constructionengineeringINEI POT.0research Jnay17laboratory__ _ _ _ _ _ _ _ _ _ _ _

tl FRACTURE CHARACTERISTICS OF TWO HIGH-STRENGTHI,LOW-ALLOY AND TWO STAINLESS STEELS

P0

byJ. K. Scott

E. P. Cox

DODG

FEl 1.E

Approve*d for public Mew~s; distribution unlimited,

The contents of this report are not to be used for advertising, publication, orpromotional purposes, Citation of trade names does not constitute anofficial indorsement or approval of the use of such commercial products.The findings of this report are not to be construed as an official Departmentof the Army position, unless so designated by other authorized documents,

4'I

DESTROY THIS REPORT WHEN ITIS NO LONGER NEEDEDDO NOT RETURN IT TO THE ORIGINA TOR

A

U NC LAS SIF NJSECURITY CLASSIPPICATION Of THIS PAGE (it.., Pat. irrtefedI

REPOR DOCMENTTIONPAGEREAD INlSTRUCTIONS_____ REPORT__DOCUMENTATION __PAGE BEF~ORE COMPLETING FORM

1, 1W~uUL .. 2. GOVT ACCESSION No. 3. RECIPIENT'S CATALOG. NUM13ER

C CERL-IR-M-ZOD i 9IETWO STAINLESSSTErtS. a.-. INTEI;M E "O

tOACTURE LARACTERISTICS OF rwo HlIGH-STRENGTH INTRI6PEFORMING ORG. 149PORT NUMBER

7. S. CONTRACT OR GRANT NUMUER(a)

~J. K. Scott* E. P. Coy.

S. RiIGOdAIZTTN AME AND ADDRESS `10. PROGRAM ELEM 11LTPROJECT, TASKCONSTRUCTION ENGIONEERING RESEARCH LABORATORY AREA& 6ORK meansP.O. Box 4005 LNW FChampaign, IL 61820 7 '... 4AI761102AT22 A2 002

I.CONTROLLINQ OFFICE NAME ANDt A111MRE5A

45

Unclassified115a. DECL ASSI F1 CATIOtN/ DOWNGRADING

SCH EDULE

1S, DII5TR BUTION STATEMENT (at thle RoPort)-

A Approved for public release; distribution unlimited.

17. OISTRIBUTION STATEMENT (of Ili. shetat en tere~d In Block 20, It different from, Repoert)

IS. SUPP~LEMENTARtY NOTES

Copies are obtainable from Nktlonal Technical Information ServiceSpringfield, VA 22151

14. KEY WORDS (C~ontinu.e or. rioera. side It necesawey and Identitly by, block numnber)

temper-embri ttl emernthydrogen-embrittlementtransgranular fracturesintergranular fractures

20. ABSTeACT (Co 1. It nw.oemy ap ki Id~n~Ily by block number)'he fracture i.haracterirtics of two high-strength, low-ailoy structural steels (ASTM

A-588 and A-242) and two stainless steels (AISI 416 and 17.4PH) were analyzed underItensile, fatigue, and Impact loading conditions. The effects of hydrogen. and temper.embrittlement on the materials' behavior when fractured under tensile and fatiguefonditions were investigated.

The structural steels were found to be unsusceptible to teniper-embrittlement. ASTM

A-588 was found to be susceptible to hydrogen-embrlttlement; A-242 wans not found Iu- -ý.DD1473 EDITION OF tMIV WSIDOSOLEY R UNCLASSIFIED /,T

DO I AM 13SECURITY CLAISIFICATION OF THIS PAGE (11ba Dael fintered)'

r!UNCLASSIFIED

SICURITY CLASSIFICATION OF THIS PAGIKOWII Date ClXlerd)

Block 20 continued.

",eptible, but this could be attributed to the abnorni .v, I .-ýhiness condilion of theas-received material. The stainless steels were found to be unsusceptible to the hydrogen.charging procedure performed in this study. Mechanical tests showed large variations invalues for the tempered stainless steel specimens, although the fracture surfaces appearedvery similar, These findings reaffirmed the generally acceptc.; concept that visual observa.tLion is not a sufficient method for determining temper.embrittlement, but must becombined with mechanical testing to reach valid conclusions.

III

LI' i 21

"UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAGE(WM1on Data Entered)

4 | 4~u I

FOREWORD

This investigation was conducted by the Metallurgy Branch, Materials and ScienceDivision (MS) of the U.S. Army Construction Engineering Research Laboratory (CERL).The study was sponsored by the Directorate of Military Construction, Office of the Chiefof Engineers (OCE), under Project 4A761 102AT23, "Structural Systems"; ScientificArea A2, "Facility Components";, and Work Unit 002, "Characterization of Fracture ofEngineering Materials," The QCDO number for this project is 2,02.001.

The OCE Technical Monitor is 1. A. Schwartz.

CERL personnel connected with the investigation were J. Aleszka, E Cox, R. Hannan,M. Khobaib, H, Lamba, and J, Scott, Dr. A. Kumar is Chief of the Metallurgy Branch, andDr. G. Williamson is Chief of MS.

COL J. E, Hays is Commander and Directoi of CERL and Dr. L. R.Shaffer is TechnicalDirector.

0.

Hill Whllt Suellue

VOC Btf SOOiN C3WNOUIC 0oilScl "

JU 'TIf ICArION .........................

.................................. ............................. ...IV................................. .. ....

3 atl

CONTENTS

DO FORM 1473 1FOREWORD 3LIST OF TABLES AND FIGURES 5

INTRODUCTION ............................................ 9ObjectiveApproachMode of Technology TransferBackround

2 EXPERIMENTAL PROCEDURE ................................. 11, Materials

Speoimen Fabrication and TestingHydrogon-EmbrittlementTemper.Ermbrittlement

3 RESULTS AND DISCUJSSION ................................... 14ASTM A.5B8ASTM A-24217.4PH416

4 CONCLUSIONS ............................................ 17

FIGURES 18REFERENCES 54DISTRIBUTION

4

TABLES

Number Page

I Chemical Comnposition of Steels 12

2 Mechanical Properties of Steels 12

3 Absorbed Impact hEnrgies of'Steels 13

FIGURES

Number Page

I Equiaxed Dimples Containing Inclusiuns, 4250X 18

2 Elongated Dimples, 3750X 18

3 Cleavage Facets, 400X 19

4 Quasi.Cleuvague Facets, 1600X 19

5 Fatigue Striations, 4000X 20

6 Intergi anular Fracture, 30OX 20

7 Mlcrustructure of ASTM A-588 Steel, 200X 21 :

8 Micrustructure of ASTM A-242 Steel, 200X 21

9 Micrustructure of AISI 416 Steel, 200X 22

10 Mlcrostructure of 17.4PH Steel, 20OX 22

I I Specimen Geometry f(-cr Tensile, Fatigue, and Charpy Impact Tests 23

12 Tensile Fracture Surface of A.588 Steel, 8X 24

13 Dimple Rupture In A.588 Steel, 150OX 24

14 Fatigue Fracture Surface of A.588 Steel, 8X 25

15 Dimple Rupture in an A.5X8 Fatigue Specimen, 75OX 25

16 Fatigue Striations in A.588 Steel, I SOOX 26

17 Tensile Fracture Surface of Hydrogen-Embrittled A-588 Steel, 8X 26

18 Cleavage Facets In ltydrogen.Embrittled A.588 Steel, 1500X 27

19 Fatigue Fractuie Surface of Hydroger-Embrittled A-588 Steel, 8X 27

FIGURES (cont'd)

Number Page

20 Dimple Rupture and Fractured Pearlite Colonies in tlydiugen-EmbrittledA-588 Fatigue Specimen. I 50OX 28

21 Intergranular Fracture and Fracture Striations in a Ilydrogen.LFmbrittledA-588 Fatigue Specimen, I50OX 28

22 Fracture Surface of an A.588 Charpy Specimen Tested at -196 0C, 8X 29

23 Cleavage Facets in an A.588 Chai py Specimen Tested at .1 960C, I OOOX 29

24 Cleavage Fracture and Dimple Rupture in an A.588 Charpy SpecimenTested at .103'C, IOOOX 30

25 Cleavage Fracture and Dimple Rupture in an A-588 Charpy SpecimenTested at 23°C, 25OX 30

26 Fracture Surface of A.588 Charpy Specimen Tested at 230C, 8X 31

27 Tensile Fracture Surface of an A-242 Longitudinal Specimen Tested at230C, I 2X 31

28 Tensile Fracture Surface of'an A-242 Transverse Specimen, 12X 32

21) Inclusion.Gencrated Dimple Rupture in an A.242 Transverse Specimen,50oX 32

30 Fracture Surface of a Hydrogen-Charged A.242 Longitudinal TensileSpecimen, l0X 33

31 Dimple Rupture in a Hydrogen.Charged A-242 Longitudinal TensileSpecimen, 60OX 33

32 Fracture Surface of a Hydrogen-Charged A-242 Transverse Tensile Specimen,lox 34

33 Eloniated Inclusions in a Hydrogen-Charged A.242 Transverse TensileSpecimen, 55X 34

34 Cleavage Fracture and Dimple Rupture ir a Hydi ogen-Charged A.242Transverse Tensile Specimen, 1400X 35

35 Cleavage Fracture and Dimple Rupture in an A-242 Longitudinal FatigueSpecimen, 60OX 35

36 Cleavage Fracture in an A-242 Transverse Fatigue Specimen, 70OX 36

37 Fracture Surface of a Hydrogen.Charged A.242 Transverse Fatigue Specimen,14X 36

6

.5 . . .- . .. l. ... .- . ..

