AN INVESTIGATION OF INTERGRANULAR OXIDATION IN STAINLESS STEEL Clarence A. Siebert Maurice J. Sinnott Robert E. Keith University of Michigan January 1954 United States Air Force Air Materiel Command Wright-Patterson Air Force Base, Dayton, Ohio WADC TR 54-120

FOREWORD This report was prepared by the University of Michigan, under USAF Contract No. AF 33(616)-3535 The contract was initiated under Research and Development Order No. 615-13, "High Temperature Alloys", and was administered under the direction of the Materials Laboratory, Directorate of Research, Wright Air Development Center, with Capt, M. J. Whitman and Lt. J. H. DeVan acting as project engineers. WADC TR 54-120 ill

NOTICES When Government drawings, specifications, or other data are used for any purpose other than in connection with a definitely related Government procurement operation, the United States Government thereby incurs no responsibility nor any obligation whatsoever; and the fact that the Government may have formulated, furnished, or in any way supplied the said drawings, specifications, or other data, is not to be regarded by implication or otherwise as in any manner licensing the holder or any other personl or corporation, or conveying any rights or permission to manufacture, use or sell any patented invention that may in any way be related thereto, The information furnished herewith is made available for study upon the understanding that the Government's proprietary interests in and relating thereto shall not be impaired. It is desired that the Judge Advocate (WCJ), Wright Air Development Center, Wright0Patterson Air Force Base, Ohio, be promptly notified of any apparent conflict between the Government's proprietary interests and those of others, WADC TR 54-120 iv

ABSTRACT Specimens from one heat of type 309 - Nb and eight heats of type 310 stainless steel were oxidized in dry, moving air for times up to 100 hr in the temperature range 1600~-2000~F* Intergranular oxidation severity measurements were made microscopically. X-ray powder patterns were made of representative scales. Visual and magnetic examinations were made of both the specimens and the oxide scales. No appreciable difference in the severity of intergranular oxidation was observed that could be attributed to differences in alloy content among the heats. In general, intergranular penetration increased with time and temperature. X-ray analysis showed Cr203, Cr203-Fe203 solid solutions, Fe203, and a high-parameter spinel phase in the scales. All of the scales examined were protective in nature, and no improvement in penetration characteristics is foreseen by making minor changes in alloy contents. PUBLICATION REVIEW This report has been reviewed and is approvedFOR THE COMMANDER: M. E. Sorte Colonel, USAF Chief, Materials Laboratory Directorate of Research WADC TR 54-120 v

TABLE OF CONTENTS Page INTRODUCTION 1 MATERIALS 2 Metallographic Examination 2 Magnetic Tests 3 EQUIPMENT 4 PROCEDURE 5 Specimen Preparation Equipment Operation 5 Evaluation of Specimens 6 Evaluation of Oxide Scales 8 Interface Movement 10 RESULTS AND DISCUSSION 11 Theory 11 Penetration Measurements 14 Interface Movement Measurements 16 Visual and Magnetic Examination 17 X-Ray Diffraction 18 Metallographic Examination 19 SUMMARY AND CONCLUSIONS 22 BIBLIOGRAPHY 25 WADC TR 54-120 vi

LIST OF TABLES Table Page 1 Composition of Stainless Steel Stock, Weight Percent 25 2 Explanation of Symbols Used in Oxidation Data Tabulations 26 3 Results of Oxidation Tests, Type 309 + Nb Alloy 27 4 Results of Oxidation Tests, Type 310 Alloy, Heat 64177 28 5 Results of Oxidation Tests, Type 510 Alloy, Heat 64270 29 6 Results of Oxidation Tests, Type 310 Alloy, Heat X11306 30 7 Results of Oxidation Tests, Type 310 Alloy, Heat X11338 30 8 Results of Oxidation Tests, Type 310 Alloy, Heat X27258 31 9 Results of Oxidation Tests, Type 310 Alloy, Heat X45558 51 10 Results of Oxidation Tests, Type 510 Alloy, Heat X46063 52 11 Results of Oxidation Tests, Type 510 Alloy, Heat X46572 52 12 Oxide-Metal Interface Movement in 100 Hours Calculated from Weight Loss Measurements 33 13 Results of Metallographic Analysis 54-35 WADC TR 54-120 vii

LIST OF ILLUSTRATIONS Figure Page 1 Photomicrograph. Type 309 + Nb Alloy, As Received 36 2 Photomicrograph, Type 310 Alloy, Heat 64177, As Received 36 3 Photomicrograph, Type 310 Alloy, Heat 64270, As Received 56 4 Photomicrograph, Type 310 Alloy, Heat X11306, As Received 56 5 Photomicrograph, Type 310 Alloy, Heat X11338, As Received 57 6 Photomicrograph. Type 310 Alloy, Heat X27258, As Received 37 7 Photomicrograph, Type 310 Alloy, Heat X45558, As Received 57 8 Photomicrograph. Type 310 Alloy, Heat X46063, As Received 57 9 Photomicrograph. Type 310 Alloy, Heat X46572, As Received 38 10 Photomicrograph, Grain Boundary Network, Type 310 Alloy, Heat 64177, As Received 58 11 Photomicrograph. Grain Boundary. Type 310 Alloy, Heat X45558, As Received 38 12 Photomicrograph. Cross Section. Type 510 Alloy, Heat 64177, As Received 38 15 Photomicrograph. Cross Section, Type 310 Alloy, Heat X11338, As Received 59 14 Line Drawing, Schematic Drawing of Oxidation Equipment 40 15 Photograph, Specimens and Ceramic Holders 41 16 Photograph, Bending Die and Specimens 41 17 Graph, Penetration vs. Depth Below Surface, Type 509 + Nb Alloy, Run 5 42 18 Graph, Penetration vs. Depth Below Surface. Type 510 Alloy, Heat 64177. Run 5 43 19 Graph. Penetration vs. Depth Below Surface, Type 510 Alloy, Heat 64270. Run 5 44 WADC TR 54-120 viii

LIST OF ILLUSTRATIONS (cont.) Figure Page 20 Graph. Penetration vs. Depth Below Surface. Type 309 + Nb Alloy. Run 6 45 21 Graph. Penetration vs. Depth Below Surface. Type 310 Alloy, Heat 64177. Run 6 46 22 Graph, Penetration vs. Depth Below Surface. Type 310 Alloy, Heat 64270. Run 6 47 23 Graph. Penetration vs. Depth Below Surface. Type 309 + Nb Alloy. Run 8 48 24 Graph. Penetration vs, Depth Below Surface. Type 310 Alloy, Heat 64177. Run 8 49 25 Graph. Penetration vs, Depth Below Surface. Type 310 Alloy, Heat 64270. Run 8 50 26 Graph. Penetration vs. Depth Below Surface. Type 309 + Nb Alloy. Run 9. Straight Section 51 27 Graph, Penetration vs. Depth Below Surface. Type 510 Alloy, Heat 64177. Run 9. Straight Section 52 28 Graph, Penetration vs. Depth Below Surface, Type 310 Alloy, Heat 64270. Run 9. Straight Section 55 29 Graph. Penetration vs. Depth Below Surface. Type 309 + Nb Alloy. Run 9. Curved Section. 54 30 Graph. Penetration vs. Depth Below Surface. Type 310 Alloy, Heat 64177, Run 9. Curved Section 55 31 Graph. Penetration vs. Depth Below Surface. Type 310 Alloy, Heat 64270. Run 9. Curved Section 56 32 Graph, Penetration vs. Depth Below Surface, Type 309 + Nb Alloy. Run 10 57 33 Graph, Penetration vs. Depth Below Surface. Type 310 Alloy, Heat 64177. Run 10 58 34 Graph. Penetration vs. Depth Below Surface. Type 310 Alloy, Heat 64270. Run 10 59 WADC TR 54-120 ix

LIST OF ILLUSTRATIONS (cont.) Figure Page 35 Graph. Penetration vs. Depth Below Surface. Type 309 + Nb Alloy. Runs 11 and 27 60 36 Graph. Penetration vs. Depth Below Surface, Type 310 Alloy, Heat 64177. Runs 11 and 27 61 37 Graph. Penetration vs, Depth Below Surface. Type 310 Alloy, Heat 64270. Runs 11 and 27 62 38 Graph. Penetration vs. Depth Below Surface. Type 309 + Nb Alloy. Runs 12 and 26 63 39 Graph, Penetration vs, Depth Below Surface. Type 310 Alloy, Heat 64177. Runs 12 and 26. 64 40 Graph. Penetration vs. Depth Below Surface. Type 310 Alloy, Heat 64270. Runs 12 and 26 65 41 Graph* Penetration vs. Depth Below Surface, Type 309 + Nb Alloy. Run 13 66 42 Graph. Penetration vs, Depth Below Surface, Type 310 Alloy, Heat 64177. Run 13 67 43 Graph, Penetration vs. Depth Below Surface, Type 310 Alloy, Heat 64270. Run 13 68 44 Graph, Penetration vs. Depth Below Surface. Type 309 + Nb Alloy. Run 14 69 45 Graph. Penetration vs, Depth Below Surface. Type 310 Alloy, Heat 64177. Run 14 70 46 Graph. Penetration vs. Depth Below Surface. Type 310 Alloy, Heat 64270. Run 14 71 47 Graph. Penetration vs. DOpth Below Surface, Type 309 + Nb Alloy. Run 15 72 48 Graph. Penetration vs. DIpth Below Surface, Type 310 Alloy, Heat 64177. Run 15 73 49 Graph. Penetration vs. Depth Below Surface, Type 510 Alloy, Heat 64270. Run 15 74 WADC TR 54-120 x

LIST OF ILLUSTRATIONS (cont,.) Figure Page 50 Graph. Penetration vs, Depth Below Surface, Type 310 Alloy, Heat X11306. Runs 16 and 28 75 51 Graph, Penetration vso Depth Below Surface., Type 310 Alloy, Heat X11338. Runs 16 and 28 76 52 Graph. Penetration vs. Depth Below Surface, Type 510 Alloy, Heat X27258, Runs 16 and 28 77 53 Graph. Penetration vs, Depth Below Surface, Type 310 Alloy, Heat X45558. Runs 17 and 28 78 54 Graph. Penetration vs. Depth Below Surface, Type 310 Alloy, Heat X46063, Runs 17 and 28 79 55 Graph. Penetration vs. Depth Below Surface, Type 310 Alloy, Heat X46572. Runs 17 and 28 80 56 Graph, Penetration vs. Depth Below Surface, Type 510 Alloy, Heat X11306, Runs 21 and 27 81 57 Graph, Penetration vs. Depth Below Surface, Type 510 Alloy, Heat X11338, Runs 21 and 27 82 58 Graph, Penetration vs, Depth Below Surface. Type 510 Alloy, Heat X27258. Runs 21 and 27 85 59 Graph, Penetration vs. Depth Below Surface. Type 510 Alloy, Heat X45558. Runs 22 and 26 84 60 Graph. Penetration vs. Depth Below Surface. Type 310 Alloy, Heat X46063, Runs 22 and 26 85 61 Graph. Penetration vs. Depth Below Surface, Type 310 Alloy, Heat X46572, Runs 22 and 26 86 62 Graph, Penetration vs. Depth Below Surface, Type 310 Alloy, Heat X45558, Runs 23 and 27 87 63 Graph. Penetration vs, Depth Below Surface, Type 310 Alloy, Heat X46063. Runs 25 and 27 88 64 Graph. Penetration vs, Depth Below Surface. Type 310 Alloy, Heat X46572. Runs 23 and 27 89 WADC TR 54-120 xi

LIST OF ILLUSTRATIONS (cont,) Figure Page 65 Graph, Penetration vs. Depth Below Surface, Type 310 Alloy, Heat X11306. Runs 24 and 26 90 66 Graph, Penetration vs. Depth Below Surface, Type 510 Alloy, Heat X11538, Runs 24 and 26 91 67 Graph. Penetration vs, Depth Below Surface, Type 310 Alloy, Heat X27258, Runs 24 and 26 92 68 Graph. Summary Penetration Frequency Curves, Type 309 - Nb Alloy, 100 Hours Duration, Effect of Air Velocity and Humidity 93 69 Graph. Summary Penetration Depth Curves, Type 309 - Nb Alloy, 100 Hours Duration, Effect of Air Velocity and Humidity 94 70 Graph, Summary Penetration Frequency and Depth Curves, Type 309 - Nb Alloy, Effect of Cold Work, 100 Hours Duration 95 71 Graph, Summary Penetration Frequency Curves, Type 309 - Nb Alloy. Effect of Time 96 72 Graph, Summary Penetration Depth Curves, Type 509 - Nb Alloy, Effect of Time 97 73 Graph, Summary Penetration Frequency Curves, Type 310 Alloy, Heat 64177. 100 Hours Duration, Effect of Air Velocity and Humidity 98 74 Graph, Summary Penetration Depth Curves. Type: 310 Alloy, Heat 64177. 100 Hours Duration. Effect of Air Velocity and Humidity 99 75 Graph. Summary Penetration Frequency and Depth Curves, Type 310 Alloy, Heat 64177, Effect of Cold Work. 100 Hours Duration 100 76 Graph. Summary Penetration Frequency Curves, Type 310 Alloy, Heat 64177, Effect of Time 101 77 Graph. Summary Penetration Depth Curves, Type 310 Alloy, Heat 64177. Effect of Time 102 WADC TR 54-120 xii

LIST OF ILLUSTRATIONS (cont.) Figure Page 78 Graph. Summary Penetration Frequercy Curves, Type 310 Alloy, Heat 64270. 100 Hours Duration. Effect of Air Velocity and Humidity 105 79 Graph. Summary Penetration Depth Curves. Type 310 Alloy, Heat 64270. 100 Hours Duration, Effect of Air Velocity and Humidity 104 80 Graph. Summary Penetration Frequency and Depth Curves Type 310 Alloy, Heat 64270, Effect of Cold Work, 100 Hours Duration 105 81 Graph. Summary Penetration Frequency Curves. Type 310 Alloy, Heat 64270. Effect of Time 106 82 Graph. Summary Penetration Depth Curves. Type 310 Alloy, Heat 64270. Effect of Time 107 83 Graph. Summary Penetration Frequency Curves. Type 310 Alloy, Heat X11306, Effect of Time 108 84 Graph. Summary Penetration Depth Curves. Type 310 Alloy, Heat X11306. Effect of Time 109 85 Graph. Summary Penetration Frequency Curves. Type 310 Alloy, Heat X11338. Effect of Time 110 86 Graph. Summary Penetration Depth Curves. Type 310 Alloy, Heat X11338. Effect of Time 111 87 Graph. Summary Penetration Frequency Curves. Type 310 Alloy, Heat X27258. Effect of Time 112 88 Graph. Summary Penetration Depth Curves, Type 310 Alloy, Heat X27258, Effect of Time 113 89 Graph. Summary Penetration Frequency Curves. Type 310 Alloy, Heat X45558. Effect of Time 114 90 Graph. Summary Penetration Depth Curves, Type 310 Alloy, Heat X45558, Effect of Time 115 91 Graph, Summary Penetration Frequency Curves, Type 310 Alloy, Heat X460635. Effect of Time 116 WADC TR 54-120 xiii

LIST OF ILLUSTRATIONS (cont.) Figure Page 92 Graph, Summary Penetration Depth Curves* Type 310 Alloy, Heat X46063, Effect of Time 117 93 Graph. Summary Penetration Frequency Curves, Type 310 Alloy, Heat X46572. Effect of Time 118 94 Graph, Summary Penetration Depth Curves, Type 310 Alloy, Heat X46572, Effect of Time 119 95 Photomicrograph. Type I Penetration. Web from Surface 120 96 Photomicrograph, Type II Penetration, Web from Pits 120 97 Photomicrograph, Type III Penetration, Rough Web 120 98 Photomicrograph. Type IV Penetration, Smooth Fissures 120 99 Photomicrograph, Type V Penetration, Developed Web 121 100 Photomicrograph, Type VI Penetration Colesced Web 121 101 Photomicrograph, Grain Boundaries Before Testing, Type 310 Alloy, Heat 64270, As Received 121 102 Photomicrograph. Penetration and Grain Structure. Type 510 Alloy, Heat 64270, 1700~F and 100 hr 121 103 Photomicrograph, Penetration and Precipitation. Type 310 Alloy, Heat X11306, 1600~F and 100 hr 122 104 Photomicrograph. Penetration and Precipitation. Type 310 Alloy, Heat X11306, 1900~F and 100 hr 122 105 Photomicrograph. Penetration and Precipitation, Type 310 Alloy, Heat X11306. 2000~F and 100 hr 122 106 Photomicrograph, Penetration and Precipitation, Type 310 Alloy, Heat X113p6. 2000~F and 10 hr 122 107 Photomicrograph, Developed Web Penetration, Type 310 Alloy, Heat X11338. 1900~F and 100" hr 123 108 Photomicrograph, Developed Web Penetration, Type 310 Alloy, Heat X1338i 1900~F and 100 hr 125 WADC TR 54-120 xiv

LIST OF ILLUSTRATIONS (cont,) Figure Page 109 Photomicrograph. Oxide-filled Pits. Type 310 Alloy, Heat X45558. 2000~F and 100 hr 123 110 Photomicrograph, Delta Ferrite, Type 309 + Nb Alloy, 1700~F and 100 hr 123 111 Photomicrograph, Sigma Phase, Type 510 Alloy, Heat X27258. 1600~F and 100 hr 124 112 Photomicrograph, Matrix and Boundary Precipitation, Type 310 Alloy, Heat X11338. 1600~F and 100 hr 124 113 Photomicrograph. Matrix and Boundary Precipitation. Type 310 Alloy, Heat X113358 19000F and 100 hr 124 114 Photomicrograph. Oxygen-rich Layer. Type 310 Alloy, Heat X27258. 1900~F and 10 hr 124 115 Photomicrograph. Concentration Gradients. Type 310 Alloy, Heat X45558. As Received 125 116 Photomicrograph. Carbide Precipitation Bands. Type 310 Alloy, Heat X45558, 16000F and 10 hr 125 117 Photomicrograph. Carbide Precipitation Bands. Type 510 Alloy, Heat X45558. 1600~F and 10 hr 125 118 Photomicrograph. Carbide Bands and Grain Refinement. Type 510 Alloy, Heat X45558, 1600~F and 100 hr 125 WADC TR 54-120 xv

FINAL REPORT AN INVESTIGATION OF INTERGRANULAR OXIDATION IN STAINLESS STEEL INTRODUCTION When a metal oxidizes the usual result is the formation of a more or less continuous and adherent layer of oxide on the exposed surface. In general, with increasing time and/or temperature of exposure to the oxidizing medium this oxide scale, which may be quite complex in its physical and chemical makeup, tends to increase in amount, becoming thicker at the expense of the underlying metal. This results in a progressive movement of the interface between the metal and the oxide layer toward the center of the metallic body. With some alloys, however, in addition to this progressive type of oxidation there is also oxidation occurring in the metal ahead of the metal-oxide interface. This latter type of oxidation has been variously known as internal oxidation or subscale formation. Internal oxidation may be further subdivided into cases in which the oxide occurs in a fine dispersion throughout the metal layer just beneath the metal-oxide interface, and cases in which the oxidation takes place principally in grain boundaries. In this latter type, the surface metal grains retain something like their original appearance even though they may become completely surrounded by a network of grain-boundary oxide. It is this last type of oxidation with which this report is concerned. Compared with the total volume of literature concerned with the problem of metallic oxidation, the portion devoted to the study of intergranular oxidation is small. Intergranular oxidation becomes important in applications where thin sections are exposed to air at high temperatures. Under these conditions, premature failures may result. The conditions favoring intergranular oxidation have never been clearly defined; consequently, very little is known about the mechanism or mechanisms by which this type of oxidation occurs. The present program arose from the need of the U.S. Air Force for information concerning intergranular oxidation in high-alloy stainless steels at very high temperatures. The alloy of principal interest is the type 310 analysis (25%.Gr20%Ni), although there is also interest in other compositions. There were four objectives of this research programs WADC TR 54-120 1

