Astm stp 370 1965

Trang 2

STRUCTURE AND PROPERTIES

OF ULTRAHIGH-STRENGTH

STEELS

A symposium sponsored by theMETALLURGICAL SOCIETY OF AIME and the

AMERICAN SOCIETY FOR TESTING AND MATERIALS

Cleveland, Ohio, Oct 22, 1963

Price $11.00; to Members $7.70

Published by the AMERICAN SOCIETY FOR TESTING AND MATERIALS

1916 Race St., Philadelphia 3, Pa.

Reg U S Pat Off.

ASTM Special Technical Publication No 370

Downloaded/printed by

University of Washington (University of Washington) pursuant to License Agreement No further reproductions authorized.

Trang 3

Trang 4

F O R E W O R D

The papers in this volume were presented at a Symposium on Steels With

Yield Strengths Over 200,000 psi sponsored by the Panel on Structural

Materials for Airframes and Missiles of the ASTM-ASME Joint Committee

on Effect of Temperature on the Properties of Metals, and the Structural

Materials Committee, Institute of Metals, Metallurgical Society of AIME

The Symposium was held on Oct 22, 1963, in Cleveland, Ohio

F M Richmond, of Universal-Cyclops Steel Corp., and J W Welty, of

Solar Aircraft Co., were the chairmen of the morning session E E

Reyn-olds, of Allegheny Ludlum Steel Corp., and J J Heger, of U S Steel Corp.,

presided over the afternoon session

Trang 5

NOTE—The Society is not responsible, as a body, for the statements

and opinions advanced in this publication.

Trang 6

C O N T E N T S

PAGE Introduction 1

Relationships Between Microstructure and Toughness in Quenched and Tempered

Ultrahigh-Strength Steels—A J Baker, F J Lauta, and R P Wei 3

Discussion 23

Relationships Between Structure and Properties in the 9Ni-4Co Alloy System—

J S Pascover and S J Matas 30

Discussion 45

High-Strength Stainless Steels by Deformation at Room Temperature—S Floreen

and C R Mayne 47

An Evaluation of the 18Ni-9Co-5Mo Maraging Steel Sheet—D L Corn 54

The Metallurgy and Properties of Cold-Rolled Am-350 and Am-355 Steels—T H.

McCunn, G N Aggen, and R A Lula 78

Discussion 93

Fracture Micromechanics in High-Strength Steels—Bani R Banerjee 94

Discussion 116

The Effect of Solidification Practice on the Properties of High-Strength Steels—

C M Carman, R W Strachan, D F Armiento, and H Markus 121

Discussion 143

High-Strength Steel Forgings—H J Henning 147

Ausform Fabrication and Properties of High-Strength Alloy Steel—W W

Ger-berich, A J Williams, C F Martin, and R E Heise 154

Thermomechanical Treatments Applied to Ultrahigh-Strength Bainites—D Kalish,

S A Kulin, and M Cohen 172

Trang 7

RELATED ASTM PUBLICATIONS

Properties of Basic Oxygen and Open Hearth Steels, STP 364 (1963).

Stress Corrosion Cracking of Austenitic Chromium-Nickel Stainless Steels, STP 264 (1960).

Chemical Composition and Rupture Strengths of Super-Strength Alloys, STP 170-C

Trang 8

STRUCTURE AND PROPERTIES OF ULTRAHIGH-STRENGTH STEELS

INTRODUCTIONThe increasing demands of the military

for improved performance of structural

materials for space, land, and deep ocean

environments has resulted in an intensive

activity in the development, evaluation,

and prototype testing of a broad range

of materials including oxides, carbides,

aluminum, titanium, and even gold A

significant portion of this activity has

been devoted to high-strength steels

Recognizing the scope of this activity

and the need to assemble into one

semi-nar the more recent advances in the

development and application of

high-strength steels, the Panel on Structural

Materials for Airframes and Missiles

of the Joint Committee of ASTM and

ASME, and the Structural Materials

Committee of the Institute of Metals

Division of the Metallurgical Society

of AIME, organized this Symposium

on Steels With Yield Strengths Over

200,000 psi

Until recently, steels having yield

strengths in excess of 200,000 psi were

not considered suitable as materials of

construction, because fabrication and

inspection techniques were not

suffi-ciently sophisticated to permit full

utilization of these high strengths, which

at that time were accompanied by low

ductility and low toughness Recently,

however, major developments have

occurred not only in alloy development,

which has permitted the achievement

of higher levels of ductility and

tough-ness, but also in inspection and

fabrica-tion techniques that permit the full

utilization of higher strengths that arenow obtainable

The papers presented at this posium included descriptions of newsteels (or new concepts for makingsteels) having yield strengths in excess

sym-of 200,000 psi, and good ductility andtoughness Specifically mentioned arethe new maraging steels, higher strengthand higher toughness martensitic steels,steels strengthened by thermomechanicaltreatments, and steels strengthened bycryogenic treatments Progress has beenmade in the understanding of the illu-sive property known as toughness, andtwo papers are presented summarizingthe state of art in this area Also theeffect of melting and processing on highstrength properties, the characteristics

of specific products—namely, forgingsand fasteners—and the fabrication of thenew high-strength steels are discussed

of the transportation vehicles that thereader will be using as a personal means

of conveyance within the next 10 years

1

Trang 9

This page intentionally left blank

Trang 10

RELATIONSHIPS BETWEEN MICROSTRUCTURE AND TOUGHNESS IN QUENCHED AND TEMPERED ULTRAH1GH-STRENGTH STEELS

BY A J BAKER, 1 F J LAUTA, 1 AND R P WEI 1

SYNOPSIS

An investigation was made of a number of 0.30 and 0.40 per cent carbon alloy steels to determine the relationships between their fracture toughness properties and their internal microstructures The plane-strain fracture toughness of the steels was measured after tempering quenched material in the temperature range 300 to HOOF A thin section transmission electron microscopy study was carried out on the tempered materials.

From the fracture toughness studies it was concluded that all the materials behaved similarly and that alloying elements (carbon and silicon) had little influence on the general relationship between tensile strength and toughness in these fully hardenable steels It was found that the fracture toughness remained low at low tempering temperatures but improved rapidly once a critical tempering temperature, characteristic of the particular steel, was reached.

The microscopy study showed that major microstructural changes occurred

in the tempering range where the rapid increase in toughness was observed.