FIGURES (cont'd)

Number Page

"i" I)imple Rupture and Cleavage Fracture in a Hydrogen-CIharged A-242Transverse Fatigue Specimen, 250X 37

39) Fiacturc Surface of an A-242 Transverse Charpy Specimen Tested at .A960C,lIox 37

A 40 Cleavage Fracture in an A-242 Longitudinal Charpy Specimen Tested at,196"C, SOOX 38

41 Grain Boundary Precipitation in A-242 Ch;irpy Specimen Tested at -196 C,I OoX 38

42 Fracture Surface of an A.242 Transverse Charpy Specimen Tested at O0C,lox 39

43 Fracture Surface of an A.242 Transverse (,harpy Specimen Tested at 23°C,loX 39

44 Fracture Surface of alL A-242 Longitudinal Charpy Specimen Tested atI 21"C, lOX 40

45 Cleavage Fracture in an A-242 Longitudinal Charpy Specimen Tested at230C, 50OX 40

46 Dimple Rupture and Cleavage Fracture in an A-242 Transverse CharpySpecimen, 50OX 41

47 Fracture Surface of a Solution Ileat-Treated 17-4PllTensile Specimen, 12X 41

48 Dimple Rupture in a Solution Heat-Treated I 7.4PH Tensile Specimen, 700X 42

49 Fracture Surface of a 17-4PH Tensile Specimen Aged at 4820C, 12X 42

50 Inclusions in a 17-4PH Tensile Specimen Aged at 538"C, 350OX 43

51 Dimple Rupture and Microvoid Coalescence in a 17-4Pii Tensile SpecimenAged at 454'C, 700X 43

52 Fracture Surface of a 17,4PH Hydrogen-Charged Tensile Specimen, I OX 44

53 Dimple Rupture in a 17-4PH Hydrogen-Charged Tensile Specimen, 1OOX 44

I 54 Fracture Surface of a Solution Heal-Treated 17-41'Pi Fatigue Specimen, 9X 45

55 Quasi-Cleavage Fracture in a Solution Heat-Treated 17.4PH FatigueSpecimen, 7OOX 45

56 Fracture SuT.ace of a 17-4P11 Fatigue Specimen Aged at 4820C, lOX 46

*1i

..................................

FIGURES (contd)

Number Page

57 Quasl.4leavage Fracture in a 17-4PH Fatigue Specimen Aged at 482°C,I 00X 46

58 Brittle Inclusiun in a 17.4P0 Fatigue Specimen Aged at 538CC, 2500X 47

59 Fracture Surface of I7.4PH Solution Heat-Treated Charpy SpecimenTested at -196'C, loX 47

60 Fracture Surface of 17.4P11 Solution lieat.Treated Charpy SpecimenTested at 121C'C, IOX 48

61 Fracture Surt..ce of an As.Quenched 416 Tensile Specimen, lOX 48

62 imntple Rupture and Microvoid Coalescence In a Quenched 416 TensileSpecimen, 70OX 49

63 Fracturu Surface of a 416 Tensille Specimen Tempered at 593'C, 11 X 49

64 Fracture Surface of an As-Quenched 416 Fatigue Specimen, lox 50

65 Quasi-Cleavage Fracture In an As-Quenclied 416 Fatigue Specimen, 1400X SO

66 Fracture Surface of As.Quenched 416 Charpy Specimen Tested at 230C,lox 51

67 Dimple Rupture and Cleavage Fracture in an AM.Quenched 416 CharpySpecimen Tested at 230C, 50OX 51

68 Fracture Surface of a 416 Charpy Specimen Tempered at 954*C andTested at 23"C, lOX 52

69 Dimple Rupture in a 416 Charpy Specimen Temperedi at 954"C andTested at 230C, IOOOX 52

70 Dimple Rupture and Cleavage Fracture in a 416 Charpy SpecimenTempered at 593 0C and Tested at -196"C, 50OX 53

71 Intergranular Fracture in a 416 Charpy Specimen Tempered at 593'Cand Tested at -196"C, IOO0X 53

.... ....

FRACTURE CHARACTERISTICS OF Many common structural metals fracture under

'TWO HIGH-STRENGTH, LOW-ALLOY monotonic load in a ductile fashion by microvoidAND TWO STAINLESS STEELS coalescence, Microvoids are small, discontinuous voids

which nucleate at grain boundaries, second-phaseparticles, or other sites where strain discontinuitiesexist. As the applied load increases, the microvoidsgrow, coalesce, and eventually form a continuous

1 INTRODUCTION fracture surface which exhibits numerous cup-likedepressions called "dimples"; this is referred to as"dimpled rupture," which is generally associated with

ObJactive ductile failure.The objective of this study is to use scanning elec.

tron miscroscopy (SEM) to establish and characterize The shape of' these dimples is strongly influencedthe nature of fractures in engineering materials. This by the orientation of the major stress axis to thecharacterization is to be accomplished by laboratory rolling direction (texture) of the material. Equiaxedsimulation of those types of fracture modes and dimples (Figure 1) result under local conditions ofmaterial embrittlements most commonly encountered uniaxial tensile stress, while elongated dimples (Figurein in-service failures, This report discusses a segment of' 2) result from failure caused by shear stress. Failurethe study covering fracture characteristics of two under this loading mode is expressed as maximumhigh-strength, low.alloy structural steels and two shear stress criterion or Tresca failure, Dimple sizestainless steels, depends on the number of fracture nucleation sites,

grain size, nslcrostructure, and the relative ductilityApproach of the metal,

The fracture characteristics of two high-strength,low-alloy structural steels (ASTM A-588 and A-242) In polycrystallne body-centered cubic (bee) metals,and two stainless steels (AISI 416 and 17-4PH-) were macroscopic cleavage fracture propagates throughanalyzed. A-588 and A-242 have ferritic-pearlitic grains, changing directions as it crosses subgrain boun-microstructures, AISI 416 is a martensitic stainless daries or passes from one grain to another. Cleavagesteel, and 17-4PH is a precipitation-hardonable, mar. fractures (Figure 3) which are usually associated withtensitic stainless steel. The materials weri fractured brittle failure, occur along a well-defined crystallo.under tensile, fatigue, and impact loading conditions, graphic plane within a grain; in ferritic steels whichThe effects of hydrogen- and tomper-embrittlement have a bet, crystal structure, this plane has the typeon the materials' behavior when fractured under tensile 100 orientation. The change in orientation betweenand fatigue conditions were investigated, grains and the imperfections within grains usually

produce easily distinguished markings on the fractureMode of Technology Transfer surface. A cleavage fracture propagating across grains

The information obtained from examination of forms arrays of cleavage steps or "river patterns."simulated failures will be combined with data gained These river patterns are rootlike networks of cleavagefrom analysis of in-service failures to provide the basis facets propagating on different levels.'for a reference manual to be used in future failureanalysis. In precipitation-hardened martensitic steels, the size

and orientation of the available cleavage planes may beBackground poorly defined because within the grains of the priorFracture of Steels austenite, expected (true) cleavage planes were replaced

Fracture surface features can be divided into two by smaller, ill-defined cleavage facets, which arecategories according to the mode of fracture: trans. referred to as quasi-cleavage planes (Figure 4).1granular (through the grains and across grain boundaries)or lntergranular (around the grain boundaries). Tr-ns- 1 R. Honda, International Conference on ractare, Sendai,granular fractures can occur by void coalesL...ce, Japan (1965); and J. R. Low, Jr., B. L. Averbach, et al.,rupture, cleavage, or fatigue. Intergranular fractures Fracture (John Wiley, 1959), p 163.occur by grain boundary separation either with or 'Fractography and Atlas of P'ratoiraphs, ASM Metalswithout rrdcrovoid coalescence, Figures I through 6 liandbook, Vol 9, 8th Edition (American 3ociety for Metulsillustrate these types of fractures. IASMI, 1974), p 65.