(1) to determine the effect of temperatures between 1600~ and 2000~F on intergranular oxidation; (2) to examine the effects of alloy composition on intergranular oxidation; (3) to determine the nature of the penetrating material in areas of intergranular attack; and (4) to devise methods of reducing or eliminating intergranular oxidation, The oxidizing medium used in the experimental work was dry air under controlled flow conditions, During the past year, research has been concentrated on objectives (1) and (2). MATERIALS The stainless steels used in this investigation consisted of specimens from one heat of type 309 + Nb steel and from eight heats of type 310 steel. The specimens of type 309 and two of type 310 steel were from special heats (64177 and 64270) and were supplied by the General Electric Company, The specimens from the remaining six heats of type 310, selected to provide as wide a compositional variation as possible within the specification limits, were from commercial production heats and were supplied by the United States Steel Corpany, The chemical analyses of these heats are presented in Table I. All the material was in the form of cold-rolled and annealed sheet stock, the thickness of which varied from heat to heat but was in the range from 0.04 to 0.07 in. Metallographic Exaination Photomicrographs of each of the nine heats in the as-received condition are presented in Fig. 1-9. These photomicrographs are intended to provide a standard of reference for photomicrographs taken following testing. In the asreceived photomicrographs, the plane of polish was the plane of the sheet. Differences between the structures in the plane of rolling and in cross section will be enumerated and discussed in the Results and Discussion section of this report. The grain size of the type 509 + Nb heat (Fig. 1) was extremely fine. Examination at 1OOOX showed that there was no grain-boundary carbide network, but that there was a considerable quantity of randomly distributed carbide, Heat 64177 (Fig. 2) had a duplex grain structure, and examination at 1000X after etching revealed an almost continuous grain-boundary carbide network (Fig, 10). This duplex structure persisted throughout the thickness of the sheet. To determine whether this structure was the result of an unstable condition, a WADC TR 54-120 2

specimen of this heat was re-solution-treated at 19000F for 20 minutes and water-que.nched. It was then polished and etched in a concentrated Muraakami's reagent. The only appreciable change resulting from the re-solution treatment was found to be slight coalescence of the carbides. Furthermore, this heat showed grain-boundary penetration in the as-received condition, an example of which appears in Fig. 12, Although the other heats showed what may have been extremely slight amounts of initial penetration, the amount occurring in heat 64177 was appreciable, A typical edge from one of the other heats (heat X11338) is shown in Fig. 13 for comparison. A duplex structure also occurred in heat 64270 (Fig. 5), although it was less pronounced than in heat 64177. The grain boundaries etched easily, and at 100OX a boundary network similar in appearance to that in heat 64177 was apparent. In etching the as-received steels, it was found that certain of the heats were very difficult to etch, notably the type 309 + Nb, X11506, X27258, X45558,- and, to a lesser extent, X46063. The procedure finally adopted with these heats was to etch them electrolytically with 10% chromic acid or chemically with a 1:1:1 mixture of nitric acid, hydrochloric acid, and water. When viewed under oblique light, the resulting surfaces showed the grains fairly well. Examination of all these difficult-to-etch heats at 100OX showed complete absence of any grain-boundary carbide networks, and the typical grain-boundary appearance is shown in Fig, 11, Heat X11306 (Fig. 4) showed a rather fine, uniform grain structure, Heat X11338 etched easily and looked very similar to heat 64270, but had a more uniform grain size (Fig. 5). Again examination at lOOOX showed an almost complete grain-boundary carbide network. Heat 27258 (Fig, 6) had a relatively large grain size, no carbide network, and in addition exhibited a fairly large amount of a randomly dispersed phase, presumably carbide. Heat X45558 (Fig. 7) was similar to heat X27258, but had a smaller grain size and less of the dispersed phase. Heat X46063 (Fig. 8) showed a large grain size, similar in structure to heats X27258 and X45558, although it etched somewhat differently. There was little dispersed phase, and again no trace of a boundary network. Heat X46572(Fig. 9) was more like heats 64270 and X11338 in appearance, except that the grain-boundary network was incomplete. Magnetic Tests Magnetic tests of the type to be described in detail in the Procedure section of this report indicated the presence of a very slight amount of magnetism (much less than the amount regularly classified as "weak" in the routine test) in all the as-received materials. Etching with Murakami's reagent showed nothing except the networks and/or dispersed phases revealed by the other etchants. Actually, these particles were too small for definite identification using Murakami's reagent, WADC TR 54-120 3

and thus, on the basis of that etch alone, could possibly have been at least partly composed of delta ferrite. Since the particles showed up with the other etches, however, it is fairly certain that the particles were principally carbides. Therefore, any delta-ferrite (or sigma phase, which should also be revealed by Murakamii" reagent) could have been present in the as-received material only in very small amounts, although the slight magnetism would seem to indicate the presence of some delta-ferrite. EQUIPMENT The oxidation equipment used in the investigation consisted of four tubetype laboratory furnaces. A schematic drawing of such a furnace with its associated air and power supply is shown in Fig. 14. Air at approximately 90 psig is first passed through a porous A1203 filter to remove any entrained oil, sediment, or othei foreigh matter. The air then passes through a pressure-reducing valve and then through two drying towers, the first filled with activated alumina and the second filled with phosphorus pentoxide. A valve is provided to bleed off a sample of the dried air for a dewpoint determination at suitable time intervals. The air is delivered to a manifold from which four side streams are withdrawn, one to each of the four horizontal tube furnaces. Flow to each furnace is controlled by a needle valve, and is metered by a flowmeter of the rotameter type in conjunction with a manometer. The air is then delivered to one end of a 2-in, sillimanite tube in the furnace, This tube is concentric with a similar, but' shorter, 3-in,'tube, The chromel heating wire is wound about the latter tube and imbedded in alundum cement. There is a 5-in, layer of vermiculite insulation contained in an aluminum shell around the heating coils. The furnace draws approximately 12 amperes at 220 volts. The 2-in, tube is packed for approximately one-third its length with broken pieces of ceramic, the purpose of which is to provide a large surface area to preheat the incoming air, A thermocouple well extends into the furnace as close as possible to the location of the specimens, which are in the center portion of the 2-in. tube and are supported on ceramic fixtures consisting of combustion carbon-analysis boats with their covers cemented in an inverted position (Fig, 15). The uniformtemperature zone of the furnace, in which as many as three of these boats are placed, is approximately 8 in, long. The temperature is controlled by means of a Foxboro controller with the control thermocouple imbedded in the furnace windings. Considerable difficulty was encountered in the early part of the program from an inablility to maintain a sufficiently low dewpoint throughout a 100-hr run using only the alumina dryer, Although the form of alumina used contained a cobalt salt intended to indicate when the alumina became useless, it was found that the aluimina ceased to adsorb an appreciable amount of moisture long before the color change took place. Furthermore, the alumina dryer alone was seldom capable of producing air with a dewpoint lower than -20~F, whereas it was felt that a dewpoint of -40~F would have been more desirable. Accordingly, a stainless-steel vessel was constructed to hold alternate layers of phosphoric anhydride and glass wool. This dryer was connected in series following the alumina dryer between runs 13 and 14, The alumina was replaced with regenerated material every lO0 hr; the phosphoric anhydride dryer, once filled, has lasted over 1500 hr without attention. With this combination of dryers, it was possible to maintain a dewpoint of -35~ to -60~F consistently. WADC TR 54-120 4

PROCEDURE Specimen Preparation Rectangles 1/2 x 1 in. were first scribed on the sheet stock. Single small punch marks were then placed on one side of the sheet in one corner of each rectangle, thus permanently identifying the sides of the specimens relative to the original sheet. The specimens were then cut from the sheet using a shear. Shearing was found to be superior to bandsawing because it resulted in a neater edge and also gave fewer scratches on the faces of the specimens, which were tested without further surface treatment other than cleaning. Specimens were given code numbers which identified their original positions in the sheets. These numbers were recorded on the individual envelopes in which the specimens were stored, no markers being placed on the specimens themselves. Equipment Operation Before starting an oxidation run, the furnaces were brought to temperature and the air was allowed to flow through them for a period of 2-4 hours to establish thermal equilibrium. An air flow rate of 30 ft/min at the furnace temperature was selected as standard. Usually, one furnace was operated at each of the temperatures 16000, 1700~, 1800~, and 1900~F, and one specimen of each of three heats was assigned to each furnace at random. In the later runs, in which all the furnaces were operated at 2000~F, specimens from all nine heats, including some duplicates, could be run simultaneously. Prior to charging, the specimens were cleaned with acetone and placed in the ceramic fixtures described in the Equipment section of this report and illustrated in Fig. 15. During the runs, readings of furnace temperatures, air flow rates, air pressure, and room temperature were taken twice a day. Barometer readings and dewpoint readings of the furnace air were taken once a day, and the furnace temperature controllers were balanced once a day. Runs during this year's program were of 10, 50, and 100-hour duration, except for one 50-hr run and one 77-hr run. At the conclusion of the runs, the specimens were quenched into distilled water in stainless-steel beakers, in order to avoid the loss of any scale, to arrest any phase transformations which might have taken place in the oxides on slower cooling, and to preserve the carbide distributions existing at the test temperatures. The contents of the beakers were then filtered through coarse filter paper and washed with acetone and then with ether to accelerate the drying the loose scales. After the filter papers had dried, their contents were bottled. The total elapsed time between the quenching and the completion of the bottling operation was about 1 hr. Any loosely adhering scale was gently scraped off the surfaces of the specimens with a surgical scapel and combined with the bottled scale. WADC TR 54-120 5

Evaluation of Specimens The visual appearances of the specimens were recorded, after which a lengthwise strip approximately 1/8 in. wide was cut from each specimen and discarded, in order to assure freedom from edge effects in the subsequent measurements. A 90~ bend was then accurately put into each of the specimens perpendicular to the sawed edge using the specially designed die illustrated in Fig. 16. At this point, qualitative magnetic tests were made on the specimens. To make these tests, a specimen was placed on a glass plate resting on the bent portion, both "legs" extending upward. The specimen was then set in motion by a gentle tap, and, while it oscillated, an Alnico magnet was brought near but not touching, the specimen and parallel to one of the flat faces. If there was no noticeable attractive effect or change in periodicity of the motion, the specimen was classified as "not detectably" magnetic. If the reverse was true, the specimen was allowed to come to rest and the magnet was again brought up parallel to a flat face. If the specimen moved violently so that the face touched the magnet, it was classified as "strongly" magnetic. If the specimen could be made to move slightly, although not sufficiently to touch the magnet, it was classified as having "medium" magnetism. If a specimen proved to belong to neither of these latter two classifications, it was recorded as being "weakly" magnetic in view of the magnetism shown in the first test, when the specimen was already in motion. Obviously, this test is qualitative, and too much emphasis must not be placed on the classifications themselves. Their only importance is the fact that almost all the specimens showed increases in magnetism after oxidation, as will be discussed subsequently. Following the magnetic tests, the specimens from each heat were mounted, several-to the mount, in bakelite with the sawed sides exposed. An untested specimen from the same heat was included in each mount. Considerable experimentation was carried out before a satisfactory polishing technique was developed. The technique finally adopted consisted of grinding the sawed edge of each specimen smooth and beveling it very slightly before mounting. Rough polishing was done under water using 240-grit silicon carbide paper. Finer polishing was done using 400and 600-grit silicon carbide papers, also wet, which were mounted on metallographic wheels. Final polishing was done on a low-nap metallographic wheel impregnated with 2-micron diamond dust. The resulting polish was not scratch-free, but it was adequate. It was not possible to use such relatively soft abrasives as Gamal, Linde "B", and C-RO to obtain a smoother surface, since with these abrasives the oxides and inclusions present tended to tear out. A finer, 1/2-micron, diamond polishing compound was also tried, but it proved excessively difficult to keep the wheel free from contamination, and its use was stopped. All the early specimens, and many of the later ones, were examined at 100OX in the unetched condition using bright field, dark field, and polarized illumination. The latter two contributed little to the investigation. Unusual areas were noted for future reference and/or photographed. After polishing and metallographic examination, measurements were made of the extent of intergranular penetration, which was clearly visible in properly WADC TR 54-120 6

polished unetched specimens, To make these measurements, a desk microscope with a 4-mm objective lens, a travelling stage, and a Bausch and Lomb grain-size eyepiece was used, Magnification was approximately 500X, The edge of a specimen cross section was aligned with one edge of the square A.ST.M. No. 6 grid of the eyepiece, and the entire straight portion of the specimen edge was traversed. The bent portion of the specimen was not read (except in Run 9). For convenience, scratches had been previously made on the polished surface to mark the beginnings of the bent portion of the specimen, and readings were made up to a scratch. Initial and final stage positions were recorded for each straight portion. As a traverse was made, the number of intergranular fissures ending in each of the six rows of the grid parallel.to the specimen edge was counted using a desk counter, and was recorded. This procedure was repeated for the other three straight portions of each specimen (both sides of both straight "legs" - a total of four), During the course of the project, penetration measurements were made by five different technicians. Considerable pains were taken to see that satisfactory agreement was attained among the technicians reading the same specimens before they were allowed to proceed to specimens which had not yet been measured. Checks were made from time to time by having different technicians repeat a specimen counted by one of the other technicians. It was found that readings of the number of penetrations in a given surface made by different technicians usually agreed within about 5% on all but the shallowest group of fissures, those between 0 and 0ooo00653 in. in depth, On these shallow penetrations, operator variability ran as high as 30%. This lack of agreement in the shallow group is understandable when it is considered that many of the fissures in this group were of the order of 0O0002 in. or less in depth, the order of magnitude of the surface roughness. In addition, even slight polishing scratches and tears can produce effects likely to be mistaken for shallow penetration, especially in specimens in which there is little true penetration. For these reasons, too much confidence must not be placed in the actual magnitudes obtained for penetrations less than 0o00063 in. deep, The figures for penetration in the four portions of a single specimen were added and divided by the total length traversed (from 1.1 to 1,4 in.) to put data from different specimens on a comparative basis. The sizes of the squares of the eyepiece grid had been previously determined, by calibration with a stage micrometer, to be 0,00063 in. on a side. Using this calibration, a frequancy-vs.depth curve could be constructed for each specimen by plotting a bar chart and then drawing a smooth curve to fit the data. Curves constructed by this process appear in Figs* 17 through 67. Theoretically, any curve obtained by this method would tend to be biased toward the shallow side,- even if the counting were completely accurate, because the probability of any fissurets ending precisely at the plane of polish, and thus the probability of its being observed in its true depth, is quite small, To condense the data further, it was decided to examine the behavior of two parameters of the frequency-vs.-depth curves. The parameters, N and X, are defined as follows: s 7 N = J n(s)ds Y E ni (1) i=l WADC TR 54-120 7

and s 7 X = fsn(s)ds = ani, (2) 0 i=l N N where ni is the number of fissures per inch of specimen surface in the ith group, ai is the midpoint of the ith group measured from the surface, and s is the depth coordinate. The data were taken in seven groups: the six levels of the grid and a seventh class, rarely occuring, consisting of fissures which extended past the sixth grid level and which were arbitrarily assigned a group midpoint one grid level (0.00065 in.) beyond the sixth level, It can be seen that the parameter N is simply the total number of fissures per inch of all depths, The parameter X is a mean penetration depth; more precisely, X is the first moment of the penetrationvs,-depth curve with respect to the y-axis (or specimen surface) divided by the total number of fissures, Alternatively, X may be considered as being the depth coordinate of the centroid of the penetration-vs.-depth curve, Having obtained these two parameters for each specimen, the penetration data could be concisely summarized. One word of caution is necessary, however, in the interpretation of the meaning of the X parameter. Referring to the summation form of Equation (2), it will be noted that it is impossible to have a value of X less than the midpoint of the first data group, Suppose, for example, that a specimen contains penetration fissures of the first group only. The quantity rn then becomes equal to N, and can be taken outside the summation sign, where it cancels with the N in the denominator, leaving only al, the first group midpoint. In the present case, the midpoint of the first group was 0.00032 in. Therefore, a number of the specimens are recorded as having X - 0.00032 in. in which the penetration was actually. much less than this average depth. The 0,00032 in, is not a lower limit in the size of fissure detectable, but is, rather, a consequence of the group size selected. The question may legitimately be asked, why was not a smaller group size chosen? There are two answers to this question: first, that when these standards were set up, only a small fraction of the specimens had been examined, and there was no way of knowing what sizes of penetration would be encountered later in the program; and, second, that any number of groups more than about six becomes excessively clumsy for an operator to use in obtaining data. In retrospect, it seems that the group size chosen was about ideal, After penetration measurements had been cnarpleted, specimens were etched and again examined metallographically, photomicrographs being taken of significant areas. Evaluation of Oxide Scales The physical appearances of the oxides were recorded after thay had been bottled. Magnetic tests were then run, which were somewhat similar to those on the specimens themselves. The magnet was placed in cbnt~aet with the outside of a bottle containing an oxide sample at a point nearest the sample itself. The bottle and magnet were then rotated, without moving the magnet relative to the bottle, until the weight of the oxide overcame the attractive force of the magnet and the oxide WADC TR 54-120 8

dropped to the bottom of the bottle. Oxides which could not be attracted by the magnet at all were classified as "not detectably" magnetic. Oxides which dropped back when the bottle wall was approximately 90~ to the horizontal were classified as "weakly" magnetic. Oxides which were attracted by the magnet sufficiently to' allow 180~ rotation of the bottle, suspending them under the magnet, were classified as "medium" or "strongly" magnetic the latter classification being used when oscillatory movement of the magnet outside the bottle produced pronounced movement of the oxide particles within the bottle, This test is even more qualitative than the tests of the metallic specimen, since it is influenced by the amount and density of the oxide, the oxide particle size, and the smoothness and cleanliness of the inside of the bottle. It is valuable, nevertheless, as an indication of the presence of significant quantities of magnetic components in the scales. Following the visual and magnetic examinations, oxides were pulverized in an agate mortar and pestle, mixed with Duco cement, and rolled out into thin cylinders between two glass microscope slides. The diameters of these cylindrical specimens were approximately 0.2 mm, and the cross sections were quite uniform. Specimens were mounted and centered in a 5753Blmm-.diameter Norelco x-ray powder camera, utilizing a Straumanis-mounted film, X-ray exposures were made on a Norelco water-cooled diffraction unit using V-filtered Cr K, radiation. A standard exposure time of 2 hr was selected, and all- exposures were of this length except two, which required somewhat longer times. One series of four samples was run using a Norelco 114.6-mm camera, which required an exposure time of 10 hr, After the films had been processed, traces were made of them on a Leeds and Northrup Knorr-Albers type of microphotometer. These traces were then measured using a ruler of adequate length. This method of measurement was used in preference to an optical comparator, first because it is believed to be more accurate, and second because it gives quantitative values for diffraction line intensities. Interplanar spacings were computed from the distances measured on the microphotometer traces in the same manner as they would have been from distances measured on the film, a shrinkage correction being included. This technique is well-known, and will not be elaborated here. Comparison was then made with interplanar spacings of the compounds expected to be present, as made available by Yearian and Radavich (1). For precise identification of the compounds present, which were solid solutions, it developed that additional corrections for such factors as the finite specimen size and small errors in specimen placement had to be applied to the computed interplanar spacings. This was done for the cubic spinel phase by plotting the value of the lattice parameter calculated from each back-reflection spinel line against the function 1/2 (cos2 G/sin Q + cos2 Q/0), as proposed by Taylor and Sinclair (2) and recommended by Henry, Lipson, and Wooster (3). This particular function linearizes the error 4-dependence somewhat better than other similar functions that have been proposed, according to the above authors. The straight line obtained in this plot was extrapolated to obtain the y-intercept, which represents the errorfree value of the lattice parameter. For the rhombohedral Cr20O-Fe203 solid-solution phase, the same method was attempted, as described by Henrry, Lipson, and Wooster (3)r but there was too much uncertainty in the line positions to obtain more than +10 atomic percent accuracy in the composition. Since this was hardly better than estimating compositions from the interplanar spacing data, extrapolations were not used in reporting the compositions of the rhombohedral phase. WADC TR 54-120 9