At low tempering temperatures the defect structure of the as-quenched tensite remained unchanged, and continuous films of carbide were formed in the boundaries of the martensite In the critical tempering range the carbide films were spheroidized and the defect structure of the matrix removed or modified by recovery processes On the basis of these observations, it was concluded that the low fracture toughness of the steels in a lightly tempered condition was due to their high defect densities and the presence of carbide films at boundaries Only when these features were removed or modified did toughness increase.

mar-In recent years there has been a grow- loy steels with carbon contents in the ing demand for materials of very high range of 0.3 to 0.5 per cent Quenched strength for aerospace applications such and tempered low-alloy steels already

as rocket motor casings This demand has have many uses both as structural stimulated research aimed at the de- rials and, at higher strength levels, as ma- velopment of ultrahigh-strength steels, chine parts When their yield strengths that is, steels with useable yield strengths are raised to the strength level men-

mate-of more, than 200,000 psi; and an im- tioned, there is the major problem mate-of portant part of this research has been the maintaining an adequate level of tough- study of quenched and tempered low-al- ness that will meet the design require-

i Technologists, U S Steel Corp., Applied mentS Pkced UP°n them'

Research Laboratory, Monroeville, Pa The present investigation was carried

3

Trang 11

out to determine the fracture-toughness

capabilities of these steels at the

ultra-high-strength level and to attempt to

relate their properties to microstructure

The investigation was based on three 0.40

per cent carbon steels and one 0.30 per

cent carbon steel, and the effects of

both chemical composition and

temper-ing treatment on plane-strain fracture

toughness were examined

To study the relationship of

micro-structure to fracture toughness, a parallel

metallographic investigation was made

using thin-section transmission and

replica electron-microscopy techniques

were hot-rolled to a 1-in plate after bebeing heated to 2300 F Steel D was cast

as a 300 Ib, 8 by 8-in ingot that was sequently reduced by rolling to a 1-in

sub-plate

Oversize 0.505-in.-diameter tensionspecimens were rough-machined from theplate or bar materials, austenitized for

£ hr at 1700F and oil-quenched Thespecimens were then tempered for times

up to 4 hr in the temperature range 400

to HOOF and air-cooled After heattreatment, the specimens were finish-machined to 0.505-in.-diameter tensionspecimens

*.L CO1 Mn 0.74 0.77 0.74 0.84

VEPOSI

PER P

0.019 0.009 0.005 0.007

TIONS CEN1

S

0.026 0.008 0.006 0.009

OF SI Si

0.27 0.27 1.60 1.59

^EELS Ni 1.79 1.16 1.87 2.04

INVES Cr 0.89 0.73 0.83 2.04

3TIGAr Mo 0.26 0.26 0.37 0.51

FED, V

0.10 0.055

Al 0.044 0.03 0.092 0.055

" Consumable Electrode Vacuum Remelted.

Materials and Method:

The chemical composition of the four

steels investigated is shown in Table 1

The composition of steels A and B is

within the specification limits for AISI

4340 steel, except that steel B is slightly

deficient in nickel The lower phosphorus

and sulfur content of steel B, compared

to that of steel A, is probably a result of

the consumable-electrode vacuum-remelt

process by which it was produced Steel

C is a high-silicon modification of AISI

4340, with a small vanadium addition,

and it corresponds to the material

com-monly designated as 300M Steel D is an

experimental laboratory-made material

Steel A was received as l^-in.-diameter

annealed bar stock, and test specimens

were machined directly from this Steels

B and C were received as 7- and 5-in

blooms, respectively, and these blooms

Notched tension specimens were pared by rough machining 6-in.-longcylinders of 0.8 or 1.05-in.-diameter fromthe plate or bar material These speci-mens were given heat treatments identi-cal with those given the smooth tensionspecimens After heat treatment, thespecimens were machined to either 0.750

pre-or 1.00-in diameter and notched Thedepth of the machined notch was chosen

so that after a subsequent introduction

of a 0.030-in.-deep circumferential tigue crack, the area of the remaininguncracked material would be 50 per cent

fa-of the gross cross-sectional area Fatigueprecracking was accomplished in an en-gine lathe, the tailstock being offset tointroduce a bending moment

Specimens for transmission electronmicroscopy were taken from the broken

tension specimens Small pieces | by \ by

0.1-in thick were cut out and ground on

Trang 12

BAKER ET AL ON MICROSTRUCTURE AND TOUGHNESS

TABLE 2—MECHANICAL PROPERTIES—FRACTURE TOUGHNESS

2706 261 242

21 23 45 39

48

53

51 105 114 117 179

YS,

ksi 214

223

222 212

193

Steel B

TS,

ksi 296

263

256 238

208

Sic

in.-lb/in.2 /48 152 43

(51

\55 129 125 142 139

206

193

212 209

269

255

242 232

Sic in.-lb/in.2

48 45

(52

\50 /49

\53

55

[78 J73 [59

!08 99 100 107 214

YS,ksi

205

212 215

179

184 174 170

2556

242

230fr

225 196 193

Sic lb/in.2

in.-88 90

87 91 83 /89

a YS = yield strength; TS = tensile strength.

6 Extrapolated value.

a wet belt to a thickness of about 0.020

in These specimens were then chemically

polished in a warm solution of 50 per cent

phosphoric acid (H3PO4) and 50 per cent

hydrogen peroxide (H202) to reduce their

thickness to less than 0.005 in This

0.005-in material was electropolished

successively in 1:10 mixture of perchloricacid and methyl alcohol and a solution ofchromic acid in acetic acid to preparethin foils suitable for transmission elec-tron microscopy (I).2 Formvar-carbon

2 The boldface numbers in parentheses refer

to the list of references appended to this paper.

Trang 13

FIG 1—Effect of Tempering Temperature on Strength and Fracture Toughness of Steels A and B.

FIG 2—Effect of Tempering Temperature on Strength and Fracture Toughness of Steels C and D.

replicas for electron microscopy were

prepared both from heat-treated material

and from the fracture profiles of the

broken notched rounds The fracture pro-"

files were obtained by coating the

frac-6

ture surfaces with electrodepositednickel, sectioning perpendicular to thefracture surface, and polishing the profile

by standard metallographic preparationmethods

Trang 14

RESULTS

Mechanical Properties:

Table 2 is a compilation of the strength

and toughness properties of the steels

examined The effect of tempering

treat-ment on yield strength, tensile strength,

and plane-strain fracture toughness is

shown in Figs 1 and 2 By comparing the

properties of steels A and B with those of

steels C and D at a given tempering

temperature, the influence of silicon and

It is important to note the effect ofnotch acuity on plane-strain fracture-toughness evaluations The effect of tem-pering temperature on plane-strain frac-ture toughness shown in Figs 1 and 2 can

be obtained only with fatigue-precrackedspecimens To emphasize this point, acomparison is made in Fig 3 between thenotched-tensile strengths obtained withfatigue-precracked specimens and thoseobtained with specimens having ma-chined notches with a 0.0015-in root

FIG 3—Schematic Showing Effect of Notch Acuity on Fracture Toughness Test Results.

