5000 reaction and the fast flux (> 4 Mev) was obtained from the “Fe(n,p)°“Mn transmutation. The average values over the length of the test specimens were 4 x 1013 neutrons cm~? sec-! thermal and 3 x 10%® neutrons em-2 sec~? > 4 Mev The tubes were measured with a profilometer at AT before and after testing. The ex-reactor specimens were measured at ORNL with a microm- eter and the in-reactor specimens were measured with an optical com- parator. The techni es I . used by both installations were in reasonably good agreement, but the ORNL measurements have been used in all graphs. The tubes were somewhat oval before testing and this made it difficult 10 to obtain accurate measurements of strain at the strains of 1% or less. The tangential stresses were calculated from the standard thin-wall formula - Fa 9 T 2t (1) where Oy = tangential stress, P = internal gas pressure, d = outside diameter, and t = wall thickness. The wall thickness, t, is probably the largest source of error in this calculation since it varied by up to 0.0002 in. (£2%). EXPERTMENTAL, RESULTS The test results obtained on the three lots of material are summarized in Tables 2 and 3. These same data are plotted in Fig. 8 as the logarithm of the tangential stress (g ) versus the rupture life. Within the accuracy of the data, there do not sgem to be any differences between the lots of material. Hence, we can consider the data as two sets — irradiated and unirradiated. The large effects of small doses on the rupture life are somewhat surprising. For example, in test 5034R where the stress was 26,000 psi there was a reduction in rupture life of about an order of magnitude although the dose at failure was only 7 x 1017 neutrons/cm2. The bulk helium content in specimen 5034R (Table 3) at the time of its failure was only about 35 ppb. The irradiated and unirradiated curves are approximately parallel. The tangential rupture strains are compared in Fig. 9 as a function of rupture life. Most of the unirradiated specimens have fracture strains from 7 to 10% although there are a few with strains as low as 4 to 5%. The irradiated specimens exhibit tangential strains from a few tenths of a percent to about 2%, The trend seems to be that of increasing fracture strain with increasing rupture life (or decreasing stress). There are two points for irradiated specimens that appear to be anomalous. 11 Table 2. Results on Ex-Reactor Tube Burst Test at 760°C Hoop or Rupture Tangential Tangential Test Specimen Tangential Life Rupture Strain Number Number Stress (hr) Strain Rate (psi) (%) (%/nhr) Heat Number 281-4-0143 5918 5069% 50, 400 ~ 0.1 17.2 5929 5057 35,900 2.2 6.7 3.1 6174 5097 29, 800 4.2 7.7 1.8 5927 5068 24,100 12.7 4.6 0.36 6173 5098 21,100 15.5 9.6 .62 5922 5048 18,600 63.7 8.9 0.14 5936 5072 14,900 153.2 8.2 0.053 5932 5062 12,400 788.5 11.1 0.014 5983 50’78b 9,930 2160.0 10.6 0.0049 6291 5088 14,900 140.0 9.3 C.066 Heat Number 5911 5919 6025 38,200 0.9 8.4 9.4 5933 6029 27,100 5.0 3.9 0.78 5921 6019 19,400 30.3 9.4 0.31 5925 6020° 14, 500 61.2 0.8 0.014 5935 6026 14,500 343.5 9.0 0.026 6193 6018 14,500 100.2 6.4 0.064 6139 6014 11,600 334.7 7.6 0.023 5931 6031 9,690 2974.5 4.6 0.0016 Heat Number 5911R 5928 5127R 36,600 0.4 .1 23 5996 5126R 31,400 5.85 4,4 0.75 5916 5057R 26,300 19.8 8.2 0.42 5923 5049R 19,700 87.5 8.9 0.10 5024 5050R 19,700 86.9 8.2 0.095 5984 5130R 15,900 142 .4 6.2 0.044 5926 5128R 13,200 372.5 7.8 0.021 5997 5123Rb 10, 500 1731.7 6.8 0.0039 6292 5124R 15,900 182.8 4o 0.024 STube split. Ppged 2300 hr in argon at 760°C. CTube not fractured; leak