9

Fatigue fracture resultis t'ronl con~trinuous iiiIwr(J.opic I lyd roger emobritt icinvit produces a sharp l oss inprgeso of* a crack caused by the appiicatimin tit a ductility, this loss is miost severe at roomi temperaturecyclic load. The mechanism of lwigue crack Initiation arid slow strain rates. The fatigue lives of steeIs subjectedis believed to involve slip plane fracture caused by to electrolytic hydrogeri-charging' or a high pressurere'petitive reversing of* the operative slip s~ystets; on hydrogen atntoisplicie5 have shown significant reduc-the suifauce of' the metal. 3 (C rack growth caused by tions, The miode of huilure of a hydrogen-embrittledrepetitive loading sometimes rc~ults in a f'racture sample depends on such variables as type of material,surface which exhibits closely spaced f'atigue stria- mcethod of' loading, anid envivronment, martensitlc stoolsLions or parallel markings us shown Iin Figure 5. Each are ttw most susceptible,tatigue striation represents the advance of a crack frontduring one loading cycle, ihe striualions may be absent Many theories concerning the mechanism of hydro.or may differ In appearance depending on such variables gen embrittlement have been proposedl, but no oneas type of mauterial, level anid f'requency of' applied theory has been able to account for more than half threstress, anid environment, experimeuntal results, The hydrogen embrittlement

theory proposed by Zaffa' was based on aturrtlc hydro.Intergranular fracturt: caused by temlper-enmbrittle- gen d iffusing through lthe metal lattice, precipitating in

itent often results from segregation of impurity atoms Internal voids as molecular hydrogen, and creating highalong tit(- grain boundaries (Figure 6). T'his tegrugatioit pressures. It was assumed that high pressures in the voidscauses the grain boundaries to be extremely weak, combined with lthe externally applied stress fracturedresulting in separation by a low energy fracture, At lthe metal, Diffusion of hydrogen to the voids couldhigh magnifications the f~racture surface reveals progres- explain the strain rate and temperature dependecne ofsive fracture along lthe embrittled grain boundaries of' hydrogen embrittlement, i~e., hydrogen eirbrittlementthe prior austenite grains, This f'racture inechanism appears to be a tate process. However, this theorygenerally occurs only in quenched and tempered requires a regular array of pre-existing internal voidsmaterials. along with a source of hydrogen, a requirement that is

Inconsistent with the results of studies on the structureIlydrogen-Embritlemc'nt of'SteeI of hydrogen-ambrittled steels,6 Petchti rejected the

Hydra en-enbiblttdement has receivud considerable idea that cracking Is propagated by Internal hydrogenattentioni since hydrogen Is easily introduced into pressure. fie sugge,;ed that adsorption of hydrogen on

metals by malting, casting, welding, corrosion, and the surfaces of miucrocracks or voids reduces the surface .electroplating. Most investigations of hydrogen embrit- free energy, resulting in a decrease In the energy neededtiement have beer. performed under sustained-load or for crack propagation. He further suggested thatslow strain rate tensile test conditions, however. Uttle although plastic deformation may produce many dis-

research has been published on the fatigue properties connected ndtcrovracks, the microcracics do not reducetof hydrogen-embrittled steels, fracture stress significantly in the absence of hydrogen.

However, when hydrogen is present, It diffuses into theThe degree of' embrittiemnent generally increases region of* the advancing tip and is adsorbed at the crack

with Increasing hydrogen content and has the greatest tip, thereby reducing the energy required to propagate

effect on the high tensile strength iron-base alloys.

Aluminum Alloys," AC7'A Metailurgica, Vol 11 (1963), p 703; Vol 57 t t957), pp 682-697; W. Beck, Electrochemical Tech-

Stress Iatigu.,,Philosophical Magazine, Vol2 (1962), p 847. Smith,UBritsht Weldingt Journal, Vol 14 (1967). pp 493-502.4p. Cottorill, ,,,he Hydrogen Enibrittlement or Metals," 6W. A. SplIzig, P. MI. Talda, and R. P. Wei, "'Fatigue -Crack

Progrersipe Materials Science, Vol 9, No. 4 (1961); A. S. Total- Propagation and Uractographic Analy'sis of IS Ni (250) Nfarag-man and A. J. NMe~vily, Jr., Fracture af Structural Materials Ing Steel Tested In Argon adHydrogen Environments,"(Ouhn Wiley, 1967); 1. M. Blernstein, "The Rote ol' liydiogo~i Engineering Fiacture Mechanics, Vol I (1968), pp 155-165.in the inibrittlemntri 01iroin and steel,"' Materials Science and 'C. A. zafrfe, journal o/ Iron and Steel institute, Vol 154,E~ngineering, Vol 6, No. 1 (1970), pp 1-19; W. Beck, L. J. No. 123 (1946).Jankowski, and P. [Uiher, Hydrogen Stress Cracking of'High. I.STelmradAJ.M viyJ.,FcteoftueStrength Steels, NADC-MA-7 140 (Naval Air DevelopmentA.STtemad.J.M viJrFcueofSr-('enter, 197 1); and Hydrogen Embrirtlernment Testing, ASTM tural Materials (John Wiley, 1967).STP543 (American Society rut Testing and Materials IASTMJ , ON. J. Petch, "The Ductile Uracture or Polycryitalline-lron,"19741. Philosophical Magazine, Vol 1 (1956), pp 136-19 1,

10

- ,,,~,,QJ

the C_.sck- I ownulai 0 lhas prupowd that ionic h ydroge n, ifr a fhiort time and quenrching to room temperaturewhich i, evenly distributed througiout the rnetal ':1ititiiates the e nhrittleiment.lattice, ts harmless because its concentration is so small.The criticald factor is the segregation of hydrogen, It is generally believed that the equilibrium segrega-which, under an applied stress, diffuses to regioins of' tion of various impurities to prior ut'stenite graintiaxial stress near pre-existing voids in the steel, Thus, boundaries is the fundamental meal, sin of temperonly hydrogen in the stressed region of the lattice or at earbrittiement I.ow and his associat demonstrateda crack tip is responsible for hydrogen embrittlemervt. the influence of specific impurities s. at as antimony,Also, since hydrogen embrittlerneit is observed in tin, phosphorus, and arsenic and alloying elements suchtransition metals having vacancies in the third subshell, as nickel and chromium in promoting embrittlement.Troiano hypothesized that hydrogen in the rtressed Marcus and Palmbergt I found that wrnen fractureregion of the atomic lattices near the voids gives up its occurs along prior austenite grain boundaries in low.electr•ons to the third subshells of the host atoms, alloy steels, significant amounts of antimony, tin, andfilling the .acancies in the third band, ind thereby phosphorus (IO0 to 500 times the bulk concentration)decreasing tile binding energy of cohesiveness of the are present on the grain boundaries. The presence ofatoms in thie lattice in this region. One flaw in this both nickel and chromium appears to .ause moretheory is thut specimens slowly cooled In a hydrogen segregation of antimony, tin, or phosphorus to theatmosphere or cathudicaily charged at room tempera. grain boundaries than when either is present alone,ture contain cracks even without externally appliedstress. Recent experiments by Ohtanl t suggest that a

central feature of temper embnittlemaent is the redistri-It should be noted that the deleterious effects butior of solute during carbide precipitation. The

associated with hydrogen can be removed by outgassing study showed that eliminating carbide precipitation inor "baking" the material for a short time in a tempera. antimony, and phosphorus-doped alloys eliminated thetire range of 10J°C to 3000C.' At this temperature remaining embrittlement resulting from equilihriumrange hydrogen ieaddy diffuses to free surfaces and segregation. Ohtanl concludes that embrittlement isescapes as diatomic hydrogen. caused by the presence of impurities in the carbide.

ferrite interfaces resulting from piling-up of the impurityTemper Embrftrlenwnt ahead of a growing carbide,

Temper embrittlement is a serious problem asso.ciated wittt quenched and tempered steels whichresults in a serious loss in ductility and toughness; itis also sometimes manifested by a loss in corrosion, 2 EXPERIMENTAL PROCEDUREres- stance and a shift in the brittle-to-ductile transitionto higher temperatures.' Temper embrittlementgenerally occurs in low.alloy or stainless steels when Materialthey are tempered (reheated after austenitizing and The steels used in this study were 1 1/2 in.(j.81 cm)quenching) at temperatures ranging from 35O0C to thick plates of ASTM A-588 and A-242 and 5 1/8 in.550 0C. 3 Subsequent reheat treatment above 600WC (13.02 cm) and 5 5/8 in.(14.29 cm) rounds of 17-4PH

and AISI 416, respectively. Types A-588 and A-242 are10 A. Troiano, "The Role of lydrogen and Other Intersiftials high-strength, low-alloy structural steels generally used

in the Mechanical Behavior ot Metats," Transactions of the where weight savings or added durability are important.American Society for Metals (ASM), Vol 52 (196U). p 52.