chromium in retarding the rate of

tempering can clearly be seen

All four steels show a similar type of

behavior in the effect of tempering

treat-ment on plane-strain fracture toughness

In each case, the fracture toughtness

re-mains low and almost constant at low

tempering temperatures but undergoes

a sharp increase in a critical tempering

range characteristic of the particular

steel The actual level of fracture

tough-ness at low tempering temperatures

de-pends on the individual steel

composi-tions Those with lower phosphorus and

sulfur contents have a higher level of

toughness

radius The tests were carried out on anexperimental 0.40 per cent carbon, low-alloy steel with a composition similar tothat of 300M steel With machinednotches, the variation of notched tensilestrength with tempering treatment fol-lows a variation similar to that of yieldstrength, which suggests that this type

of behavior is controlled by the initiation

of a crack at the root of these relativelydull notches Apparent values of Slc com-puted on the basis of these results wouldindicate a more complex variation of 9icwith tempering treatment than is ac-tually the case Using fatigue precrackedspecimens, the actual variation of gtc is

Trang 15

8 STRUCTURE AND PROPERTIES or ULTRAHIGH-STRENGTH STEELS

characterized by a region of almost

con-stant and low fracture toughness at low

temperatures and a region of rapidly

increasing fracture toughness above a

critical tempering temperature

The results shown in Figs 1 and 2

indi-cate that the critical tempering

tempera-tures, above which fracture toughness

increases rapidly, are dependent on the

alloy content of the steels Silicon has a

particularly strong influence in that, for

steels A and B, containing 0.25 per cent

the relationships exhibited by steels Band C in Fig 4 These steels differ in sili-con content, having 0.25 and 1.50 percent, respectively, but it is apparent thatthe higher silicon content of steel C hasnot led to any improvement in the bal-ance of fracture toughness to strengthover that attainable in steel B

The systematic relationships shown inFig 4 created interest in the associatedmicrostructural changes, and conse-quently an electron metallographic study

FIG 4—Relationship Between Plane-Strain Fracture Toughness and Tensile Strength for Four

Steels.

silicon, the critical temperature is 600 F,

whereas for steels C and D, containing

1.50 per cent silicon, the critical

tempera-ture is 900 F

So that the four steels can be compared

at equal strength levels, the data shown

in Figs 1 and 2 are replotted in Fig 4

with fracture toughness as a function of

tensile strength The sharp change in

fracture-toughness behavior is again

ap-parent: above a tensile strength level of

about 240,000 psi the steels have low

fracture toughness, whereas at lower

tensile strengths the toughness increases

A further conclusion can be drawn from

was made on these materials after varioustempering treatments It was possible toconsider all four steels as a group despitetheir varying composition, and themetallographic study was further facili-tated by the fact that each steel exhibitedtwo distinct levels of fracture toughnessabove and below the 240,000-psi strengthlevel It was possible, therefore, to predictthe critical tempering-tfemperature rangewithin which important structuralchanges were likely to occur Thus, while

it was still necesasry to examine thestructures developed throughout thewhole tempering sequence, an intensive

Trang 16

BAKES ET AL ON MicsosTRuorasE AND TOUGHNESS 9

Pic, 5—Electron Micrographs of Steel D Specimens Anstenitized at 1700 F and Oil-Quenched

Trang 17

10 STKUCTUSE AND PKOPERTIES OF ULTRAHIGH-STKENGTH STEELS

(top) X 50,000 (bottom) X 150,000

FIG 6—Electron Micrographs of a Steel D Specimen Tempered 1 hr at 600 F.

Trang 18

BAKER ET AL ON MICROSTRUCTURE AND TOUGHNESS 11

investigation of microstructure could be

concentrated on the narrow range likely

to be significant

Microstructure:

The metallographic studies showed

that all four steels underwent similar

microstructural changes during

temper-ing To avoid unnecessary repetition, a

detailed discussion of the effects of

tem-pering on microstructure will be given for

steel D only; discussion of the results for

steels A, B, and C will be restricted to

those parts of the temperature sequence

in which important changes occurred, or

in which their behavior differed from

that of steel D

The martensitic structure of steel D is

shown in Fig 5 The martensite has a

predominantly plate-like morphology,

with rrfany of the plates lying in parallel

groups The plates are about %n wide and

about 4/i long Their thickness is difficult

to estimate accurately because of the

uncertainty in knowing the angle at

which a plate meets the foil surface; but

the thinnest plates measured were about

1000 A Two types of lattice defect are

present within the plates, namely, a high

dislocation density and many microtwins

The dislocation density is about 10"

lines/cm2, which is comparable with that

of heavily cold-worked metals (2) In

many areas of the foils, the dislocation

density is too high to resolve the

in-dividual dislocations, and a mottled type

of contrast results instead The

micro-twins are closely spaced at about 200 A

and are as narrow as about 100 A in

width The twinned structure observed is

similar to that found in high-carbon,

plain-carbon steels (3)

In addition to dislocations and

micro-twins, the martensite contains many

small precipitate particles as shown in

Fig 5 (bottom) The precipitate particles

are small rods, approximately 50 A in

diameter, that usually lie parallel to the

twin-boundary direction, which suggeststhat they may nucleate in the twinboundaries In addition, there are a num-ber of smaller particles that lie across thetwins and often link together the twin-boundary particles These smaller parti-cles have a ragged appearance and areprobably nucleated on dislocations Otherparticles are formed at the martensiteboundaries, and these also connect withadjacent particles in the matrix

The precipitate particles observed are

a feature of the martensite produced inthick-section, oil-quenched material, andthey do not appear in thin-section, water-quenched specimens This fact wouldsuggest that the particles are created byauto tempering during quenching; since

the M g temperature of steel D is tively low (about 550 F), the particlesare probably e-carbides, as this is thecarbide formed at low tempering temper-atures (4) However, it was not possible

rela-to identify them by selected-area tron diffraction using either thin foils orextraction replicas; in the thin foils, theprecipitate volume fraction was too small

elec-to produce an identifiable diffractionpattern, whereas with extraction replicasthe density of particles was insufficient toobtain a pattern The difficulty experi-enced in extracting these particles,coupled with the fact that they are small

in size and general shape, suggests thatthey may have a high degree of coherencywith the matrix

Tempering steel D at 600 F resulted inthe formation of further quantities ofprecipitate The preferred sites for pre-cipitate formation were the martensiteplate boundaries and the twin bounda-ries The structure is shown in Fig 6

(top), where the martensite boundary

pre-cipitate can be seen to constitute an most continuous film The precipitate atthis tempering temperature had sufficientvolume fraction to produce an electron-diffraction pattern, and it was identified

Trang 19

12 STRUCTURE AND PROPERTIES OF ULTRAHIGH-STRENGTH STEELS

FIG 7—Electron Micrographs of a Steel D Specimen Tempered 4 hr at 1050 F.

