An Investigation of the Plastic Fracture of

AISI 4340 and 18 Nickel--200 Grade

Maraging Steels
T. B. COX AND J. R. LOW, JR.

The mechanisms of plastic fracture (dimpled rupture) in high-purity and c o m m e r c i a l 18 Ni,

200 grade maraging steels and quenched and tempered AISI 4340 steels have been studied.
Plastic fracture takes place in the maraging alloys through void initiation by fracture of
titanium carbo-nitride inclusions and the growth of these voids until impingement results
in coalescence and final fracture. The fracture of AISI 4340 steel at a yield strength of 200
ksi (1378 MN/mm 2) occurs by nucleation and subsequent growth of voids formed by fracture
of the interface between manganese sulfide inclusions and the matrix. The growth of these
inclusion-nucleated voids is interrupted long before coalescence by impingement, by the
formation of void sheets which connect neighboring sulfide-nucleated voids. These sheets
are composed of small voids nucleated by the cementite precipitates in the quenched and
tempered structures. The sizes of non-metallic inclusions are an important aspect of the
fracture resistance of these alloys since the investigation demonstrates that void nuclea-
tion occurs more readily at the l a r g e r inclusions and that void growth also proceeds more
rapidly from the larger inclusions. Using both notched and smooth round tensile specimens,
it was demonstrated that the level of tensile s t r e s s triaxiality does not effect the void nu-
cleation p r o c e s s in these alloys but that increased levels of triaxial tension do result in
greatly increased rates of void growth and a concomitant reduction in the resistance to
plastic fracture.

J ~ E C E N T L Y , increased use has been made of high-
strength alloy steels for critical structural applica-
tions in naval v e s s e l s , air craft, and aerospace ve-
hicles. Unlike the low-strength ferritic steels in
which unstable brittle fracture occurs by cleavage,
these high-strength alloys may fracture in a brittle
manner by dimpled rupture due to highly localized,
severe plastic deformation. 1 While brittle fracture
of steels by cleavage has been extensively studied
and many of the m i c r o s t r u c t u r a l features which affect
cleavage have been determined, very little r e s e a r c h
has been done on the plastic fracture process (fre-
quently r e f e r r e d to as dimpled rupture, fibrous
fracture, or ductile fracture) since, in the past, it
generally occurred in a very ductile manner without
catastrophic results.
There are three generally recognized stages of
plastic fracture: void initiation, void growth, and void
coalescence. It is usually observed that the initiation
of voids at second-phase particles within the metal
matrix is the first step in the plastic fracture proc-
ess. 2 The subsequent growth and coalescence of these
voids during further loading eventually leads to final
fracture. An extensive study using copper alloys with
various second-phase particles including metals, non-
metals, and voids by Edelson and Baldwin s showed
that the ductility of the aggregates increased as the
volume fraction of any of the second-phase particles
w~s reduced. The improvement in ductility was ex-
t r e m e l y rapid with reductions in volume fractions of
T. B. COX is Metallurgist, Code 2811 Metals Division, Naval Ship
Research and DevelopmentCenter, Annapolis, Md. 21402. J.R.
LOW,Jr., is Professor, Department of Metallurgy and Materials Science,
at Carnegie-Mellon University, Pittsburgh, Pa. 15213.
Manuscript submitted July 9, 1973.
second-phase at and below about 0.01, a level ap-
proximately that found in most c o m m e r c i a l alloys. It
has been demonstrated that lowering the sulfur con-
tent of pearlitic steels can result in substantial im-
provements in the energy level of the upper Charpy
shelf, 4 and reduced sulfur levels in quenched and
tempered steels has also been shown to improve f r a c -
ture toughness. 5 While the practical method of im-
proving the fracture toughness of the high-strength
steels by reducing the impurity levels is based upon
these observations, very little is understood concern-
ing the mechanisms of the plastic fracture process in
the high-strength alloys.
The objectives of this investigation were to describe
the mechanisms by which several common high-
strength steels fracture plastically, with the belief
that through a better understanding of the fracture
p r o c e s s , it would be possible to suggest means of
improving the resistance of these alloys to fracture.
Determination of the reasons for the brittle nature
of plastic fractures when occurring under high levels
of triaxial tension was also an objective of this re-
search, together with an explanation for the generally
observed fact that the maragtng steels exhibit much
higher fracture toughness values than the quenched
and tempered steels when compared at equal strength
levels.
MATERIALS
The materials used in this investigation included
both high-purity and commercial heats of AISI 4340
steel and 18 Ni, 200 grade maraging steel provided
in plate form. The chemical analyses of the four al-
loys are presented in Table I. The major differences
to be noted are the substantial reductions in impurity