IIA. S. Tetelman and A. J. McEvOy, Jr., Fracture of 141. R. Low, Jr., D. F. Stein, A. M. Turkalo, and R. P.

Structural Materials (John Wiley, 1967). Lafarci, Transactions of AIME, MT6TD, Vol 242 (1968),1IHeat Treating, Cleaning, and Yinithing, ASM Metals p 14-24.

Handbook, Vol 2, Bth Edition (1964), p 245; and R. r. Ault, IS H, L. Marcus, Jr. and P. W. %atmberg, "Effect of SoluteR. B. Holtmann, and J. R. Myes,lHeat Treatment oy'a Marten. Elements on Temper Embrittlement of Low Alloy Steels,"sitit Stainless Ste-l for Optimum Combination of Strength, Temper Embrittl.ment of Steels, ASTM STP 499 (ASTM,Touga•aess, and Stress C'orrosion Rerlitance, Technical Repoat 197 1), pp 90-103,AFML-TR.6B.7 (Air Force Materials Laboratory, April 1968). 1tHl, Ohtani, H, C, FenX, and C. J. McMahon, Jr., "Now

3J. R. Low, Jr., Faecture of Engineering Materials (ASM, Information on the Mechanism of Temper Embrlttlement1964), p 127; and C. J, McMahon, Jr., Temper Embrittlement of Alloy Steels," M~tallurgical Transactions, Vol 5 (1974),In Steel, ASTM STP 407 (ASTM, 1963), p 127. pp 516-5 I8.

I1

Their inicroatructures (Figures 7 and 8) consist of* The chemical compositions, mechanical properties,

regions of ferrite and pearlitc. Type 416 is designed for and Impact energies of the steels are shown in Tables 1,use in ftr",-machining stainless rnid heat-resisting steel 2, and 3, respectivelIywire and bars where optinium inachinability, generalcorrosion resistance, and high temperature service are Specimen Fabrication and Testingrequirad. The microstructure (Figure 9) is mairtensitic, Figure I11 shows the specimen geometry for theType 17.401i is used for hot-rolled and cold-finished tensile, impact, and fatigue tests. The A.242 specimensage-hafidening stainless And heat-resisting steel bars were machined with some specimens' longitudinal axesand shapes where corrosion resistance and high strength parallel to the roiling direction and some axes perpen-at room and elevated temperatures are required, It Is dicular to the rolling direction. The A-588 specimensprecipitatiun-hardenable and has a martensitic micro. were machined with their longitudinal axes parallel to

structure (Figure 10). the rolling direction.

Tsbla IIChemical Composition of Steels

ASTIS ASTM AlitEmetA-588, % A-242,% 416,% 17-011, %

C0.10-0.19 0.15 max. 0.15 min. 0.07IMn 0.90-1.25 1.00 max, 1.25 1.0

P0.04 max. 0.15 max. U.06 0.040.05 max, 0.0 max. 0.15 min. 0.03

Si 0.15-0.30 -1.0 1.0Cr 0.40-0.65 -- 12-14 15.5-17.5

MU ---- *- 0.60 (Opt.) --

NI02 0.002 ii.--- 3.0-5.0

V U.02--03.10 -- --

Table 21Mechanical Properties ot Steels

Tondle Strength, Yield Stroupt,Sitee ka (iPs) ki (Ma)

A-588,as-rolled 90,0 (621) 58.4 (403)A-588, hydrogen-embrlttled 88.8 (612.7) 59.2 (408.5)A-588. rcheat-treated 83.2 (574.1) 57.6 (397.4)

A-242, as-rolled, longitudinal 76.0 (524.4) 48 (331.2)A-242, as-rolled.transverse 75.2 (518.9) 48 (331.2)A-242, hydrogen-embrittled, longitudinal 76.0 (524.4) 49.6 (342.2)A-242, hydrogen-embrittled, transverse 72.8 (502.3) 47.2 (325.7)A-242, reheat-treated, longitudinal 73.6 (507.8) 48.8 (336.7)A-24 2.reheat- treasted. transverse 68.0 (469.2) 42,4 (292.6)

4 16, quenched 212.0 (1462.8) 168 (1159.2)416, tempered at 5930C. 110.4 (761.8) 92 (634.8)416, tempered at 316*C 193.6 (1335.8) 147.0 (1915.7)416, tempered at 9540C 192.8 (1339.3) 120 (828.0)

416, tempered at 483 0C 186.4 (1286.2) 144 (993.6)416, hydrogen-embrIt tied 2140C (1479.4) 134.4 (927.4)

17.4PH, xulutlun heist-treated 152 (1040.8) 97.6 (673.5)17-4PH, hydr uten-embrh ttled 152.8 (1054.3) 104.0 (717.6)17.4011:aged at 4820 C 164.8 (1137.1) 150.4 (1037.8)17-4PH aged at 5 38"C 164 (1131.6) 153.6 (1059.8)17.4011 aged at 4340 C 184.8 (1275.1) 156 (1076.4)

12

The ten~.ie tests were conducted at room tempera- for A-5hh and A-242 specimens were conducted at5ture at %train rates between 0.01 in./in.fmin, (0.01 cru! 10 cycles/sec in a tension/tension sinusoidal mode at

-~m/sin~ ~d 006 n./i.Imn, 0.0 cmcm/ In roomn temperature (23*C) using a SO-kip MITS unit.using a 50-kip NITS elect rohydraulic testing machine. The fatigue tests for the 416 and 17-011 specimensThe impact specimens, which were the Charpy V. were conducted using a Dynatup Chiarpy Precrackcr.notched type, were tested at A tange of temperatures The specimens were given an initial load of approxi-by using temperature control baths. The fatigue tests mutely 450 lb (204 kg) and cycled at a rate of 1400

II Tale 3cpm until fracture.

Absorbed Impact Enler lies of Steel Hydrogen EmbriftternentTestng bob.I!aieuy, To induce hydrogen emb rittlement, specimens were

Specimen Tempersture,'C ft-lb QOuke) cathodically charged in a solution of 10 weight percent112S0 4 and 0,3 weight percent AS2 0 3.- The AsO 3 was

A-588 .196 3.0 (4.1) used to promote absorption of hydrogen during cathod -A-388 -103 5.0 (6.8) Icpour in. The cathodic charging was conductedA-5 88 _30 13.0 11 7.6) at a2ur est f6m/n n 09 eon oAVA-5lI8 23 24.5 (33.2) a or cst f6m/n n(.3m/m)o

12 hours.A-242, longitudinal .196 0.5 (.7)A-242, longitudinal 0 4,0 (5.4) Temper Embrittlemrent

A-22, ongtudna 23 8.0(108) Since A-588 and A-242 are normally used in theA.22,inmgtudnal121 74,(10,3) hot-rolled condition, temper embrittlement is not a

A-242, transvorhe -196 0.0 (0.0) serious problem. However, several specimens wereA-242, transverse 0 2.0 (2.7) slowly cooled fromt the austenitizirig range in an attemptA-242, transverse 23 6.5 (9.8) to embrittle them.A-22, transverse 121 17.ý5 (23.7)

17.4PH, solution heat-treated .196 3 3.U (44.7) The 416 stainless steel specimens were placed in a17.4'2l soulln hat-teatd ~ 66. (9.2) preheated 538'C furnace, heated to 954*C, hield for

17.4PH, Nolutiot, heat-treated 23 73.5 (99.7) 30 minutes, then oli-quenched, This is a solution heat17.4Pl, solution hc.;t-tueated 121 8210 (121,2) treatment during which all precipitating elements are

taken Into solid solution. Four groups of specimens weret74)llagd at 8 9 , 00 then tempered according to the following schedules:

17-0P1aged at482 C 0 9,0 1212)

17-401l aged at 4820C 23 31.0 (42.0)17-01-1 agod at 4a20C 121 62.0 (84.1) 1. Place in preheated 593*C furnoce, temper for I17.40ll aged at 538"C -196 2.0 (2.7)17-4111-1aged lt 538a C 123 623.0 (341.2) 2. Place in preheated 3160C. furnace, temper for I

i7-411 gedat 4H C121 62. (L.i, hour, oil quench.