The use of longer exposure times would have been beneficial in increasing the accuracy of the x-ray data, since longer exposure would have resulted in denser lines in the back-reflection region. In a number of cases, there was insufficient oxide to prepare an x-ray specimen by the usual means, In such cases, the specimen was omitted, but if it should later develop that any of these specimens are of critical importance, patterns can be obtained using methods of specimen preparation suited to smaller amounts of material. Interface Movement In reporting the penetration data, it would be desirable to refer the penetration depth always to the original metal surface. Early in the experimental work, attempts were made to measure the specimen thickness losses directly by mounting an untested specimen from the same heat and nearly the same location in the sheet stock along with the specimens from a given heats After polishing, several thickness measurements were made of each specimen with a travelling microscope,. the object being to subtract these values from the thickness of the untested specimen. It was found, however, that the interface movement was of the order of magnitude of the experimental error and normal thickness variation in the sheet, so that it was not possible to obtain consistent results by this method. Attempts were then made to carry out weight-loss measurements on specimens given the same oxidation cycle as the specimens used in the regular procedure. Specimens were weighed and their dimensions were measured before and after oxidation and after descaling treatment and brushing, The descaling treatment consisted of making the specimen the cathode in a 10/ H2SO4 electrolyte as described by Upthegrove and Murphy (4) and Siebert (5). A current of 10 amp was passed through each specimen for 1 min. The treatment was not successful in removing any but interference-film types of scale, although it is known to be very effective for the scales found on lower-alloy steels. In an effort to remove the scale further, the specimens were brushed with a rotary wire brush attached to a hand power tool. This treatment succeeded in removing all but the pit-type oxide withour removing any appreciable amount of metal. Density measurements were made on specimens from each heat by measuring the loss of weight in water. Knowing the density and the weight loss, the movement of one metal-oxide interface could be calculated by the formula: A s = A w (3) (4ab - c) d where A s is the interface movement,A w is the weight loss, a and b are the dimensions of the sheet faces of the specimen, c is a correction for the rounded corners of the specimen, and d is the density of the metal. The results of these interface measurements are presented in Table 12, and are discussed more fully in the Results and Discussion section of this report. It should be noted here, however, that the interface movements were, as the early thickness measurements indicated, quite small. In view of the uncertainty in the interface movements arising from the small amounts of oxide remaining on the surfaces and from the roughness of the surfaces, it was decided to report the penetration data relative to the existing metal-oxide interfaces. WADC TR 54-120 10

RESULTS AND DISCUSSION Theory The general mechanism of oxidation of metals has been broken down into several steps for purposes of fundamental study. Expressed briefly, these steps are as follows: (1) Diffusion of oxygen to the surface, (2) Adsorption of the oxygen on the surface, (3) Dissociation and ionization of the oxygen, (4) Inward diffusion of the oxygen, through any scale that is present, (5) Outward diffusion of metal ions and electrons, (6) Reaction between oxygen and metal ions to form oxide. In any-given case, this mechanism may be complicated by the existence of multiple valence states for the metal ions, presence of minor elements, existence of multilayered oxide scales, volatility, cracking, or spalling of the scales, and the like. Any one of the basic steps may prove to be the slowest, or rate-controlling, step of the process. For an excellent review of the present state of knowledge in the entire field of oxidation the reader is referred to the recent book by Kubaschewski and Hopkins (6). In the case of an alloy which is subject to intergranular oxidation, it seems reasonable that the grain boundaries must be attacked by oxygen ions reaching the metal-oxide interface, either by means of cracks in the scale layer or by true diffusion through the scale. Since the stainless steels as a group have dense, adherent scales, it would seem that oxygen must reachthe metal-oxide interface by true diffusion in these steels. Davies,$Simnad, and Birchenall (7) in their study of the scaling of pure iron have shown that of the three oxides of iron, FeO grows almost entirely by outward diffusion of Fe ions, Fe304 grows principally (80%) by outward diffusion of Fe ions, and Fe2Os grows almost entirely by inward diffusion of O ions.- Thus it would appear that if the scales formed on the steel contained appreciable amounts of either Fe304 or Fe2O3, there would be a possibility of the existence of mobile 0 ions at the metal-oxide interface. Hickman and Gulbransen (8), Boren (9), Warr (10) and Radavich (11), in work on the oxidation of stainless steels of composition similar to the ones used in this investigation, in no case report presence of FeO in the scale, but do report the- presence of Fe304, Fe203, Cr203, and a Cr-rich spinel phase. FeO might be present, but it would have to be in the form of an extremely thin layer. Radavich also reports the presence of a solid solution of Fe203 and Cr2O3 in some cases and notes that such a phase could easily be mistaken for either pure component, especially by reflection electron diffraction, since the two compounds have complete solid solubility, the same crystal structure, and only slightly different lattice parameters. In the present investigation all these compounds except for the Fe304 were found in the scales. As for the role of Cr203 in 0 diffusion. Cr203 is an oxide in which the ionic transport rates are very slow (12). The low ionic transport is associated with good refractory characteristics, which WADC TR 54 -120 11

have come to be recognized as necessary for formation of a protective oxide such as Cr203 (13). Randall and Robb (14) have shown that Cr203 and a high-parameter (about 8.43 kX) spinel are associated in high Cr and Cr-Ni steels with a protective type of scale, while a second type of scale containing a low-parameter (about 8.37 kX) spinel and no excess Cr203, though more tenacious than the first type, was not nearly as protective. It would thus appear that the key oxide scale component for oxidation resistance is Cr20O3 Turning now to the question of intergranular oxidation, the differences that may exist between the volume diffusion coefficient and the grain boundary diffusion coefficient of a metal were pointed out by Turnbull (15), who carried out a mathematical analysis of an idealized model, and on the basis of the results of this analysis showed that grain boundary diffusion coefficients may be 105 times the corresponding volume diffusion coefficients or more, This difference is basically the result of the nature of the grain boundary itself. Although there is not complete agreement at the present time as to the detailed structures and properties of grain boundaries, it is generally agreed that at least those grain boundaries for which there is a large angular difference between the lattices on either side of the boundary are regions of relatively large disorder on an atomic scale. This means that the mean interatomic distance in a large-angle grain boundary must be greater than the mean interatomic distance in the lattice itself (since the lattice represents a close-packed array). One would therefore expect a larger diffusion coefficient in the grain boundary than in the lattice. Looking at the diffusion process from the thermodynamic point of view, the grain boundary is a region of high strain energy relative to the lattice as a result of its more or less random packing. Therefore the activation energy for diffusion of foreign atoms can more easily be supplied in the vicinity of the grain boundary. Research soon to be published (16) indicates that the grain boundaries have a very important part in the solid-state diffusion of Bi in Cu, Ni in Fe, and Ni in Cu, the latter two systems having previously been considered to be classical examples of volume diffusion. In short, the major role played by grain boundaries in diffusion processes is only beginningto be fully realized, and comes about as a result of the nature of the boundaries themselves. Thus far, we have seen how, in the case of oxidation of stainless steel, a certain amount of 0 in the ionic form can reach the metal-oxide interface by transport through the scale, the amount depending on the scale composition and, of course, on the scale thickness. Once the 0 ions are at the metal-oxide interface, a plausible argument has been presented to show that the ions which do not immediately react to form oxides will tend to diffuse preferentially into the grain boundaries of the metal. Once in a grain boundary, an 0 ion can do one of two things. It may at some point leave the boundary and diffuse out into a grain, where it ultimately becomes contiguous with that fraction of the 0 ions that diffuses from the metaloxide interface to form the 0-rich layer that ultimately becomes part of the scale. The other alternative for the 0 ion diffusing down a grain boundary is that it may react with.a cation and other 0 ions if necessary in the boundary to form an oxide. The role of the cations, especially the minor alloying elements, is a particularly complex one. The general rules of behavior of alloying elements have been summarized by Lustman (13), and are as follows: WADC TR 54 -120 12

(1) Alloy elements tend to become concentrated in the layer of scale adjacent to the metal, either as oxides in the case of elements more easily oxidizable than the base metal, or as metal islands in the case of elements less easily oxidizable than the base metal. (2) Marked improvement in oxidation resistance requires that the oxide of a more easily oxidizable alloying element be refractory (which implies that it is a poor ionic conductor, such as Cr203) and that it be practically the only component of the innermost scale layer, (5) Little, if any, improvement in oxidation resistance is to be expected from the addition of alloying elements less easily oxidizable than the base metal. Considering the relative thermodynamic stability of the various oxides formed from the elements in stainless steel, one finds that SiO2 is the most stable, followed in order of decreasing stability by MnO, Cr203, FeO, and NiO. On the basis of thermodynamics, then, one would expect to find Si oxidizing first, then Mn, and so on. Indeed, Baeyertz (17) demonstrated metallographically the prior oxidation of Si and Cr in 1936. Unfortunately, however, the role of alloying elements is more complex than is indicated by thermodynamics. Investigators using the most advanced techniques available, including electron diffraction and electron microscopy (18), believe at present that the order in which alloying elements will oxidize cannot be predicted in general from the thermodynamic data. Recent work by Gulbransen and McMillan (19) on 80%o ni-20*Cr alloys and by Radavich and Yearian (20) on ferritic stainless steels of the 400 AISI series points to the possible importance to the alloy oxidation resistance of a thin layer of SiO2 at the metal-oxide interface. Baeyertz showed that Si also precipitates as SiO2 and as glassy silicates within the oxide-rich metal layer near the interface, Mn occurs as a major component of the high-parameter spinel phase in the scale, which has been identified as being MnCr204 and has been studied in detail by Longo (21). In addition, Mn probably occurs below the metal surface as particles of a silicate. Cr, as has been mentioned previously, occurs in the scale as free Cr203, as the Cr-rich spinel, and in solid solution with Fe203. Some Cr is present in the intergranular oxides also, as Baeyertz has shown. Neither Ni nor Ni oxide has been reported as a separate phase in the scale from 18*Cr-8*Ni, 25*Cr-12%Ni, 25%Cr-20%oNi, or 35%Ni-15%Cr steels in the investigations of Hickman and Gulbransen (8), Boren (9), Warr (10), Radavich (11), and McCullough, Fontana, and Beck (22). The last named investigators did report the presence of small amounts of Ni in their scales, as determined by chemical analysis, however. In the present investigation, a limited program was carried out to determine whether any substantial amount of Ni was present in the scales by means of a qualitative dimethylglyoxime test following Na2CO3 fusion of the scales. The results of these tests were negative, confirming that if Ni is present, at least in the scales tested, it is present only in very small amounts. The dimethylglyoxime tests could be carried out on only a few of the largest specimens of scale. C appears to be depleted in the metal near the surface, as will be discussed later, but it is difficult to see how C can influence the oxidation resistance of low C alloys such as those used in this investigation, since its oxides are gaseous and presumably escape in some manner through the scale. In summary, the process of oxidation in stainless steels, as well as in other alloyed materials, is accompanied by diffusion from the metal towards the metal-oxide interface of those alloying elements less noble than the base metal, and by an increase in concentration WADC TR 54-120 13

of alloying elements more noble than the base metal in the metal layer beneath the interface and probably some subsequent diffusion of these elements away from the interface. Summarizing the probable mechanism of intergranular oxidation, mobile 0 ions, reaching the metal-oxide interface by diffusion through the external scale, diffuse preferentially into the grain boundaries which intersect the surface, though appreciable 0 volume diffusion also occurs and some 0 diffusion may take place from the grain boundaries into the grains. Simultaneously, alloying elements present in the metal diffuse both toward the metal-oxide interface and at least at the lower temperatures toward the grain boundaries, where chemical reaction takes place between them and the 0 ions, producing an intergranular oxide which may be different in overall composition from the external scale. As time proceeds, the intergranular oxidation product at any given point may be consumed by the advancing external scale layer. Another effect of time which would be accelerated at high temperatures would be changes that might result in the 0 transport in the external scale due to its changing composition with time. Such changes could conceivably produce an oxide layer at the interface that grows principally by cation transport, thus reducing the free 0 ion concentration at the interface and preventing further grain boundary oxidation. On the other hand, however, an oxide layer that grows by principally cation transport, in steels at least, would imply a nonprotective type of scale, and under such conditions the scaling rate of the steel would become excessive. Penetration Measurements The technique by which the curves of number of intergranular cracks vs. depth below the oxide-metal interface were arrived at was explained in detail in the Procedure section of this report. The curves obtained by this method appear in Figs. 17 through 67. The basic type of penetration curve was found to be a curve which we have called a "decay" type, monotonically decreasing to zero fissures per in. with increasing depth. With increasing time and/or temperature, the mean depth of this curve tended to increase and a maximum in the curve tended to develop and move toward greater depths. The explanation for this behavior is twofold. First, the number of grain boundary sites for oxygen penetration is finite, and as the process continues, a state is reached where the existing penetration fissures have grown deep while relatively few new fissures are being formed. The second factor that tends to deplete the number of small fissures is the progressive surface oxidation. The behavior of the 1900*F penetration curve for type 310 alloy, heat X11338, Run 16 (lOOhr) as shown in Fig. 51 should be mentioned. This specimen showed some very deep penetration, photomicrographs of which appear in Figs, 99, 107, and 108. A number of the fissures were of greater depth than the six-row grid used in making the count of the fissures, and were recorded only as being beyond the grid. When the data were assembled, an arbitrary group interval equal to that of the six regular groups was assigned to these fissures. Since there was an appreciable number of the fissures, the resulting curve has an abnormal shape. If the deep fissures were subdivided into groups, the curve would no doubt be of normal shape. Since this particular specimen was the only instance in which the shape of the curve was affected by the excessively deep penetration, it was not considered worth while to alter the routine of measurement. WADC TR 54-120 14

In each metallographic mount, one untested specimen was included as a control. Originally, it was planned to use this control specimen to obtain a value of the original thickness, from which thicknesses of the tested specimens were to have been subtracted to obtain thickness losses. This proved to be impractical because of the small magnitude of the thickness losses, but it was found in the meantime that the control specimens gave an indication of the effect of surface roughness and of polishing on an otherwise undamaged surface. Surface irregularities gave rise to a background count of small artifacts on the control specimen which varied somewhat in different mounts from the same heat and were presumably indicative of the amount of error included in the fissure counts of the tested specimens. The fact that the penetration curves of some of the control specimens are the same order of magnitude as the curves for some tested specimens reflects the uncertainty in the count of the very small penetrations, as explained in the Procedure section. It is to be observed that the control penetration curves almost always decay at a rate much faster than tested specimens,which results from the fact that only rarely was anything seen on a control specimen that would be interpreted as being a fissure extending as deep as the second grid level. The maximum fissure depths, except for the one specimen from heat X11338 previously referred to, were never found to be over 0.005 in. in depth, and usually were much less. As is apparent from referring to Fig, 17 through 67, an analysis of the intergranular penetration characteristics of the alloys which was based on the detailed shapes of the penetration curves themselves would be a sprawling one which would be very difficult for the reader to assimilate. Therefore, the two parameters N, the penetration frequency, and X, the mean penetration depth, as referred to in the Procedure section, were adopted which summarize the essential features of the penetration curves. The behavior of these parameters with temperature are presented in Figs. 68 through 94. In several early runs (Nos. 5, 6, 8, 9 and 13), the air velocity through the furnaces was varied and the duration of the runs was held constant at 100 hr. These runs are summarized in Figs. 68 and 69, 75 and 74, and 78 and 79 for the 309 + Nb alloy and heats 64177 and 64270 of type 310 alloy, respectively. Since the humidity was also a variable in these runs, the results are inconclusive, and the penetration data are included only for the sake of completeness. Run 7, and also Run 25, were discarded because of temperature control difficulties during the oxidation period. One run, No. 9, was carried out to determine the effect of cold work prior to exposure. The penetration results appear in Figs. 70, 75, and 80. Cold work was introduced into the specimens by making a sharp angle bend in each specimen. After the specimens had been oxidized, quenched, mounted, and polished, penetration readings were made on the bent sections as well as the straight sections, the lengths of the bent sections being estimated by difference between the overall specimen length and the measured lengths of the straight sections. These results showed that there were fewer penetration fissures in the cold-worked areas, but that the penetration present was somewhat more severe than that in the straight sections. There was no catastrophic effect of the cold work, most of the residual stress having relaxed out WADC TR 54-120 15

before appreciable oxidation occurred. The case of a continuously applied stress during the oxidation period would be a different matter, however, and would likely lead to substantial increases in the penetration depth. The effect of time on penetration for the 509 + Nb, 64177, and 64270 alloys is presented in Figs. 71 and 72, 76 and 77, and 81 and 82. In general, the penetration became deeper and the fissures more numerous with time, although there was considerable overlapping. There does not appear to be any marked difference in the number of fissures occurring in the three alloys, and, if there is any difference in the penetration depth, it would be that the depth in the type 309 + Nb alloy is slightly less than in the two type 310 alloys. Figs. 83 through 94 show the effect of time an the penetration in the remaining six heats of type 310 steel. Heat X11306 (Figs 83 and 84) showed severe penetration, both from the standpoint of number and depth of fissures. Of the other five heats, all showed penetration frequencies of the order of those of the 309 + Nb, 64177, and 64270 heats. Except for the previously-mentioned 1900bF specimen of heat X11338 from Run 16 (Fig. 85), penetration depths were all somewhat lower that those found in the 309 + Nb, 64177, and 64270 heats. Of the five heats, the penetration depth of heat X45558 was least by a narrow margin, with heat X46063 next best. This ranking is probably not significant, since the differences among these heats are of the order of the experimental error. It is also noteworthy that in the six alloys, the magnitudes of the penetration frequencies at a given temperature did not usually increase in the order of increasing times. In the case of the penetration depths, however, magnitudes at a given temperature did in general increase in the order of increasing times. The values obtained for penetration depths appear to be more significant than the values for penetration frequency, since the depths are not as strongly influenced by the uncertainty in the count of the smallest fissures, Interface Movement Measurements The results of the measurements of the metal interface movements with temperature after 100-hr oxidation treatments are reported in Table 12. The method by which these measurements were made was outlined in the Procedure section of this report. As the results in Table 12 show, measurements made on specimens which had a powdery gray black surface appeared to be low relative to the values that would have been expected on the basis of the specimens of light gray appearance, on which it was possible to remove all of the scale. Warr (10) gives oxidation attack rates for type 309 and type 310 steels, gathered from the literature. At 1600~, 1800~, 2000~, and 2200~F, these figures are 0.004, 0.02, 0.05, and 0.06 in./yr for type 309 steel and 0.007, 0.02, 0.03, and 0.05 in./yr. for type 310 steel. When these figures are recomputed for 100 hr oxidation time, the results are somewhat higher than the experimentally-measured figures presented in Table 12, but are of the same order of magnitude. The highest measured value was about 0.0003 in., and the highest calculated value (type 309 alloy at 2000~F) was about 0.0006 in. No measurements were made on 10- or 30-hr specimens, but they would be expected to show correspondingly smaller interface movements. The size of the interface movements after 100-hr oxidation explains the reason that it was not possible to obtain consistent data by subtraction of measured thicknesses from the thickness of a control specimen —the interface movements were too small. In view of the variability of the apparent WADC TR 54-120' 16

interface movements calculated from the loss in weight measurements, it was decided to present the penetration data relative to the existing metal-oxide interfaces, rather than to make any attempt to correct for the interface movement. Visual and Magnetic Examination The results of the visual examintaion of both the specimens and the oxide scales are presented in Tables 3 through 11. A key to the abbreviations used is supplied in Table 2. Where more than one appearance is recorded for a specimen, the abbreviations are ranked in descending order of the proportion of the surface having that appearance. Great variations in the visual appearance of spedimens and scales were observed, and a fuller description of the more or less arbitrary classifications adopted for their systematization is in order. Powdery gray black - This type of specimen surface appearance, when it constituted the majority of the specimen surface, was accompanied by small quantities of free scale loosened in quenching. Although it was possible to scrape some of the powdery oxide off the specimen, a tough, continuous, adherent film always remained which could not be removed without damage to the underlying metal. Powdery gray black with orange substrate - If a specimen like that described above showed a red-orange color after scraping, it was given this classification. This color may have been partly due to Fe2O3, but in some cases, at least, was due to an interference film beneath the powdery gray black scale. The x-ray results are inconclusive on the former possibility. Shiny gray black - This appearance was not found as often as the powdery gray black and appeared to be associated with oxidation times of 30 hr or less. The adherent scale causing this appearance was very thin and often discontinuous, showing smooth light gray areas beneath, and had almost a metallic sheen. Black and scoriated - Occurrence of this appearance was unpredictable. Specimens having this appearance looked almost as though the oxide had once been liquid. The surfaces were checked, and individual segments of scale were plain and shiny. The scale was extremely dense, hard, and adherent. Light gray - This type of appearance was associated with a loose, flaky scale which often came off entirely on quenching. The appearance of the specimen was a very smooth light grays Light gray with interference colors - Specimens of the preceding basia appearance often showed orange, blue, or magenta interference films in swirling patterns. Light gray with black pits - In some cases, the light gray surface was pitted, the pits containing an adherent gray-black material. Often the light gray areas around the pits showed interference colors. Black pits with light gray - This classification differs from the previous one in degree, the pits being the predominant feature. WADC TR 54-120 17