in system. Table 3. Results of In-Reactor Tube Burst Tests at 760°C Hocp or Tangential b Thermal Neutron Location of Tube Tangential Strain, % Rupture Dose at Helium Strain Fracture from Experiment Number Stress - Life Rupture Produced Rate Fnd of Specimen Number ; (hr) 2 (at. fraction) (%/hr) P ORR (psi) ORNL AT (neutrons/cm?) (in.) Heat Number 5911 6002 19,400 0.65 0.35 8.2 1.86 x 1018 9.0 x 1078 0.079 3 172 6003 15, 500 0.51 0.65 22.8 3.28 y 1018 1.6 x 10-7 0.022 21/8 172 6004 13,550 0.86 0.72 46.0 6.62 x 1018 3.2 x 1077 0.019 2 172 6005 11, 600 0.93 1.01 133.1 1.92 x 10%? 9.0 x 1077 0,0070 1 3/4 172 6006 9,700 1.33 1.41 244.3 3.52 x 10%? 1.6 x 1076 0.0055 17/8 172 6007 8,730 0.96 1.8 572 8.23 x 101° 3.4 x 107 0.0017 3 172 6008 7,750 1.33 1.54 632 9,10 x 101° 3.7 x 10-8 0.0021 11/2 172 Heat Number 5911R 5034R 26,200 0.62 0.67 0.7 7.3 x 1017 3.5 x 1078 0.89 2 3/4 171 5036R 20,900 0.21 1.33 4.7 1.3 x 1018 6.1 x 108 0.045 3 171 5037R 15,700 0.84 1.60 25.5 4.3 x 1018 2.1 x 1077 0.033 21/2 171 5046R 4 13,100 1.90 2.01 83.2 1.3 x 1019 6.2 x 1077 0.023 1 3/4 171 5039R4 11,800 0.37 0.31 109.5 1.6 x 10%° 7.5 x 1077 0.0034 3 171 5040R 10, 500 6.3 3.88 145.8 2.2 x 101° 1.0 x 10-° 0.043 2 171 Heat Number 281-4-0143 4504 24,900 0.39 0.68 1.25 8.4 x 10%7 2.2 x 1078 0.31 2 171 4550 19,900 4.5 2.64 2.25 9.7 x 107 2.6 x 1078 2.0 3 171 4588 17,400 0.11 0.52 6.7 1.6 x 1018 4.2 x 1078 0.017 2 3/4 171 4596 14,900 0.53 0.40 5.7 1.4 x 108 3.7 x 1078 0.093 2 171 4602 12,400 0.64 0.60 56.3 8.8 x 1018 2.3 x 1077 0.011 3 172 4616 11,200 0.43 0.69 105.9 1.6 x 1019 4.1 x 10-7 0.0041 21/2 172 4621 9,790 1.00 0.87 88.3 1.3 x 101° 3.4 % 1077 0.0113 3 172 50714 10,000 2.1 1.95 409.6 5.9 x 10%9 1.4 x 107° 0.0051 1 3/4 178 4639 9,000 2.8 2.1 505.1 7.4 x 1019 1.7 x 107° 0.0055 1 3/4 178 5090 g, 500 1.7 1.37 930 1.3 x 1020 2.8 x 1076 0.0018 1 3/4 178 5053¢ 8,000 1.5 1.85 1125 1.6 x 1020 3.1 x 1076 0.0013 178 50945 7,000 0.9 0.48 1125 1.6 x 10%0 3.1 x 1078 0.0008 178 50667 12,000 1.2 0.59 40.6 1.4 x 102 2.9 x 10-6 0.030 1 3/4 178 ) = 4 x 1012 neutrons cm—? sec™ ¥, o' Reactor was at power about 5 hr before the specimens were stressed. o4 Measured from end where pressurizing tube attached. Split. o °Did not fail during reactor cycle. -fHeld at temperature for 950 hr before applylng stress. ct TANGENTIAL STRESS (1000 psi) -~ Fig. 8. 60 04 1 10 100 RUPTURE LIFE (hr) ORNL—DWG 67-3534 HEAT NO. 5811R 5911 281-4- 0143 IN-PILE EX-PILE e < 4 A . a IRRAD. CONDITIONS $=ax10" oy T =760°C 1000 10,000 Stress-Rupture Properties of Hastelloy N Tubes at 760°C. (The irradiated tubes were exposed for about 5 hrs before the stress was applied. ) - Fig. 9. ORNL-DWG 67-4936 12 | R v T 11T ] T ] IN-PILE EX-PILE HEAT NO. ! . o 5¢11R f 1 & A 5ei g n o 28t-4-0143 IRRAD. CONDITIONS | 10— ¢=ax10"mw i 7=750°C ° | - o | Q “ 0 [aY Al © o 8 i { o] 0 A 3 ! -7 : : : x A 5—) 8(?) ., 6 < - = o | =5 T - l 3 | - (?). l a lL T 3 D L P . » . " h ! | A A f { | ' o [a[[F] ® 4 . A m " .' 4 1 0 s | i Qf { 10 {00 1000 {0,000 RUPTURE LIFE {hr] Tangential Strains for Hastelloy N Tubes Tested at 760°C. 14 Another interesting correlation is shown in Fig. 10. If the strain at fracture is divided by the time to fracture, a creep rate is obtained. At 760°C Hastelloy N exhibits little, if any, primary creep. TFor a thin- walled tube with the test being terminated when the first leak occurs, there would probably be no tertiary creep. Hence, the creep rate obtained is very close to the minimum creep rate that would be obtained from a standard creep curve. The creep rates obtained in this manner are listed in Tables 2 and 3 and plotted in Fig. 10. The data scatter about a common line, independent of whether or not the specimens were irradiated. Hence, the minimum creep rate does not appear to be influenced by irradiation. 