17.4PH aged tit4S40C .196 0.0 (0,0)17-01-1 aged at 454*C. 23 5.0 (6.8) 3. Place in preheated 954"C furnace, temper for I17-0P11 agcd at 454 C 121 28.0 (24A) hour, slow cool.

416, quenched .196 0.0 (0.0) '416, quenched 23 3.0 (4.1) 4, Place in preheated 483

0C furnace, temper forIA hour, slow cool.

416, tempered at $93 aC -196 1.5 (2.0)41,tmee t5930C 0 2035 (27.8) Schedules I and 2 are normalteprncodins

416, t :mperedatr5930C 23 18H.0 (24.4) tceue n smlt osbeemperitlng citonsl.416, t mpered at 5930C 121 29.0 (25,18) sheules. 3ad4smlt osbeebitigcni

426, tempered at 3 160C -196 0.5 (3)416, tempered at 316*C 0 4.0 t 5.4) The l7.4PH specimens were solution heat-treated426, tempered at 316 0C 23 3.5 (475 by a 5380C preheat followed by belogheated to 1038 0C,

426,temere et160C222 10. (1.6) held for 30 minutes, then oil-quenched. Three groups

416 tempered at 9 54 0C 23 12.5 (17.,0 of specimens were age-hardened using the following416, temper-ed at 403C 23 3.0 (4.1) heat treatments:

13

1. Place in 482"C preheated furnace foi I hour, it contained various mixed modes of fracture.Figure 20air cool. shows areas of dimple rupture interspersed with areas

of fractured pearlite colonies. The combination of2. Place in 538*C preheated furnace for I hour, fatigue striations and intergranular fracture shown in

slowly cool. Figure 21 is somewhat uncommon; inteigranularfracture occurred in the regions that had a preferred

3.. Place in 454"C preheated furnace for 1 hour, grain-boundary fracture path, in this case probably theslowly cool, result of hydrogen accumulation in the grain boundaries,

[I Schedule I is a normal age-hardening treatment, The cleavage fracture in the tensile specimen andschedules 2 and 3 are embrittling treatments. intergranulat fracture in the fatigue specimen show

that A-588 is susceptible to hydrogen embrittlement.Both mechanisms indicate a loss in ductility.

3 RESULTS AND DISCUSSION Figure 22 shows the fracture surface of an A-588Charpy inmpact specimen tested to failure at .196"C.

Summaries of the mechanical tests conducted in this A higher ntagnification of' the surface (Figure 23)study are shown in Tables 2 and 3, shows that failure occurred by cleavage.Numerous riverA....patterns are found on the facets, For the Charpy specS.ASTMA-688 Steel men tested at .103*C, the fracture surface consisted

The fracture surface of the A-588 control tensile primarily of cleavage facets with some interspersedsample is shown in Figure 12, Failure occurred by dimpled regions (Figure 24). When fracture occurred atinclusion-generated dimple rupture. Figure 13 shows -I 80C, the interior surface showed primarily cleavagevarious dimple sizes and several inclusions, The fracture facets; at the edges, however, regions of mixed cleavagesurface of the reaustentizted and slowly cooled A-588 and d(mples appeared. At room temperature (23*C),tensile specimen was very similar to that of the as-rolled one area in the interior of the surface failed by cleavage,sample, The mechanism of failure was also inclusion- however, most of the specimen had failed by a com-

generated dimple rupture, bination of cleavage and dimple rupture (Figure 25),which gave the surface a fibrous texture (Figure 26), M

Fatigue and tensile overload regions can be easily(identified on the fracture surfaces of the af a tolled ASTM AF242 Steel(Figure 14)nand reheatutreated A-588 fatigue spedimens, Figure 27 shows the fracture surface of the At242Failure in the tensile overload regions occurred by tensile specimen machined with its long axis longitudi-dimple rupture which initiated at inclusions (Figure nal to the rotling direction of the steelFailure occurred15). The fatigue regions consisted of parallel fatigue by inclusion.generated dimple rupture; extensivestriations Indicative of the step-wise, cyclic progression secondary cracking and some necking also occurred.of the crack front across the fracture surfaces (Figure The fracture surface of the A-242 tensile specimen16). machined with its long axis transverse to the rolling

direction is shown in Figure 28. Some neckar occurredThe resits of both the SEM examination and the before failure, A higher magnification micrograph

mechanical testing indicate that tomper-embrittlement (Figure 29) shows the mechanisms of failure to bedid not occur in A-588 steel, inclusion.generated dimple rupture. Since the plane of

fracture is parallel to the roiling direction, flattenedThe hydrogen-embrittled A.588 specimen (Figure inclusions can be easily seen on the fracture surfaces.

17) shows considerably less necking before failure thanthe control specimen (Figure 12). Further indication of The reheat-treated specimens' fracture surfaceshydrogen embrittlement is confirmed by a higher were similar to the fracture surfaces of the as-rolledmagnification micrograph (Figure 18) which shows specimens. Failure iii both the longitudinal and trans.large cleavage facets, The river patterns show the verse specimens occurred by inclusion-generateddirection of crack propagation within each facet. dimple rupture.

"The fracture turface of the hydrogen.embrittled The fracture surface of the hydrogen-chargedfatigue specimen (Figure 19) was quite complex since longitudinal A.242 tensile specimen was slanted

14

L9

towards a shear lip on one side (Figure 301. The specimens. The elongated inclusions provided weakmechanisms of failure were dimple rupture (Figure 3 1) zones through which cracks could more readily props-and some cleavage. The trant'verse hydrogen-charged gate when the long axis of the inclusion was in thetensile sample (Figure 32) shows elongated invisions same plane as the fracture,(Figure 33), Failure occurred by a combination ofdimple iupture and cleavage (Figure 34). The results of the SEM examination and the me-

chanical testing indicated that neither temper. norThe unembrittled A.242 fatigue specimens failed by hydrogen-embrlttlement produced any reduction in

a combination of cleavage and dimple rupture (Figure toughness in A-242. The absence of detectable hydro-35) in the tensile overload region and by cleavage gen.embmittlement is surprising, but can be explained(Figure 36) in the fatigue region. The fracture surfaces by the results of the Charpy tests (Table 3), The impactof the rolhat-treated fatigue specimens closely resem. energies of the as-received A-242 specimens werebled those of the as.rolled fatigue specimens, abnormally low, possibly because improper mill

treatment had already caused embrlttlement. SinceThe hydrogen.charged fatigue specimens produced additional embrittlement could not occur, no effect

similar fracture surfaces. The transverse specimen in from hydrogen charging was observed on the A-242Figure 37 shows the long cracks formed as a result of test specimens.elongated inclusion, Failure occurred in both specimens ,by a combination of dimple rupture and cleavage M7.4PHM Steel(Figure 38). The fracture surface of the 17.4PH solution heat-

treated tensile specimen is shown in Figure 47, FailureThe fracture surfaces of the A-242 Charpy specimens resulted from dimple rupture and microvold coalescence