METALLURGICALTRANSACTIONS VOLUME 5, JUNE 1974-1457

levels in the high-purity heats compared to the com- compact tension specimens and bend specimens in
m e r c i a l heats. accordance with ASTM Standard E399-72. The f r a c -
Since it was desired to investigate the fracture be- ture toughness specimens were tested in the TL ori-
havior of these alloys at approximately the same entation (notched through the plate thickness with the
strength level, they were given heat treatments de- direction of crack propagation in the p r i m a r y rolling
signed to produce yield strengths close to 200 ksi direction). The tension test specimens were cut from
(1378 MN/mm2). Both AISI 4340 alloys were austeni- the plates with the tension axis lying in the plane of
tized at 843~ (15500F) for one hour and oil quenched. the plate and perpendicular to the p r i m a r y rolling
Due to the differences in carbon contents, the tem- direction, that is, with the direction of the tension
pering treatments were different with the c o m m e r c i a l axis in the same orientation as the principle normal
AISI 4340 being tempered at 4350(2 (815~F) for one hour s t r e s s in the fracture toughness specimens.
and air cooled while the high-purity alloy was tem- Reference to Table II reveals that the yield strengths
pered at 427~ (800~ for one hour and air cooled. of the alloys are in the general range of 200 ksi (1378
The specially produced high-purity 18 Ni, 200 grade MN/mm2). The values of true strain to fracture (re-
maraging steel was solution annealed at 843~ (1550~ duction in area) indicate that the ductility of these
for one hour and water quenched, then aged for three alloys is substantially improved by high-purity melt-
hours at 4540(2 (850~ and air cooled. The c o m m e r - ing, and that the general level of ductility is higher
cial grade 18 Ni steel was solution annealed at 8990(2 for the maraging steels than for the quenched and
(1650~F) for two and one-half hours and air cooled, tempered AISI 4340 alloys. Also, the level of fracture
followed by aging at 4460(2 (835~F) for six hours and toughness is greater in the maraging steels than in
air cooled. the AISI 4340 steels even when considering the dif-
The room temperature mechanical properties re- ferences in yield strengths. Within a given alloy sys-
sulting from these heat treatments are given in Table tem, reduction of the impurity levels improves the
II. The tensile data were determined using standard fracture toughness.
geometry, smooth, round tensile specimens; and the
fracture toughness testing was c a r r i e d out using both
NON-METALLIC INCLUSIONS
As indicated above, past investigators have noted
that in many instances non-metallic inclusions are
Table I. Chemical Composition of High Strength Steels (Wt Pct)
the sites for void nucleation. In o r d e r to define the
Commercial High Purity Commercial High Purity exact nature of the non-metallic inclusions in these
AISI 4340 AISI 4340 18 Ni 18 Ni four alloys, a metallographic investigation was c a r -
C 0.43 2 0.01 0.38 2 0.01 0.021 2 0.004 0.002 + 0.002
ried out. F o r each of the four alloys, metallographic
Mn 0.78 2 0.02 0.71 2 0.02 0.04 2 0.001 <0.02 + 0.01 sections were examined from each of three ortho-
P 0.010 • 0.001 0.001 2 0.001 0.003 2 0.001 0.002 • 0.001 gonal directions, the axes being defined by the rolling
S 0.013 2 0.001 0.004 2 0.001 0.008 2 0.001 0.004 2 0.001 directions and the through thickness direction in each
Si 0.27 2 0.02 0.29 2 0.02 0.04 2 0.01 <0.03 2 0.02 of the plates.
Ni 1.7320.02 1.81 2 0 . 0 2 18.320.1 18.1 20.1
Polished sections of both c o m m e r c i a l and high-
Cr 0.76 2 0.02 0.80 2 0.02 < 0 . 0 4 2 0.02 0.05 2 0.02
Mo 0.25 2 0.01 0.25 2 0.01 4.23 2 0.05 4.16 + 0.05 purity AISI 4340 alloys revealed that the predominate
Al(total) 0.049 2 0.003 0.023 2 0.002 0.080 2 0.002 0.062 2 0.002 inclusion type in these alloys appeared as a gray
Co - 0.010 2 0.01 7.85 2 0.1 8.03 2 0.1 ellipse in each of the three perpendicular directions.
Ti - <0.005 2 0.002 0.19 2 0.01 0.23 2 0.01 Generally, the inclusions were observed to be l a r g e r
N2 0.008 :t 0.001 0.003 2 0.001 0.005 2 0.001 0.004 • 0.001
in the commercial AISI 4340, and the major axes of
O2 502 10ppm 23 2 3 ppm 302 10ppm 23 • 3 ppm
Fe Balance Balance Balance Balance
the ellipses seemed more elongated in the rolling di-
rection of the commercial alloy. Examination of
these inclusions, using a scanning electron m i c r o -
scope equipped with an X - r a y energy-dispersive
analyzer, showed that the ellipsoidal inclusions in
Table II. Room Temperature Mechanical Properties of High Strength Steels both the commercial and high-purity AISI 4340 al-
(All Values Represent Means of Duplicate Tests)
loys were much richer in manganese and sulfur than
Commercial High Purity Commercial High Purity the matrix while small i r r e g u l a r particles generally
AISI 4340 AISI 4340 18 Ni 18 Ni found within the ellipsoids were shown to contain
calcium and aluminum. Based upon these X - r a y data
Yield Strength 0.2 204.8 204.6 193.5 189.9
pet Offset (ksi)
and their appearance in the optical microscope as
(1 ksi = 6.89 MN/mm 2) compared with descriptions from previous inclusion
Ultimate Tensile 221.6 218.0 197.7 198.5 studies, 6 these gray, ellipsoidal inclusions in the AISI
Strength (ksi) 4340 alloys were identified as manganese sulfides
(1 ksi = 6.89 MN/mm 2) (MnS); and the dark particles internal to the sulfides
True Strain to 0.287 0.515 0.747 1.005
as some product of the deoxidation p r o c e s s upon which
Fracture
Fracture Toughness 67.9 97.2 113.0 149.1 *
the sulfides are nucleated.
Kle(ksi V'~) The most prevalent non-metallic inclusion type in
(1 k s i V ~ = 1.10 both of the 18 Ni, 200 grade maraging alloys appeared
MNV-~m2) as sections of cubes and exhibited a pink color in the
*KQ,invalid test due to undersized specimens. All specimens were nominally optical microscope in all three views. The cubes did
1 in (2.54 em) thick. not appear to have been plastically deformed during
the hot-rolling operation, and the cubic sections ap-
1458-VOLUME 5, J U N E 1 9 7 4 METALLURGICAL TRANSACTIONS
p e a r e d to have a p p r o x i m a t e l y the same size d i s t r i b u -
tion in each of the three views of a given alloy. The
cubes did a p p e a r to be l a r g e r in the c o m m e r c i a l 18
Ni m a r a g i n g s t e e l . C o m p a r i s o n of these optical
m e t a l l o g r a p h i c o b s e r v a t i o n s with those of a previous
investigation7 together with X - r a y data from the scan-
ning e l e c t r o n m i c r o s c o p e showing that the cubes were
r i c h in titanium identified these inclusions as titanium
c a r b o - n i t r i d e s , Ti(C ,N).
A quantitative m e t a l l o g r a p h i c investigation using
light m i c r o s c o p y was undertaken to d e t e r m i n e the ex-
act nature of the inclusion populations in the four a l -
loys. Based on the definition of the inclusion shapes
indicated above, e s t a b l i s h e d quantitative m e t a l l o -
graphic techniques were used to d e t e r m i n e the num-
b e r s , spacings, s i z e s , and volume f r a c t i o n s of the
inclusions in the a l l o y s . The quantitative m e t a l l o -
graphic a n a l y s i s for the manganese sulfide e l l i p -
soids is due to DeHoff and Rhines a'~ and the a n a l y s i s
for the titanium c a r b o - n i t r i d e cubes is due to Hull
and Houk. 1~ The r e s u l t s of the quantitative m e t a l l o g -
raphy a r e s u m m a r i z e d in Table III where the l i m i t s
given for each r e p o r t e d value r e p r e s e n t plus and
minus one s t a n d a r d deviation. While a few inclusion
types other than the manganese sulfides and titanium
c a r b o - n i t r i d e s were o c c a s i o n a l l y o b s e r v e d , they were
too few in number to be s t a t i s t i c a l l y significant and
a p p a r e n t l y did not affect the f r a c t u r e p r o c e s s as d e m -
o n s t r a t e d below.
It should be noted in Table III that the number of
inclusions p e r unit volume (Nv) is g r e a t e r for the
high-purity alloys than for the c o m m e r c i a l l y produced
p l a t e s , but that inclusion s i z e s (d,, d2, ~/3 r e p r e s e n t
the a v e r a g e d i a m e t e r s of the e l l i p s o i d a l manganese
sulfides and ~ the average cube edge of the titanium
c a r b o - n i t r i d e s ) and volume f r a c t i o n s (Vv) a r e sub-
s t a n t i a l l y l a r g e r in the c o m m e r c i a l alloys. Note also
that the volume f r a c t i o n s of inclusions in the m a r a g -
ing s t e e l s a r e l a r g e r than in the AISI 4340 alloys even
though the m a r a g i n g alloys a r e tougher (see Table II).
The c e n t e r - t o - c e n t e r spacings of inclusions (~) follow
the i n v e r s e r e l a t i o n of the number p e r unit volume,
namely the i n t e r p a r t i c l e spacings a r e g r e a t e r in the
c o m m e r c i a l alloys than in the high purity alloys. As
may be seen from the r e s u l t s , the v a r i a b i l i t y of the
v a r i o u s m e a s u r e m e n t s is quite high. However, t r e a t -
ment of the data using s t a n d a r d s t a t i s t i c a l methods
for hypotheses concerning means shows that the
above conclusions concerning n u m b e r s , s i z e s , s p a c -
ings, and volume f r a c t i o n s of the inclusions a r e
s t a t i s t i c a l l y meaningful with 90 pct confidence.
Table III. Results of Quantitative Metallographic Investigation of Inclusions in High Strength Steels
Commercial
AISI 4340
N v (inclusions/cm a) 4.9 X 106 + 2.6 X 10 6
cll or 9 ~ m ) 9.7 + 5.8
d2 ~ m ) 7.3 + 6.3
da (/zm) 5.5 + 1.4
~. Oam) 33 + 6
Vv (pet) 0.14 -+ 0.05
Note: 4" are average diameters of ellipsoidal manganese sulfide inclusions; ~ is the average cube edge of titanium carbo-nitride inclusions;}, is the average center-to-center
spacing between inclusions; Vv is the volume fraction of inclusions.
METALLURGICAL TRANSACTIONS
FRACTOGRAPHY
A f r a c t o g r a p h i c study was c a r r i e d out on the four
alloys to define the nature of the f r a c t u r e s u r f a c e s .
The f r a c t u r e s u r f a c e s f r o m plane s t r a i n f r a c t u r e
toughness s p e c i m e n s , smooth, round t e n s i l e b a r s ,
and mildly notched t e n s i l e b a r s were examined for
each alloy. All of the f r a c t u r e s u r f a c e s on the f r a c -
ture toughness s p e c i m e n s w e r e rough and fibrous in
a p p e a r a n c e when viewed at low magnification. Both
the smooth and notched round t e n s i l e b a r s for a l l a l -
loys exhibited a cup and cone type f r a c t u r e with the