Trang 20

BAKER ET AL ON MICROSTRUCTURE AND TOUGHNESS 13

as 6-carbide The twin-boundary

precipi-tate at this stage of tempering is about

500 A wide and must have incorporated

several twins during its sidewise growth

At high magnification, both the

martens-ite boundary and the twin-boundary

precipitates exhibit a substructure of fine

parallel striations (Fig 6 (bottom)), which

terial, do not appear in the temperedstructure and must therefore dissolve asthe larger particles grow

On tempering at higher temperatures,

up to 900 F, increasing quantities of bide were formed, but above 1000 F sev-eral important changes occurred

car-Figure 7 (top) shows the 1050-F

tem-FIG 8—Electron Micrograph of a Steel D Specimen Tempered 4 hr at 1050 F (X 100,000).

may be due to fine-scale faulting within

the carbide structure The c-carbide on

the twin boundaries probably retains

some coherency with the matrix at this

stage of tempering, and the faulting

could be ihe result of a strained mode of

growth Carbides are known to contain

strains when formed at low tempering

temperatures (5) The small precipitate

particles linking the twin boundaries,

which are a feature of quenched

ma-pered structure, and it can be seen thatthe carbide precipitates are discretespheroidized particles at the martensiteboundaries, rather than continuous films

The carbide formed at this temperingtemperature was identified as iron car-bide (Fe3C) The original martensiteboundaries can still be seen between thecarbides, and the martensite retains itsplate-like morphology

A further important structural change

Trang 21

at 1050 F was that the initial, randomly

tangled dislocation mass showed evidence

of some recovery for the first time In

Fig 7 (bottom), which is a higher

mag-nification area than that of Fig 7 (top),

groups of regular, fine-mesh, dislocation

networks can be seen within the

marten-site The formation of these dislocation

planation for the spheroidal morphology

of the carbide precipitate formed Since

Fe3C has a lower density than iron, amass transfer within the matrix is neces-sary if the Fe3C particles are to growand change shape rapidly This masstransfer requires the diffusion of ironatoms within the matrix and this, in

FIG 9—Electron Micrograph of a Steel C Specimen Tempered 1 hr at 800 F (X 80,000).

networks indicates that dislocation climb

and consequent recovery are occurring at

this tempering temperature Since

dis-location climb is a vacancy-controlled

process, the changes in dislocation

ar-rangement also indicate that vacancy

generation and migration must be rapid

at this temperature

The availability of vacancies at this

tempering temperature provides an

ex-turn, is controlled by vacancies Hence,major changes in carbide morphology canonly occur at a temperature sufficientlyhigh to provide a large supply of vacan-cies Below this temperature, the carbidesare restricted to easy nucleation sites,such as boundaries, and a slow andrelatively strained type of growth alongthe boundaries results

The spheroidization of the martensite

Trang 22

BAKER ET AL ON MICROSTRTJCTURE AND TOUGHNESS 15

boundary precipitates is accompanied by

a similar spheroidization of the twin

boundary precipitates, as shown in Fig

8 In addition, it can be seen that, as the

twin boundary precipitate spheroidizes

and leaves a region of matrix, the traces

of the twin boundaries tend to disappear

also This disappearance of the twin

The microstructural changes in steel Cfollowed a similar sequence to those inSteel D The martensite contained all thefeatures already discussed for steel D,and at tempering temperatures up to

800 F, progressively thicker precipitates

of carbide were formed both at the tensite boundaries (Fig 9) andwithin the

mar-FIG 10—Electron Micrograph of a Steel C Specimen Tempered 1 hr at 800 F (X 140,000).

boundaries suggests that the twin

bound-ary and matrix-carbide interface are

closely coupled at this tempering

tem-perature The coupling probably comes

about by the incorporation of the twin

boundary into the precipitate interface

as the precipitate loses coherency during

growth Once incorporated, the twin

boundaries disappear as the carbides

withdraw from a region of the matrix

twins (Fig 10) Again, as in steel D, therewas no change in dislocation arrange-ment up to this temperature At 1000 F,however, spheroidization of the Fe3Cformed along the martensite boundariesand twins occurred, and dislocation re-covery resulted in network formation

The spheroidization of the carbides atthe martensite boundaries is shown in

Fig 11 (top); the corresponding changes

Trang 23

STRUCTURE AND PROPERTIES or ULTRAHIGH-STRENGTH STEELS

FIG 11—Electron Micrographs of Steel C Specimens Tempered 1 hr at 1000 F.

Trang 24

at the twin boundaries are shown in Fig

11 (bottom).

In steels A and B, the changes in

microstriicture followed a similar

se-quence to that observed in steels C and

D, but the changes occurred at lower

tempering temperatures At the lowest

tempering temperature, a continuous

film of e-carbide was produced at the

martensite and twin boundaries, and at

vated dislocation climb results in theformation of networks which are, in ef-fect, low-angle boundaries As thenetworks develop, the martensitic matrix

is divided into many small subgrains ofsimilar orientation With further temper-ing, the subgrain boundaries migrate tobecome higher angle boundaries and, ul-timately, a well-developed fine-grainedstructure is developed within the mar-

FIG 12—Electron Micrograph of a Steel A Specimen Tempered 1 hr at 600 F (X50,000).

higher tempering temperatures (600 F)

the e-carbide was replaced by Fe3C (Fig.

12) At 800 F the boundary Fe3C began

to spheroidize (Fig 13 (top)), and the

matrix showed signs of the changes

re-sulting from dislocation recovery (Fig 13

(bottom).

The structure shown in Fig 13

(bot-tom) illustrates an early stage in the

microstructural changes produced in the

matrix as dislocation recovery and

migra-tion occurs Initially, thermally

acti-tensite matrix The creation of the

sub-grain structure shown in Fig 13 (bottom)

is thus the initial stage in the lization of the steel

recrystal-At higher tempering temperatures, therecrystallization process proceeds morerapidly Figure 14 shows the advancedstage reached at 1000 F

Fracture Profile Metallography:

Figure 15 shows replica micrographstaken from the fracture-surface profiles of

Trang 25

FIG 13—Electron Micrographs of a Steel A Specimen Tempered 1 hr at 800 F.