Oxide appearances varied in color from metallic gray to dull black, and from fine powders to flakes sometimes almost as large as the specimen faces, The magnetism of the specimens and the oxides varied from those with no detectable magnetism to one or two specimens in which the magnetism was sufficiently strong so that the specimens could be suspended from the magnet and a few oxide specimens which could be made to follow a magnet beneath the sheet of paper on which they rested in the same manner as iron filings. It was found that specimen magnetism was usually concomitant with scale- magnetism. There was no definite segregation of magnetic components in the one to the exclusion of the other. It should be remembered that the presence of ferromagnetism does not necessarily imply the presence of bodycentered iron, since all of the transition metals have appreciable values of saturation magnetization and a number of their oxides (ferrites) are also magnetic. The scale magnetism may in some cases be due to islands of Ni, but no such islands were observed metallographically, and in some cases, the finely-divided oxide appears to be magnetic in its own right. Scale magnetism is not in general due to the presence of the magnetic oxide of Fe, Fe304, since in no case was this oxide found by x-ray analysis (although not all of the oxide specimens were x-rayed). Strong specimen magnetism is associated to a degree with the occurrence of a light gray specimen surface, In Table 12 are recorded the specimen and oxide appearances of three duplicate runs the purpose of which was to obtain weight loss data. These runs duplicate Runs 15, 16, and 17, and comparison with descriptions from these runs shows that there is considerable variability in the specimen and oxide appearances even in specimens which are as nearly duplicates as we could possibly make them. Furthermore, there seems to be little correlation of specimen and oxide appearance and magnetism with time or temperature of oxidation, or with scale composition as revealed by x-ray analysis. The only explanation of the visual and magnetic examination data presently at hand would be that they depend upon the rapidity with which the specimens are quenched from the high temperatures. Although in no case was the time between extraction of the specimen from the furnace and its submersion in the distilled water greater than 7 sec, it is true that some specimens stuck to the holding fixtures and had to be pried loose, which took more time than if the specimen did not stick. X-Ray Diffraction The scaling of high alloy steels has been extensively studied by a research group at Purdue University and their findings (1, 9 -10, 11) have proved most valuable in the interpretation of the x-ray powder patterns of scales which were obtained. For types 309 and 310 stainless steels, the Purdue group reports scales after 100 hr oxidation at temperatures from 1600~ to 2200OF consisting of Cr203, rhombohedral solid solutions of Cr203 and Fe203 having complete solid solubility, and a spinel the lattice parameter of which varied from 8.310 to 833550 kX in type 309 steel and from 8.323 to 8.429 kX in type 310 steel. The low-parameter spinel was identified as an Fe-Cr oxide, and the high-parameter spinel as an Mn-Cr oxide. Individual scales always contained the spinel except the type 309 alloy at 2200~F. All but a few of the scales contained one solid solution, and a few contained two solid solutions of differing compositions. Most of the scales contained either CrOs3 or a solid solution high in Cr203, Tyrpe 310 scales were found to contain two layers, WADC TR 54-120 18

The x-ray results obtained in the present program are presented as part of Tables 5 through 11, with a key to the symbols used being given in Table 2. Agreement with the above-noted findings is good in the sense that the compounds found were the same, the only exception being occurrence of Fe203 in this investigation, Since the accuracy in specifying the percentage composition of the Cr203Fe203 solid solution was estimated to be about + 10 atomic percent, compounds reported as being. Fe203 may well have been the same as FeOs3-rich solid solutions reported by the Purdue group. There were no clear trends of scale composition with time or temperature. It was found that the heats differed among themselves in the compositions of the scales formed under the same time and temperature conditions. Warr (10) reported similar behavior in duplicate specimens. Either Cr203 or a Cr203-rich solid solution was present in every scale analyzed. The spinel phase was widely distributed, and the lattice parameters of the spinel were 8.37 to 8.48 kX, with most of the values being from 8.41 to 8.43 kX, All of the scales analyzed were of the protective type, by Randall and Robb's criteria of the presence of Cr203 and high-parameter spinel. This includes the scales found on heat X11306, which showed much more severe penetration than the rest of the heats. It was observed that the two type 310 heats in which penetration was slightly less deep than the rest, heats X45558 and X46063, seemed to show less tendency towards Fe203 or solid solutions in their scales than the other heats. It seems likely that, since all of the scales were protective, not much improvement in penetration resistance can be hoped for by making small changes in the alloy compositions in order to obtain an exceptionally protective scale, Metallographic Examination The results of the metallographic examination of the specimens following testing are presented in Table 13, and include, in addition to penetration appearance, changes in micro-structure which took place in the alloys during the oxidation cycles. For convenience in presenting the data, the metallographic appearances of the intergranular penetration encountered were classified in seven arbitrary groups. Examples of six of these groups appear in Figs, 95 through 100, Type 0, which consisted of penetration fissures so small that they could not be definitely typed as belonging to any of the other six groups, was omitted. Referring to Table 15, the very shallow Type 0 penetration was found at low temperatures of oxidation, 1600~F and sometimes 1700~F after 100-hr exposure. At the other extreme, Type VI penetration (Fig. 100) occurred at the highest temperature, 20000F, especially at the. longer times. Type VI, which we have called "coalesced web" penetration, is characterized by a coalescence of the intergranular oxide into discrete particles spaced along the grain boundaries This effect is doubtless the result of relatively high atomic mobility at the high temperature. At intermediate temperatures, there is considerable variation in the precise appearance of the penetration. Some heats, notably heats 64177 and X46063, show Type III penetration consistently, a rough-edged web. The other heat show no specific trends, and there were no penetration characteristics that could be associated with composition. Figs, 101 and 102 show the typical appearance of the subsurface metal associated with penetration. In all of the heats except heat 64177, which showed penetration in the as-received condition (see Figs, 12 and 13), the specimen surfaces were initially smooth, and etching (Fig. 101) revealed the grain boundaries extending all the way to the surface. After oxidation, the etched structures resembled Fig. 102o As far, as could be seen, penetration took place along grain boundaries, but not along twin boundaries. A prominent characteristic was the WADC TR 54-120 19

occurrence of a layer in which no structure was revealed between the surface and a point beyond the ends of the visible penetration fissures. This layer gives the appearance of having an altered composition from that of the original alloy, The region is likely high in oxygen content, though not high enough for oxide formation, Consequently, there is a possibility that the layer is decarburized, Furthermore, since no Ni was observed in the scales, the region may be Ni-rich, Below the subsur~face layer, the typical grain structure and carbide precipitation were observed, Figs 103 through 106 show the appearance of heat X11306 in the etched condition after oxidation. Heat X11306 behaved in a unique manner in two respects: its penetration was far greater, both in depth and number of fissures, than any of the other alloys; and its metallographic appearance differed from that of the other heats. Large quantities of precipitate appeared throughout the matrix following testing which could be observed without etching. It was also interesting to note that recognizable penetration fissures disappeared after 100 hr at 2000~F, as shown in Fig. 105, but were still present after 10 hr at the same temperature and at all times at lower temperatures. Another example of penetration which looked unusual was the penetration occurring in the 100-hr specimen from heat X115538 tested at 1900~F, which was abnormally deep and of the developed web type. Photomicrographs of this specimen appear in Figs. 99, 107, and 108. Characteristics of the penetration included the presence of islands of intergranular material. Small particles also appeared in the metal adjacent to the intergranular islands, and seemed to be associated with them. More general matrix precipitation was revealed by etching, No explanation can be offered at this time for the occurrence of this unique type of penetration in this particular specimen and, to a lesser extent, in the specimen of the same heat tested for the same time at 2000~F. In the heat X45558 specimen tested for 100 hr at 20000F, several examples of a deep pit type of penetration were noted, as shown in Fig. 109. This type of attack may only represent points where grains had fallen out as a result of being encircled by oxide, and in any event, does not appear to be of great significance, since these pits were not abnormally deep. Several other structural features were observed which it would be appropriate to note here, although it is not known whether they have any appreciable effect on the penetration. The first two of these are the occurrence of delta ferrite and sigma phase. The temperatures at which these constituents were observed are tabulated in Table 13. Identification was made by means of Marakami's reagent. This reagent, when used concentrated, as described by Burgess and Forgeng (21), colors sigma light blue and ferrite yellow. Representative photomicrographs are presented in Figs. 110 and 111. It appears that all of the heats tested were subject to one or the other of these constituents, but that there was only one instance (heat 64177) where even a tentative identification of both constituents could be made in the same heat. Neither of these constituents was found in the region near the metal surface in any case, but tended to occur throughout the remainder of the cross section. No features of the observed penetration could be related to the presence of either delta ferrite of sigma phase. A check was made to determine the effects of the 30-hr oxidation treatments on the grain sizes and carbide distributions in the alloys. An electrolytic chromic acid etching treatment was used, and the results appear in Table 15, On series of WADC TR 54-120 20

specimens oxidized for 100 hr, heat X11338, was etched to reveal grain size, and representative photomicrographs at 1600o and 19000F appear in Figs. 112 and 113 at 250X. Reference is also made to the as-received structure as shown in Fig. 5 at 100X. The appearance of the specimens at 17000 and 18000F was the same as at 16000F, but at 19000F, agglomeration and probably some solution of the carbides was noted, along with a slight increase in grain size, At 2000~F, grain boundaries could not be revealed by the electrolytic chromic acid etching treatment, This is taken as meaning that all of the carbides in the grain boundaries had gone into solution, Precipitation, agglomeration, and re-solution of carbide with increasing temperature was the typical behavior of all the materials, The temperature at which the grain boundaries could no longer be distinguished varied somewhat, and was lowest in heat X11306 by some 2000F. As the photomicrographs show, however, (Figs. 103-106) a second phase in the matrix remained. It was true in all of the heats that all of the second phase or phases did not go into solution within the 100-hr period. Fig. 114, taken using oblique light, shows in a striking manner the distinct surface metal layer mentioned previously. In fact, a second intermediate layer etching darker than the matrix is also apparently present. The increased thickness of the surface layer in the vicinity of a grain boundary is also clearly visible, though the grain boundary itself is not visible for some distance further back into the metal. This photomicrograph illustrates the greater activity of the grain boundary relative to the bulk of the material even prior to the formation of a penetration fissure in the boundary. When certain of the specimens were originally etched in chromic acid, a structure was developed consisting of many long parallel stringers which appeared black under the microscope. Examination under oblique light at high magnification revealed that these stringers were troughs gouged in the metal surface by severe overetching, When the specimens were repolished and etched using a very low current density, a structure consisting of bands of carbide particles was revealed. Careful etching of the as-received heat X45558 and examination under oblique light showed the existence of depressions (Fig. 115), accompanied by several pits along the depressions where particles had etched out. These depressions are interpreted as being concentration gradients in the matrix. Following exposure for 10 hr at 1600~F, carbide particles had precipitated as shown in Fig. 116, and the depressions no longer appeared on etching. Figs. 117 and 118, taken at lower magnification, show that the bands of carbides were sometimes associated with a refinement in grain size, and sometimes not. It was observed that the bands tended to disappear, presumably by diffusion, with increasing time and/or temperature. Cross section specimens from all nine of the heats were examined to determine whether banding could be observed, A second series of cross section specimens at right angles to the first series were prepared and examined, since it was believed that the stringers would lie in the rolling direction and would hence be revealed in one series of specimens or the other if they were present at all. Results of this examination appear in Table 13. The particles are identified as being carbides because the bands were not found near the surfaces on the oxidized specimens. It was reasoned that, if the particles were either oxides or nitrides, they would tend to be concentrated at the surface, rather than away from it. No penetration effects were observed that could by attributed to the presence of the carbide bands. WADC TR 54-120 21

SUMMARY AND CONCLUSIONS 1. Specimens from eight heats of type 310 stainless steel and one heat of type 309 stainless steel were oxidized in moving air at temperatures of 16000, 17000, 18000, 19000, and 2000'F for times of 10, 30, and 100 hr, Following the oxidation periods, specimens were quenched into distilled water, 2. Visual and magnetic exa.minations were made on the oxidized specimens and the oxide scales, It was found that the visual and magnetic characteristics of the specimens and scales had no relation to the resistance of the alloys to intergranular penetration or to the composition of the scales, 35 Specimens were mounted and metallographically polished, and counts of the number and depth of the intergranular fissures were made on each specimen, Curves were constructed plotting number of fissures vs. depth. For the purpose of analysis of the data, these curves were described in terms of two parameters, N and X, the total number of fissures and the mean depth of fissures. It was found that N was not a particularly useful parameter, due largely to uncertainty in the numbers of very shallow fissures. The parameter X was less affected by this uncertainty, however, and it was determined that the depth of the penetration fissures increased with time and with temperature. In addition, there was no appreciable difference in penetration characteristics among the heats tested, except in heat X11306. Since the heats were selected to show a wide spread in the percentages of alloying elements within the specification: limits, it can be concluded that minor modifications in alloy content have no effect on intergranular penetration. There was no appreciable difference between the depth of penetration in the type 309 + Nb alloy and the type 310 alloys, 4, Measurements were made of the rate of external scaling, or volume oxidation. These measurements showed the rates of scaling to be small and of the order of those reported by other investigators for the 25loCr-20%Ni analysis, and can be expressed as a recession of the metal surface amounting to 0,0005 in. or less per 100: hr. 5. X-ray powder patterns were taken for many of the oxide scales. Phases composing the scales always included Cr203 or a solid solution of Cra03 and Fe2Os3 Most of the scales also contained a high-parameter spinel (8,43 kX) believed to be essentially MnCr204, No separate phases identifiable as containing Ni were observed in the scales by x-ray diffraction, nor was Ni found to be present by qualitative chemical analysis of a limited number of specimens of scale. All of the scales analyzed can be classified as being protective, since they contained Crz03, and there is little chance therefore, that minor compositional changes would result in scales which would give better resistance to intergranular penetration. This is in agreement with the conclusions reached on the basis of the measurements of penetration depths. 6. The effect of prior cold working was to increase slightly the depth of the intergranular fissures. The conclusion is drawn that severe stressing of the metal is not particularly harmful provided that this stressing is done at temperatures below those at which oxidation of the steel takes place and that the stressing is discontinued before elevating the temperature. WADC TR 54-120 22

BIBLIOGRAPHY 1. Yearian, H. J. and Radavich, J. F., Dept of Physics, Purdue University, Unpublished data. 2. Taylor, A. and Sinclair, H. B., Proc. Phys. Soc, 57 (1945), 126. 3. Henry, N. F. M., Lipson, H., and Wooster, W. A. The Interpretation of X-Ray Diffraction Photographs, Macmillan, 1951 (London). 4, Upthegrove, C. and Murphy, D. W., Trans. Amer. Soc. Steel Treat., 21 (1933), 93. 5. Siebert, C. A. and Upthegrove, C., Trans. Amer. Soc. Metals, 23 (1935), 187. 6. Kubaschewski, 0. and Hopkins, B. E., Oxidation of Metals and Alloys, Academic Press, 1953 (New York). 7. Davies, M. H., Simnad, M. T., and Birchenall, C. E., Trans. Amer. Inst. Metall. Engrs., 191 (1951), 889 and J. Met, 5 (1953), 1250. 8. Hickman, J. W. and Gulbransen, E. A., Trans. Electrochem. Soc. 91 (1947), 605. 9. Boren, H. E., Jr., M. S. Thesis, Dept. of Physics, Purdue University, August,1950. 10. Warr, R. E., M. S. Thesis, Dept, of Physics, Purdue University, August, 1951. 11. Radavich, J. F., Ph. D. Thesis, Dept. of Physics, Purdue University, May, 1953. 12. Hauffe, K. and Block, J., Z. Phys. Chem., 198 (1951), 232. 13. Lustman, B., in Amer. Soc. Metals, Metals Handbook, 1948 ed,, 223. 14. Randall, E. C., M. S. Thesis, Dept of Physics, Purdue University, 1950. Robb, P. F., Dept. of Physics, Purdue University, Unpublished research. 15. Turnbull, D., in Amer. Soc. Metals, Atom Movements, 1951, 129. 16. Yukawa, S. and Sinnott, M. J,, Dept. of Chem. and Metall. Engrs., University of Michigan, Unpublished research, 17. Baeyertz, M., Trans. Amer. Soc. Metals, 24 (1936), 420, 18. Gulbransen, E. A., Phelps, R. T., and Hickman, J. W., Ind. Eng. Chem. (Anal. Ed.), 18 (1946), 391. 19. Gulbransen, E. A. and McMillan, W. R,, Ind. Eng, Chem, 45 (1953), 1734. WADC TR 54-120 25

20. Radavich, J. F. and Yearian, H. J,i Dept. of Physics, Purdue University, Undated report. 21, Longo, T. A., M. S. Thesis, Dept, of Physics, Purdue University, August, 1953. 22. McCullough, H,. M., Fontana, M. G., and Beck, F. H., Trans. Amer. Soc. Metals 43 (1951), 404. 23. Burgess, C, 0. and Forgeng, W. D., Trans. Amer. Inst. Metall. Engrs. 131 (1958), 277. WADC TR 54-120 24

0 TABL;E 1 COMPOSITION OF STAINLESS STEEL STOCK, WEIGHT PERCENT I ro Heat No. C Ni Cr Si Mn P S Mo Cu Co W Nb 309 + Nb 0.08 15,39 22,64 o046 2,531 0,012 0.008 - - 0.82 64177 0.15 16.96 24,03 0.55 0.42 oo018 0.008 0,055 0,15 0,Dl <0.01 64270 0.12 19.14 22.30 0,.43 0.50 0.025 0,008 0.042 0,10 0,01 <0.01 h X1306 0,07 20.33 25.48 0.78 1.41 0,023 0,0135 -- - X11558 0.085 20.08 24.28 0.83 1.55 0.029 0.029 0,14 0,25 - - - X27258 0,073 20.81 24,64 0.75 1.58 0.025 0.008 0,15 0.530 X45558 0.069 21,.56 24.71 0.58 1.54 0.025 0.011 0.12 0.27 X46063 0.059 20.67 24.28 0,31 1.57 0,024 0.014 0.12 0,25 - - X46572 0.089 19.48 24,55 0.67 1.60 0.021 0.022 0.14 0,534

TABLE 2 EXPLANATION OF SYMBOLS USED IN OXIDATION DATA TABULATIONS SPECIMEN APPEARANCE: PGB - Powdery gray black PGB w/ OS - Powdery gray black with orange substrate SGB - Shiny gray black B and S - Black and scoriated BP w/ LG - Black pits with light gray surface (mostly pits) LG - Light gray surface LG w/ BP - Light gray with black pits (mostly light gray) LG w/ IC - Light gray with interference colors MAGNETISM: ND - Not detectable W - Weak M - Medium S - Strong OXIDE APPEARANCE: Quantity: N - None S - Small amount M - Medium amount L - Large amount Flake Size: P - Powder SF - Small flake (<1/32 in.) MF - Medium flake (1/32 - 1/8 in.) LF - Large flake (>1/8 in.) Color: G - Gray B - Black GB - Gray-black OXIDE COMPOSITION: S.S. xQ - Cr203- Fe203 solid solution with Q j: 10 atomic percent Cr203 HPS - High-parameter spinel a = p - Lattice parameter of spinel in kX units * - Lattice parameter determined from two-point extrapolation Insuf. - Insufficient oxide for x-ray pattern W - Weak pattern M - Medium pattern S - Strong pattern WADC TR 54-120 26