c e n t r a l regions being rough and fibrous in a p p e a r -
ance at low magnifications.
Two stage, c e l l u l o s e a c e t a t e - p l a t i n u m shadowed
carbon r e p l i c a s were taken f r o m the a r e a s of f a s t
f r a c t u r e initiation on the K i c f r a c t u r e s u r f a c e s and
from the c e n t r a l a r e a s of n o r m a l rupture on the
smooth and notched t e n s i l e f r a c t u r e s u r f a c e s for each
of the a l l o y s . The r e p l i c a s were examined using an
e l e c t r o n m i c r o s c o p e , and based upon extensive ex-
amination of s t e r e o p a i r s , it was found that the fea-
t u r e s on the f r a c t u r e s u r f a c e s were qualitatively the
s a m e for each of the three f r a c t u r e t e s t s (Klc; smooth,
round tension; and notched, round tension) of any
given alloy. All the f r a c t u r e s were e n t i r e l y by d i m -
pled rupture (plastic f r a c t u r e ) .
Quantitative m e a s u r e m e n t s of the s i z e s and s p a c -
ings of d i m p l e s on many f r a c t u r e s u r f a c e s r e v e a l e d
that the a v e r a g e spacings of l a r g e d i m p l e s c o r r e -
sponded well with the c e n t e r - t o - c e n t e r spacings of
the predominate n o n - m e t a l l i c inclusions in each of
the a l l o y s . F u r t h e r m o r e , these m e a s u r e m e n t s con-
f i r m e d the o b s e r v a t i o n that the f e a t u r e s on the f r a c -
ture s u r f a c e s were the same for the three s p e c i m e n
g e o m e t r i e s for any given alloy. Thus, it is suggested
that the m i c r o s t r u c t u r a l p r o c e s s e s which lead to f r a c -
ture a r e the same for all three t e s t s in these p a r t i c u -
l a r alloys even though the level of c o n s t r a i n t ( s t r e s s
t r i a x i a l i t y ) i n c r e a s e s substantially when going f r o m a
smooth, round s p e c i m e n to a notched, round s p e c i -
men to a fatigue c r a c k e d plane s t r a i n s p e c i m e n . It is
i m p o r t a n t to note that s t e r e o s c o p y was used exten-
sively to d e t e r m i n e the exact nature and extent of the
d i m p l e s on the f r a c t u r e s u r f a c e s . I n c o r r e c t conclu-
sions can be r e a c h e d concerning the f e a t u r e s on f r a c -
ture s u r f a c e s if p l a n a r m i c r o g r a p h s a r e used alone
without the aid of a s t e r e o v i e w e r , as d e m o n s t r a t e d
by B r o c k . 36
An example of a typical f r a c t o g r a p h taken f r o m the
18 Ni, 200 grade maraging s t e e l s is p r e s e n t e d in F i g .
High Purity Commercial High Purity
AIS14340 18 Ni Maraging 18 Ni Maraging
6.8 X 106 + 4.6 X 1 0 6 2.9 X 106 + 1.5 X 106 5.4 X 106 + 5.1 X 106
4.5 + 4.1 8.6 + 6.4 3.0 + 3.8
4.9 + 4.2 - -
3.3 -+ 0.2 - -
29 + 7 39 + 7 32 -+ 10
0.06 + 0.03 0.21 + 0.08 0.09 + 0.05
V O L U M E 5, J U N E 1 9 7 4 - 1 4 5 9
1 and shows the fracture surface to be almost com-
pletely covered by large equiaxed dimples approxi-
mately 10 to 20 /~m in diam. At the positions marked
" X " on the fractograph, examples of impressions of
shattered carbo-nitride particles may be seen at the
bottoms of dimples. Stereoscopic examination of these
shattered carbo-nitrides revealed that the pieces fit
together much like a jigsaw puzzle. Although a few-
carbo-nitrides were observed to be cracked in un-
strained specimens, no separation of broken pieces
nor noticeable clustering of inclusions was observed
before straining. The conclusion is that the shatter-
ing occurred at small plastic strains and substantial
separation of the broken pieces of the carbo-nitrides
observed on the fracture surfaces occurred during
the fracture p r o c e s s .
The particular carbo-nitride marked " Y " in Fig. 1
exhibits clearly a cleavage fracture pattern, sug-
gesting that the associated dimple was nucleated by
the separation of the halves of the fractured non-
metallic inclusion. At the position marked " R " in
Fig. 1, evidence of surface rumpling of the dimple
walls is seen. These wavy, linear features are
reminiscent of the intense deformation markings on
the free surfaces of tensile specimens which have
been strained well into the plastic region. The im-
plication is that the dimple wails were free surfaces
during a portion of the plastic deformation of the
matrix and that large amounts of plastic deformation
occurred after void initiation. This latter point is
further substantiated by the large separations ob-
served between pieces of the shattered carbo-nitrides
on the maraging fracture surfaces, in many instances
Fig. 1--Electron fractograph from 18 Ni maraging steel
(shattered earbo-nitrides indicated by X; cleavage fracture
pattern on carbo-nitride indicated by Y; deformation mark-
ings indicated by R).
1460-VOLUME 5, JUNE 1974
representing strains several times the observed mac-
roscopic strains at fracture.
A representative fractograph illustrating the fea-
tures observed on the AISI 4340 fracture surfaces is
presented in Fig. 2. The AISI 4340 fracture surfaces
were composed of widely spaced large dimples 5 to
15 ~tm in diam. (as marked with the letter " L " in Fig.
2) separated by extensive areas of very fine dimples
one-to-two orders of magnitude smaller. Both com-
mercial and high-purity AISI 4340 alloys exhibited ex-
tensive surface rumpling on the large dimple walls as
was observed in the maraging steels. The feature
marked " X " within the large dimple in Fig. 2 is the
impression of the void nucleating particle. The repli-
cated shapes of these particles were generally ellip-
soidal. This observation combined with the fact that
the large dimple spacings on the fracture surfaces
corresponded with the spacings of the non-metallic in-
clusions in the AISI 4340 alloys demonstrates that the
large dimples were nucleated by manganese sulfide
inclusions. Careful examination of a number of the
nucleating sulfides at the bottoms of large dimples
suggested that the void had been formed by decohesion
of the sulfide-matrix interface since no fracture pat-
terns were observed on the nucleating sulfides. Some
evidence in the form of deformation markings and
steps on the inclusion surfaces suggested that the
sulfides had undergone plastic deformation during
straining.
Examination of the areas of fine dimples at high
magnification showed that many of these small dim-
ples contain the impression of a nucleating particle.
The absence of any fracture patterns on these nu-
Fig. 2--Electron fractograph from AISI 4340 steel (large
dimple indicated by L; nucleating particle indicated by X).
METALLURGICALTRANSACTION,.;
cleating p a r t i c l e s suggest that the s m a l l voids were
nucleated by decohesion of the m a t r i x from the s m a l l
p a r t i c l e s . In o r d e r to identify the nucleating p a r t i -
c l e s of the s m a l l d i m p l e s , a s e r i e s of e x t r a c t i o n
r e p l i c a s w e r e taken f r o m the f r a c t u r e s u r f a c e s of the
AISI 4340 alloys and r e s u l t e d in the e x t r a c t i o n of
many fine p a r t i c l e s a s s o c i a t e d with the s m a l l d i m p l e s ,
as i l l u s t r a t e d in F i g . 3 (the a r r o w s indicate e x t r a c t e d
p a r t i c l e s ) . Using s e l e c t e d a r e a e l e c t r o n diffraction,
these s m a l l nucleating p a r t i c l e s were found to have
the c r y s t a l s t r u c t u r e of cementite (Fe3C).
In an attempt to define the f r a c t u r e paths in these
a l l o y s , f r a c t u r e d t e n s i l e halves w e r e nickel plated,
sectioned longitudinally, and polished to the midplane.
R e p r e s e n t a t i v e m i c r o g r a p h s from the AISI 4340 and
18 Ni m a r a g i n g alloys a r e p r e s e n t e d in F i g . 4. The
f r a c t u r e p r o f i l e of an AISI 4340 s p e c i m e n (Fig. 4(a))
shows the f r a c t u r e to be quite i r r e g u l a r and to fol-
low the boundaries of the m a r t e n s i t e laths. This ob-
s e r v a t i o n of the f r a c t u r e path following the m a r t e n -
sitic b o u n d a r i e s is not too s u r p r i s i n g since a p p r o x i -
m a t e l y 90 pct of the f r a c t u r e s u r f a c e a r e a of each
AISI 4340 s p e c i m e n was covered with s m a l l d i m p l e s
nucleated by cementite which p r e c i p i t a t e s p r e f e r e n -
t i a l l y at m a r t e n s i t e lath boundaries. I~ In c o n t r a s t , the
f r a c t u r e p r o f i l e s of the 18 Ni, 200 g r a d e m a r a g i n g
s t e e l s , as shown in Fig. 4(b), exhibit no a p p a r e n t ten-
dency of the f r a c t u r e to follow p a r t i c u l a r m i c r o -
structure features.
The g e n e r a l p r o c e d u r e f o r the m e t a l l o g r a p h i c in-
vestigation was to load a t e n s i l e s p e c i m e n to some
p r e d e t e r m i n e d level of s t r a i n (all r e f e r e n c e to s t r a i n
in this paper should be i n t e r p r e t e d as: e = ln(Rg/R~)),
unload and section the s p e c i m e n longitudinally, ex-
posing the longitudinal midplane n o r m a l to the p r i -
m a r y rolling d i r e c t i o n to m e t a l l o g r a p h i c examination.
It was o b s e r v e d on the longitudinal midplanes of f r a c -
t u r e d t e n s i l e s p e c i m e n s that the p r e s e n c e of l a r g e
voids was not l i m i t e d to the f r a c t u r e path, but r a t h e r
voids extended away from the f r a c t u r e s u r f a c e s in the
d i r e c t i o n of the t e n s i l e axis. To provide m e a s u r e -
ments which offer an a v e r a g e of s t r e s s conditions
found in the neck (which v a r y with position in the
neck) yet a r e p e r t i n e n t to the f r a c t u r e p r o c e s s , an
a r e a c e n t e r e d on the point of i n t e r s e c t i o n of the mini-
mum d i a m e t e r of the neck and longitudinal b i s e c t o r
of the s p e c i m e n was m e t a l l o g r a p h i c a l l y examined.
The d i m e n s i o n s of the a r e a v a r i e d f r o m alloy to alloy.
These dimensions were chosen with r e f e r e n c e to the
f r a c t u r e d t e n s i l e halves so that the longitudinal di-
mension of the a r e a extended to include all voids of
s u b s t a n t i a l size away f r o m the f r a c t u r e s u r f a c e , and
the t r a n s v e r s e dimension of the a r e a was taken as the
width of the a r e a of n o r m a l rupture on the f r a c t u r e
s u r f a c e , i.e., the s h e a r lips w e r e not included. Within
t h e s e a r e a s in the c e n t e r s of the t e n s i l e necks, the
n u m b e r s , s i z e s , and locations of voids w e r e tabulated
f o r v a r i o u s l e v e l s of s t r a i n .