Trang 26

steel C, tempered at 600 (top) and 1000 F

(bottom) It can be seen that, for material

in the 600-F tempered condition, the

fracture path tends to contain many more

flat facets than that for material

tem-pered at 1000 F; and these facets are

bide In the higher-temperature temperedmaterial the profile shows evidence ofgreater plastic deformation and tearingalong the fracture path, and the fracturepath shows little tendency to follow anydistinct microstructural feature

FIG 14—Electron Micrograph of a Steel A Specimen Tempered 1 hr at 1000 F (X60,000).

often oriented parallel to the carbides at

the martensite boundaries or twin

bound-aries The fracture surface of the steel

tempered at 1000 F has a more roughened

appearance, and there is no obvious

cor-relation between the fracture path and

the microstructure It appears from the

fracture profiles that cracks propagate in

the low-temperature tempered material

along a succession of short, almost

straight paths and that these paths tend

to follow suitably oriented boundary

car-DISCUSSIONThe mechanical property results showthat both tensile strength and fracturetoughness are closely related to temper-ing treatment In each of the steels in-vestigated, there is a a decrease in tensilestrength as the tempering temperature isincreased, but the fracture toughness re-mains low until a critical temperaturerange is reached within which it increasesrapidly For the lower-silicon steels A and

Trang 27

FIG 15—Fracture Profiles of Steel C Tempered 1 hr at 600 and 1000 F (X 10,000).

B, the fracture toughness improves for

temperatures above 600 F, whereas in the

higher-silicon steels C and D the increase

is retarded until a temperature of 900 F

is reached

The metallographic structures showthat the sharp rise in fracture toughnessoccurs at a temperature range in whichmajor structural changes take place Themost important of these changes are (1)

Trang 28

the spheroidization of carbide precipitate

particles, and (2) the reduction or

redis-tribution of the defect structure of the

martensite

It is possible to correlate the changes

in mechanical properties with the

struc-tural changes by considering

qualita-tively the possible influences of

micro-structure on properties In the

marten-sitic condition, the steels exhibit a type

of microstructure in which yielding and

general ductility are likely to be very

limited The martensite is tetragonal

be-cause of the carbon contained in the

lat-tice, and the transformation produces a

high dislocation density In addition, the

many twin boundaries present will act as

barriers to dislocation movement

On tempering at low temperatures,

little relief from the embrittling features

can be obtained; the metastable

martens-ite begins to decompose and the

tetra-gonality is consequently reduced

How-ever, the carbon liberated is utilized in

the formation of almost continuous films

at the martensite boundaries and twin

boundaries These films appear to act as

preferred paths for crack propagation

through the structure, either by

provid-ing adjacent weak zones in the matrix or

by fracturing themselves Consequently,

while the tensile strength of the

marten-site falls, there is no accompanying

in-crease in fracture toughness

In the critical tempering range, several

changes occur that lead to improved

toughness First, continuous films of

car-bide are no longer produced at the

mar-tensite and twin boundaries; instead,

dis-crete rounded particles are formed there

Second, the dense dislocation array is

modified by recovery so that larger areas

of defect-free matrix are developed

Third, the twin boundary structure

dis-appears with the spheroidization of the

twin boundary carbides All these

changes are likely to lead to improved

toughness However, the over-all

densi-ties of carbides and dislocations remainrelatively high, so that the tensilestrength does not fall sharply, as thetoughness increases The phenomenon isessentially one of redistribution of micro-structural features rather than of theircomplete removal

While the observations on fracturepath emphasize the embrittling role ofboundary carbides, the potential tough-ening influence of dislocation recoverymay be equally valuable By freeing largeareas of the matrix from their defect con-tent and by generating a fine-grainedsubstructure, the dislocation changes arelikely to produce some beneficial effect ontoughness

It has been shown that silicon has astrong influence in retarding the majortempering reactions that lead to the de-velopment of high fracture toughness

As already discussed, the tural changes observed as carbide sphe-roidization and dislocation recovery arevacancy - controlled processes and canonly occur at temperatures high enough

microstruc-to provide rapid vacancy formation andmigration If silicon retards tempering,

it would seem, at first sight, that it mightexert its influence by inhibiting the neces-sary vacancy generation or mobility Amore likely explanation is provided by(1) the well-known fact that silicon doesnot form a simple carbide in steels, be-cause of the instability of silicon carbidewith respect to iron carbides, and (2) therecent observation that silicon is in-corporated into the e-carbide formedduring the early stages of tempering (6)

The presence of silicon in these carbides

is likely to reduce the rate at which theycan grow and hence to delay the processesthat will lead eventually to spheroidiza-tion

As long as the carbide dispersion mains stabilized and finely dispersed, andthe carbides are small enough to retainsome coherency with the matrix, the dis-

Trang 29

22 , STRUCTURE AND PROPERTIES OF ULTRAHIGH-STRENGTH STEELS

location array will have little freedom to

climb and recover Thus, the carbide

persion could, in effect, stabilize the

dis-location arrangement and prevent any

general migration

CONCLUSIONS

From the investigation of the changes

in mechanical properties that occur on

tempering and the accompanying

changes in microstructure, it is possible

to draw the following conclusions:

1 In the steels investigated, the

plane-strain fracture toughness shows a rapid

increase within a critical range of

temper-ing temperatures Below this range, the

fracture toughness remains almost

con-stant

2 There is a close relationship

be-tween mechanical properties and

micro-structure; fracture toughness, in

particu-lar, is very sensitive to structural changes

3 Good fracture toughness develops

only when certain important structuralchanges have occurred

4 The microstructural changes thatare of major importance in the improve-ment of fracture toughness are (a) theelimination of embrittling carbide films at

boundaries by spheroidization, and (b)

the redistribution and removal of the tice defect structure by recovery proc-esses

lat-5 The retarding influence on ing exerted by silicon delays the develop-ment of good fracture toughness by in-hibiting the essential microstructuralchanges necessary for its improvement

temper-6 Retarding the tempering process bythe addition of alloying elements doesnot lead to improved fracture toughness

at a given strength level While alloy tions enable higher strengths to be re-tained at higher tempering temperatures,the balance between strength and tough-ness is not improved

addi-REFERENCES

(1) P M Kelly and J Nutting, "Techniques

for the Direct Examination of Metals by

Transmission in the Electron Microscope,"

Journal of the Institute of Metals, Vol 87,

1958-59, p 385.

(2) A S Keh and S Weissmann, "Deformation

Substructure in Body-Centered Cubic

Metals," Electron Microscopy and Strength

of Crystals, Interscience, 1963, p 231.

(3) P M Kelly and J Nutting, "The

Morphol-ogy of Martensite," Journal of the Iron and

Steel Institute, Vol 197, 1961, p 199.

(4) B S Lenient, B L Averbach, and Morris

Cohen, "Microstructural Changes on

Tem-pering Iron-Carbon Alloys," Transactions,

American Society for Metals, Vol 46, 1954,

p 851.

(5) E V Kurdjumov and L Lyssak, "The Application of Single Crystals to the Study

of Tempered Martensite," Journal of the

Iron and Steel Institute, Vol 156, 1947, p 29.