T A 3 L E 5 RESULTS OF OX I D A T I ON T E S T, T Y E 5 09 + Nb ALLOY 0.08% C, 15.39% Ni, 22.649 Cr, 0.82% Nb, 0.46% Si, 2.31% Mn, 0.012% P, 0.008% S Run Duration, Air Velocity, Dew-point, enetration Temp., Specimenide Scale Fo r f/in F Curve,.No. of Mean Fissure Flake Composition No. hr ft/min ~F Fig. No.'F Appearance Magnetism ssure/in. Dth in 3 Amount Color Sie Magnetim -ray) Fi No. Ffssures/l.n DePth in.xlO Size x-ray 5 100 40 < -20 17 1600 I w/ BP - 222. 0.455 M GB SF D IS w/ IC 1700 PGL C 1011.1 0.692 L G LF D - 1800 Gw/ -C 699.8 1.525 M GB LF ND 1900 LG w/ IC - 719.0_ 1.700 L B MF M - 6 100 20 - 20 1600 LG w/ IC - 9-.0 0.332 M GB ND - 1700 LG w/ IC - 480.9 0.368 M GB LF ND 1800 LG w/ BP - 723.3 0.403 S GB MF M POB 1900 LG - 869.1 1.307 S GB P M 8 "So 50 S0215 25 6000from +15 LG w/ BP 8 100 30 to +20 -23 1600 LG w/ BP t435.9 0.371 M B P ND 1700 B and S - 423.3 0.344 L GB MF M 1800 B and S 321.6 0.486 L GB MF M 1900 LGwv/ IC - 386.8 1.117 L GB MF M - Straight Section 9 100 30 to - 60 26 1600 PGB - 611.4 0.544 S GB P M 1700 PGB - 5+7.5 0.513 S GB P M 1800 wI - 837.8 0.936 S GB MF M 1900 LG - 852.6 1.499 L GB MF S Curved Section 29 1600 - - 376.9 0.683 - - - - 1700 - - 553.3 0.727 - - - 1800 - - 438.6 1.605 - - - - 1900 - - 623.0 1.319 - - - - - from'45 10 50 30 to 40 32 Control - - 280.6 0.343 - - - - - to +10 BP w/ L1 1600 BP w LG M 938.0 0.459 M GB P-SF M PGB 1700 PB M 772.2 0.774 L GB MF ND 1800 LG w/ IC S 1016.0 0.369 L GB LF M 1900 LG w/ IC S 859.6 0.572 L G LF M - from -45 to -22 1 30 30 to -22 35 Control - - 1066.1 0.361.. 6 PGB - 862.4 0.408 S B P ND Insuf. 1600 10w NP LG w/ BP 1700 O B - 1288.0 0.487 S B P-SF M Insuf. Ioo w/ w Cr203(S) 1800 Lw - 1023.4 0.455 L GB MF S Pe2O,(M) HPSW Cr203 (S) 1900 G w/ IC - 1071.6 0.567 L GB MF S FO (W) 10 w/ BPUN a=8.43* from -22 12 10 30 to -16 38 Control - - 274.1 0.320 - - - 1600 PGB W 318.8 0.320 S B SF W Insuf. LG w/IC 1700 PGB W 818.4 0.323 S B P ND Insuf. 18o00 P/ M 900.6 0.330 M GB P-SF S r203() LG BPS(M) 1G w/ BP 1900 PGB M 719.6 0.365 M B P-SF M S.S.x50 13 100 40 from 0 41 Control - - 113.0 0.320 to +20 1G w/ IC 1600 BP w/ LG ND 785.4 0.336 L B SF ND LG w/ BP PGB w/ OS SOB 1700 PSw/ M 831.4 0.336 L G MF S LG w/ BP 1800 LG W/ BP M 778.3 0.580 L G LF S 1900 LG / IC M 775.7 0.449 L G LF S _ 14 77 30 o -45Control - - 340.3 0.329 - to -35 LG w/ BP 1600 B and S W 409.4 0.334 S B P M PGB LG w/ BP 1700 PGB W 949.0 0.328 S B P M B and S PGB B and S 1800 B ad SC W 1125.7 0 510 M GB P-SF W BP w/ LG B and S 1900 PGB M 851.2 0.843 L GB P-SF S BP w/ LG 15 100 30 from 5 47 Control - - 146.6 0.339 - - - to -40 Cr203(S) 1700 BP w/ LG M 1242.6 0.363 S B P M Fe205(S) KPS(W) Cr203(S ) Fe203(M) 1800 LG w/ IC S 475.2 0.529 L G LF S PS(M) a8. 45* cr203(S) 1900 LG w/ IC M 504.5 0.496 L G LF S HPS(M) a=8. 45 PGB 26 10 30 -38 38 2000 LG S 1456.2 0.562 M GB MF 8 PGB w/ OS 27 30 5 3to- 35 2000 BP w/ Lp M 1100.6 0.686 S B P M P -120 27 WADC TR 5I-120 27

TABLE 4 RESULTS OF O X I D A T I ON TESTS, TYPE 310 ALLOY, HEAT 6 4177 0.1%5 C, 16.96% Ni, 24.05% Cr, 0.55% Si, 0.42% Mn, 0.018% P, 0.00% S, 0.03355 Mo, 0.15% Cu, O.O01 Co, O0.01O% W Penetration Specimen Oxide Scale Run Duration, Air Velocity, Dew-point, Peoetration Temp., Specimen Oxide iale No. hr ft/min OF Curve, F Appearance Magnetism No. of a. Depth, F O Amount Color Flake Magnetism Co ayposition Fig. o. ize(x-ray) 5 100 40 <-20 18 1600 PGB - 530.7 0.836 M GB P-sF W 1700 PGB - 222.6 0.817 8 GB P - 1800 PGB - 130.9 0.630 M GB P-SF ND - 1900 BP w/ LG - 536.6 0.856 S GB P ND - 6 100 20 - 21 1600 PGBw/ OS - 341.6 0.743 8 GB P D - 1700 PGB w/ OS - 451.5 0.730 S GB P ED - 1800 PGB w/ OS - 495.6 0.705 S (0 P W - 1900 PGB - 440.0 703 8 GB P ND - LG 8 00 0 fro +15 24 1600 G/ - 438.4 0.749 S GB P M - to +20 LG w/ BP 1700 PGB - 470.5 0.752, B P M - 1800 LG BP - 435.2 0.983 M GB MP - 1900 LG w/ - 280.4 0.730 S B P M - 1. w/ IC Straight Section 9 100l 0 from <-60 27 1600 PGB - 421.2 0.773 M B P M - to -30 1700 PGB - 556.2 0.883 8 B P M - 1800 PGB - 443.2 0.771 8 B P W - 1900 PGB - 461.7 0.709 s GB P M - Curved Section 30 1600 - - 194.5 1.032 - - - - - 1700 - - 205.5 0.72 - - - - 1800oo - - 227.6. 931 - - - - 1900 - - 180.2 o0.82 - - - - 10 50 30 from 450 33 Control - - 525.7.455 - - to +10 1600 PB ND 589.0 0.770 8 B P ID 1700 p / LG 675.6 0.697 s B P-SF M PGB 1800 LG w/ IC ND 662.9 0.740 M B P-F S - BP v/ LG 1900 PGB ND 883.9 o.812 S B P M - 11 30 30 r 45 56 Control - 460.2 0.547 - - - to -22 1600 BP w/ L - 757.0 O.548 S B P ED Insuf. PFB 1700 P/B 100 4.2 0.556 8 B P M Insuf. 1800 BP v/ - 666.4 0.341 8 B P ED Inauf. 1900 BP w/ L - 882.1 0.505 8 B P ED Insuf. 12 10 30 ro -22 39 Control - - 447.6 0.428 - - to -16 1600 POB v/ 08 ND 490.9 0.563 8 B 8F W Inauf. PGB 1700 LG w/ BP ND 548.5 0.551 8 B P M Insuf. BP w/ LG LGC w/ BP s.s.x50(S) 1800 LG w/ IC ND 572.7 0.538 M B P-SF 8 S.S.x25( PGB Cr2O, (s) 1900 LG v/BP ND 655.2 0.586 M B P-SF 8 Fe20o (N) LG w/ IC HP8 (M) _ ______________________________________a " 8.42* 13 100 40 from 0 42 Control - - 608.3 o.468 - - - to +20 1600 PGB w/ 8 ND 1099.1 0.528 8 B P M 1700 ND 1076.6. 547 8 B P M 1800 POB w D 991.5 0.787 M GB P-SF M Lo w/ IC 1900 PGB w/ OS ND 1222.3 0.698 L B 8F N LGX w/ BP 14 77 0 from -45 45 Control - - 217.8 0.408 - to -55 1600 PGB/ M 544. 0.683 8 B P W BP w__w 1700 PGB W 822.1. 564 8 B P M 18o00 W 625.2 0.648 B P M - 1900 PBP L W 671.6 0.652 8 B P M FOB from -55 15 10 0 30 o 48 ontrol - - 634.6 0.557 - - - 1600 PGB M 895.6 0.710 S B P M Cr 1700 PGB M 1147.5 0.802 8 B P M T #. 1800 BP L 1275.0 0.842 8 B P M 8.S.x50 1 900 / N W 8M 1283.9 0.766 M B P M SS..X50 26 10 0 -3 59 2000 PGB 90.2 0.6 B P M - 27 30 30 to -3J 36 200 C 8 1281.3 0.774 L G M S WADC TR 54-120 28

TABLE 5 RESULTS OF OXIDA T IO N TESTS, TYPE 310 ALLOY, HEAT 64270 0,12% C, 19.14q Ni, 22.30% Cr, 0.435 Si, 0.50% Mn, 0.025% P, 0.008% S, 0.042% Mo, O.10% Cu, 0.01% Co, <0.01% W,Penetration _ Secimen I Oxide cale Run Duration, Air Velocity, Dew-point, enetration Temp., Specimen Oxide Scale o. hr ft/min F Fig. o. Apearance Magnetism NFi fi eath in. 3 Amount Color e Maoneti. - f.ig. No. Fi.sure./in, Depth, in.x103 (x-r.. 5 100 40 (-20 19 Control - - 56.2 0.508 1600 PGB - 269.4 0.605 B P ND - PGB 1700 LG - 295.3 0.573 L GB P-SF M LG w/IC 1800 PGB - 147.3 ~.593 8 GB P-SF ND 1900 LG w/BP - 131.4 0.666 8_ B P ND - 6 100 20 - 22 1600 PGB w/OS - 225.9 0.386 8 GB P w - 1700 PGB w/OS - 370.4 0.439 S GB P-8F 8 - 1800 PGB w/OS - 323.8 1.127 s GB P ND - 1900 PGB v/OS - 223.3 1.063 8 GB P W - 8 100 30 om +20 25 1600 rPB - 328.0 0.387 S GB P-SF M to+20 1700 PGB - 373.2 0.345 s B P M - 1800 PGB - 460.6 0.411 B P M - PGB 1900 w/BP - 232.0 0.416 s B P M - PGB w/OS Straight Section 9 100 30 from<-60 28 1600 PGB - 216.0 0.370 S GB P M to -30 1700 PGB - 350.3 0.431 8 B P M - 1800 PGB - 284.1 0.469 M B P M 1900 PGB - 305.6 0.503 M B P M Curved Section 31 1600 - - 68.8 0.389 - - - 1700 - - 109.5 0.808 - - - - 1800 - 93.1 0.511 - - - - - 1900 - 108.9 0.511 - - - - 10 50 30 from -45 34 Control - 442.5 336 - - - - to +10 1600 PGB ND 500.4 0.390 S B P S PGB 1700 BP w/LG ND 445.4 0.482 M B P-SF S LG w/IC B and S 1800 BP w/LG ND 604.7 0.808 L B P-SF S LG w/IC 1900 PGB ND 784.3 0.615 L B P-SF S 11 30 f0 ro-45 37 Control - - 106.0 0.351 - - to-22 PGB 1600 BP v/LG 533.4 0.332 S B P M Cr203 LG w/IC PFB 1700 PGB /IC 579.2 0.359 S B P-SF M Insuf. o1800 UBP- 597.0 0o.485 S B P-SF S Insuf. BP1900 w/IC. y1900 BP C w - 791.3 o.777 M B P-SF S Insuf. BP w/nG 12 10 r0 from -22 12 ~10 30 fro -22 40 Control - - 242.4 0.324 to -16 1600 PB w/OS W 457.9 0.345 S B P-SF M Insuf. 1700 PGB ND 453 3 0.373 8 B P W Insuf. 1800 PGB ND 945.0 0.359 S GB SF W Insuf. PB 1900 GB W 845.0 0.376 M B P-SF M S.S.x50 13 100 40 from -30 43 Control - - 380. 028 - - - to +20 1600 PGB w/08 ND 660.1 0.355 s B SF M 170 PGB w/OS ND 739.4 0.430 M B P M P&B w/08 1800 andS ND 769.1 0.509 M B P M LG w/BP LG v/IC 1900 FB andS W 967.5 0.603 L B P-SF M - PGB w/OS 14 77 30 from -45 46 Control - - 361.9 0.323 to -35 1600 PGB W 940.2 0.369 s B P M - 1700 PGB M 986.4 0.362 S B P M - 1800 G /B W 870.2 0.499 8 B P M - 1900 PGB ND 1104.1 0.426 S B P M - v/LG 15 100 30 from 55 49 Control - - 354.1 0.325 - - - to -40 1600 PGB M 621.0 0.374 8 B P M S.S.x50 LG L- s.s.x5o(s) 1700 and S 8 752.0 0.784 L B M S 5 1800 PGB M 927.9 0.508 S B P M Cr20((S) 1900 BP /L w 1154.7 0.512 M B P w ss.x5(s) eo 203?(W) 26 10 30 -38 40 2000 PGB W 438.2 0.467 S B P M B and S 27 30 30 from -46 BP w/LC to27 -8 37 2000 PGB S w/ M 1411.8 0.794 S B P 8 B and S WADC TR 54-120 29

TABLE 6 RESULTS OF OXIDATION TESTS, TYPE 310 ALLOY, HEAT Xi 1 0 6 0.07% C, 20.33% Ni, 25.48% Cr, 0.78% Si, 1.41% Mn, 0.0253 P, 0.013% S Penetration Specimen Oxide Scale Run Duration, Air Velocity, Dew-point C, Curve, Temp., ean Fissure Flaeposition No. hr ft/min ~F urve ~F Appearance Magnetism Fissures/in. Depth, in.xlO3 Amount Color ize Magnetism (x-ray) ft/s Pig. No. I, I Fissures/in. D in.x10 (x-ray) 16 100 30 from 50 Control - - 318.0 0.326 - - - to -42 Cr203 (S) 1600 LG w/ BP S 1106.1 0.421 M B P-SF ND Fe2O3(M) HPS IM) Cr203(S) 1700 LG w/ BP S 620.0 1.560 L G ND S.S.x5(M) HIPS (M) a=8.48* Cr20j (S) 1800 LG w/ IC W 631.6 1.631 L G LF ND PS S) a=8.42 PCGB w/ OS HPS (S) 1900 BP w/ LG W 1791.0 2.099 M B P ND a=8.42 B and S Cr203(M) 21 30 30 rom 50 56 Control - - 219.6 0.322 to -42 16oo SGB M 1393.3 0.407 S B P M Insuf. 1700 Lw IC M 600.0 0.456 M GB SF W S.S7(S) LG w/BP HPS (M) 1800 PGB w/ OS M 1340.0 0.921 S B P M C2S) Cr2QCS) 1900 LG w/ BP M 1305.8 1.585 M B P-SF ND HPS (M) LG w/ IC a=8.44* 24 10 30 from -52 65 Control - - 178.8 0.323 to -44 LG 1600 LGw/ BP M 548.0 0.343 M B P-SF M PGB LG w/ IC 1700 LG w IC W 667.5 0.641 M GB P-MF M LG wl BP 1800 L0 w/ IC M 1031.2 0.839 M B P-SF M LG W/ BP 1900 LG w B M 1148.8 1.247 M B P-SF W LG w/ BP 26 10 30 -38 65 2000 PGB M 2590.5 0.716 S B P W - from -46 SF 10 27 30 30 56 2000 Bw/ M 2589.7 0.998 M B P-SF S to -38 B and S 28 100 30 fom _0 50 2000 BP w/ LG M 2134.5 0.962 S B P M to -40 TABLE 7 RESULTS OF OX I D A T I ON T E S T S, TYPE 310 ALLOY, HEAT X 1 3 3 0.085% C, 20.08% Ni, 24.28% Cr, 0.83 Si, 1.53 Mn, 0.029 P, 0.029 S, 0.25 Cu, 0.14%Mo Penetration S ecimen Oxide Scale Run Duration, Air Velocity, Dew-gTint, ur, ep., pp No of Meon Fissure F ake Composition No. hr ft/min Fig. No. F ppeFissures/in. Depth, in.xl t Color Size agnetism (x-ray) 16 100 30 from 52 51 Control - - 36.6 0.404 BP w/ LI HPS(M) 1600 Po B M 592.1 0.381 M GB M ND a=8.37* L, S.S.x25(41 S.S.x85(S) 1700 L u/ IC M 363.3 0.592 L G MF ND HPS(M) S.S.x95(S) 1800 L uw/ IC M 202.9 0.323 L G MF ND HPS(M) S~ ad0S.S. x90(6 ~~~~~~~~LB and S ND HPS(M) 1900 LG w BP M 1364.7 1.765 L B P-SF M PS( tL u/ IC 21 30 30 frto -42 57 Control - - 378.3 0.325 - - - - ( StoOS -42 ~~~~~~~~~~~~~~Cr203(S) 1600 SGB W 845.1 0.324 S B P MN PS(M) LG a=8.44* 1700 M 1030.1 0.335 S B P S 2P(M) 1700 LG lPS(M) LJ w/ IC Cr203(S) 18 M 100 5.7 0.54 M GB MF ND HPS(S) LG w/ BP a=8.42* Cr203(S) a=8.44 1900 BG w/ IC M 118.2 0.857 L B M W _ 24 10 0 from -2 66 Control - - 762.7 0.344 - - - 21 30 30 to -44 SOB 1600 I M 1165.1 0.365 S B P M o, W/ IC SGB 1700 to w/ IC M 1266.1 0.417 S B P-Si M LG!/ BP SGB 1800 LG w/ IC M 1417.5 0.420 S B P-SF M BaP t LG 1900 LG w IC M 1186.4 0.593 L B MF W 26 10 30 66 2000 ls/ IC S 699.2 0.734 L GB HF S ^ ---- ~ ^ 58 ^ 220000Q 3 ND 1327.' 4 0.869 S B P S from -50 28 100 30 Do: 51 2000 Ba/to M 1425.4 0.707 5 B P 5 - WADC TR 54.-1 2 50