were shiny and faceted --indicating brittle fracture, The (Figure 48). A small amount of necking occurredCharpy specimiens tested at liquid nitrogen temperature before failure.(-1960C) were quite flat, and showed no lateral con.traction (Figure 39), Both the longitudinal (Figures 40 The fracture surfaces uf thu tensile specimens whichthrough 46) and ti insverse specimens failed by cleavage; were quenched and subsequently age-hardened atin some places the fracture propagated along grain either 482*C, 454'C, or 5380 C were similar, Character.boundaries and the iractire surface reveals the presence Istic illustrations of these tensile fractures are asof some type of precipitao(Figure 40). The amount of follows: (1) Figure 49 indicates that some neckinglateral contraction on the fracture surface increased as occurred before failure, with shear lips developingthe testing temperature increased from OC (Figure 40) on all sides of dte fracture (in other specimens thisto room temperature (Figure 43), to 121C(Figure 44), phenomenon was not consistent, since shear lips oc.Failure occurred at all temperatures by cleavage curred only on one side); (2) Figure 50 shows extensive(Figure 45) in the center of tht specimens, and by a secondary cracking iround large inclusions; and (3)combination of cleavage and dimple rupture in the Figure 51 indicates that fallure occurred by the mech-shear lips (Figure 46), In all cases the Charpy impact anisms of dimple. rupture and microvoid coalescence.energy was much lower than normal. Even at roomtemperature the energy value was only 8.0 ft-lb ( 10.8 1). The hydrogen-charged 17-4PH tensile specimen is

shown in Figure 52. The fracture surface slants towardsThe effect of rolling direction on the fracture a prominent shear lip on one side. The mechanism of

surfaces of die A-242 specimens could be seen in the failure was dimple rupture (Figure 53).low magnification of fractugraphs (Figures 27, 28, 30,32, and 33), where elongated cracks resulting from The fracture surface of l-,c solution heat-tteatedrolled inclusions were noticeable in the transverse 174PH fatigue specimen was very flat (Figure 54), as,specimens. Although some elongation of cleavage evidenced by the square outline of the fracture. Failurefacets and dimples could be observed at the higher occurred by quasi-cleavage (Figure 55). The fracturemagnifications, the microstructures of the longitudinal surfaces of the 17-4PH fatigue specimens age-hardenedarid transverse specimens were very similar, and the at 482*C, 4540C, or 5380 C and hydrogen-charged aremechanisms of failure were the same, The effect of all quite flat, as exemplified in Figure 56. Failurerolling direction on the mechanical properties of the occurred typically by quasi-cleavage in all three speol-A.242 specimens was significant, as shown by the mens as illustrated in Figure 57, The 4820C and 538*CCharpy V-notch energy values for the transverse specimens had dark areas on the low magnification

. . .

micrographs, which at higher magnification were as-quenched specimens. Dark areas on the low magni-'hown to be unusually flat regions, as shown in Figure fication microugiphs of other speciniens,416 specimens58. The appeawance of this area indicated that fracture tempered at 482°C, and the hydrogen-charged 416occurred through a very brittle material or interface, specimen, were found to be very flat regions whereprobably soime kind of inclusion, as shown in the fast fracture had evidently occurred through brittlernicrostructure of Figure 58, inclusions. Similar areas were observed in some of the

17-4PH specimens.The fracture surfaces of the solution heat-treated

17,4PH Charpy specimens tested at liquid nitrogen, The 416 as.quenched Charpy specimens were fairlyOoC, room temperature, and 121C were very similar, brittle. At room temperature, only a very small amountThe amount of lateral contraction and the size of the of shear failure was present (Figure 66), Failure occur-shear lips increased with testing temperature and the red from a combination of cleavage and dimple rupturematerial became more ductile as exemplified in Figures (Figure 67). The 416 Charpy specimens which had59 and 60. The liquid nitrogen specimen failed by a been tempered at 482*C and 316'C developed fracturecombination of dimple rupture and cleavage in the surfaces very similar to those of the as.quenchedcenter and by dimple rupture in the shear lip regions, specimen, Failure occurred by cleavage at the lower IThe reuamning solution heat.treated specimens railed testing temperatures and by a combination of cleavageentirely by dimple rupture. and dimple rupture at room temperature and 121, C.

The specimens slowly cooled from 954*C or tempered

The solution heat-treated and ag'-hardened 17-4PH at 593*C were more ductile than the other 416 CharpyCharpy specimens were considerably more brittle than specimens. Extensive secondary cracking occurred onthe solution heat treated Charpy specimens. The speoS- the 954"C specimen (Figure 68), and failure occurredmerns ttcd at liquid nitrogen temperature showed by dimple rupture (Figure 69). The fracture surfacefractures which were very fiat; no lateral contraction of the Charpy specimen tempered at 593 0C and testedoccurred .nd failure was by cleavage. Results of the at liquid nitrogen temperature had a very fibrousCharpy V-notch test indicated an increase in absorbed appearance, The mode of fracture was quite complex;energy as woll as lateral contraction. The appearances most of the surface failed by a combination of cleavageof th ftracture surfaces of the specimens aged at 538 0C and dimple rupture (Figure 70), but areas of inter.and 482"(: wre very similar, but fracture surfaces of granular fracture were also observed (Figure 71), Thethe specimvnr kiged at 454%C were less ductile than the specimens tested at higher temperatures failed by4820 C and 983"C specimeni. dimple rupture,

AI16 416 5tel Neither of the stainless steels was found to beThe as-quenched 416 tensile 3pecimen is shown in susceptible to hydrogen.embrittlement, The effect of

Figure 61 after testing. The fracture surface slanted the tempering heat treatments on the stainless steelstowards a large shear lip on one side, and some necking is quite complex and, therefore, difficult to define.occurred prior to failure. The mechanisms of failure The tempering procedures can alter the yield strength,were dimple rupture and microvoid coalescence; some ultimate strength, and/or the ductility of a materialsecondary cracking occurred (Figure 62), The fracture simultaneously, As the tensile, yield, and impact datasurfaces of the tempered and hydrogen-charged tensile in Tables I and 2 show, a wide variety of values wasspecimens were very similar to that of the as.quenched obtained from the various tempering schedules.specimen, More extensive secondary cracking occurredin the specimen tempered at 5930C (Figure 63); very Even though the mechanical property values for thelittle cracking occurred in the 316*C, 4820 C, and tempered specimens varied greatly, the fracture surfaceshydrogen-charged specimens.The mechanisms of failure, were very similar, Intergranular fracture occurred in aiLe., dimple rupture and microvoid coalescence, were 416 Charpy specimen which had been tempered atthe same for all the tensile specimens, 593*C, but only when tested at liquid nitrogen temper.

atures. Specimens tempered at 3160C showed cleavageThe fracture surface of the as.quenched 416 fatigue at higher test temperatures and Charpy fracture energies

specimen is shown In Figure 64. The very fiat fracture were also much lower. The specimens which weresurface was caused by quasi-cleavage (Figure 65). The slowly cooled from 959"C or tempered and slow-cooledfracture surfaces of both the teratpered and hydrogen- from 4830C w6re teated at room temperature and hadcharged specimens closely resembled that of the fracture surfaces composed mostly of dimpled rupture.

16

4 CONCLUSIONS mechanical properties of A-242 steel, The fracturesurfaces of transverse specimens contained elongatedinclusions, but no change in fracture mechanism from

1, The structural steels ASTM A.588 and A.242 that of the longitudinal specimens was observed.were found by SEM examination and mechanicaltesting to be unsusceptible to temper-embrittlement. 5. The stainless steels, 17.4PH and 416, were found

* to be unsusceptible to the hydrogen.charging procedure2. ASTM A.588 was found to be susceptible to,•r performed in this study,hydrogen.embrittlement, Cleavage fracture in thetensile and intergranular fracture In the fatigue speoo.men indicate a loss in material ductility as a result of 6. The effect of tempering on the stainless steels isthe hydrogen. charging procedure, difficult to define. Even though the mechanical test

results showed large variations in values, the fracture3. ASTM A-242 was found to be unsusceptible to surfaces of the as.quenched and various tempered

hydrogen.embrittlement in this study, Since the specimens were very similar, The appearance of theaabnormally low fracture surface Is therefore not a sufficient methodas-received A.242 material wsinanborlylw

toughness condition, however, the effect of hydrogen for determining temper-embrittlement, but must be,harglng on the material would be masked by this combined with mechanical testing to reach validcondition, A high.strength, low.alloy structural steel conclusions,such as A-242 in a quenched and tempered conditionwould ordinarily be susceptible to hydrogen.embrittle. 7, The effect of testing temperature on the Charpyment, as the similar A-588 steel was shown to be. specimens of all the steels was seen in a change from

low.energy, brittle fracture (primarily cleavage) at low4. The orientation of the specimens with respect to temperatures to higher.energy, ductile fracture (pri." the rolling direction was shown to affect primarily the martly dimple rupture) at elevated temperatures.

17

......... -.

Flgipw 1. Equt~ed dimples containing inclusions, 4250X.