(6) B G Reisdorf, "The Tempering teristics of Some 0.4 Percent Carbon Ulcra-

Charac-high-Strength Steels," Transactions, AIME,

Trang 30

B R BANERJEE1—I am happy to

have the opportunity to discuss this

interesting paper The authors are to be

commended for their fine structural

characterizing of the final

vacancy-recovery-steps in these steels, and indeed

they confirm our own findings in this

respect

However, the authors' conclusions—

based on their fracture-property data

on these steels—tend to contradict much

that has been firmly established on these

steels over the past several decades To

suggest that the toughness of 4340 or its

modifications remain unchanged upon

tempering up to any temperature up

through 600 to 900 F, as the authors

allege (Figs 1 and 2 in their paper), is

contrary to the vast body of

metal-lurgical literature accumulated over the

years through much careful

experi-mentation with these materials

The authors have used fatigue-cracked

circumferentially notched round tension

specimens, and determined the critical

strain-energy release rate, or

crack-extension force values, presumably using

the Irwin approximation But this

approximation does not give the true

glc, but actually a lower bound of

9ic In order for this lower-bound value

to reasonably approach the true values,

the notch must always be sufficiently

sharp, and the plastic zone size must

remain small enough, relative to

speci-men geometry; furthermore, at the

lower toughness regions, loading axiality

becomes quite critical Considering all

1 Manager, Basic Research and Applied

Physics, Research Division, Crucible Steel

Company of America, Pittsburgh, Pa.

these factors, a sufficient number ofdatum points may be critical in revealingtrue trends in behavior

Therefore, I have taken the liberty tocalculate opening-mode fracture-tough-

ness values, KIC , from the authors'

glc data on their air-melt 4340 (Steel A)

on which they present the most completeset of datum points These calculated

KM values are plotted in Fig 16, along

with other recent literature data2 andsome Crucible research data3—all ob-tained on fatigue-cracked circumferen-tially notched round tension specimens

of air-melted 4340 The dotted line,drawn through these points, immediatelysuggests two significant trends which runcontrary to the authors' conclusions:

1 The toughness of air-melted 4340increases throughout the entire temper-ing range of 200 to 1000 F, rather than

"a region of about constant low ture toughness at the lower temperingtemperatures, and a region of rapidlyincreasing toughness," as suggested bythe authors A similar, continued tough-ness increase upon tempering air-melt

frac-4340 was also found by plane-stresstension tests performed with center-notched, fatigue-cracked sheet speci-mens.4

2 A discontinuity in the tempering curve is suggested in the

toughness-2 W A Backofen and M L Ebner, lurgical Aspects of Fracture at High Strength

"Metal-Levels," Watertown Arsenal, WALTR

310.24/5-4, May, 1963.

3 Unpublished results, Research Division, Crucible Steel Company of America, Pitts- burgh, Pa.

Trang 31

500-F (400 to 700 F) embrittlement behavior in both fatigue-cracked and

region brittle-boundary specimens, broken

The authors attribute their unique either under impact or slow-bend

condi-conclusions on fracture-toughness de- tions

pendence upon tempering and to the The l-in.-diameter round bars, which

FIG 16—Dependence of Plane-Fracture Toughness Upon Tempering of Air-Melt 4340.

employment of fatigue cracks in their

tests, thereby rejecting all prior

Charpy-impact literature data as inapplicable

However, in Fig 17 are plotted

fatigue-cracked Charpy-impact data on

con-sumable electrode vacuum-arc-melted,

4335-V from the recent literature,5 which

again confirms the 500-F embrittlement

the authors have used for their fracturetests, may, upon oil quenching, introducesufficient autotempering (due to theslack-quench effect) in the embrittling

8 "Investigation of Fracture Toughness in

High-Strength Alloys," Progress Report No 8,

WADD Contract AF33(616)-8165, ManLabs,

Inc., Cambridge, Mass., May, 1963.

Trang 32

DISCUSSION ON MlCROSTRUCTURE AND TOUGHNESS 25

FIG 17—Dependence of Sharp-Crack Charpy Impact Upon Tempering of Vacuum Arc-Melted

4335-V.

FIG 18—Influence of Notch Acuity on K lc Results for 300M Steel.

range of 4340 steels that recovery from tively lower toughness values observed

these structures may not be obtained y the authors (Fig 16)

until relatively high tempering tempera- In view of these, I would like to

Trang 33

FIG 19—Comparison of K\ c Results From Several Sources.

new test schemes, we should cautiously

proceed to acquire a sufficiently large

body of data through different test

approaches before attempting to broadly

interpret results which violate a

sub-stantial, established body of literature

gained through general experience

A J BAKER, F J LAUTA, AND R P

WEI (authors)—Banerjee's comments

consist of a series of imputations

con-cerning the validity and interpretation

of the fracture toughness data presented

in our paper The comments appear to

be based on a combination of false

assumptions and a general disregard for

the principles of fracture toughness

testing To reply to them adequately, it

is necessary to deal with them in

se-quence and at some length

The fracture toughness parameter,

QIC , used in this paper, is clearly denned

by the Irwin-Orowan-Griffith mechanics analysis6 and requires noelaboration here glc designates thecritical strain energy release rate for theplane-strain opening-mode fracture It isrelated to the plane-strain stress-in-

fracture-tensity factor, KIC , by

where E = elastic modulus and v =

Poisson's ratio The practical ments for valid determinations of glc

require-or K lc were rigorously observed in ourexperiments Fatigue precracking was

6 G R Irwin, "Fracture Mechanics," tural Mechanics, Pergamon Press, New York,

Trang 34

DISCUSSION ON MlCROSTRUCTURE AND TOUGHNESS 27

employed to ensure adequate notch

acuity

Specimen sizes were chosen to insure

adequate constraint; for example,

1-in.-diameter specimens were used for glc

measurements on Steel A (AISI 4340

steel) tempered at 800 F when

f-in.-diameter specimens were found to be

inadequate for this tempering treatment

Axiality of loading was good and

intro-duced errors of less than 10 per cent in

the 9ic values It should be noted that

poor alignment will accentuate any

embrittling phenomenon that may be

present, and consequent effects, such as

a purported 500-F embrittlement, are as

likely to be observed under these testing

conditions as any other Since the need

for fatigue precracking of specimens has

been recognized only recently, a

disagree-ment with previously published results

is expected A typical example is

pro-vided by Fig 18, comparing the

ap-parent KIC results reported by Carmen

et al7 with the authors' KIC data for

300M steel, which is Steel C in this

paper Such discrepancies are bound to

persist, moreover, if K^ c results are

com-puted and reported in complete

viola-tion of one or more of the fracture

mechanics requirements Banerjee's data

for AISI 4340 steel tempered at 1000 F,

shown in Fig 16, are an obvious example

of such violation

Regarding the specific comparison

between the KIC results of Backofen

and Ebner8 and those reported in this

paper for AISI 4340 steels, Fig 1, it was

suggested in our paper that comparisons

be made at equal tensile-strength levels,

and that proper allowances be made for

7 C M Carmen, D F Armiento, and H.

Markus, "Plane Strain Fracture Toughness

Measurements of High Strength Steels,"

Jour-nal of Basic Engineering, ASME, March, 1963,

p 87.