TABLE 8 RES U L T S OF O X I D A T I ON TESTS, TYPE 3 10 ALLOY, HEAT X 2 7 2 5 8 0.0735 C, 20.81 Ni, 24.645 Cr, 0.75% Si, 1.58% Mn, 0.025, 0.008% S, 0.30g Cu, 0.15% Mo Penetration Specimen Oxide Scale Run Duration, Air Velocity, Dew-point, Temp., Ho. of Mean Fissure Flake Composition No. hr'ft/mineF FCurve, F Appearance Magnetism Fisures/in Depth, in.xlO3 Amount Color Sie Magnetism x-ray) 16 100 30 fom -52 52 Control - - 501.7 0.442 - - PGB 1600 BP / M 13704 05 B M 1Cr203(S) S.S.x25(4 IG w/ BP Cr203(S) 1700 PGB w/ OS W 779.3 0.566 L B MF ND HPS(M) PGB a=8.43* PGB w, OS 1800 BP w W 1010.5 0.661 S B P-SF ND iS LO w/ M a=HY3 LG w IC LG w/ BP LG w/ IC C S 1900 BP v/ LG W 1235.4 0.665 M G P-SF ND C B and S LG 21 30 0 from 58 Control - - 463.8 0.321 to -42 SGB Cr203(S) 1600 PGB W 801.8 0.321 S B P M Fe20 (M) LG HPS(W) SGB 1700 IGr W 1203.5 0.320 S B P M Insuf. BP w/ LG Cr203(S) 1800 PGB w/ OS W 900.9 0.84 S B P M Fe203(M) 1IG HPS(W) PGB w/ OS S.S.x75( 1900 B and S M 894.1 0.527 S B P M SFepO(M) 24 10 30 from -52 67 Control - - 985.8 0.329 to -44 SGB 1600 G / M 1214.6 0.347 S B P-SF S LG w/ IC PGB w/ OS 1700 LG w/ BP ND 1300.0 0.584 S B P-SF S LG w/ IC 1800 BP W/ LG W 1147.8 0.410 M B P-SF W LGw IC 1900 LG w/ IC S 911.1 0.343 M B MF W - BP w/ LG PGB w/ OS 26 10 30 -38 67 2000 PGB S 1269.7 0.526 M GB MF S - B and S - from -46 2000 BFBP w/ LG 27 30 30 from -46 58 2000 PGB w/ OS M 1214.8 0.800 S B P S _____________________to -30____8 _B and S BP w/ LG 28 100 30 from -5 52 2000 B and S W 956.2 0.645 S B P S ___________" "PGB w/ OS TABLE 9 RESULTS OF O X I D A T I ON TESTS, TYPE 310 ALLOY, HEAT X 4 5 5 5 8 0.069% C, 21.56% Ni, 24.71% Cr, 0.58% Si, 1.54% Mn, 0.025% P, 0.011% S, 0.27% Cu, 0.12% Mo Penetration Specimen Oxide Scale Run Duration, Air Velocity, Dew-point, urve Temp. No. of Mean Fissure Flake Composition No. hr ft/min F Figv. No. F Appearance Magnetism Fissures/in. Depth, in.xlO Amount Color Size MBgetism (-ray) 17 100 30 fro -55 53 Control - - 955.0 0.352 to -38 PGB 1600 LG M 748.1 0.338 S B P ND Insuf. LG w/ IC PGB 1700 BP w/ LG M 1145.6 0.337 S B P W Insuf. LG w/ IC 1800 BP w/ G M 1303.3 0.486 S B P S Insuf. PB w/ OS PGB HPS (S) 1900 BP / LG M 1124.6 0.514 M B P ND a 8.41 LG / IC Cr203 (M) from -46 22 10 30 to -42 59 Control - - 252.7 0.322 1600 w/ S W 450.0 0.320 S B P M CFe2 (M) SOW/ ICFe203 (M) SGB 1700 PGB w/ OS M 1164.1 0.323 S B P M Insuf. LG w/ IC 1800 SGB W 1137.6 0.334 S B P M Insuf. 1900 PGB w/ 0S M 943.8 0.367 S B P M r2g03 (S) fr________________LG / ICS (M) 23 30 30 ro -5 62 Control - - 88.9 0.320 - - to -41 1600o SIB 748.5 0.320 S B P M - PGB w/ OS 1700 SGB - 878.4 0.334 S B P M - LG w/ IC PGB w/ 0O 1800 SGB - 662.3 0.370 S B P M - BP w/ LG 1900 BP w/ LG - 679.6 0.372 S B P M - PGB w/ OS 26 10 30 -38 59 2000 SGB M 841.3 0.398 S B P S - 27 30 30 to -38 62 2000 Bw/ LG ND 872.9 0.490 S B P M - from -50 B and N 28 100 30 to 40 53 2000 BPw/l0 M 1075.6 0.590 M B P-SF S WADC TR 54-120 51

TABLE 10 RESULTS OF O X I D A T I ON TESTS, TYPE 310 ALLOY, HEAT X 4 6 0 6 3 0.059% C, 20.67% Ni, 24.28% Cr, 0.31% S1, 1.57% Mn, 0.024% P, 0.014% S, 0.25% Cu, 0.12% Mo Penetration T Specimen Oxide Scale Run Duration, Air Velocity, Dev-point, Penetration Tm -- Specimen -- Oxide Scale No. hr t/min F CuNre, F c Magnetism N o Me nF ue Unt, Color e Magneti Cmp io N. h tei F Fig. No. Fissureszin.'thSize X= 17 100 0 om55 54 Control - - 728.9 0.354 - - - PGB 1600 LG v/BP W 1010.0 0.360 S B P-SF ND rPS(s) 1700 w/ W 1024.9 0.487 M B P-SF ND HPg(M) a —8.42* — W /in D e t h IC n l Omut Cr2 3(-) 18 00 w W 1116.6 0.622 L B MF ND fHS(M) LG w/ BP a=8.46 Cr2O3(S) 1900 L / BP W 806.7 0.582 L B MF ND HPS(M) toLG8 a=8.42* 22 10 30 frto -4 60 Control - - 269.7 0.330 - - - - to -42 SGB 1600 PGB w/ OS M 553.2 0.332 S B P M Insuf. LG w/ IC 1700 LSG 1C M 565.1 0.321 S B P-SF M 2O3() 1 LG 700 ISFe2O3(M) 1800 SGB M 560.2 o.386 S B P W C20P(S) mZ w/ ICB SGB LG3 v/ 1C Cr03(N) 1900 LG w/ BP M 635-. 0.402 M B P-SF W HPS( S) PGB w/ OS a=8.42* 2 30 5 from -52 23 30 30 to -41 63 Control - - 65.5 0.320 1600 SGB W 194.9 0.334 S B P-SF M - SGB 1700 LG w/ IC ND 287.3 0.354 M B P-SF M LG w/ BP BP w/ LC 1800 PGB w/ OS W 425.0 0.593 M B P-SF M - LG w/ IC LC v/ IC 1900 SGB S 366.1 0.443 M B P-SF S LG w/ BP 26 10 30 -58 60 2000 tow/NP 27 30 0 from -46 63 2000 PGB I W 1150.0 0.588 S B P W to -38 PGB ~~~28 100 ~ ~ ~from -50 BP w/ 14 28 100 30 to -40 54 2000 B and M 343.4 0.437 M B P-SF S TAB LE 11 RESULTS OF O X I D A T I O N TESTS, TYPE 3 10 A L L O Y, HEAT X 4 6 5 7 2 0.089% C, 19.48% Ni, 24.35% Cr, 0.67% Si, 1.60% Mn, 0.021% P, 0.022% S. 0.34% Cu, 0.14% Mo Bun Duration, Air Velocity, Dew-point, Penetration | Specimen Oxide Scale No. hr t/min *F Curve, ppearance Magnetism No of Mean Fissure Color e Composition Fig. No. Fissures/in. Depth, in-x1O Size (x-ray) 17 100 30 ro -55 55 Control - - 422.1 0.339 PGB Cr20,(S) 1603 L, w/ IC W 520.1 0.358 M B P-SF M HPS(9) LG a=8.43* BP w/ 1f Cro (S) 1700 w/ M 781.3 0.361 M B P-SF M HPS(a) PGB a=8.45* BP / L Cr203(S) 1800 w/ L W 686.4 0.631 M B P-SF M HPS(M) to w/ BP a=8.42 B and S HPS(S) 1900 w/IC M 1220.2 0.634 M B P-MF S a-8.46 LG w/ BP S.S. x25(M) 22 10 50 from -46 61 Control - - 145.0 0.0 - - - 22 1o 30 - 38 6w/ BPto-42 SGB 1600 PGB w/ OS M 496.6 0.322 S B P W Insuf. LG w/ BP PGB w/ OS 1700 SGB M 495.0 0.359 S B P M Cr20(S) LG w/ IC S2B Cr2O0(S) 1800 LG w IC M 366.4 0.327 M B P-SF M HPS(M) to w/ NP a-8.42* PGB w/ OS LG v/ IC Cr0,o(S) 1900 SGB M 488.3 0.405 M B M W EPS(a) LG w/ BP a-8.42* 23 30 30 from -52 64 Control - - 299.1 0.322 - SGB 1600 PGB w/ OS - 630.7 0.323 S B P M - LG w/ IC SOB 1700 PG w/IC - 611.1 0.346 S B P-SF M - FOB w/ OS BP w/ LG 1800 Btw IC - 571.7 0.442 M B P-SF W - NP w to 1900 L w IC - 432.2 0.608 M B P-SF W 26 10 30 -38 61 2000 w/ I S 1211.4 0.800 L OB MF S - 27 30 30 tfr - 46 64 2000 PGB M 1434.0 0.7353 B P 5 - ~28 iDD 5 fr 3D 55 2000 B and S M 1084.5 0.749 M B P-SF S - WADC TR 54-120 32

TABLE 12 OXIDE-METAL INTERFACE MOVEMENT IN 100 HOURS CALCULATED FROM WEIGHT LOSS MEASUREMENTS Measured Total Interface Specimen Oxide Scale Run Alloy Densi ty, F Weight Loss, Movement, Appearance A nt Color Flake Magnetism No. gm/c percent in.x5 AppearancSize 20 309+Nb 7.94 Control 0.01 0.2 - - - 160 0.10.1.1 w/ P S B P M BP LG 1700 0.79 9.2 BP w/ LG GB P- M LG 1800 1.32 14.8 L/ BP M G MF S BP wI LB L& w/ IC 1900 1.90 21.7 G w EIC L G-GB MF S 20 64177 7.90 Control 0.01 0.1 - - - 1600 -0.05 -0.8 PGB w/ oS - 1700 0.01 0.1 PGB v/ O B P M 1800 0.08 1.2 BP w/ S B P M 1900 0.02 0.3 BP0 /LG 8 B P W 20 64270 7.93 Control 0.01 0.2 - - - 1600 -0.04 -0.6 PGB v/ - 1700 -0.03 -0.4 PGB w/ 08 8 B P M 1800 0.24 3.8 PGB w/ 08 S B P M 1900 0.08 1.3 PGB 8 B P M 18 X11306 7.83 Control 0.04 0.4 - - - - - 1600 0.54 5.7 La M GB SF M 1700 0.99 10.3 V/ IC L G MF ND 1800 0.74 7.4 w/ U M B P ND 1900 1.94 20.5 LG M B PSF ND LG v/ BP 28 2000 1.81 19.0 WGv/C MBP MG 18 X1338 7.90 Control 0.02 0.2 1600 0.31 3.6 L / BP M B WM M 1700 0.69 8.0 / L B LF W 1800 1.12 12.8 IL w/ c L GB MF ND PGB w/ 08 1900 2.09 2.4 BP w/ LG 8 B P-SF LG 28 2000 0.52 6.0o BP d/ S B P S 18 X27258 7.89 Control 0.01 0.2 - - - BP v/ LB 1600 0.21 2.8 LG w/ IC M B P-S W PGB w/ OS 1700 0.29 3.8 G vw BP 8 B P-SF M 1800 0.95 12.6 G w L B MF ND BP w/ LG 1900 0.33 4.4 PBv w/o s B P-SF N 28 2000 0.12 1.6 PGB ld8 8 B P S B and S LG v/ BP 19 X45558 7.90 Control 0.00 0.0 PGB 1600 -0.02 -0.2 LG S B P M LG w/ IC 1700 0.45 4.9 B G M B PSF W 1800 0.79 8.7 LG w/ IC M B MF ND PGB w/ OS 1900 0.46 5.0 BP w/ LG M B P M LG PGB w/ 08 28 2000 0.14 1.6 BP w/ LG S B P ND B and S 19 X46063 7.93 Control 0.00 0.0 SGB 1600 0.27 3.3 w/ BP S B P-F M wa v/ IC LG w/ IC 1700 0.51 6.2 LG M B P-SF W 1800 1.23 15.2 LGw B M B P-NP W 1900 2.23 26.6LG / I GB Mw ND 19 X46572 7.89 Control O.01 0.2 - - - SGB 1600 0.14 1.8 PB S B P-SF W LG 1700 0.48 6.2 M B P-SF W 1800 1.10 14.6 / C L GB M ND 1900 0.86 11.2 PGB M B PSF M PGB w/ Os WADC TR 54-120 35

Q TABLE 15 i' RESULTS OF METALLOGRAPHIC ANALYSIS 0o 30-hr Runs 100-hr Runs (Unless Otherwise Specified) Alloy 100 hr Runs Chromic Acid Etch Penetration Types (see Figs. 95-100) hr Type Banding a Phase 6 Ferrite Results Control 1600~F 1700~F 1800~F 1900~F 2000~F 309 + Nb Not 1600~F No Detectable G.S. Change. S 0 S IV M IV L IV M II M II 10 Detectable 1700~F Carbides Agglomerate. MIII 3' 64177 Small 1600~F G.S. Increase with Temp. F 0 M III M III M III M III M III 10 Amt. 1700~F Carbides Agglomerate. No MIII 80 1800~F Grains Visible at 2000~F. 1900~F S 64270 Not Detectable G.S. Increase with Temp. S 0 0 M I S II S III SIII 10 Carbides Agglomerate. No MVI 50 2000~F Grains Visible at 2000~F. X11506 Small 1600~F No Detectable G.S. Change. S 0 S IV M III M IV See Figs. 105Amt. 1700~F No Grains Visible at 1800, 106 for 1600~F 1800~F 1900, 2000~F. Carbides Ag- and 1900~F and 1900~F glomerate. 20000F. X1133558 Small 1600~F G.S. Increase with Temp. S 0 S IV S IV L I L M III 10 Amt. 1700~F Carbides Agglomerate. No MII 50 1800~F Grains Visible at 2000~F. SV 100 1900~F

Q TABLE 13 (cont,) i_ _ o 3 _g _0 — 4. o0 Alloy o F3 0-hr Runs 100-hr Runs (Unless Otherwise Specified) Alloy 100 hr Runs Type Ba'nding a Phase 6 Ferrite Chromic Acid Etch Penetration Types (see Figs. 95-100) Results Control 1600~F 1700~F 1800~F 1900~F 2000~Flhr X27258 Yes 1600~F Band of Small Grains. M 0 S III M III M I LI S II 10 1700~F Increase G.S. with Temp. M I 30 1800OF Carbides Agglomerate. No LVI 100 19000F General G.S. Change. No Grains Visible at 20000F. X45558 Yes 1600~F Band of Small Grains In- S 0 0 S III M II S II S VI:10 1700~F crease G.S. with Temp. M VI 30 1800~F MVI 100 1900~F General G.S. Increase X46063 Not 1800~F No Detectable G.S. Change. M 0 0 S III S III M III S VI 10 Detectable 1900~F Carbides Agglomerate. S VI 30 No Grains Visible at 2000~F. MVI 100 X46572 Yes 1800~F No Detectable G.S. Change. S 0 S II S I S III S III M I 10 1900~F Carbides Agglomerate. S III 30 M VI 100 S = small penetrations M = medium penetrations L = large penetrations F = fine penetrations

:.r;s,' \ i.'.!, -~~~ ~~~~ x~,.I:~."; ~.:~.': i~ (~~7S~~''~~~~~~~~~~~~~~~~~~~~~~~'. r.:; Li L Type 309+ Nb Alloy. F ip. 2. Type 310 Alloy, Heat As Received. lOOx. Plane of 64177. As Received. lOOx. Rolling. Etch: Electr. Plane of Rolling. Etch: Chromic Acid. Electr. Chromic Acid. ": ~.... ~''4" fYI::~ ~ ~~~~~ ~~~~~~~~~~~: "'"':r'I' "'";:;-.-,/'' ~' ~~~~~~~~~~~~~~~~~~ *,v.i'~~~~~~i ~' —,~1;,~c~j~.,.,/' -..'~-. ~ -'..:,:... ~ ~I..: V~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ I Vii /,~i ~,~,,.,~,,.,.~. CT 4i'"" */ i,:~' "~,~.__/~-~r~r'"A )'.L' s'" ~~~~~~~~.... Fi.3 Type 310 Alloy, Heat Fig. 4. Type 310 Alloy, Heat 64270. As Received. lOOx. X 11306. As Received. lOOx. Plane of Rolling. Etch: Obl. Light. Plane of Rolling. Electr. Chromic Acid. Etch: Electr. 1:1:1 HC1, HNO3, H2O. WADC TR 54-120:56

''V. *'^ / o-'\ ^ ^''V....j.: v,~~~~~~~ o, /',:^ \^^..:.-:...,::,.'..,'.. ~.::.:,. ^ ^ *U.. -^ A'-"'-*'- *'... (.~~~~~~~~~~~~~. Pl e.. of R. E., g. Plane... Rol,.I ~...~.','... ~''.,.,'., o.''. U,~~~~~~~~~~~~~~~~~~~~~ ~ ~ ~ - o-.~. Eletr Chrmi Aci.Ech El r 1:1: HC1, / oV..'*':'..~ ~'.:'". "' \".'" I'." ".'.:i~~~~i L >;:: _,... <' r \.~m s c,.., Fl:.7 Type 310 Alloy, Heati. 8. Type 310 Alloy, Heat X 11338. As Received. lOOx^.X 272658. As Received. lOOx. O.Lit.Plane of Rolling. Ec:Ol ih.Plane of Rolling.Eth I~et.CrmcAi.Etch: Electr. 1:1:1 HCl,,Eet.Choi cd HNO^,~~~~~~N HO.. C T- 54-12r~3'2 i'~"' f': *.. ~,...:...,......_, ~ ~~~~~~~~~~~~~~~~~~iI... b~~r /,.\.. ~~S.) x " 4':O. X 6. A. lx Obl. Light..,P o l. Etch: __.~. —:~:::'i:::: Etch:~. Eet. 1:: Hl. lect. ChoncAi. ~i.5 ye30AllyDCt i.6 TRp 54120 A loy Ha -~~ ~ ~~~:lectr. ~ Chromic Aci. Eth Electr. lZ~ H1......:', "..'.~,. /'_, i ~ ~ ~ ~:,...':~~~~~~~~~~~..-..~.. 1~~~~~~~~~~~~~~~~~~~~,.. ~. /: ~;,...~'.':.',,:',..,.;, F~~~~~~~~~~~...~g l rT e ~o8 g 0~~y e )~ 45558. As Receivea. lO~~~~~~~~~~~x. X 46063. As Receive~. lO~~~~~~~~~~~x. O. i gt TPlae of Rollo Ka ing. Plaue of Rollug. HEtc Eth lcr lllH lect~r. Chromic Acid c Eet. H1 H:O3 ~~~~~~H0, H20. WADC TR 54-~~~~1205

.X 46 572. As Received 1O~x. work.. p o, He-at Etch: hlectr. Maromic Acid. Plane of Rolling. Etch: ^'/iv 2 I/ */'\* _s _ 1 K. K e. - /'i F ~9.. " *'' i....*'.... ".''', " "'.'*\'*""''"'"',I"..r... I _, Etch: lectr. romic Acid. Plane of Rollin. Etc SA R5-2 ~ Fig. 11. Grain Boundary. Type Fig. 12. Type 310 Alloy, Heat 110 Alloy, Heat X 45558. As 64177. As Received. lOOOx. Received. lOOOx. Obl. Light. Cross Section. Unetched. Plane of Rolling. Etch: Electr. 1:1:1 HC1, HN0., H2 7ADC TR 54-120 53

Fig. 13. Type 310 Allov, Heat X 11338. As Received. 100Ox. Cross Section. Unetched..'ADC TR 54-120 39

FILTER REGULATING ~.___ ~VALVE U | AIR'2 P3 05 = I RY1 DRYER 90 psig'-3 TO ~' P R E S S U R E F U R N A C GAUGE DEW PT. o |INDICATOR ROTAMETERS CONTROL THERMOCOUPLE SPECIMENS ALUNDUM MANOMETER------— r ~ / > ~ CEMENT7 ~'^ ~~~~~~~~~~ ^ | -^ —---— INDICATING o= | | —-| \VERMICULITE / | THERMOCOUPLE CONTROLLER BROKEN CERAMIC 220 V. AC G SCHEMATIC DRAWING OF OXIDATION EQUIPMENT FIGURE 14

Fig. 15. Specimens and Ceramic Holders. Fig. 16. Bending Die and Specimens. WADC TR 54-120 41

zi7 0?T-itg 91 OaVMiv'Nj Nn' AO11V qN+60~ 3dA1 "33VdnlS M0-138 Hld3C'SA NOIV01.l3N3d-LI'9I S3HONI'NOILV.L3N3d dO H.d3I 90 aoo'o kg9'o so'o0 0J'o o 100,0 o / 001 \/ OOz....___., / 00\!~00 e:1000091.00100 % ^c 0 0~ ----— 009 ----- 009........ ____ —------------------- 0 0 ~.....- I___ -- I.II.-m.....- —.- 006