PROCEDURE FOR INVESTIGATING THE VOID INITIATION

PLASTIC FRACTURE MECHANISM
As was i l l u s t r a t e d above, the f r a c t u r e s u r f a c e s of
As d i s c u s s e d above, it was o b s e r v e d that for any the 18 Ni, 200 grade m a r a g i n g s t e e l s a r e composed a l -
one of the four a l l o y s , the f r a c t u r e f e a t u r e s in the r e -
gion of fast f r a c t u r e initiation on Kic s p e c i m e n s were
the s a m e as those f r o m the c e n t r a l regions of n o r m a l
r u p t u r e on smooth and mildly notched, round t e n s i l e
s p e c i m e n f r a c t u r e s u r f a c e s . These o b s e r v a t i o n s lead
one to speculate that the m e c h a n i s m s of p l a s t i c f r a c -
t u r e a r e at l e a s t qualitatively the s a m e for the three
s p e c i m e n g e o m e t r i e s in these p a r t i c u l a r a l l o y s . While
p r e c r a c k e d f r a c t u r e toughness s p e c i m e n s a r e meant
to s i m u l a t e flawed m a t e r i a l in s e r v i c e , they a r e ex-
pensive to produce, and it is g e n e r a l l y quite difficult
to a r r e s t the f r a c t u r e p r o c e s s once it has begun,
e s p e c i a l l y in h i g h - s t r e n g t h m a t e r i a l s . Thus, it was
decided to study the initiation and p r o g r e s s i o n of
p l a s t i c f r a c t u r e in smooth and mildly notched, round
t e n s i l e s p e c i m e n s with the belief, based on the above
f r a c t o g r a p h i c o b s e r v a t i o n s , that the o b s e r v e d p r o c -
e s s e s would be c o m p a r a b l e to those taking place at the
tip of a c r a c k .
The dimensions and geQmetries of the two t e n s i l e
s p e c i m e n types a r e given in Fig. 5 together with t h e i r
orientation in the plate m a t e r i a l s . The p a r t i c u l a r
g e o m e t r y of the mildly notched or p r e - n e c k e d s p e c i -
men was chosen b a s e d on the previous work of Sachs TM
in o r d e r to a s s u r e f r a c t u r e initiation in the c e n t e r of
the s p e c i m e n and to c l o s e l y match the s t r e s s and
s t r a i n s t a t e s found in the smooth, round s p e c i m e n s .
These conditions w e r e d e s i r a b l e in o r d e r to d e t e r -
mine the effects of i n c r e a s e d t e n s i l e t r i a x i a l i t y by
c o m p a r i n g the o b s e r v a t i o n s made with the smooth
s p e c i m e n s with those made with the m i l d l y notched Fig. 3 - - E x t r a c t i o n r e p l i c a f r o m f r a c t u r e s u r f a c e of AISI
specimens. 4340 s t e e l (arrows indicate e x t r a c t e d cementite particles).

METALLURGICAL TRANSACTIONS VOLUME 5, JUNE 1974-1461

most entirely of large dimples. The presence of
cleaved carbo-nitride inclusions lying in the bottoms
of the large dimples together with the good c o r r e l a -
tion between inclusion spacing in the matrix and the
dimple spacing on the fracture surfaces led to the
supposition that the dimples (voids) in the maraging
steels are nucleated by the carbo-nitride inclusions.
The metallographic investigation at 1000X of the se-
ries of strained and sectioned tensile specimens con-
f i r m s this hypothesis. In both high-purity and com-
m e r c i a l 18 Ni maraging alloys and for both smooth
and notched tensile specimens for which straining
(a)
(b)
Fig. 4--Transmission micrographs of replicas from profiles
of smooth tension fractures. (a) AISI 4340 steel; (b) 18 Ni
maraging steel.
1462-VOLUME 5,JUNE 1974
was interrupted before fracture, the cracking of
carbo-nitride inclusions was observed directly as
the exclusive origin of voids in the maraging alloys.
It was observed that the l a r g e r inclusions f r a c -
tured first, i.e., at lower strains, with the process
of cracking continuing as ,the strain increased with
smaller and smaller inclusions beginning to fail.
Although some inclusions were fractured before
straining, it was the general observation that the
cracking of the carbo-nitrides occurred after loading
the specimens into the plastic region and at s t r e s s
levels above 200 ksi (1378 MN/mm 2) which c o r r e l a t e s
well with the results of Floreen and Hayden. 13
The fact that the carbo-nitrides fail by cleavage in-
dicates that they are brittle, which was further con-
firmed by microhardness tests using a Tukon tester
with a 136 deg diamond pyramid indentor and 5 g
load. Many of the inclusions fractured during inden-
tation with several fragments being dislodged during
the shattering of the inclusions. The very brittle na-
ture of the carbo-nitrides provides a basis for an
explanation of the observed fact that the l a r g e r inclu-
sions fail at lower strains (applied stress). As f i r s t
developed by Weibull, 14 the fracture strengths of
brittle solids depend upon the size of the solid being
loaded. Simply put, the argument is that the l a r g e r
the volume of brittle material, the g r e a t e r is the prob-
ability of finding l a r g e r size flaws in the structure
which lead to fracture at lower s t r e s s e s . Although it
is impossible to quantitatively apply the statistical
approach to the fracture strengths of the carbo-
nitride inclusions in the maraging steels for lack of
sufficient data, the Weibull theory does qualitatively
predict that the l a r g e r the inclusion, the lower the
s t r e s s at which it will fail.
In o r d e r to study quantitatively the void initiation
p r o c e s s , the number of inclusions present in the cen-
tral a r e a s of the strained tensile specimens were
counted together with the number of voids present
and the number of inclusions with associated voids,
i.e., cracked inclusions and those associated with
well developed voids. These tabulations permitted
o~;,o. ~ -I'T -I'~ .~ o~,~.
I- I.ooo" =!~ 1.25o" = TM
: ' :.0 '
o) SMOOTH ROUND TENSILE SPECIMEN
Rodius of notch .
R.oo~o. ~,o./ ~,.o2~o.
b) NOTCHED ROUND TENSILE SPECIMEN
ORIENTATION OF ~ Primary rolling direction
TENSILE SPECIMENS / -~ Secondary rolling direction
(Tensile axisl
Thru thickness direction
(1" = I inch 9 2 . 5 4 cm.}
Fig. 5--Shapes and o r i e n t a t i o n of t e n s i l e s p e c i m e n s
METALLURGICAL TRANSACTIONS
calculation of the percent of inclusions which have were nucleated by manganese sulfide inclusions. The
voids associated with them at each level of strain, a results of the metallographic study of strained and
quantity which is a measure of the extent of void sectioned tensile specimens confirms this hypothe-
nucleation. The results for the percent of inclusions sis. It was observed that the large voids are initiated
in the maraging steels which had voids associated almost exclusively by separation of the sulfide-
with them are presented in Fig. 6. matrix interface in both specimen types in the AISI
These plots indicate that void initiation is a con- 4340 alloys.
tinuous p r o c e s s with increasing strain. Note that in
the c o m m e r c i a l grade steel, 19.3 pet of the carbo- F F F
nitrides were fractured before any load was applied. ~100 ~ ~ ] ] ~ " ' " ' [ F
The largest inclusions present were generally the
ones which were fractured in the a s - r e c e i v e d condi-
tion. At any given level of strain, the percentage of
inclusions with voids was l a r g e r for the c o m m e r c i a l
grade alloy, undoubtedly due to the l a r g e r size inclu- o~
50
752!
1
z_
sions in the c o m m e r c i a l ahoy. Also, the percentage
of inclusions with voids at a given strain was much
l a r g e r in the notched specimens than in the smooth, / S o l - - o r commercial purity
/ Open symbolsfor high purity
and the rate of void initiation was much g r e a t e r for
the notched specimens. o. .o'[ I I I I I I I 1
0.10 0.20 0,50 0.40 0.50 0.60 0.70 0.80 0.90 1.00
The differences between the behavior in the smooth TRUE STRAIN
and notched specimens could be due to either the higher Fig. 6 - - P e t of inclusions with voids as a function of true
level of applied s t r e s s in the notched specimens at strain for 18 Ni maraging steel ("F" indicates value taken
from only one-half of a fractured specimen).
any given strain or to the g r e a t e r level of tensile
s t r e s s triaxiality found in the notched specimens. To
illustrate these points, consider the s t r e s s - s t r a i n 550
I I I I I L I
curves for the c o m m e r c i a l purity 18 Ni maraging 9 0 SMOOTH SPECIMENS