40 ORNL-DWG 67—-3533R ? ° | IN-PILE EX-PILE HEAT NO. . o 5914R A A 591 u o 281-4-04143 IRRAD, CONDITIONS $=4x10° av 7 =760°C TANGENTIAL STRESS (1000 psi) 000t 0.01 04 1 10 MINIMUM CREEP RATE (% / hr} Fig. 10. Variation of Minimum Creep Rate with Stress Level for Hastelloy N Tubes. Figure 11 shows the random nature of the failures in the irradiated specimens. There appear to be no systematic problems with temperature control nor any marked influence on rupture life due to the range of dose received by each specimen. Two specimens were aged prior to testing in an effort to determine whether thermal treatment produced any deleterious effects on the proper- ties. These specimens were aged for 2300 hr at 760°C in argon and then tested at 760°C. The data, shown in Table 2, indicate no severe effects. However, the slight ductility reduction of heat 5911R (Test 6292) may be y / 7 6 2" OE)E) /72 -6 175 . 2t (2t 2" 12" e 12" 7 4" ol S g\ g g g 5 g S 2 3 © © © © 2 2 3 + < < / PRESSURIZING TUBE ! | / ’ 7 Pl pl[|2 pli[3 Pl||a Pl |5 PG Pl |7 pll|8 Pl |o P10 ’ I 3 / B o bl Ll bbb oo b oot ob oo - T P 2 2 194 0 | ‘ " /, e L / FURNACE AREA / o / 1Yo L Lol bk / MOUNTING PLATE .020 S.S. / 2 o] L] i et . Fig. 11. REACTOR FACE AT s e VI ////////////////////////W/MW/////W// JI,J/4 S.S. TUBE Schematic Drawing of Experiment ORR-172 Showing the Locations of the Failures. ¢T 16 real and should be studied further before using this particular lot of tubing. A subject of importance is a comparison of the behavior of the tubular specimens with that of wrought specimens of the same material. Specimens of heat 5911 were tested as both tubes and small rod specimens with a gage section of 1 x 1/8 in. diameter. The details of the work on the rod specimens were reported previously.14 The heat treatments of the rods and the tubes differed slightly, but probably not enough to be significant. The small rod specimens were irradiated to a thermal dose of 2.3 x 1020 neutrons/cm2 at 760°C and subjected to postirradiation creep testing at the same temperature. The differences in stress state between the tubes and the rods must be considered. Weil EE.Ei'ls showed that the effects of end restraint on the properties of tubes are negligible for length to diameter ratios greater than about two for a material with a strain hardening exponent of about 0.3, Since our tubes had a length to diameter ratio of about eight and were very thin walled, they are assumed to have been exposed to a two-dimensional stress described by: . Pa I, (axial stress) = T o, (tangential stress) = Pa 0 2 O (radial stress) = QO , (2) A comprehensive study of stress state was made by Kennedy et g&.ls on Inconel at 816°C. This work will be drawn on extensively to predict 14H, E. McCoy and J. R. Weir, An Evaluation of the Effects of Irra- diation on the Mechanical Properties of Two Vacuum Melted Heats of Hastelloy N (to be published). 15N. A. Weil, M. A. Salmon, and C. J. Costantino, "Approximate Burst Strength of Thin-Walled Cylinders with Hemispherical Caps," ATAA (Am. Inst. Aeron. Astronaut.) J. 1, 2088 (1963). 160, R. Kennedy, "The Effect of Stress State on High-Temperature Low-Cycle Fatigue," pp. 92-107 in Symposium on Fatigue Tests of Aircraft Structures: Low-Cycle, Full-Scale, and Helicopters, Am. Soc. Testing Mater. Spec. Tech., Publ. 338 (1963). 