900 - 800.. 600' lll - 500 400 IL 0 0 0. 500 1600*F 200 1600*F 1900F 1700'F 100 I 800*F' * 0 0.001 0.002 0.003 0.004 v.005 01 06 DEPTH OF PENETRATION, INCHES FIG.18- PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64177. RUN 5. WADO TR 54-120 43

'1-t O;T-ts HtI OCIVM'* Nn'Oa.Zt9 1V3H'AO1V 012 3dAl'33V.lUn M0138 Hld3a SA N011Vi13N3d-61 91I S3HONI'NOllV1I3N3d JO Hld3a 90'o. -0' 0' i0 m, 0',0 de*0061 "^^ f0 Jo0 00'. olooI:1.,~0091 --. 009,m m 00~ -------------------------------------- 0 0 9~~~~~~~~~~~~~~~~~C: ------------------------------------- OOZ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~:;I _____ —-------------------— 0 0 *. * ii i i - 0 0 6~~O

800 | - - 700....... 600 - z 0- 500 6: 400 U) I~I \ 3 500 200 co tI /I 600> O 1 0 0.001 0.002 0 003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG.20-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 309Nb ALLOY. RUN 6. WADO TR 54-120 45

900 - - — vl 800....... 700 --- 600 w 40 00: ( 100 20 0 19300 F 160. 200 ~~~1900*~F\~ 8 0 100 4 —--,1800'F 0 0.001 0.002 0.003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG. 21-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64177. RUN 6. WADC TR 54-120 46

Lt OZT-t VIj oavYM'9 Nn' OLZf9.L3H'AO1ltV 01 3dAl'330dnlS M0138 Hld3a'SA NOli0Vi3N3d-ZZ'91J S3HONI'NOIlVU.3N3d iO Hld3a 900'0 00OO'O 00'0 100'0 ZOO0 000 0 -—,-6 Ai— I. 0061- I\ 0 9 ------------------— \ I1 o og 001 \~ 0000 0 OO1 0n,__,,___!00G00 ------------ -. -- 006

a<r onT-tlg u.L oavm *8 Nnf'A011O qNI-60~ 3dA.'33V-d3S M0138 Hld30'SA N011V13N3d-~Z'91I S3HONI'NI0V1113N3d JO Hld30 90'0 O v'O ~00'0 ZOO' 100'0 0 Xo 0 b0 91 - 0061 I 0008 I -" 001 1\ 003. I "! I o ------------------------------------ 0 0 9 OO" 008

*8 NnY'LLI t9 I3H'AOTI 01~ 3dA ~33VidUns M0138 Hld30'SA N011UVI3N3d -Z 91I S3HONI'NOIIV113N3d JO Hld30 90 MIo Mo' o goo'o, o000 0 0\^ \J~0061 l i { v.0081's\.'t {, l I0I0I I s il o J*0091-.,-,,OL I.000,oo, -00l 009 ~OL 009 ------ - ---- ---- - --- - — 006 ------ ------ ------ ------ ------ ------ 002i

900 I 800 Bo 600 100 w \ DEPT ____ - 400 0 3:1' 01700 F $ W 00 I 200 Sii \'I 100' I 19001F9 0 0.001 0.002 0.003 0.<04. 0.( O06 DEPTH OF PENETRATION, INCHES FI0. 25-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY,HEAT 64270. RUN 8. WADO TR 54-120 50

900j. goo 700 _ 600:,1600*F 0T 500 300 I 54Q \\ O iI I' 200: *F 100 i i i \ 17009F 0 0.001 0.002 0.003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG.26-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 309+Nb ALLOY. RUN 9. STRAIGHT SECTION. WADO TR 54-120 51

900 | - 800 700 600 z 5L00 19000F I 1 8 IIII 1600*F,/\ 300 1700F OI, -- 0.001 ll O.4 0200 0k.l O8 DEPTH OF PENETRATION, INCHES FIG.27-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64177. RUN 9. STRAIGHT SECTION. WADC TR 54-120 52

900 800 Soo 700 600 z 0 co I \ a 400 0 700F z 300 200,, 1900"F DEPTH OF PENETRATION, INCHES FIG.28-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64270. RUN 9. STRAIGHT SECTION. WADC TR 54-120 55

M^g; OZT-TS T t OAIVY'N01133S 03A:nO 6 Nn A011OT qN 460 ~ 3dAl 3Ovidns M0139 Hid3a'SA NOIIva13N3d-6Z'91d S3HONI'NOUVY3N3d 40 Hld3Q 900 _ 0 00 OOo1000 0 600l11 ~l.000" ~.:l. 001 0 *1 iQ L 00 m

900 - 800 700 600 z 0 500'L 400 0 300 200 O 0.001 0.002 0.003 0.< O4 i.05. 06 DEPTH OF PENETRATION, INCHES FIG.30-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 3700OALLOY,HEAT 64177. RUN 9. CURVED SECTION. WAD D TR -120 55 1900"F |~ ~'~0"0 0.001 0.002 0.003 0.004 O005 "'06 DEPTH OF PENETRATION, INCHES FIG.30-PENETRATION VS. DEPTH BELOW SURFACE. WAg WADC TR 54-120. 55'

9S oez-tgs Hitg oav NO113S a3^Alno'6 Nfl'OLZ t9 lV3H'A01V01OI~ 3dAl 3330VAlns M0138 Hld3a SA NOIiVl13N3d-1~'911 S3HONI'NOI11Vt3N3d.O Hld3Q 90 _>o OO 00,0 L oo I A.OOZLI00 61 ------------------------ 003 --------------- oot | 00~ 0 -I1 OOL ------------------------— 0 z...008 ------- ----- ------------- I IIIII..006 ------------------------------------ ~ ~~~~~~I II OIO I I II

900 h B o \'I 800 --..........!'1 i 600 z00 0.0 I' 40 D \ \P1700 FIDT PENETRATION, BES. 500'1 U I- 30I 9 A WADC TR 54-120 57 | $00 o 1900'F 200 --:4 00 --- Goo-T~I l -------- -------- 0 0.001 0.002 0.003 0.004 0.00. 06 DEPTH OF PENETRATION, INCHES FIG.32-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 309'Nb ALLOY. RUN 10.;;ADO TR 54-120 57

900 "..- - 700' CONTROL 600.O ____1 ___ I \o W o.T 3 100ALLOY —---'6- 1. 200 wb. ~~~I* \ ~~-a O170 600F 100i i3 30~0~~~TY 10 A TI —------ 61- 7 — R-N-10. D TR 5-1800F8'00 1600OF \ \ V s1900'F 0 0.001 0.002 0,003:.004 0005 0.006 DEPTH OF PENETRATION. INCHES FIG.$33-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64177. RUN I0. WADC TR 54-120 58

900' 800..... 700 600 -, — -- 0 300 w e 0. 0 i 300 -............ 200 \1600"F 100 \ ^^\ %V 1900*F CONTROL \K " 0 0 0.001 0.002 0.003 0.004 0.005 0.0 06 DEPTH OF PENETRATION, INCHES FIG. 34-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64270. RUN 10. WADC TR 54-120 59

900 - - 800 800 ---—'-' —600 i. I 1 z 0 ia _ I. 500 --.... trt D I, I (I). T I3 +Nb A RNI 2 3 300 - Zo ~2 | {-a1800.F 300 _____'_. 1600F 70\ CONTROL > 2000F 0 0.001 0.002 0.003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG. 35-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 309+Nb ALLOY. RUNSII 27. WADC TR 54-120 60

900 - - - 800 | 700,,\L 3o 600 O..... T 3i A _1700F _ 500 _____ 0L~~~~B 1600F c ioo 200 -— 1900F'" 400... 4 0 I 800F 200 1 \ 100,..... CONTROL 2000% F 0 0.001 0.0.004' 0 O" 06 DEPTH OF PENETRATION, INCHES FIG.36-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64177. RUNS II 8 27. WADC TR 54-120 61

900 20000F \ 700 F 600 00F z - 1600*F it I w. LL 00 i 300 6 200,o,',.t \ ID 100 8 CONTROLN s' 0 0.001 0.002 0.0503 0.,,04 0.(5,,0, 6 DEPTH OF PENETRATION, INCHES FIG.37- PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64270. RUNS II & 27. WADC TR 54-120 62

900' oo800 I I 8 00 _ _ _ _ _._ _. _ _ I 700 I 600 I.... z ^-| —1 CONTROL ( 500.. - 17000F 1 400 flC l-j -1900 F I 300 i I | \12000F 100 u ---- 0 0.001 0.002 0.003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG. 38-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 309+Nb ALLOY. RUNS 12 8 26. WADO TR 54-120 63

900 - - 800 700 \2000*F 600 a. 500 ~ 0r \\ u- 400 U. 0 300, 200 1600eF V1700F'100 CONTROL \'... ~O1900'F 0 0.001 0.002" 0003 o0.004 o.5.oo06 DEPTH OF PENETRATION, INCHES FIG.39-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64177. RUNS 12 & 26. WADO TR 54-120 64

7-0 800*F 1700F 6 600;* I:: 50 _00,'I' I 200 - 100 1900'F CONTRO. 0 0.001 0.002 0.003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG. 40- PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY. HEAT 64270. RUNS 12 826. WYADO TR 54-120 65

900 - 800....... 600 I, 500 700 *_& 1700*F 300...: ~f-I 900OF 200 1800*F 00 - - \ 00... F. 200 - 4V ---------------------------- 6 0o001 0.002 0.003 0. 04 0.005 0" 06 DEPTH OF PENETRATION, INCHES FIG.41- PENETRATION VS. DEPTH BELOW SURFACE. TYPE 309+ Nb ALLOY. RUN 13. WADC TR 54-120 66

L9 OZT-irS I OYCVM'~I Nn'LLIfr9 1V3H'AO1lV 01~ 3dAl'33oVinS M0138 Hld30 SA NOIiVU13N3d-Zf' 91I S3HONI'NOIV1,3N3d.0 Hld30 90 >0 90 —0 tO'.0O.. Zo00'O 100'0 0 ~\ %\%*J -'1.0091 I IIT a S1 I\I I I " \10o1N0O -'.00I -4-I- 00 00~ c I\ o *, \\ \ \ I*'\ 0 t I\I I:\l 1 I I \ 006'.C l.) 00i 9 rVi ipl...00601 x 009 OS' -------—' - -006

900 800 1600~F 700 600 500 -O 18001 F 200 T 3A, H.00I 10 T 470 R 1 C NTROL. 54-120 68 0 0.001 0.002 0.003 0.004 0.005.006 DEPTH OF PENETRATION, INCHES TYPE 310 ALLOY, HEAT 64270. RUN 13. WADO TR 54-420 68

900 - | - 800..-4 —------ JT"- 170000F'I 1600-F 600 _ ~ 1800F 1t U)a a 50 C \.. -------- w \\ \ u,^' ~ti! 1 * r!i I 1900'F U) I U) I. n. 400 o 200. 1I DEPTH OF PENETRATION, INCHES TYPE3 O-9-+Nb ALLOY. RUN 14. 2ADO TR 54-120 69 WADG TR 54-120 69

900 800 Boo.... 700 __ —-- -- 700 0 I1700'F' 500 _ _ cn * i 400 z 300 1600 F 200 * I \ 4100 CONTROL\ 0 ao001 0.002 0,003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG. 4-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64177. RUN 14. WADC TR 54-120 70

1L OZT-tGC, OaYM *f' Nnl OLZ t9 LV3H'AO11 01~ 3dAl *3DoYlnS M0138 Hld3O'SA NOI.LV13N3d-9fr'9,I S3HONI'NOIJVU13N3d JO Hld3a 900'0 900'0 100'0 ~000 Q00'0 1000 0.600,^I.00 91 t ——'-"' —.................. 0001 I.oI 0, \'! I J~000 1 I1 lll|tSo iil d.006 I' —-4lti i 0 009'nouiNOO' OL 008

900 --'S 700 - l O 600'~- 50 0............... VS................ 5 00 40 i 11 IL. 400,._-_-, —----- 100 200..... i \ i' * 00-1600 CONTRd~' 1800o 0 ~00.00.0 I I5 0I0 I I I I.06 DEPTH OF PENETRATION, INCHES FIG.47-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 309TNb ALLOY. RUN 15. WADC TR 54-120 72

900 *7.t- -- 1900"F 600 -- S' \.1700*F 9. 500 w.1600~F \ 1800F 300.__ _ ~ \ \ \\ 200 CONTROL' 100 -V 100 I_____',\ _ __ \ \\ DEPTH OF PENETRATION, INCHES TYPE 310 ALLOY, HEAT 64177. RUN 15. WAPO TR 54-I20 75

900 i' 800 I 700'I ~~16006",n It * ~ I w Ijl o- 500 I i I (I: 0 0 0 DI \O0 DEPTH OF PENETRATION, INCHES FIG.49-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT 64270. RUN 15. WADO TR 54-120 74

9 00 - -- Soo,,I00' _ _ 700. 1 —;I 600 _ —---- I \ 20 2000'F 0:...... 3 1! ______ 1001 16h- CONTRlI, 900FF 2 400.... - mm.~in~mm 00*F. \ 0 0.001 0.002 0.003 0. 4 0. 006' 06 DEPTH OF PENETRATION, INCHES FIG.50-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT X11306.RUNS6a& 28. WADO TR 54-120 75

900l i 700 - 600 __ _ _ _ _... \i' o. 500. -....-............... s wiL.__\ 1700*F coW & 5'CONTROL (_ 400 0 Z A \2000'F 300 1'It mi i ii 200 f. \ 11900'F Il 800'F 0 0.001 0.002 0.003 0.004 0.00.006 DEPTH OF PENETRATION, INCHES FIG. 51-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT XI 1338. RUNS 16 a 28. WADO TR 54-120 76

900 _ | i q', I; —* —1600'F *: -1700F 700........ 600 5 |\ S V —1900F 9. 50Cv 0 _______ U) I II; -1800' =! * \ 2 00 100oo CONTROL5, \% 9 0.001 0.002 0.003 0o.(4 -5,o 00,' 06 DEPTH OF PENETRATION, INCHES FIG.52- PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT X27258.RUNS 16 & 28. WADC TR 54-120 77

900- -:i 700 1i 8005 I.i i 1 -It~16001I I40 I _______I 800_ _ 7t. 500- I --- 600 Ah' I 100 70 CONTROL I - 1900'F 2'100 -- \ ---— F__ TYPE 310 ALLOY,iEAT X45S58.RUNS 1728. TR 54-1200F i00 - CONTROL.... ~0 0.00t 0.00. 0.005 0.0-4 0.06 O. DEPTH OF PENETRATION, INCHES FIG.53- PENETRATION VS. DEPTH BELOW SURFACE. TYPE310 ALLOYHEAT X45558. RUNS 178 28. WADC TR 54-120 78

6L ozi-tg VI OVm'8E LI SNni'~909q X 1V3H'AO11V 01~ 3dAl 330V:lnS MOl3g Hld3a SA N011VU13N3d-tg'91d S3HONI'NOl0VUi3N3d JO Hld30 90 )o ip oo oO Q00o0 100l 0 O. I0'.OLN0O \* \. ----—'1 0001 \I 4.40091 c d.-009. 1 _ 1 __0,.'I _\ I' 0 ----------— i.y! \ ooo IS. o i.00003 I \|......!0o9 IV i.0061 r 1?',..............009

900 iii Soor... 600 700 *16000F ~t Il 1soo.ma _ _ _ _ _ _ _ _ 10 (in at. -1700F 200 "-';I....F -44:...-..'100:V —---- 100 \'*' ".__ _ _OOOF I000-F CONTRO\ 0 0.001 0.002 0.003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG.55-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT X 46572. RUNS 17 8 28. rvADC TR 54-120 80

900 - i_ 1600'F 800 V00 — 17000F I i 1-2000F 700 iI /., 600 I - z i I iD t 4500 I30 -01 I \ w A \ I P Co Ii i', I 200 I: \ I', 300 6 - - CONTRO. *\ 0 0.001 0.002 O.05 0.004 0.00. 06 DEPTH OF PENETRATION, INCHES FIG. 56-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY,HEAT XI 1306. RUNS 21 8 27. WADC TR 54-120 81

900 ll — 18700:F 100 1600F -l --:CONTROL 700 IL- 1 I\2000o I!'vJ 600 ------— __ - -_____ "o \ lI IW, I. 500' II00 400 -l i — 1 —-------— _ 0 I\\ I/ I. o o0.001 0.002 o0.005 r..6-04-. 6 4 DEPTH OF PENETRATION, INCHES FIG.57-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT XI 1338. RUNS 21 &27. WADO TR 54-120 82

900 CONTROL 800. 1600eF v: --- 1700'F |1 1800F!m 700 600 z \ _A -' 2000~F 0. 500'AJ /( VT 1900~F -L. 400 1I\:500 3 / - _ _ o!ii 1800F 2004!00 0 3 - -l — 0.001 0.002 O. —3 —--— ______ 0.006 DEPTH OF PENETRATION INCHES FIG.58- PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT X27258. RUNS 21 27. WADC TR 54-120 85

900 800 700 ---- 1 600 oi z IL5I00 1600"F I 1700F1 (n I — 800-F aU) B~e 1900F 6 400' ---- CONTROL I 300 1900~F "J2lJL^1800'F 1 800 F j"\\ \ 1600F bCONTROL \ \12000'F 17000F'\ 0 0.001 0.002 0.003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG.59-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY,HEAT X 45558. RUNS 22826. WADO TR 54-120 84

900'G._ "16-00 F 1800F 700 - 600 1 _ —-------- 8L 500, — 85 w H ---- CONTROL c400I\ z 100.. ADC TR 54-120 5

98 OZT-lf I1 OQDYM'9Z9 ZZ SNnI ZLgS9fX 1V3H'AO11V 01~ 3dAl'33ovlns M0138 Hid3'SA NOlvu13N3d - 1991.I S3HONI'NOIIV.L3N3d JO Hld30 900'0 9000Q t00' 0 ~00'0 000'0 100'0 0 00. I0-\ I I \ t e 009 J\o O IS!0061' 00 9 4.0091\00o --- - 006

900i i | — i_ 800 70060 0 z I |-2000'F -05 I 5. 4004 —--- FL 400 U. 3 300 8 1800IF!2-& a 1900' 1600'F I1700F 100 CONT OL 0 0.001 0.002 0.003 0.004 0.005 0.006 DEPTH OF PENETRATION, INCHES FIG.62-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY, HEAT X 45558. RUNS 23 827. WADC TR 54-120 87

900 \ 2000F 800 700 I........ - I \ 600,_..... z — 1800*F 0. 500 w' U) I I rr~ 1^- 1700F' \i.......... L 40C 0 1600"F 300... 200 6 GCONTROL 100 1900"F 0 0001oo 0.002 6 0.04 O. 0.. 06 DEPTH OF PENETRATION, INCHES FIG. 63-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOY,HEAT X46063. RUNS 238 27. WADC TR 54-120 88

600 _ —---- S 8500, "1800F 700 ---------------- 1600'F I 7TOO0F 600.......:XZ: LD CONTROL 5' 54-12000*F ccS 500-,,.... C ONTROL 20 1 100 DEPTH OF PENETRATION,INCHES FIG. 64-PENETRATION VS. DEPTH BELOW SURFACE. TYPE 310 ALLOYHEAT X46572. RUNS 23827. WADO TR 54-120 89

06 ozT-1Ts I oGav'92 9 fZ SNnU'90211 X 1V3H'AOtIV 012 3dAX 3oViuJnS MO038 Hld30'SA NI01V13N3d-S9' 91 S3HONI'NOI.VWL3N3d 40 Hld30 90;0.. 0 g200' 10 000 0 j.40091 I * I001 I OOI I IJ --------------- Ht'{ -- r -- l Adooz........................ I I'~, J "-'l 1 --— l-^ -tt?|~~~~0-0 i OOL................. I I I 1 1 O0,.oo,,, ~ /.:00I l,~\-' 009 I i, 006

T6 OZT —^S Hl; OGVJ'9Z e f SNn'8~~1 IX IV3H'AOTIV 01~ 3dAi *330Vans M0138 Hld3a'SA N0IVlU13N3d-99'91J S3HONI'NOIlVU13N3d JO Hld30 900'0 g0000' OOO ~00' ZO 00'0 0 00 0 001 \;OOL.I 4.00991 00Z l,0 I I e iX00 00~ c rrn --------'^ - 00 m _ _ I _ _- _ _ I l I00 4.10006! 00 -....... 006 ------------------------------— 1,- 0 6

z6 OZI-irS dO oIeM'9Z e Z SNnY'8egZi X 1V3H'AO'lV 01~ 3dAJ *30VAUnS M0138 H.d3'SA N0IVUL13N3d-L9'9)IJ S3HoNI'NOVlJ.1N3d JO Hld30 l0 10'0 ~00,01 ^T~~~ ^^\.*^10M1NOO.009 1,o,09_.^ ____I.,001 100081,. L <OL I - 009l- -006 009! __.._._,_,.l~-~,~'" —" —-"'

RUN NO. AIR VEL.,fthnin DEW PT.,F 0 —--— 0 5 40 <-20 6 20 -20/+20 ~0- — 0 — 8 30 +17/+20 9 Str. 30 -60/-30 (-" —,-X - 13 40 -30/+20 (0 w cn w 1500 U. 0 U'~x 5 00 —- -- -... —e —— I CONTROL 1600 1700 1800 1900 2000 TEMPERATURE *F FIG.68-SUMMARY PENETRATIOR FREQUENCY CURVES. TYPE 309+Nb ALLOY. 100 HOURS DURATION. EFFECT OF AIR VELOCITY AND HUMIDITY.