steel presented in Fig. 7. The solid curves represent ~ o && NOTCHED SPECIMENS
325
the s t r e s s - s t r a i n relationships determined using =E True stress-strain curve Solid symbols from specimens
E
tched specimens loaded to fracture
single specimens loaded to fracture (solid data points), Open symbols from specimens
while the open data points represent the s t r e s s and Z 30r unloaded before fracture
==
strain conditions of interrupted tests which were sub- r
sequently sectioned for metallographic examination.
27.=
As can be seen in the figure, the data from the inter-
rupted test specimens fall on the same s t r e s s - s t r a i n =< Tr~:r :~eog:lh~
curves, assuring that the specimens all have the same 25C
flow c h a r a c t e r i s t i c s . This same observation was also UJ
made for all other alloys in this investigation. Also
2 2 .=
included in Fig. 7 is the uniaxial flow curve (dashed ~ uJ~nO Bridgrnon Correction
curve) as determined using the Bridgman correction
for .necking. zs At any given strain, the differences be- 20C
L I 1 I I I I [
tween the solid curves and the uniaxial flow curve 0.10 020 0.30 0.40 0.50 0,60 0.70 0.80 0.90
roughly correspond to the levels of triaxial tension TRUE STRAIN
in the specimens. Thus, as may be readily seen in Fig. 7--Tensile curves for commercial purity 18 Ni marag-
Fig. 7, at any given strain the level of applied s t r e s s ing steel.
and the level of triaxial tension are much g r e a t e r in
a notched specimen than in a smooth. 'oo /
By plotting the percentage of inclusions with voids
against the applied tensile s t r e s s , as in Fig. 8, it was ~ 60
found that the data for both smooth and notched speci-
4O
mens fall on the same curve for each alloy. Since the 2O COMMERCIAL PURITY
level of triaxiality varies from smooth to notched
specimen at any level of applied s t r e s s , Fig. 8 dem- m 2;o s;o 5~o 4oo
onstrates that the level of tensile s t r e s s triaxiality 9 SMOOTH SPECIMEN
has no measurable effect on void nucleation in these J 9 NOTCHED SPECIMEN
maraging alloys. Rather, void nucleation is dependent
~, '00 / t j~ t A ";t ""
upon the level of applied tensile s t r e s s . ii
The fracture surfaces of the AISI 4340 alloys were O&
8~I
60 9
shown above to be covered by two populations of
dimples. Good correlation between the spacings of 4~
20
t A
", HIGH PURITY
manganese sulfide inclusions in the matrix and large , I ~
dimple spacings on the fracture surfaces together 250 300 3 0 400
AVERAGE APPLIED TENSILE STRESS (ksi)
with examples of characteristic inclusion impressions (I ksi =6,89 MN/rams)
in the dimples led to the supposition that the large Fig. 8--Pet of inclusions with voids as a function of applied
dimples on the fracture surfaces of the AISI 4340 tensile stress for 18 Ni maraging steel.

METALLURGICAL TRANSACTIONS V O L U M E 5, J U N E 1 9 7 4 - 1 4 6 3
 As was o b s e r v e d f o r the 18 Ni m a r a g i n g s t e e l s , in i n t e r e s t to know exactly how the s m a l l voids a r e nu-
this c a s e also the l a r g e s t inclusions nucleate voids cleated. Although the nature of the i m p r e s s i o n s of
f i r s t , i.e., at l o w e r s t r a i n s . As s t r a i n i n g p r o c e e d e d , the c a r b i d e p a r t i c l e s on the f r a c t u r e s u r f a c e s point
s m a l l e r and s m a l l e r inclusions would then begin to to a mechanism of i n t e r f a c e f a i l u r e , since they ex-
nucleate voids so that the nucleation p r o c e s s was hibit no f r a c t u r e p a t t e r n within t h e m s e l v e s , the point
continuous as s t r a i n i n c r e a s e d with p r o g r e s s i v e l y is not conclusively decided by f r a c t o g r a p h y . In o r d e r
s m a l l e r inclusions acting as nucleating s i t e s for voids. to e s t a b l i s h the exact m e c h a n i s m of void nucleation
No void initiation at the s u l f i d e - m a t r i x i n t e r f a c e was at the c a r b i d e s , a s e r i e s of thin foils were p r e p a r e d
o b s e r v e d unless the m a t r i x had flowed p l a s t i c a l l y . from n e a r the f r a c t u r e s u r f a c e s of broken AISI 4340
Chao and Van Vlack 16'17 have shown that d e f o r m a - s p e c i m e n s . Evidence was found, using the e l e c t r o n
tion is concentrated in manganese sulfide p a r t i c l e s m i c r o s c o p e , thatthe s m a l l voids in the AISI 4340
which a r e embedded in a f e r r o u s m a t r i x during s t r a i n - s t e e l s a r e nucleated by decohesion of the c a r b i d e -
ing. F u r t h e r m o r e , Van Vlack et al. ~8'19 r e p o r t that m a t r i x i n t e r f a c e . F i g . 10 p r e s e n t s two o b s e r v e d ex-
p l a s t i c d e f o r m a t i o n in manganese sulfide o c c u r s by a m p l e s of voids f o r m e d by s e p a r a t i o n of the c a r b i d e -
p l a n a r slip which r e s u l t s in s t r a i g h t slip steps on m a t r i x i n t e r f a c e . The voids a r e indicated by the
suitably oriented f r e e s u r f a c e s as confirmed in this a r r o w s , and the c a r b i d e s by the l e t t e r " C " . Selected
investigation by m i c r o h a r d n e s s indenting numerous a r e a e l e c t r o n diffraction confirmed these p a r t i c l e s
sulfides in the AISI 4340 a l l o y s . These o b s e r v a t i o n s to be cemenfite.
provide a p o s s i b l e explanation for the o b s e r v e d effects
of inclusion s i z e on nucleation of the l a r g e voids in
VOID GROWTH
the AISI 4340 a l l o y s . As the m a t e r i a l is loaded, m o r e
p l a s t i c deformation o c c u r s in the manganese sulfide As the s t r e s s e s a r e i n c r e a s e d beyond the l e v e l s at
inclusions than in the m a t r i x . P l a n a r slip p r o v i d e s which void nucleation f i r s t o c c u r s in these a l l o y s , the
r a t h e r long, n a r r o w slip bands within the inclusions voids which have been initiated at n o n - m e t a l l i c in-
which a r e blocked at the i n c l u s i o n - m a t r i x i n t e r f a c e , clusions grow by some m e c h a n i s m involving p l a s t i c
r e s u l t i n g in s u b s t a n t i a l s t r e s s concentration a c r o s s flow of the m a t r i x . Examination of the m i d p l a n e s of
the boundary and r e s u l t a n t f r a c t u r e t h e r e . Z e n e r 2~ many s t r a i n e d and sectioned t e n s i l e s p e c i m e n s r e -
has shown that the local s t r e s s at the tip of a blocked
slip band is p r o p o r t i o n a l to the square root of the I00 F F
length of the band. The a v e r a g e length of slip band is I I I eOSmoofh Sper . . . .
obviously r e l a t e d to the size of the manganese sulfide.
~=L~ ~ ,~ Solid symbols represent _
This t r e a t m e n t p r e d i c t s that the local s t r e s s at the "~-t f / commercio,purity -
boundary i n c r e a s e s with the size of the sulfide; and ~I /" / Open symbols represent
thus the boundaries at the l a r g e r inclusions should i Ill I
fail f i r s t , as has been o b s e r v e d .
The s t r a i n e d and sectioned s p e c i m e n s of AISI 4340 0~ / /
w e r e examined m e t a l l o g r a p h i c a l l y at 1000X, and the
p e r c e n t a g e of inclusions with voids in the c e n t r a l r e -
gions of the s p e c i m e n s w e r e r e c o r d e d at v a r i o u s
s t r a i n s . The r e s u l t s a r e plotted in Fig. 9. The c u r v e s
,i 0,10 0,20 030 0.40 0.50 0.60
TRUE STRAIN
indicate that the initiation p r o c e s s is continuous as Fig. 9 - - P c t of i n c l u s i o n s with v o i d s a s a f u n c t i o n of t r u e
s t r a i n i n g p r o c e e d s until the s p e c i m e n f r a c t u r e s o r a l l s t r a i n f o r AISI 4340 s t e e l ( " F " i n d i c a t e s v a l u e t a k e n f r o m
of the inclusions have nucleated voids. Note that, by only one-half a fractured specimen).
c o m p a r i s o n with F i g . 6, the s t r a i n s to f r a c t u r e for
both a r e much lower for the smooth and notched ten-
sile s p e c i m e n s than for both m a r a g i n g s t e e l s . The
data show that in the c o m m e r c i a l p u r i t y alloy, voids
a r e initiated at lower s t r a i n s than in the h i g h - p u r i t y ,
and the s l o p e s of the plots in Fig. 9 indicate that the
r a t e of void initiation is higher for the c o m m e r c i a l
purity AISI 4340. Both of these o b s e r v a t i o n s a r e un-
doubtedly the r e s u l t of the l a r g e r s i z e s of inclusions
found in the c o m m e r c i a l s t e e l . The p e r c e n t a g e of
voids at a given s t r a i n and the r a t e s of void nucleation
for the notched s p e c i m e n s a r e much g r e a t e r than for
the smooth. This fact is again due to the higher
s t r e s s e s in the notched s p e c i m e n s at a given s t r a i n .
It was found that the data for both smooth and notched
s p e c i m e n s fell on the s a m e curve when plotting the
p e r c e n t a g e of inclusions with voids against the ap-
plied tensile s t r e s s indicating, as was o b s e r v e d for
the m a r a g i n g s t e e l s , that t h e r e i s no effect of s t r e s s
t r i a x i a l i t y on void initiation.
While it was p r e v i o u s l y e s t a b l i s h e d that the s m a l l Fig. 1 0 - - V o i d i n i t i a t i o n b y d e c o h e s i o n of c a r b i d e - m a t r i x i n -
d i m p l e s on the f r a c t u r e s u r f a c e s of the AISI 4340 t e r f a c e in AISI 4340 s t e e l s t r a i n e d p [ a s t i c a U y ( a r r o w s i n d i -
s t e e l s a r e nucleated by c e m e n t i t e p a r t i c l e s , it is of c a t e v o i d s and " C " i n d i c a t e s c a r b i d e ) .