8 W A Backofen and M L Ebner,

"Metal-lurgical Aspects of Fracture at High Strength

Levels," Watertown Arsenal, WALTR 310.241

5-4, May, 1963.

differences in sulfur and phosphoruscontents, since they influence fracturetoughness Backofen and Ebner's results9

for this steel are replotted in Fig 19using the tensile strengths reported byShih et al.10 With sulfur and phosphoruslevels intermediate between Steels A and

B, the steel exhibits a plane-strain ture toughness intermediate betweenthat of Steels A and B The agreementwith our finding is, in fact, good

frac-A direct comparison between K IC

results and Charpy impact test results,

as suggested by Banerjee, should not beattempted, since the bases of the two

tests are quite different Whereas K^ c

characterizes the plane-strain mode fracture, the Charpy impact testmeasures the energy required to fracture

opening-a mildly-notched specimen in opening-a mode fracture with other attendantenergy losses Similarly, Kc or gc resultsfor sheet- or plate-type specimens arealso strongly influenced by the modes offracture, plane-stress versus plane-strain,

mixed-and comparisons with K IC values should

be made with caution The continued

increase of K c values with increasingtempering temperature, 300 F to 700 F,indicated by Banerjee had been ob-served previously by the authors.11

However, this behavior reflects cipally a change in fracture mode, asindicated by a change from 20 per centshear in the fracture for a 300-F temper-ing treatment to 100 per cent shear for

prin-9 Kic recalculated using Irwin's relationship;

Kic = 0.414 ff n\/D, where <rn is the nominal

notch tensile strength and D is the specimen

diameter Bueckner's relationship, used by Backofen and Ebner, uses a constant of approximately 0.46.

10 C H Shih, B L Averbach, and Morris Cohen, "Some Effects of Silicon on the Mechani-

cal Properties of High Strength Steels,"

Trans-actions, American Society for Metals, Vol 48,

1956, p 86.

11 R P Wei, "Fracture Toughness Testing

in Alloy Development," Fracture Toughness

Testing and Its Applications, ASTM STP 381

Am Soc Testing Mats., 1964, to be published.

Trang 35

tempering treatments above 600 F, in

O.lOO-in.-thick specimens

A 500-F embrittlement, or martensite

temper embrittlement per se, was not

indicated by our data, and Backofen

and Ebner8 reached the same conclusion

in their study The embrittlement

indi-cated by the discusser, Fig 16, results

in Charpy impact energy are regarded

as evidence of a significant change infracture toughness

Concerning the question of tempering in the 4340 steels (owing

auto-to the use of large section size specimens),

we must emphasize that essentially allthe fracture tests on the 0.40 per cent

FIG 20—Effect of Specimen Size on Fracture Toughness Test Results.

from an indiscriminate averaging of two

distinctly different sets of data, and the

use of an alternately contracted and

expanded scale for the tempering

tem-peratures Furthermore, the impact

and bend test data, Fig 17, do not

provide convincing substantiation of a

500-F embrittlement, unless the

slow-bend test results are ignored and the

minor changes (1 to 2 ft-lb variation)

carbon steels (A, B, and C) were ried out using f-in.-diameter specimens,and not l-in.-diameter specimens as thediscusser states Only in the case ofSteel A, tempered at 800 F, and Steel C,tempered at 1050 F, were l-in.-diameterspecimens used, since additional con-straint was required All the steelsstudied possess high hardenability andare fully hardened in a 1-in section

Trang 36

DISCUSSION ON MlCROSTRUCTTJRE AND TOUGHNESS 29

Figure 20 shows KI C data derived from

both f-in.- and ^-in.-diameter specimens

of Steel A.12 The close agreement

be-tween the two sets of values refutes the

suggestion that autotempering can have

12 F J Lauta, Unpublished data, U S Steel

Corp., Applied Research Laboratory,

Trang 37

RELATIONSHIPS BETWEEN STRUCTURE AND PROPERTIES

IN THE 9NI-4CO ALLOY SYSTEM

BY J S PASCOVER1 AND S J MATAS1

SYNOPSIS The effects of alloy content and heat treatment on the strength and toughness of 9Ni-4Co alloy steels are discussed in detail The strength was determined with standard tension specimens, while toughness was evaluated with various accepted, sharply notched specimen types.

Of the alloying elements used, carbon acted as the strongest embrittler but was necessary as a strengthener Silicon and the carbide-forming elements decreased toughness, particularly at higher carbon contents Hence, the use of silicon at other than residual levels was undesirable The addition of carbide- forming elements in appreciable amounts needed careful consideration Nickel was used in substantial amounts to improve toughness, while cobalt was added

to inhibit retained austenite formation consequent upon the high nickel tent Cobalt enhanced self-tempering effects, particularly for the lower carbon varieties.

con-Heat treatment strongly influenced the properties as a result of the tendant changes in the microstructure Isothermal transformation to lower bainite enhanced toughness in the 0.45%C, 9%Ni-4%Co high performance (HP) 9-4-45 steel The toughness was superior to that of tempered martensite at equivalent strengths The properties of upper bainite were not as at- tractive as those of lower bainite Mixed structures composed of bainite and martensite lowered toughness.

at-The two commercial grades of the 9Ni-4Co alloy system, HP 9-4-25 and

HP 9-4-45, had a better combination of strength and toughness than other carbon-strengthened steels The self-tempering characteristics of the HP 9-4-25 grade and its inherent toughness contributed to high joint efficiences and formability in the as-welded condition No post-heat treatment of the welded joint was required.

The need for materials of high sistance to premature failure owing to strength-to-weight ratio is unquestion- pre-existing flaws Resistance to corro-

able Strength per se, however, is not a sive environments and to cyclic

fatigue-sufficient criterion for the use of a ma- type loading, as well as amenability to terial in many applications For highly fabrication, are additional requirements, stressed structural components, the Many approaches have been used to material must also exhibit a high re- obtain steels with both high strength and

adequate reliability (1-5) 2 This paper,

1 Research metallurgist and supervisor of high

strength steels, respectively, Research Center, 2 The boldface numbers in parentheses refer Republic Steel Corp., Cleveland, Ohio to the list of references appended to this paper.