RUN NO. AIR VEL.,ft/min DEW PT.,OF e 0. —-— 0 5 40 <-20 I' 1 —— A 6 20 -20/+20 3,0 — --- -0 8 30 +1 7/+20 0" O -- Xo x 1o — 7 9 Str. 30 -60/-30 LX —--— X 13 40 -30/+20 aID I — 1.' z z.. I I I 0.5 CONTROL 1600 1700 1800 1900 2000 TEMPERATURE * F FIG.69-SUMMARY PENETRATION DEPTH CURVES. TYPE 309+Nb ALLOY. 100 HOURS DURATION. EFFECT OF AIR VELOCITY AND HUMIDITY.

RUN NO. o 1.0. cn Io / 0.5 O-.,o {i' 00 x w TEMPERATURE, eF O. 0 TYPE 309 +Nb ALLOY. EFFECT OF COLD WORK. I00 HOURS DURATION.

I6S r~ oRUN NO. DURATION, HR. ----- 0 50 > ~' —-- || 30 II 0e 11 ~~~~30' — — 1 12 10 f 5 X-l —----— X 14 77 o Z o —-- 0 15 100 cn C 1500 0cn Is 1000 - - laJ F- G 7 1 - SMA PEERTO I _ _________ 500-E, -' CONTROL / 1600 1700 1800 1900 2000 TEMPERATURE, OF FIG.71-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 309+Nb ALLOY. EFFECT OF TIME.

RUN NO. DURATION, HR. g~> e — I — 10 50 (3 2-1 —- 1 130 r o o x X —. — X 14 77 Z O 0 — 0 15 100 z I UJ 1.5 0 ZI pCONT1.R 1600 1900. 2000 CONTROL'f 1600 1700 1800 1900 2000 TEMPERATURE, *F FIG.72-SUMMARY PENETRATION DEPTH CURVES. TYPE 309+Nb ALLOY. EFFECT OF TIME.

RUN NO. AIR VEL.fthnin DEW PT.,OF ^ 0I —--— ~0 5 40 <-20 e 2 a —-— o 6 20 -20/+20 G D- — O 8 30 +17/~20 U^ 5 - - 9 Str. 30 60/-30 z X —-— X 13 40 -30/~20 Id Cl) Qf Cl) LJ a1 co cr CNTO 6500 ^ ^ 1500 —----------— ~ —--------------------- iZ cr 1000 31 -AXLHA6 7 0 OSU N 1E 2 #1' 500 5QO ---------- ----- op^ —-— ^ _-_ - ~~ —---- CONTROL'' 1600 1700 1800 1900 2000 TEMPERATU RE OF FIG.73-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY, HEAT 64177. 100 HOURS DURATION. EFFECT OF AIR VELOCITY AND HUMIDITY.

=-~c^~ | RUN NO. AIR VEL.,ftn/in DEW PT.,~F 0 -— o 5 40 <-20: n —- — A 6 20 -20/+20 -f, I 1- — {-0 8 30 + 17i+20 0 9 x i -— 7 9 Str. 30 -60/-30 cn -> = —-X 3 40 -30/+20 IOL O. 5I1 1 1 0 oZ:'C O.5_ _ - — x-, — --- -----—. CONTROL 1600 1700 1800 1900 2000 TEMPERATURE OF FIG.74-SUMMARY PENETRATION DEPTH CURVES. TYPE 310 ALLOY, HEAT 64177. 100 HOURS DURATION. EFFECT OF AIR VELOCITY AND HUMIDITY.

RUN NO. -3 < U; H ^ _______________ 0 —--— 0- 9 STRAIGHT SECTION r co o U) U. --- 9 CURVED SECTION LL O 0 x. I-' LLo CL 0 o8 ^ 0.5 LU DIr co0 cnO UL 500 (0 0(00. o: 1 50 --------------— ^ Q-0 —-- 0 1600 1700 1800 1900 TEMPERATURE, OF FIG.75-SUMMARY PENETRATION FREQUENCY AND DEPTH CURVES. TYPE 310 ALLOY, HEAT 64177. EFFECT OF COLD WORK. 100 HOURS DURATION.

=~>^~~ rRUN NO. DURATION,HR. ot:- - I| 10 50 -- 12 10 l l 1, X —-- X 14 77 z 0 -0 1 5 100 a. CO I0 ~C 1500IL CONTROL 100 1900 2000 GONT 1600 1700 1800 1900 2000 TEMPERATURE, *F FIG. 76-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY, HEAT 64177. EFFECT OF TIME.

I

RUN NO. DURATION,HR. > a — -- l O10 50 - — { 1 2 1 30?,0 0 —-0 12 10 ^ 0 o 1 X —-- X 14 77 cn I 0 —-— 0 15 100 (I. Li X 0=1.5 Q. I I —----------- l —- - 0 5 CONTROL /_ 1600 1700 1800 1900 2000 TEMPERATURE, *F FIG.77-SUMMARY PENETRATION DEPTH CURVES. TYPE 310 ALLOY, HEAT 64177. EFFECT OF TIME.

RUN NO. AIR VEL.ft/tnin DEW PT.OF --— ~ —C 5 40 <-20 ul --— n 6 20 -20/t 20 D — -D 8 30 t 17/+20 0 ^ —- ^ 9 Str. 30 -60/-30 z x- — X 13 40 -30/+20 IL (0 uIj c, 1500 —------------------------------- C) IL IL 0C 00 CD 0 do~~~~~~~~~~~~-o Z * —-- - -----— " - 500 ~~~~~~~~ — I~~~~~~~~~~~~~~~~~~~~~~~~~~~ CONTROL 1600 1700 1800 1900 2000 TEMPERATURE OF FIG.78-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY) HEAT 64270. 100 HOURS DURATION. EFFECT OF AIR VELOCITY AND HUMIDITY.

RUN NO. AIR VEL.ftAnin DEW PT.,OF I 0 —-- s5 40 <-20 p I —. —^ 6 20 -20/+20 ^~ Kn D~ —--- 8 30 +17/+t20 o 0 x ^- * -— ^ 9 Str. 30 -60/-30 cn L^I X —— x 13 40 -30/+20 I C. z _________^_____ ___________ _ _____ 1 ~ 0 Ia cr 1.5 I_ _ _ _ _ _ 0. z 4 r ~I - - 0.5 --------- CONTROL 1600 1700 1800 1900 2000 TEMPERATURE OF FIG.79-SUMMARY PENETRATIOJ DEPTH CURVES. TYPE 310 ALLOY, HEAT 64270. 100 HOURS DURATION. EFFECT OF AIR VELOCITY AND HUMIDITY.

o en|RUN NO. 0 — 9 STRAIGHT SECTION __ - 3 o 0n.L - -- 9 CURVED SECTION LL 0 0 x 1.0 wz 0 U) 0 Oo: 0 0 E 0D 0 II a 1600 1700 1800 1900 TEMPERATURE, OF FIG.80-SUMMARY PENETRATION FREQUENCY AND DEPTH CURVES. TYPE 3 10 ALLOY, HEAT 64270. EFFECT OF COLD WORK. 100 HOURS DURATION.

RUN NO. DURATION,HR. -- -a I 10 50 g,- ~' --—'"| 1 30 > 0- - 12 10 ----- o z X-. —-X I 4 77 0. 0) w )( —— IC- 1500 Co 500 (/) IiiL 0- so* 50 ~.-o_ —--— ____________________. _- I- _________ CONTROL// 1600 1700 1800 1900 2000 TEMPERATURE, OF FIG.81-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY, HEAT 64270. EFFECT OF TIME.

RUN NO. DURATION,HR. Ah, I 10 50 -3 --- S II 30 - 10-0 12 10 0 X. —' -X 4 77 U) I l_0 05 100 ____________ ______________ I I1.0 U' 0.5 o^I l 1 1 0 CONTROL'1600 1700 1800 19002000 TEMPERATURE, OF FIG. 82- SUMMARY PENETRATION DEPTH CURVES. TYPE 310 ALLOY, HEAT 64270. EFFECT OF TIME.

: RUN NO. DURATION,HR. 3 I -0 0 — 16 100 r n, —-2 21 30 l/ 0 m D — 24 10 CA c 1500 —-- U.IL CO 1000 0 50 46/ -- A/ CONTROL 1600 1700 1800 1900 2000 TEMPERATURE, OF FIG.83-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY, HEAT X11306. EFFECT OF TIME.

I? ~RUN NO. DURATION,HR. O —-.-Z-O 16 100 M -- ~~~21 30 (I O —— D- 24 10 Lrn I3 C) -~~~0 2 4^ 1 \ U z a. I 8 ^ 1.5 F0~ OL LU z CLw 1.0 a. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ N jLU z 0.5 -- __ __ __ __ __ __ _ _ _ _ __ _ _ _ __ _ _ _ __ _ _ _ _ CONTROL 1600 1700 1800 1900 2000 TEMPERATURE, OF FIG.84-SUMMARY PENETRATION DEPTH CURVES. TYPE 310 ALLOY,HEAT X11306.EFFECT OF TIME.

ZE>^~~ rRUN NO. DURATION,HR. ~8 0 — 0 16 100'-3 r 1 —- E 21 30 wO 0 o.f CONTRL 1600 1700 1800 1900 IJ UJ c 100 FIG.85-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY, HEAT X 11338. EFFECT OF TIME.

I%~ I ~~~RUN NO. DURATIONHR. 0 — 01 16 100 I V --- 21 30 N 0 2 CD0 -- 24 10 I Pw 1.5 z ____________ —-----------— ^ ~ —-- I 0 ~1)!a 1.5 Iz'a 1.0 0.5 CONTROL 600 1700 1800 1900 2000 TEMPERATURE, "F FIG.86-SSUMMARY PENETRATION DEPTH CURVES. TYPE 310 ALLOY,HEAT X 11338. EFFECT OF TIME.

gJla RUN NO. DURATIONHR. 0 00 616 100,, —'~ —- 21 30 -- 0 cn PI L. QW O [3 —- 2:4,I 00 Jw CONTROL 1600 1700 1800 1900 2000 TEMPERATURE, ~F FIG.87-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY, HEAT X27258. EFFECT OF TIME. TYPE 310 ALLOY, HEAT X 2:7258. EFFECT OF TIME.

I6r |~ ~RUN NO. DURATION, HR O —— 0O 16 100, ~ 1- - 21 30 0 D- --— 0 24 1 0.W I-. 0 w C i, ] I1. 5._.. 00.5 CONTROL 1600 1700 1800 1900 2000 TEMPERATURE, OF FIG.88-SUMMARY PENETRATION DEPTH CURVES. TYPE 310 ALLOY, HEAT X27258. EFFECT OF TIME.

I^ ~RUN NO. DURATIONHR. o 0 —-— O 17 100 f IA — -- 22 10 I-a 0 I ^~ D —- — D 23 30 CL w LjJ cr -cn 1500 LL U. 0 y00om —--- ^^ —--— ^.............. ID z od 500 CO NTROL 1600 1700 1800 1900 2000 TEMPERATURE, OF FIG.89-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY, HEAT X45558. EFFECT OF TIME.

:~)r~> | RUN NO. DURATION, HR. O -0 -— 0 17 100 o ~ 1 — — A 22 10 -O R) 0 u3) 0- - 23 30 I 0 ____________ z I" C> 0.5 1==-== --------- E LU CONTROL 1600 1700 1800 1900 2000 TEMPERATURE, ~F FIG.90-SUMMARY PENETRATION DEPTH CURVES. TYPE 310 ALLOY, HEAT X45558. EFFECT OF TIME.

I^ ~ RUN NO. DURATIONHR. S 0 —-- ~ ~ 17 100 ^'a r 22 10 o j z 0 — 23 30 UJ W C, UJ (0 150 CD U. IL 0 0 / W 5 00 -I.. - I _ _ __ ^ 500 \^''=^ ^ —^- ---- Jr^^ l~~~~~~~ —I T-' - L^' CONTROL 1600 1700 1800 1900 2000 TEMPERATURE, OF FIG.91 - SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY, HEAT X46063. EFFECT OF TIME.

RUN NO. DURATON, HR.:, 0 - 0 001 17 I00 0 >X ~ -- — 2a 022 10 I ----- 2 3 30 al L5 0 z I0.5 ujS~ -- - CONTROL /S 1600 1700 1800 1900 2000 TEMPERATURE, ~F FIG. 92-SUMMARY PENETRATION DEPTH CURVES. TYPE 310 ALLOY, HEAT X46063. EFFECT OF TIME.

%~?~ |I ~ RUN NO. DURATION, HR. 08 O —— 0 I 017 100 ~0 ~'1 ~- 22 10 iL 0.1 Z1~0 ~ - 23 30 U. 0 c C 100 o- PNRTN EQ CCU S T' D- CONTROL 1600 1700 1800 1900 2000 TEMPERATURE, ~F FIG.93-SUMMARY PENETRATION FREQUENCY CURVES. TYPE 310 ALLOY,HEAT X46572. EFFECT OF TIME.

~> rRUN NO. DURATION,HR. o 0 —O —-- 17 1700 O r - — a 22 10 r', 0 23 3-*- 0 2l: LJ cr -r A.t z 1.5 —........ 0 I0.b CONTROL 1600 1700 1800 1900 2000 TEMPERATURE, F FIG.94-SUMMARY PENETRATION DEPTH CURVES. TYPE 310 ALLOY, HEAT X46572. EFFECT OF TIME.

Web from 3urface. TyPe 309+Nb Web from Pits. T pe 710 Alloy, Alloy. 1900OF and 100 hr. 750x. Heat 64270. 1.800F and 100 hr.. * Cross Section. Unetched. 750x. Cross Section. Un.etched. eough Web. Type 310 Alloy, Heat Smooth Fissures. Type 310 Alloy, 6417Alloy. 1900F and 100 hr. 750x. Heat X 11364270. 100F an 100 hr. Cross Section. Unetched. 750x. Cross Section. Unetched. TADO TR 54-120 120 Fig. 97. Type III Penetration. Fig. 98. Type IV Penetration. Rough Web. Type 510 Alloy, Heat Smooth Fissures. Type 310 Alloy, 64177. 1900~F and 100 hr. 750x. Heat X 11558. l&00~F and 100 hr. Cross Section. Unetched. 750x. Cross Section. Unetched. WADC TR 54-120 120

\" d,,' m-. *'4 Fig. 99- Type V Penetration. Fi~. 100. Type VI Penetration. Developed Web. Type 310 Alloy, Coalesced Web. Type 310 Alloy, Heat X 11338. 1900~F and 100 hr. Heat X 11338. 2000~F and 100 hr. 750x. Cross Section. UnetcLed. 750x. Cross Section. Unetched. _* *,1 ~,, v *..': * \,,.'. - t 64270. As Received. 750x. 64270. 1700F and hr. 750x. Sectio.: Elc...E.t:.E..ec'._-,.,=,.w.. ~~''*.^ I'-~.-,r omi c Acid -. C' Fig. 101. Grain Voundaries Fi.. 10 TPenetration and Grain before Testinb. Type 310 Alloy, Structure. Type 510 Alloy, eat Heat 64270. As r aceived. 750xr. 6ea 270. 17001. 0F an d 10 r. 750x. 5ross Section. Etch. Electr. Cross Section. Etch: Electr. Chromic Acid. Chromic Acid. WADC TR 54-120 121

y"; —'-**'*A''.<..'^:(. to- >' r',,. * J.' ~'' " " ^ - \.:-. r \ --- " 4 * N, i d * i 10. PP" a ~! r 4'-; - - 7 | ^'',. -' -/ K f f Fi.- 105. Penetration and Pre- FiIg. 104. Penetration and Precipitation. Type 310 Alloy, cipitation. Type 310 Alloy, Heat X 11506. 16000F and 100 hr. Heat X 11506. 19000F and 100 hr. 750x. Cross Section. Unetched. 750x. Cross Section. Unetched.''.'' S,,., _.,. ~ - qI,, 7K1~~~~~ ~ ~~~~~~. 4'l'j' - Se. Fit. 105. Penetration and Pre- F:i.. 106. Penetration and Precipitation. Type 510 Alloy, cipitation. Type 510 Alloy, Heat X 11306. 2000~F and 100 hr. Heat X 11506. 2000~F and 10 hr. 750x. Cross Section. Unetched. 750x. Cross Section. Unetched. vIALC TR 54-120 122

..750x Cross SectionsI. S. Electr. romic P ~)~.' 11338. 900 and 100 r. X 11338. 1900F and 100 r. SElection. CromiUnetche Acd. Etch uraamis Reaent.. -WAC TR 54-120 125 Type 310 Alloy, Heat X 45558. ~ 09+ Alloy. 1700OF" an f:'. ~O". ~' ~'.'C'o Eletr. Choi TRci-2012

*~. -11 1.:.1... -;.''. 1 1.'; " a' - *d0h. l x Crs H X1. 0 a.. / IF * *'\ / s' N' " ri /.-... //. N. j.E~ /t "-r-. " A'c','., *.,. -....:. Fig~ 1. 11.. at ri and BoundaryFig. 114. Oxygen-Rich Layer. Precipitation. Type 310 Alloy, Heat X 27258. r~ I Heat X 11336.. 1900~F and 100 hr. 1900~F and 10 hr. lOOx. 250x. Cross Setion. Etch Ma i'ross Section. Oblique L Electr. Ch c Etch: Electr. Chromic Acid. F ig'ADC TR 54-120 124''=9.-. *". Heat ~ F 1 ad 10 h l 250x Co S o E. g.,ADO T 51 (r \~~~~~~~~~~~~~~~~~~~~:-:::..:{ r~~~~, ~.

~! —'~: " ~ 0 ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~..............:: i"O~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~71................'~ ~:':~' ~:?~ L>~:O~~~~~~~~~~~~~~~~~~~~~~~~~~.................................................................:0 -:0 k.~::'~. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~........................... ~-:~ 0:~ ~..~.,~,~:~ ~~~~~~~~~~~~~~~~~~~~ ~~~~~ ~~ ~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~...........................................!:~ ~3,::;!'a:~:~l'~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~~.......i,i:'~ ~~ t~:............ 1-1 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0............ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~,..i':i,~:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~................... a:............................ k..........~:;?:::!;.......::.:..: i'? ~~~~~~~~~~~~~~~~~~~~~~~~....................................-....................................................... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~....:o:~:':~..........."~..:.:~?:~::71::':..........':!!:: i;,~,:?~: "' -,~,~ ~:.:e.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.................................. v~~~~~~~~~~~~~~~~~~~~~~~~~~~~..:?................. ~.................::.~.......;t-~ 0 i:.'~' "',.''"":?:......................:4;i(...... -o ~ -,~..~-~-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~..................-............. ~~~~~~~~~~~~~~~~~~~~~~~~~.......................... ~t 0 ":~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~......................; C It......,~ ~ ~ ~ ~:::. - -...............

UNIVESIY FICIGAN 3 9015 03524 3610