1 4 6 4 - V O L U M E 5, JUNE 1974 METALLURGICAL TRANSACTIONS

vealed that there is a progressive increase in both the
number and size of voids in the necked regions of
both smooth and notched specimens of both alloys as
the strain is increased. The increase in the number
of voids as the strain increases is the result of con-
tinuing void initiation at smaller and s m a l l e r inclu-
sions, while the increasing size of voids is the result
of progressive void growth. These two simultaneously
occurring p r o c e s s e s result in a wide spread in the
sizes of voids present on the longitudinal midplanes.
(a)
(c)
METALLURGICAL TRANSACTIONS
It was observed for both alloys that the l a r g e r
voids were located in the centers of the specimens,
with the sizes of voids generally decreasing as one
moves away from the center line. The general shapes
of the voids also change from edge to center of the
midplane at high strains. In the centers of the smooth
specimens and more generally in the notched speci-
mens, the voids show a tendency to grow in both the
direction of the tensile axis and in the radial direc-
tion (perpendicular to the tensile axis). Examples of
(b)
I Tensile
Axis
Fig. l l - - V o i d s in the c e n t e r s
of t e n s i l e s p e c i m e n s of 18 Ni
maraging steels strained
various amounts plastically.
(a) ~ = 0.18; (b) ~ = 0.58;
(c) e = 0.66.
VOLUME 5, JUNE 1974-1465
voids in the central portions of the tensile specimens reasoning behind this choice is that the largest void
which have grown during straining are presented in present would naturally be the one which had nucleated
Fig. 11. (The examples in Fig. 11 are from speci- first (at the largest inclusion) and had grown continu-
mens of 18 Ni maraging steel, but these same obser- ously ~s the strain had increased, and thus represents
vations were also made for the AISI 4340.) the best available information regarding void growth.
It should be noted that at small strains (Fig. ll(a)), The areas of the largest voids were used to calculate
the voids in the center of the specimen showed little the radii of circles of equivalent a r e a s , values of
sideways growth, rather they tend to grow preferen- which are plotted against strain for all the alloys in
tially in the direction of the applied tensile s t r e s s . Fig. 13. Only values observed before void coales-
However, as the strain increases (Fig. il(b) and cence began are plotted. The values observed for the
11(c)), the voids in the central region of the neck maraging steels agree reasonably well with those
grow both longitudinally and laterally. It was also observed by Floreen and Hayden for an 18 Ni, 300
observed that the voids which are nucleated outside grade maraging alloy. 22
the necked region, i.e., removed from the minimum The data in Fig. 13 indicate that for any given alloy
specimen diameter in the longitudinal direction, the rate of void growth is g r e a t e r in the c o m m e r c i a l -
tend to show little lateral growth even at high strains. purity material than in the high-purity material. This
Likewise, voids toward the edges of the specimens fact is due to the differences in inclusion size when
also showed little tendency for appreciable growth in going from c o m m e r c i a l - p u r i t y to high-purity alloys.
the radial direction even at high strains. These As may be seen in Table HI, the c e n t e r - t o - c e n t e r
voids lying away from the center of the neck tend to distances between inclusions (I) are l a r g e r in the
be elongated in the direction of the tensile axis and
are generally no wider than the nucleating particle.
The l a r g e r degree of triaxial s t r e s s in the center E
of the neck which develops at high strains aids the z 8(: i i i i
lateral growth of the voids. _o
AISI 4340 STEEL
Using a standard point counting technique, 21 it was
6(:-
possible to determine the area fractions of the longi-
tudinal midplanes covered by voids. Using this in-
formation together with the number of voids present
on the test section permitted calculation of the aver- ~ zc
age c r o s s - s e c t i o n a l a r e a of void intersected by the
midplane. This particular measure of void size was
r C(~ 010 020 030 040 0.50
chosen since it was observed that the voids are gen-
< 9 SMOOTH S P E C I M E N S & A NOTCHED SPECIMENS
erally quite i r r e g u l a r in shape such that the measure- ,~ Solid symbols represent commercial purity
ment of some linear dimension of the voids, e.g., z Open symbols represent high purity
0 120, ~ I I I
length or width, has questionable utility. /,8Ni MARAGING P
The values of the average c r o s s - s e c t i o n a l areas of r STEEL 4
voids intersected by the midplanes as a function of
=% / /
strain are presented in Fig. 12. Each of the curves
40 9
in the figure shows a linear initial portion where the .

average size of void increases continuously with

strain. At the higher levels of strain, the curves 0.'~ Ao ,.oo
TRUE STRAIN
(especially those for the smooth specimens) show a Fig. 12--Average cross s e c t i o n a l a r e a of v o i d a s a f u n c t i o n
rapid increase in the size of voids present. This of strain.
rapid jump corresponds with the onset of void coales-
cence and the formation of a large central cavity in
the tensile specimen. It should be noted that the t i ~ i i
i2[ ~ ~ ~ ~
rates of void growth are generally higher for the
commercial grade alloys and noticeably higher for
the notched specimens.
Actually, the average c r o s s - s e c t i o n a l area of void _o
represents the combined effects of void nucleation, 0>
growth, and coalescence such that the plots in Fig. 12 I I I / I
O. 0 0.40 0.60 0.20 0.40 0.60
represent some measure of the p r o g r e s s of the total Commercial moroging steel High purity moraging steal
fracture process. These results do indicate that at .J
i i i i i
any given strain, the most detrimental situation is 12
found in the notched c o m m e r c i a l grade AISI 4340
steel while the least damaging situation is found in
the smooth high-purity grade 18 Hi maraging steel.
~
The general o r d e r of the curves follows the observed 4
values of fracture strain and fracture toughness in 9 Smooth specimen
9 Notched specimen
these alloys, C t t I I t O~, I I I ~ I
0.20 0.40 0.60 0.20 O. 0 0.60
In o r d e r to separate the effects of void nucleation Commercial A I S 1 4 3 4 0 s t e e l High purity A I S 1 4 3 4 0 steel
and growth, the c r o s s - s e c t i o n a l a r e a of the largest S T RA I N
void present on the midplane was determined. The F i g . 1 3 - - R a d i i of l a r g e s t v o i d s a s a f u n c t i o n of s t r a i n .