17 the differences in properties under uniaxial tension and the two-dimensional stress state of a thin-walled tube. Kennedy showed that the time to failure, tr’ was predicted under various stress states by the relationship - o where B and B material constants that were obtained from the uniaxial stress-rupture curve (B = 37,500 psi and B = 5 for Hastelloy N), 0, = maximum principal stress, and T = effective stress. Based on the von Mises (distortion energy) criterion, 5 === [0, ~ 0g)2 + (0 ~ )2 + (o — 0,)21Y/2 . (4) /2 For the internally pressurized tube % = %% o, = 1/2 o4 (5) op = 0 (5) With the same maximum principal stress, the ratic of the rupture life of the rod, ti, to that of the tube, tg, is given by = 0.56 . (6) ct ct HI-EIIHDU Figure 12 compares the rupture lives of the uniaxially stressed rods and the biaxially stressed tubes. In the unirradiated condition the rupture ORNL -DWG 67-4937 (I | HEAT 591 40,000 | 760 °C 1 Pl .’5 & w ® P N : ! E 20,000 , T T = i i T - \ = I H i i T "N Q o ~ £ RUPTURE , UNIRRADIATED TUBES =1~ N TS . Ty "~ & 40,000 | -———=—- RUPTURE, UNIRRADIATED RODS _i—_ ™ \\\\ = " ——-—— 10% STRAIN UNIRRADIATED RODS 3 = g [ —---— RUPTURE, IRRADIATED TUBES e g —— —— RUPTURE, IRRADIATED RODS L T 5000 l IR P Lol : ‘x | ‘ T | 0l { 10 1000 10,000 TIME {hr) Fig. 12. A Comparison of the Creep-Rupture Properties of Hastelloy N Tube and Rods at 760°C. lives of the rod specimens were greater than those of the tubes. This is probably due to the different methods used to determine failure. Failure of the tubes was indicated by a drop in the internal pressure. This would occur when the first crack penetrated the thin tube wall and tangential strains of only 7 to 1L0% were noted at failure. The rods were strained until the specimen completely parted and strains of 30 to 40% were noted. Thus it is probably more accurate toc compare the lives of the two test specimen configurations at similar strains. The curve for the time to 10% strain for the rods is also shown in Fig. 12. This curve has a slightly different slope than that for the tube data, but at low stresses the rods reach a strain of 10% in about one half the time for the tube to fail. Thus, with similar fracture criterion for the two types of specimens, Eq. (3) seems to predict the behavior reasonably well. The irradiated rods and tubes both failed at low strains. The tubes were stressed and irradiated simultaneously and about 100 hr were required 019 neutrons/cm?. The rods were to accumulate thermal doses of 1 x 1 tested after they had been irradiated to a dose of 2,3 x 1020 neutrons/cmz. Thus the longer rupture lives of the tubes at high stresses is probably due to their lower dose. At low stresses the rupture lives of the tubes and rods are similar. It is quite likely that the irradiated specimens are brittle enough that their failure time is governed solely by the maximum principal stress; hence the tubes and rods would be expected to fail in equivalent times. Equation (3), which is based primarily on a * 19 shear-stress criterion and depends to a lesser extent on the maximum principal stress, may not strictly apply to the irradiated specimens. The variation of creep rate with stress state is given by é =é0 --GR+ 9_{) Z FL°Z 2 . _er %R T % e T 5% 2’ —_ On + G - _ € 8 Z—f R FL9R ~ > (7) (8) where A, @ are material constants determined from uniaxial data. For Hastelloy N at 760°C the values of & and A were 5 and 30,000 psi, respec- tively. By substituting the appropriate values into q. (7), we can relate the axial stain rate for the rod éR with the tangential strain