Figur 2, Elonpted dimples, 3750X.

i8

I

-- I

2" ',Sk

F..igre 3. Cleavuge facets, 400X. "

il

Figure 4, Quasi-cleavage facets, 1600X.

19

Figure S. Fatigue striatiuns, 4000X.

A, ,

Figure 6. intergranular fracture, 300X

20

-'r4

Figure 7. Mic~rostructure of' AMA~ A-588 steel, 200X.

Figure 8, Microstructure of ASTM A-242 steel, 200X.

21

I -OIAA

fO I .i

Lk

H Figure 9. MicrostruttCot'r uAISI 416 socil, ZOOX,

Figure 10. Microstructure of 17*4P11 steel, 200X.

1 22

I'lllUL E

K In

I - zIn w Uto2

23

vI Figure 13

Figure 12, Tensile tructure surface at A 588 steel, 8X Noto the largo reduction In area ofthe fracture surface and separation at eoungated inclusiuns

Figure 13. Dimple rupture in A.588 steel, 15OOX.

24

Fatigu

Figure 14. Fatigue~ rrar~Lurc surfa~ce of A-588 steel, 8X. Fatigue crack growth occurred

frorn both sides toward the center of specimen.

Figure 15. Dimple rupture In an A-598S fatigue specimen, 750X.

25

iiA

Figure 16. Fatigue striations in A-588 steel, I SOOX.

Figure 17, Tettsile fracture surface ot'hydrogen-embrittled A-588 steel, 8X. The reductionin area of the fracture surface is less than the as-rece'ived specimen.

26

River

Patterns

Figure 18. Cleavage facets In hydrogen-mbrittled A-588 steel, I SOOX,

Figre 9. atiue racuresuraceof ydrgennlbitted .58 stelBX. Faigure 2

cracks initiated and propagated from either side towards the center.

27

Fracture~d

Figure 20. Dimple rupture and fractured pearlite colonies in hydrogen-embrittled A-588I. fatigue specimen, 1 S00X.

Figure 21. Intergranulor fracture ao d fatigue striations in a hydrogen-embrittled A-588fatigue specimen, 1 500X.

2B

*~ý M ~

,T - - F ig u re 2 3

2.4

Figure 22. Fracture surface of an A-588 Charpy specimen tested at l196 C, SX. Theabsence of lateral contraction and the flat fracture surface is evidence ofa low energy, brittle fracture.

Figure 23. Cleavage facets In an A- 588 Charpy specimen tested a t -I960 C, IOO0OX.

29

A.~~I . V...IV~~~~V .L4 nA,.A . M'n .>f~llA .LIC~A .~. ... .V ~ . . . -.. ... .. ............

Figure 24. Cleavage fracture and dimple rupture in an A-588 Charpy specimen tested at*1030 C. IOOOX.

Figure 26

Figure 25. Fracture surface of A..588 Charpy specimen tested at 23*C, 8X. Note theamount of lateral contraction and the change of fracture surface near thecenter of the specimen.

30

at 23C 2j

V,4... .4 .....

Figure 27. Tensile fracture surface of an A-242 longitudinal specimen tested at 230C,1 2X. The specimen shows considerable necking.

31

Figure 29

Figure 28, Tensile fracture surface of an A-242 transverse specimen, 12X. The fracturesurface has necked down less than the longitudinal specimen and has a muchflatter fracture surface, Note cracking along elongated inclusions,

4,z,

N

Figure 29. Inclusion-generated dimple rupture in an A-242 transverse specimen, 500X.

32

I, L ..

Figure 30. Fracture surface of a hydrogen-charged A-242 longitudinal tensle specimen,l OX. The f'racture is very flat indicating lw ductility,

Figure 31. Dimple rupture in a hydrogen -charged A-242 longitudinal tensile specimen,600X.

33

Figure 33

~ Figure 34

Figure 32, Fracture surface uf a hydrogeu-churged A-242 transverse tensile specimen,IlOX, The fracture surface is flat and the limited reduction in area indicateslow ductility.

FIguaN 33. Elongated inclusions in a hydiogen.charged A-242 traiisverse tensile specimen,55X.

34

Figure 34. Cleavage fracture and dimple rupture in a hydrogen-charged A-242 transversetensile specimen, 1400X.

Figure 35. Cleavage fracture and dimple rupture In an A-242 longitudinal fatiguespecim'on. 600X.

35

Figure 36. Cleavage fracturo In an A-242 transverse fatigue specimen, 70UX.

Figure 38

Figure 37. Fracture surface of a hydrogen-charged A-242 transverse fatigue spechrncn,

36

I,4

FRgre 38. Dimple rupture, and cleavage fracture in a hydrogen-charged, A-242 transversefatigue specinlin, 25MX The cleavage region is in the fatigue crack propagation

~1 region, while the dimple rupture occurred by tensile overload on the last cycle,

Figure 39. Fracture surface of an A-242 transverse Charpy specimen tested at -196 0C,lOX. Note the flatness of the fracture and absence of lateral contraction.

37

Figure 40, Fracture surface of an A.242 transverse Chupy specimen tasted at O*C, lox, i

TI

Figure 41. Fractufe surface of an A.242 transverse Charpy specimen tested at 230r, lox.

38

, v

Figure 42. Cleavage fracture In an A-242 longitudinal Charpy specimen tested at -I96WC,500X.

GrainBoundary

k 4 Precipitation

Figure 43. Gruin houndary precipitation in A-242 Charpy specimn~n tested at -196 0C,10OOX.

39

LIN

flgwe 44. Fracture surface of an A-242 longitudinal Charpy specimen tested at 121*C.IlOX. Note the extensive lateral contraction and deteriorations at inclu.-Jons.

Figmf 45. Cleavage fracture in an A-242 longitudinal Charpy specimeon tested at 230 C.

40

Figure 46, Dimple rupture and cleavage fracture in an A-242 transverse Charpy specimen,

Figure 48

Figure 47, Fracture surface uf a eolution heat-treated i7-4PH- tensile specimen, 12X.Note the presence of shear along the outer aurfaces and the flat fracture onthe Interior of the fracture surface.

41

I I 'O.¶ ~ I.V ffN

M b

Figue 48. Dimple rupture in a solution heat-treated 1 7-4PH tensile specimen, 700X.

Figure 49. Fracture surface of a 17 4PH tensile specimen aged at ,4820C, 1 2X. Note flatraegon on the interior of the fracture surface.

42

~W.

411

Figure SO. Inclusions in a 17.4PH tensile specimen aged at 5380 C, 3SOOX.

Ire

Figure St. Dimple rupture and microvoid coalescence In a 17.4PH- tensile specimenaged at 454'C, 700X. The crack-like regions are seps~rations along elongatedinciusiuns.

43

4' 44

Figure 52, Fracture surface of a 17 4PH hydrogen-charged tensile specimen, lOX.

]Figure 53. Dimple rupture in a 17-4PH- hydrogen-charged tensile specimen, IlOOOX.

44

~p p

Figure 55. Quasi-cleavage fracture in a solution heat-treated I 7-4P1- fatigue specimen,700X.

45

~ ~3W3 '~3'J3#&4~37 ~ ~ - ~ . 7 ~ 77 ¶77 4 73A

4;46

huI1 I" IL~~

IA0

.AA

Figure 58, Brittle inclusion in a 17.4PH- fatigue specimen aged at 538*C, 2500X.

Figure 59. Fracture surface of a 117-01i solution heat-treated Charpy specimen tested at*196*C, lOX. Note the large amount of lateral contraction and the large shearregions.

47

r~~~~~~~~~~~~~t~~~~1 W,~'l MJh~M 'I 'iU .LI L.,.iwa4~.~ 1~h

qit

Figure 60. Fracture surface of a I 7.4PH solution heat-treatedl Charpy specimen tested at1210C. lox.

SFigure 62 V

* ~Figure 61. Fracture surface of as~quanched 416 tensile specimen, l OX. T[he fracture is* I relativeiy flat with little redluction In area,

48

.. .... . ...... ..

Pipre 62, Dimple rupture and nkerovoid coalescence in as-quenched 416 tensilespecimen, 700X.

Figure 63. Fracture surfuce of a 416 tensile specimen tempered at 593u,,, IlIX.

497

F~i&re64 Frctue urfccof amgqUenched 416 fatigue spgoitnenf, IMX The flat fracture

and lack of lateral contraction are evidence of low energy ffacturt,

Floure 65. QuaskitcievaS ffacture in as-qiienchad 416 fatigUd speclimen, 140OX.