1 4 6 6 - V O L U M E 5, JUNE 1974 METALLURGICAL TRANSACTIONS

c o m m e r c i a l - p u r i t y alloys. However, if the differences of a right p r i s m of e l l i p t i c a l c r o s s - s e c t i o n . Using the
in the s i z e s of the inclusions between h i g h - p u r i t y and e x p r e s s i o n developed by McClintock and the values
c o m m e r c i a l purity alloys a r e c o n s i d e r e d , it is seen of the v a r i o u s m a t e r i a l s constants f o r the alloys being
that the b o u n d a r y - t o - b o u n d a r y d i s t a n c e s between investigated h e r e , the o b s e r v e d void s i z e s were com-
n e a r e s t neighbor inclusions a r e roughly the same for p a r e d with those p r e d i c t e d . In all c a s e s the o b s e r v e d
both purity l e v e l s in a given s t e e l . Once voids nu- values were at l e a s t twice those p r e d i c t e d by the
c l e a t e at neighboring inclusions, they have a p p r o x i - McClintock a n a l y s i s . The poor a g r e e m e n t probably
m a t e l y the s a m e amount of growth in the m a t r i x r e - r e s u l t s from the fact that the McClintock a n a l y s i s
quired before impingement in both c o m m e r c i a l and a s s u m e s no i n t e r a c t i o n between voids, i.e., it should
h i g h - p u r i t y a l l o y s . Thus, the p r i m a r y difference b e - be valid only f o r the v e r y e a r l y s t a g e s of void growth
tween the alloys r e g a r d i n g growth of voids s t e m s from and for s y s t e m s with much l a r g e r spacings between
the fact that the voids nucleate e a r l i e r in the c o m m e r - inclusions than those encountered in these s t r u c t u r a l
cial s t e e l s since the inclusions a r e l a r g e r ; and be- alloys.
cause the inclusions a r e l a r g e r , the r e s u l t i n g voids
a r e born l a r g e r which produces i n c r e a s e d l e v e l s of
VOID COALESCENC E
applied s t r e s s due to the reduced c r o s s - s e c t i o n a l
a r e a and i n t e r a c t i o n of the s t r e s s fields around the As was suggested by examination of the f r a c t u r e
voids. With the i n c r e a s e d l e v e l s of local s t r e s s in the s u r f a c e s of the 18 Ni m a r a g i n g s t e e l s which were
c o m m e r c i a l a l l o y s , the voids grow m o r e rapidly. c o v e r e d with l a r g e d i m p l e s c e n t e r e d on f r a c t u r e d
These findings r e g a r d i n g the d e t r i m e n t a l effects of c a r b o - n i t r i d e i n c l u s i o n s , the m e t a l l o g r a p h i c e x a m i n a -
inclusion size a r e supported by the r e c e n t work of tion of s t r a i n e d and sectioned s p e c i m e n s of the m a r a g -
Darlington 23 who studied f r a c t u r e in e l e c t r o l y t i c iron ing alloys d e m o n s t r a t e d that the i n c l u s i o n - n u c l e a t e d
and s p h e r o i d i z e d low carbon s t e e l . voids c o a l e s c e by growing until they impinge on one
The r e s u l t s p r e s e n t e d in Fig. 13 also d e m o n s t r a t e another. The f i r s t appearance of the c o a l e s c e n c e of
that the i n c r e a s e d level of t r i a x i a l s t r e s s introduced voids in both the high-purity and c o m m e r c i a l m a r a g -
in the notched s p e c i m e n s g r e a t l y i n c r e a s e s the r a t e ing s t e e l s was o b s e r v e d to occur at s t r a i n s of a p p r o x -
of void growth. At a given level of s t r a i n , the uni- i m a t e l y 0.8 of the f r a c t u r e s t r a i n s f o r the smooth
axial flow s t r e s s (which is twice the maximum r e - s p e c i m e n s and between 0.4 and 0.6 of the f r a c t u r e
solved s h e a r s t r e s s ) is the same in both smooth and s t r a i n s for the p r e - n o t c h e d s p e c i m e n s . An example
notched s p e c i m e n s of a given alloy, as shown in Fig. of voids c o a l e s c i n g in the m a r a g i n g s t e e l is p r e s e n t e d
7, yet the s i z e s of the voids a r e much l a r g e r in the in Fig. 14. As s t r a i n i n g continues, the void c o a l e s -
notched s p e c i m e n s and the r a t e of void growth is cence p r o c e s s leads to the f o r m a t i o n of a l a r g e cen-
g r e a t e r . This difference in void growth must then
be the r e s u l t of the difference in the level of t r i -
axiality.
I n c r e a s e d levels of t r i a x i a l i t y i n c r e a s e the longi-
tudinal applied s t r e s s e s r e q u i r e d to cause g e n e r a l
p l a s t i c flow in the m a t r i x . R e s t r i c t e d flow in the m a -
t r i x will tend to concentrate the p l a s t i c deformation
at the existing voids where the s t r e s s e s a r e highest
and c o n s t r a i n t s reduced, since the s t r e s s e s n o r m a l
to the f r e e s u r f a c e s of the voids go to z e r o . Thus,
the maximum s h e a r s t r e s s i n c r e a s e s n e a r the s u r f a c e
of a void (since one of the p r i n c i p a l n o r m a l s t r e s s e s
is zero) r e s u l t i n g in i n c r e a s e d p l a s t i c deformation.
Hence, as the level of t r i a x i a l i t y i n c r e a s e s , the higher
applied longitudinal s t r e s s e s n e c e s s a r y to continue
p l a s t i c flow r e s u l t in intense deformation around the
voids r e s u l t i n g in enhanced r a t e s of void growth.
T h e r e have been s e v e r a l attempts to d e s c r i b e the
void growth p r o c e s s in the past. Many of the t r e a t -
ments involved e x p r e s s i o n s for p r e d i c t i n g void s i z e
f r o m the s t r a i n . However, for s e v e r a l of the
c a s e s , 24'25 it is difficult to c o m p a r e the r e s u l t s of
the c u r r e n t investigation with those p r e d i c t e d be-
cause the t r e a t m e n t s contain v a r i o u s a r b i t r a r y con-
stants which a r e used to make the data fit models.
On the other hand, the work of Rice and T r a c e y 26 in-
volves complex e x p r e s s i o n s of s t r a i n r a t e which a r e
not r e a d i l y i n t e g r a t e d for c o m p a r i s o n with the c u r -
rent data. The a n a l y s e s to which the data from this
investigation may be most r e a d i l y c o m p a r e d a r e those
of McClintock, 26'27 which a r e based on continuum p l a s -
t i c i t y and involve no a r b i t r a r y constants. McClintock
c o n s i d e r s an i s o l a t e d hole in a m a t r i x with the shape Fig. 1 4 - - V o i d c o a l e s c e n c e in 18 Ni m a r a g i n g s t e e l , e = 0.81.