rate of the tube ég. For the same maximum principal stress Yo = 2.4 . (9) mo‘ (M e D K3 Figure 13 shows a plot of the minimum creep rate as a function of the maximum principal stress. The straln rates appear to be equal for the two specimen geometries, indicating that the ratio in Eq. (9) is approximately 1. However, reference to Fig. 10 shows that the scatter in the minimum creep rate data is greater than a factor of 2.4 at a given stress. The parameter of prime importance in this study is the fracture strain. In general, the strain at fracture eF i1s given by F € = ¢t . (10) By combining Egs. (7), (3), and (10) the axial fracture strain for the 20 ORNL-DWG 67-4939 40,000 20,000 RADIATED AND DIATED TUBES HEAT 591t © UNIRRADIATED RODS ® IRRADIATED RODS 10,000 MAXIMUM PRINCIPAL STRESS (psi) 5000 Q.000t 0.001 0.0t Ot ! 10 MINIMUM CREEP RATE (% /hr) Fig. 13. Comparison of the Creep Rates of Hastelloy N Tubes and Rods at 760°C. rod, eR’F, and the tangential fracture strain for the tube, eg’F, can be related for the same maximum principal stress. £ = 1.33 . (1) Figure 14 compares the fracture strains of rods and tubes. Again, the scatter in the data is greater than the predicted variation. One complicating factor that has not been considered in comparing the behavior of tubes under a biaxial stress and rods under a uniaxial stress is that the stresses change differently with strain. In both types of tests the stress is based on the initial dimensions (engineering stress) and the true stress actually increases during the test. In the uniaxial case the true stress (up to necking) ¢’ is given by ’ g = O(l+€)‘- (12) In the biaxial case the true stress is given approximately by R 2t , Pd =7z (1+e) . (13) ] (1+e) 2 1 21 ORNL—-DWG 67—6640 30 ‘ ‘ J ¢ [RRADIATED TUBES (TANGENTIAL STRAIN) ©IRRADIATED RODS 10 —ficr-.fi— (AXIAL STRAIN) MAXIMUM PRINCIPAL STRESS (1000 psi) 0 1 2 3 4 S 6 7 RUPTURE STRAIN (%) Fig. 14. Comparison of the Fracture Strains for Irradiated Hastelloy N Tubes and Rods at 760°C. Hence the tangential stress increases more rapidly with strain than does the axial stress and the stress state changes with strain. This change in stress state complicates the comparison of the two types of tests. DISCUSSION OF RESULTS The data indicate an effect of neutron irradiation characterized by (a) a reduction in stress-rupture life, (b) a reduction in rupture ductility with a trend of increasing fracture strain with increasing rupture life (or decreasing stress), (c) no change in creep rate, and (d) a very low neutron dose necessary to produce these property changes. These charac- teristics are consistent with our understanding of how helium would affect the properties of a material once it was introduced by the trans- mutation of 19B. At least two excellent papers have been written which deal with the behavior of helium in metals and we shall not discuss this 22 subject in detail.17,18 mhe loB, because of its size and low solubility, is initially segregated at the grain boundaries of the metal.l’ As the 10B is transmuted, the recoil range of the helium is sbout 2 u (ref. 19) and, hence, most of the helium will lie in the proximity of the boundaries. The formation of intergranular voids under creep conditions is a naturally occurring phencmenon in most metals. 20724 Several mechanisms have been suggested for their nucleation,?® but they are generally assumed to grow by the diffusion of vacancies into the voids. The voids are thermodynamically stable?