30

Trang 38

PASCOVER AND MATAS ON PROPERTIES or ALLOY STEELS 31

however, will be confined to a discussion

of an investigation aimed toward

op-timizing the properties of the

carbon-strengthened, low-alloy, high-strength

steels by balancing the composition The

final result of this study was the

devel-opment of 9Ni-4Co-XC steels3 which

contain basically about 7 to 9 per cent

Ni, 2 to 5 per cent Co and 0.15 to 0.50

per cent C (4-8)

The 9Ni-4Co alloy system exhibits a

wide range of properties For example,

ultimate tensile strengths as high as 300

ksi and, at lower strengths, impact

energies up to 80 ft-lb were obtained

The 9Ni-4Co alloy steels are amenable

to various processing techniques It wasfound that at a given strength level asignificant improvement in toughness can

be obtained through the use of a specialmelting practice, referred to as vacuumarc remelt carbon deoxidation (VAR-CDOX) practice4 (5,8,9) With thistechnique the product of deoxidation is agas instead of a solid reaction product,

as in conventional deoxidation practice

This alloy system is also responsive tothermal-mechanical treatments (1) Inparticular, both the strength and tough-ness can be increased by hot-cold work-

TABLE 1—COMPOSITIONS OF ALLOY STEELS INVESTIGATED, WEIGHT PER CENT.

0.45 0.28 0.40 0.40 0.36 0.35 0.43 0.41 0.46 0.45 0.41

0.25 0.35 0.30 0.70 0.70 0.90 0.02 0.21 0.21 0.21 0.33

0.10 0.10 0.90 0.25 0.27 1.30 0.01 0.03 0.24 0.44 0.82

8.00 8.00 1.80 1.80 1.80 8.00 8.00 7.29 7.38 7.38

0.30 0.50 5.00 0.80 0.80 0.95 0.09 0.34 0.20 0.24 0.22

0.30 0.50 1.30 0.25 0.35 0.40 0.08 0.28 0.19 0.21 0.19

0.10 0.10 0.50 0.20 0.14 0.19 0.12 0.08 0.08 0.12

4 0 0

3^04.0 0

2.0 7

4.0!

3.6 0

0 Typical alloy composition.

The lower carbon grade, referred to

commercially as HP 9-4-25, exhibits a

Charpy impact energy of 50 ft-lb at the

200 ksi yield strength level The higher

carbon grade, HP 9-4-45, on the other

hand, will develop 25 ft-lb of impact

energy at a yield strength of 250 ksi

Furthermore, the lower carbon varieties

can be welded in the quenched and

tempered condition (200 ksi strength

level) without preheat or postheat The

higher carbon compositions, however,

must be welded in the annealed condition

and fully post-treated to achieve joint

efficiencies in excess of 90 per cent at an

ultimate tensile strength of 290 ksi

3 The 9Ni-4Co-XC steels, designated

com-mercially as HP 9-4-X, are proprietary steels

of Republic Steel Corp.

ing (l) Such treatment involves the formation of austenite prior to thetransformation of austenite to martensite(1,2) or bainite

de-Although some information is available

in the literature on the effects of alloyingelements on the strength and toughness

of the 9Ni-4Co alloy system (4,8,10), nocomprehensive study has been published

Thus, this paper is concerned with theinfluence of composition as it relates tothe strength and toughness of the9Ni-4Co alloys when they are heat-treated to a tempered martensitic struc-ture In addition, the properties of the

HP 9-4-45 alloy steel, when heat-treated

to a bainitic microstructure, are

com-4 A proprietary melting practice of Republic Steel Corp.

Trang 39

The production-sized heats weremelted as 70 or 5-ton basic, electricarc-furnace heats and poured intoelectrode molds without additions ofmetallic deoxidizers During vacuumconsumable electrode remelting into24-in.-diameter crucibles, the steel wascarbon deoxidized, yielding a very clean,

pared to those obtained by tempering

martensite

MATERIALS AND PROCEDURE

Materials:

The chemical compositions of the heats

investigated are presented in Table 1

Specimens were austenitized, oil-quenched, refrigerated, and tempered at the tempering

tem-peratures indicated.

FIG 1—Influence of Carbon on Strength of HP 9-4-X Alloy Sytem.

They represent two basic types—150-lb

air induction heats (Si-Al deoxidized) and

commercial production-sized heats

(vac-uum arc remelted carbon deoxidized)

The 150-lb heats were melted as air

induction heats Using pure materials,

two 70-lb ingots were poured, forged to

2 by 2-in billets and subsequently

straightaway hot-rolled to appropriate

plate and sheet sizes

low gas content material (5,6,9) Thevacuum remelted ingot was then forged

to a 4 by 12-in slab product and rolled (2:1 cross roll) to the appropriateplate or sheet section thicknesses

hot-Procedure:

Oversized specimen blanks were cutfrom hot-rolled product, heat-treated,and then finish-ground to appropriate

Trang 40

PASCOVER AND MATAS ON PROPERTIES OF ALLOY STEELS 33

dimensions Standard sheet tensile

speci-mens were used for evaluating sheet

product The tensile strength of plate

material was evaluated by standard

round tensile specimens To evaluate

notch toughness, several types of

speci-mens were used (11)

em-Values are for longitudinal and transverse specimens from vacuum arc remelted carbon

de-oxidized heats.

FIG 2—Influence of Carbon on Charpy V-Notch Impact Energy of HP 9-4-45 Plate Tempered

at the Temperature Indicated.

The fatigue precracked impact

speci-men employed in this study was similar

in nearly all respects to the ASTM

standard Charpy V-notch specimen The

only difference lay in the fatigue crack

introduced at the V-notch root by a

device especially designed for the purpose

(12,13) In this case, the value reported as

W/A represents the unit work to failure

in a manner analogous, but not

neces-sarily identical, to the G c value

deter-mined from notched sheet specimens

a secondary influence on strength but isvery important in controlling toughness(6,14) The effect of carbon and alloycontent on strength, ductility, toughness,and tempering response of martensitewill be discussed in detail in this section

Effect of Carbon:

The effect of carbon on the strength

of tempered martensite in the 9Ni-4Coalloy system can be seen in Fig 1 Thestrength increase with increasing

Tiêu đề	Structure and Properties of Ultrahigh-Strength Steels
Người hướng dẫn	F. M. Richmond, Chairman, J. W. Welty, Chairman, E. E. Reynolds, Presiding, J. J. Heger, Presiding
Trường học	University of Washington
Chuyên ngành	Metallurgy
Thể loại	Bài báo
Năm xuất bản	1965
Thành phố	Cleveland

Định dạng
Số trang	230
Dung lượng	9,68 MB