METALLURGICAL TRANSACTIONS VOLUME 5, JUNE 1 9 7 4 - 1 4 6 7

tral cavity. Once a cavity as long as 5 to 10 pet of
the minimum specimen diameter was formed, final
fracture occurred very rapidly.
This same p r o c e s s of void coalescence was ob-
served in the notched specimens. It was generally ob-
served that the strains at which the coalescence of
voids was first detected were smaller in the notched
specimens, approximately 20 pct of those observed
in the smooth specimens, which results in the fact
that the fracture strains of the notched specimens
are substantially less than those of the smooth speci-
mens. This early development of coalescence obvi-
ously follows from the higher void growth rates ob-
served in the notched specimens. F u r t h e r m o r e , the
duration of the coalescence process, i.e., the differ-
ence between the strain at which coalescence is first
detected and the fracture strain, was found to be much
shorter in the notched specimens, of the o r d e r of half
the values noted for the smooth specimens. The fore-
shortened coalescence stage of the voids in the notched
specimens must also be a direct result of more rapid
growth under the influence of the higher triaxial
stresses.
Thus far, it has been seen that at least qualitatively
the process of void initiation and growth from non-
metallic inclusions in the two alloy steel families
(maraging and quenched and tempered) are not essen-
tially different. However, the process of the coales-
cence of inclusion-nucleated voids in the AISI 4340
alloys differs dramatically from that in the maraging
steels. One would expect as much after considering
the fracture surfaces of the quenched and tempered
steels which are covered by two distinct populations
of dimples. Based on observations made from the
strained and sectioned tensile specimens, it is found
that the large voids nucleated at manganese sulfide
inclusions coalesce by the propagation of crack-like
features which link adjacent large voids. The first
appearance of this coalescence occurs at approxi-
mately 0.6 to 0.8 of the fracture strain for the smooth
Fig. 1 5 - - V o i d c o a l e s c e n c e in AISI 4340 s t e e l , ~ = 0.25.
1468-VOLUME 5, JUNE 1974
specimens and approximately 0.3 to 0.4 of the fracture
strain for the notched specimens. A photomicrograph
of the coalescence of inclusion-nucleated voids in AISI
4340 is presented in Fig. 15. At strains within a few
percent of the fracture strain, the coalescence of
voids in the central region of the tensile specimen
is extensive enough to form a rather large continuous
cavity spanning as much as 5 or 10 pet of the speci-
men diameter. Final fracture occurs catastrophically
after this point.
The crack-like features which link the large inclu-
sion-nucleated voids in the AISI 4340 alloys initially
have the appearance of a collection of small voids
similar to the "void s h e e t s " described by Rogers 29
which he observed during the plastic fracture of
copper. This fact is readily seen in the example pre-
sented in Fig. 16(a). As the strain is increased, the
small voids coalesce and the "void sheet" becomes
an actual crack of substantial opening, running be-
tween the large voids, as demonstrated in Fig. 16(b).
Serial sectioning of several of these "void s h e e t s "
conclusively demonstrated that they are planar fea-
tures, generally oriented at approximately 45 deg
with the tensile axis, composed of many small voids.
The metallographic observations lead to the con-
clusion that the voids nucleated by the manganese sul-
fides grow to some critical size or until a critical
spacing between them is reached. At this point, fur-
ther deformation seems to be concentrated in the nar-
row bands between the neighboring large voids. As
straining continues, the bands of intense deformation
between the voids become more distinct and the small
voids formed at the carbide particles in these bands
begin to coalesce, resulting in the formation of a
"void sheet" and subsequent crack connecting neigh-
boring large voids.
If it is assumed that the nucleation of small voids at
the carbide precipitates could be avoided without af-
fecting the rates of growth of the inclusion-nucleated
voids, as determined from Fig. 13, it is possible to
ascertain whether or not improvement in the strains
to fracture could be realized. Using half the average
inter-inclusion spacing as the approximate radius of
the large voids at fracture if void sheet formation
were avoided and the slopes (growth rates) from Fig.
13, the predicted fracture strains for the AISI 4340
alloys are considerably higher than those observed
in practice. F o r example, the observed fracture
strain for smooth specimens of the high-purity alloy
was 0.52 and the predicted fracture strain, assuming
no void sheet formation is 0.90. Although this treat-
ment is highly problematic since changing the carbide
distribution to avoid the nucleation of small voids will
undoubtedly change the flow c h a r a c t e r i s t i c s of the
matrix, it does indicate that improvement in the re-
sistance to plastic fracture can follow from r e s t r i c t -
ing the formation of void sheets which result in the
premature coalescence of the inclusion-nucleated
voids.
This premature coalescence in the AISI 4340 alloys
also explains the observed differences in toughness be-
tween the AISI 4340 alloys and the 18 Ni, 200 grade
maraging steels. The particles of intermetallic com-
pound which strengthen the maraging steels are one
to two o r d e r s of magnitude smaller than the carbides
METALLURGICAL TRANSACTIONS
 (a)
(b)
in the quenched and tempered steels. Roesch and
Henry 3~ have demonstrated that these strengthening
precipitates in the maraging alloys do nucleate voids
when they are coarsened and this results in a concur-
rent drop in toughness as measured by impact ener-
gies. These observations suggest that it may be pos-
sible to prevent or delay the nucleation of voids at
the cementite particles by suitably refining the c a r -
bides in the quenched and tempered structure, thereby
improving the fracture toughness.
SUMMARY OF RESULTS
The microstructural aspects of the plastic fracture
processes in AISI 4340 and 18 Ni, 200 grade marag-
ing steels have been defined by studying strained and
sectioned, smooth and notched, tensile specimens
together with the fracture surfaces of tensile and
fracture toughness specimens. The interruption of the
growth of the inclusion-nucleated voids in the quenched
and tempered AISI 4340 steels by the formation of
void sheets is the p r i m a r y reason that the quenched
and tempered steels are generally less tough than
the maraging steels at the same strength level. It
METALLURGICALTRANSACTIONS
I Tensile Axis
Fig. 16--Coalescence of voids by void sheet formation in
AISI 4340 steel strained plastically. (a) r = 0.25; (b) 9 = 0.27,
was observed that void nucleation occurs at lower
strains at the large inclusions and that void growth
takes place more rapidly from the l a r g e r inclusions,
thus reducing the resistance of the alloy to fracture.
By comparing the metaUographic results from smooth
and notched tensile specimens, it was demonstrated
that void initiation is independent of the level of ten-
sile s t r e s s triaxiality, but that void growth rates are
increased substantially by increasing the level of tri-
axial tension. Increased growth rates result in more
rapid coalescence of voids and premature fracture,
explaining the mechanism by which notches and flaws
in high strength structural materials lead to brittle
f r a c t u r e s even when the microscopic fracture mode
is dimpled rupture (plastic fracture).
The investigation indicates that the most important
m i c r o s t r u c t u r a l features governing the plastic f r a c -
ture of these alloys are the void nucleating, second-
phase particles. The toughness of these alloys may
be most readily improved by decreasing the number
and size of non-metallic inclusions. Although recently
reported work by other investigators s~-3s indicates
that at high strength levels (250 to 320 ksi) fracture
toughness is insensitive to steel cleanliness, the f r a c -
VOLUME 5, JUNE 1974-1469
ture mode exhibited at these high strength levels in
severely notched Kic specimens is mixed, consisting
of plastic fracture and cleavage 31 and thus the results
reported herein are not directly applicable.
Thus far, the approach taken to improve the f r a c -
ture toughness of the high strength steels has been to
use special melting practices to reduce the levels of
impurity elements. The practices employed to pro-
duce the higher purity levels are generally quite ex-
pensive and add substantially to the cost of the final
product. The current investigation suggests that in-
clusion size is an important factor in the fracture
p r o c e s s ; and that if the average size of inclusions
were reduced at any given level of cleanliness, im-
provements in toughness could be realized. As ob-
served for the manganese sulfide particles, these
particles frequently nucleate on deoxidation products
present in the melt. This observation suggests that
it might be possible to suitably seed or otherwise in-
c r e a s e the nucleation of inclusions for a given im-
purity level of a melt in o r d e r to produce a finer dis-
persion of non-metallic inclusions in the final product.
Thus, it might be possible to attain adequate tough-
ness with higher impurity levels provided the sizes of
inclusions were reduced.
As the quenched and tempered steels are generally
of lower alloy content than the maraging steels and
hence more economically produced, it would be bene-
ficial to investigate means of improving their tough-
ness in relation to the maraging steels. As pointed
out in the current investigation, the major difference
in the fracture behavior of these two different types
of steels is the void sheet p r o c e s s of coalescence of
inclusion-nucleated voids observed in the AISI 4340
alloys. The voids of these sheets are nucleated by
the coarse eementite particles which precipitate at
the martensite lath boundaries. It is felt that by re-
fining the size of the carbides and dispersing them
more generally throughout the matrix, void initiation
at the carbides could be delayed or avoided with a re-
sultant improvement in toughness. Both t h e r m o m e -
ehanical treatments 34 and repetitive austenitizing
treatments 3s have the ability to refine the carbide nu-
cleation during tempering. An investigation of the ef-
fects of these treatments on the structure and fracture
behavior of AISI 4340 would be helpful for our under-
standing of the plastic fracture p r o c e s s in the quenched
and tempered steels.
ACKNOWLEDGMENTS
The findings presented herein were part of a dis-
sertation submitted by T. B. Cox in partial fulfillment
1470-VOLUME 5, JUNE 1974
of the requirements for the degree of Doctor of Philos-
ophy at the Carnegie-Mellon University. This r e s e a r c h
was sponsored by the National Aeronautics and Space
Administration under Research Grant NGR 39-087-
003.
REFERENC ES
1. J. R. Low, Jr.: Prog. Mater. Sci., 1963, vol. 12, no. 1, pI). 106.
2. C. P. Sullivan: Welding Research CouncilBulletin No. 122, 1967.
3. B. I. Edelson and W. M. Baldwin, Jr.: Trans. ASM, 1962, vol. 55, pp. 230-50.
4. A. G. Franklin and W. J. McG. Tegart: J. Iron Steellnst., 1964, vol. 202, pp.
588-92.
5. A. J. Birkle, R. P. Wei, and G. E. Pellissier: Trans. ASM, 1966, vol. 59, pp. 981-
90.
6. R. Kiesslingand N. Lange: Non-Metallic Inclusions in Steel Part 11, p. 97,
Iron and Steel Institute, London, 1966.
7. T. Boniszewski and E. Boniszewski: J. Iron Steellnst., 1966, vol. 204, pp. 360-
65.
8. R. T. DeHoff and F. N. Rhines: Quantitative Microscopy, p. 128, McGraw-
Hill, New York, 1968.
9. R. T. DeHoff and F. N. Rhines: Trans. TMS-AIME, 1961, vol. 221, pp. 975-82.
10. F. C. Hull and W. J. Houk: Trans. AIME, 1953, vol. 197, pp. 565-72.
11. A. J. Baker, F. J. Lauta, and R. P. Wei: ASTMSTPNo. 370, 1965, pp. 3-28.
12. A. W. Dana, E. L. Aul, and G. Sachs: NACA Tech. NoteNo. 1830, Cleveland,
Ohio, 1949.
13. S. Floreen and H. W. Hay&n: Met. Sci. Z, 1970, vol. 4, pp. 77-80.
14. W. WeibuU: lgeniors Vetenskaps Akadamien, Hondlinger No. 151, Stockholm,
1939.
15. P. W..Bridgman: Large Plastic Flow and Fracture, p. 9, McGraw-Hill, New York,
1952.
16. H. C. Chao and L. H. Van Vlack: Trans. TMS-AIME, 1965, vol. 233, pp. 1227-
31.
17. H. C. Chao and L. H. Van Vlack: Mat. Res. Stan., 1965, vol. 5, pp. 611-13.
18. L. E. Wood and L. H. Van Vlaek: Trans. ASM, 1963, vol. 56, pp. 770-72.
19. H. C. Chao, L. Thomassen, and L. H. Van Vlack: Trans. ASM, 1964, vol. 57,
pp. 386-98.
20. C. Zener: E(asticity and Anelasticity, p. 133, University of Chicago Press,
Chicago, 1948.
21. E. E. Underwood: Quantitative Stereology, p. 29, Addison-Wesley,Reading,
Mass., 1970.
22. S. Floreen and H. W. Hayden: ScriptaMet., 1970, vol. 4, pp. 87-94.
23. H. Darlington: Ductile b~acture Under Axisymmetric Stresses in Electrolytic
Iron and Spheroidized Low-Carbon Steel, Ph.D. Thesis, Lehigh University,
1971.
24. J. Gurland and J. Plateau: Trans. ASM, 1963, vol. 56, pp. 442-54.
25. P. Feltham and A. S. Beyron: Phil. Mag., 1966, vol. 13, pp. 311-16.
26. J. R. Rice and D. M. Tracey: J. Mech. Phys. Solids, 1969, vol. 17, pp. 201-17.
27. F. A. McClintock: Ductility, p. 255, ASM, Metals Park, Ohio, 1968.
28. F. A. McClintock: J. Appl. Mech., 1968, vol. 35, pp. 363-71.
29. H. C. Rogers: Trans. TMS-AIME, 1960, vol. 218, pp. 498-506.
30. L. Roesch and G. Henry: ASTMSTP453, 1969, pp. 3-35.
31. J. J. Hauser and M. G. H. Wells: Air Force Mat. Lab. Tech. Rep. AFML-TR-
69-339, 1970.
32. J. J. Hauser, M. G. H. Wells, and I. Perlmutter: J. Vac. Sci. Tech., 1972, vol. 9,
p. 1339.
33. L. F. Porter: Z Vac. Sci. Tech., 1972, vol. 9, pp. 1340-44.
34. M. J. May and D. J. Latham: Toward 1reproved Ductility and Toughness, p.
157, Climax Molybdenum Development Company, Japan, 1971.
35. R. A. Grange: Trans. ASM, 1966, vol. 59, pp. 26-48.
36. D. Brock: Eng. Fract. Mech., 1970, vol. 1, pp. 691-95.
METALLURGICAL TRANSACTIONS