® when 2Y 9= r (cos 0)° where: = applied stress, = gurface tension, = radius of the void, and o B < Q | = angle between the applied stress and the normal to the plane of the boundary in which the void lies. Thus, the surface tension provides a driving force for the void to collapse. If a stress is applied that is large enough to balance the surface tension, 17D. R. Harries, "Neutron Irradiation Embrittlement of Austenitic Stainless Steels and Nickel Base Alloys," J. Brit. Nucl. Energy Soc. 5, 74 (1966). = 188, Russell, "Inert Gas Bubbles in Irradiated Solids," J. Australian Inst. Metals 11, 10 (1966). H. P. Meyers, Aktiebolaget Atomenergi (Sweden), Report AE-53 (May 1961). 7. N. Greemwood, D. R. Miller, and J. W. Suiter, Acta Met. 2, 250 (1954) . —_— = 21p. Mclean, J. Inst. Metals 85, 468 (1956-57). 2. Shinoda, T. Sano, and T. Sakurai, J. Japan Inst. Metals (Sendai) 24, 818 (1950). W. Davies and B. Wilshire, Trans. Met. Soc. AIME 221, 1265 (1961). Rhines and P. J. Wray, Trans. Am. Soc. Metals EZT 117 (1961). Davies, J. Inst. Metals 87, 119 (1958-59). o Balluffi end L. L. Seigle, Acta Met. 5, 449 (1957). = ® B 23P. 24F. 25P- 26p, 23 the vold will be stable. If an even larger stress is applied, the bubble will grow as the supply of vacancies allows. The presence of helium com- plicates this process in at least two ways. The helium atoms can agglomer- ate and serve as the void nucleus. The helium will also produce some internal pressure in the void and thus reduce the stress that must be applied to stabilize the void. 1In this case the equation above becomes _ 2y (0 +P) = r (cos 82 Thus, the mechanism of creep fracture is not altered by the presence of helium. If the stress is low and the temperature high, failure will occur by the linking up of the intergranular voids. The strain in this case will be a/d, where a is the void spacing and d is the grain diameter.?’ At higher stresses and lower temperatures, bulk deformation predominates and the intergranular void formation process will become of less impor- tance. The presence of helium will help nucleate and stabilize the voids so that their growth will be important at higher stresses and lower temperatures than normal. The effects of helium content on the somewhat simplified process Just described are not clear. The expected trend would be that increasing concentrations would help nucleate and stabilize more voids. Since the growth of voids under creep conditions is a natural phenomenon, it is difficult to determine the minimum quantity of helium necessary to be effective and the maximum quantity above which saturation of the effect should ocecur. However, the data in the present study showed that large effects were noted when the helium content was a few parts per billion (atomic) and that these effects did not seem to increase greatly as the helium content increased to a few parts per million (atomic). The observation of increasing fracture ductility with decreasing strain rate (decreasing stress) is quite interesting and has been obser- ved for several lots of Hastelloy N. This can be rationalized from the fact that the size of void that will grow decreases with decreasing 275, H. Cottrell, "Structural Processes in Creep," J. Iron Steel Inst. (London) Special Report 70, 1 (1961). 