4 so

Figure 67

Figure 66, Fracture surface of us-quonched 416 Charpy specimen tested at 23C I OX.

UU

IR ýi

Figure 67. Dimple rupture and cleavage fracture in as.quenched 416 Charpy specimentested at 23"C, 500X.

f5

K Figure 69I Figure 68, Fracture surface of 2 416 Charpy specimen tempered at 954'C and tested at

230C, l OX,

A'LA

*Po

Figure 69, Dimple rupture In a 416 Charpy spetiaton tempered at 954*C and tasted at2310C,IO~OX.

52

0 0,,

'Figure'70. Dimiple rupture and cleavage fracture in a 416 Charpy spocirnon tOmpered atS93*C and tested m -196 0C, 50OX.

rigure 71. In torgranular fracture in o. 416 Charpy 4pecirnen tempered at 5930C and testedgt .1960C 1000Y

53

REFERENCES Laird, C. and G. C. Smith, "Crack Propagation in HighStress Fatigue," Philosophical Magazine, Vol 2(1962), p 847.

Ault, R. T., R. H. Holtinann, and J, R. Myers, Heat Low, J. R., Jr., Fracture of Engineering MaterialsTreatment of a Mas-tensitic Vtainless Steel for (ASM, 1964),p 127.Optimum Combination of Strenrth, Toughness,and Stress Corrosion Resistance. Technical Report Low, J. R., Jr., B, L. Averbach, et al,, Fracture (JohnAFML-TR-68-7 (Air Force Matevials Laboratory, Wiley, 1959).April 1968).

Low, 3. R,, Jr., D. F. Stein, A. M. Turkalo, and R.P.Beck, W., Electrochemical Technology, Vol 2 (1964), Lafarci, Transac ions of AIME, MT671B, Vol 242

pp 74.78. (1968),pp 14.24.

Beck, W,, E, J. Jankowski, and P. Fisher, Hydrogen Ma,'nas, H. L. and P. W. Palmberg, "Effect of SoluteStress Cracking of High-Strength Steels, NADC- Elements on Temper Embrittlement of Low AlloyMA-7140 (Naval Air Development Center, 1971), Steels," Temper Embrittlement of Steels, ASTM

STP 499 (ASTM, 1971), pp 90.103.Bernstein, 1. M., "'The Role of Hydrogen in the Embrit.

tle,.,,-, t of Iron and Steel," Materials Science and McMah',?i, C. J., Jr., Temper Embrittlement in Steel.Engineering, Vol, No, 1 (1970), pp 1. 19. A." M STP 407 (ASTM, 1968), p 127,

Cotterill, P., "The Hydrogen Embrittlement of Metals," Ohtani, H., H.C. Feng,and C. J. McMahon, Jr., "New In.Progressive .ý'aterials Science, Vol Q, No. 4 (1961), formation on the Mechanism of Temper Embrittle.

ment of Alloy Steels," Metallurgical Transactions,Forsyth, P. J. E,, "Fatigue Damage and Crack Growth Vol 5 (1974), pp 516.518.

in Aluminum Alloys," Acta Metallurgica, Vol 11(1963), p 703, Petch, N. J., "The Ductile Fracture of Polycrystalline.

Iron," Philosophical Magazine, Vol 1 (1956),Fractography and Atlas of Fractographs, ASM Metals pp 186-191.

Handbook, Vol 9, 8th Edition (American Societyfor Metals [ASM1, 1974), p 65. Schwen, G., G, Sachs, and K, Tonk,ASTMProceeadint,

Vol 57 (1957), pp 682.697.Harrison, J. D. and G. C. Smith,British WeldingJournal,

Vol 14 (1967), pp 492.502. Spitzig, W. A., P. M. Talda, and R. P, Wei, "Fatigue-Crack Propagation and Fr&ctographic Analysis of

Hleat Treating, Cleaning, and Finishing, ASM Metals 18 Ni(250) Maraging Steel Tested In Argon andHandbook, Vol 2, 8th Edition (1964), p 245. Hydrogen Environments," Engineering Fracture

Mechanics, Vol 1 (1968),pp 155-165.Hill, M. N. a.,d E. W. Johnson, Transactions of the

American Institute of Mining, Metallurgical and Tetelman, A. S. and A. J. McEvily, Jr., Fracture ofPetroleum Engineers (AIME), Vol 215, No. 717 Structural Materials (John Wiley, 1967).

: (1959). Troiano, A., "The Role of Hydrogen and Other Inter.

Honda, R,, International Conference on Fracture, stitlals In the Mechanical Behavior of Metals,"Sendal, Japan (1965). Transactions of the American Society for Metals

(ASM), Vol 52 (1960), p 52.Hydrogen Embrittlement Testing, ASTM STP543

(American Society for Testing and Materials Zaffe, C. A., Journal of' Iron and Steel Institute,[ASTM], 1974). Vol 154, No. 123 (1946).

54

CERL DISTRIBUTION

ComAander USA Engr Division, EuropeDANCOM StIT-EUR ATTN: Titchrical Librarylop 46 European Div. Corps o~f EngineersAnO New York 0911U APO New York 09757

US Military Acadenmv USA Engr Division, New EnglandATTN: Dept of Mechanics ATTm: L0rrATYN:- Library AT0996 Chief, NECDD-T

West~Wltam PonMA196* 02154

CThie of EN-AIner 2) USA tngr Division, HuntsvilleATMN DAEN-ASIL S ATTN. Library (2)ATYh: 0AEN-NCDL PO Box 1600 West StationATIth DAE-CE-RO *iNotsville. AL 35807

jep WAS the 2031 USA Engr Division, Lower Mississippi ValleyiAHD 01 Arr~i Library

Commander P0 sag 80US Army Scionce and technology Vicksburg, MS 39180

center - Far East OfficeAPO San Francisco, CA 96328 USA Engr Div. Ohio River

ATTNt LibraryUSA tngr District, Naw Orleans PO Box 1159AIIM: Library Cincinnati, OH 45201AIIM: Chief, LMNED-OG UAEg iiin ot etaPO Box 60267 UAEg iiin ot etaNew Orlean, LA 70160 AITTN. Library536 S. Clark StUSA Eng' Distric~t, Littl* Rock Chicago, IL 60605ATIN-: Chief, Eiigr DivPO Box 876 USA Engr Division Missouri RiverLittle Rock, AR 72203. ATTN: Library (21PO Box 103 Downtown Station

U SA Ingr District. 101s4 Omaha, MB 68101

AT0 sox ar USA Engr Division, Southwestern

Tulsa. OK 71001 ATTM: Library,Iait Tower BlIdg. 1200 Main StUSA Engr District, Albuquerque Dajl1ýs, TX 75202ATIM: Library A ii nrCne/RP0 Box 1680AFCvl'renr/RAlbuquerque. )M 87103 Tyndall AFI, Fl. 3240l1

USA Engr District, San Francisco Chief, Naval Air Systems CommnandATTN: Chief, Engr Div AT"TN: Library211 Main St Dept of the NavySan Francisco, CA 94105 WASH DC ýLO360

USA Engr Dist, Sacrafflentn 04vficer in ChargeATTN: Chief, SPKED-D Civil Engineer In, Laboratory660 Capitol Mall Naval Con$truto Battalion CenterSacramento, CA 95814 ArTTN Library (Code LOSA)Port Huenmene CA 93043

USA Engr District, JapanLbayATTN., Library irvAPO San Francisco. CA 96343 Transportation Research Board

2101 Conytitution AveUSA Engr District, Por~tland WASH CC ?0418ArTYN Chief, DB64PO box 2946 Library of CongressPortland, DR 97201 Exchange and Gift Divis9ionA1TM;ý Federal Documents Section (2)USA Engr District, Seattle WASH DC 20540ATTN: Chief, NPSCOPO BOA 31i5 Dept of Yrgnuportation LibrarySeattle, WA 98124 Acquisitiont section (SR) TAD.491.1

400 7th Stpont SWUSA Engr District, W04l Wa04 WASH DC 20590AT14: LibraryA TIN: Chief. Enqr Div Defernt Docianeentation CenterBftyG0,', CitY-Lounty Airpornt ATTM: (CA (12)Ia 611Ala, WA 99362 Cameron Station

Alemandriti, VA 22314USA Enogr District, AlaskaATrh: Library Enginenring Societies Library011I.- NPADE-R 345 East 47tn S treatPC son 7002 Mew Yook, NY 10111Anchorage, AK 99501