24 stress. This would probably result in an increase in the spacing of the voids, a, and the strain, a/d, would become larger. The spacing could also be decreased by the coalescence of bubbles. Hull and Rimmer?8 showed that the creep behavior of copper (in the absence of irradiation) could be explained on the basis of an increase in the void spacing with decreasing stress. SUMMARY AND CONCLUSIONS We have determined the properties of Hastelloy N tubes in- and ex- reactor at 760°C. The rupture life was reduced by about a factor of 10, even under conditions where the integrated thermal neutron dose was only of the order of 1018 neutrons/cmz. The tangential rupture strains decreased from values of 7 to 10% for the unirradiated tubes to 0.1 to 2% for the irradiated tubes. The fracture strains for the irradiated tubes were a minimum for those that failed in short times and increased as the rupture life increased. Although the fracture strains were affected markedly by irradiation, the creep rates of the tubes were not altered. These cobser- vations can be rationalized on the basis of a mechanism of radiation damage involving helium produced by the thermal 10g (n,w) reaction. The results for the tubular specimens are compared with those obtained for rod specimens on the same heat of material. The results are in reason- ably gcod agreement. The small strains and the inherent experimental errors involved with the irradiated samples make it difficult to detect any effects of the two different stress states on the fracture strains. ACKNOWLEDGMENTS The authors gratefully acknowledge the assistance of several other persons at ORNL in this study. V. G. Lane — Ex-Reactor Tests J. W. Woods — Supervised construction and running of in-reactor experiments 28p. Mull and D. E. Rimmer, Phil. Mag. 4, 673 (1959). 25 Metals and Ceramics Reports Office — Preparation of manuscript Graphic Arts — Preparation of drawings H. R. Tinch — Metallography C. R. Kennedy — Review of manuscript J. L. Scott — Review of manuscript We alsc are pleased to acknowledge the assistance of Atomics International in supplying the materials used in this program and the support of the Division of Space Nuclear Systems of the ARKC. S e e apge i 1-3. 4=5, 6—25. 26. 27, 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42, 43~45, 46. 77 . 78. 79. 80. g1. 82. 83. 84. 85. 86. 87. 88. g89. 90. 91. 92. 93. 94—108. 27 ORNL-TM-1206 INTERNAL DISTRIBUTION Central Research Library 47. H. Inouye ORNL — Y-12 Technical Library 48. P. Kasten Document Reference Section 49. C. R. Kennedy Laboratory Records 50. R. T. King Laboratory Records, ORNL RC 51. A, P. Litman ORNL Patent Office 52. E. L. Long, Jr. G. M. Adamson, Jr. 53. H. G. MacPherson S. E. Beall 54—58. H. E. McCoy, Jr. D. Billington 59, C. J. McHargue E. E. Bloom 60. A. J. Miller E. G. Bolhmann 61. A. R. Olsen G. E. Boyd 62. P. Patriarca R. B. Briggs 63. M. W. Rosenthal D. Canonico 64, H. C. Savage E. L. Compere 65, J. L. Scott J. E. Cunningham 66. C. E. Sessions J. H. DeVan 67. J. Stanley J. H Frye, Jr. 68. J. 0. Stiegler R. Gelbach 69. G. M. Slaughter D. G. Harman 70. D. B. Trauger W. 0. Harms 71-75. J. R. Weir M. R, Hill 76. J. W. Woods N. E. Hinkle EXTERNAIL, DISTRIBUTION G. G. Allaria, Atomics International J. G. Asquith, Atomics International D. F. Cope, RDT, SSR, AEC, Oak Ridge National Laboratory H. M. Dieckamp, Atomics International J. L. Gregg, Bard Hall, Cornell University F. D. Haines, AEC, Washington C. E. Johnson, AEC, Washington W. L. Kitterman, AEC, Washington W. J. Larkin, AEC, Ozk Ridge Operations A. B. Martin, Atomics International D. G. Mason, Aftomics International G. W. Meyers, Atomics International D. E. Reardon, AEC, Canoga Park Area Office J. M. Simmons, AEC, Washington S. R. Stamp, AEC, Canoga Park Area Office R. F. Wilson, Atomlcs International Division of Research and Development, AEC, Oak Ridge Operations Division of Technical Information Ekten81on