-

Available online at www.sciencedirect.com

__
ScienceDirect
JOURNAL OF IRON AND STEEL RESEARCH, INTERNATIONAL. 2007, 14(6): 63-67

Effects of Applied Stresses on Martensite Transformation in AISI4340 Steel

ZHAO Hong-zhuang"*, Seok-jae LEE', Young-kook LEE', LIU Xiang-hua' , WANG Guedong'
(1. State Key Laboratory of Rolling and Automation, Northeastern University, Shenyang 110004, Liaoning, China;
2. Department of Metallurgical Engineering, Yonsei University, Seoul 120-749, South Korea)

Abstract: This study aims at the experimental analysis of the transformation induced plasticity (TRIP) phenomenon.
Experiments are conducted in which martensite is allowed to grow under the influence of a series of externally applied
stresses. The magnitude of the applied stresses is less than 67 .% of the yield strength of austenite uy ( T ,). Since
there is no obvious difference between the transformation plasticity under tension and the compression for the lower
applied stresses, only compressive stresses are applied. The results confirm that the transformation plasticity is pro-
portional to the applied stress if the latter does not exceed 67% of a y ( T , > .The TRIP-strain, the kinetics, and their
dependence on the applied stresses are studied. The comparison between calculated results and experimental results
shows that the model accurately describes the phenomenon.
Key words: applied stress; TRIP; martensite transformation; low-alloy steel

Transformation induced plasticity ( T R I P ) can
be defined as the anomalous plastic strain that oc-
curs when material transformation is affected by an
external stress lower than the yield stress of the par-
ent phase (weak phase). T R I P strain consists of not
only a plastic contribution (" Greenwood-Johnson"
effect), but also a contribution owing to the strain
caused by the transformation shear component of the
martensite ( o r bainite) variants ("Magee" effect).
Both contributions have been
T h e briefest equation of describing the Green-
wood-Johnson mechanism result for a low, con-
stant, and uniaxial stress is as followsL51
etP=K p(e) CT - (1)
where K is a material parameter ( T R I P coefficient),
and 9 ( € 1 is a normalized function governing the
TRIP kinetics. Different forms of this function have
been proposed to describe the TRIP phenome-
n ~ n ~ In~ fact,
. ~ ~Eqn.
. ( 1 ) can be used to explain
Magee effect, and parameter K may include the two
effects mentioned aboveL3].
T h e purpose of the present study is to study the
effects of externally applied stresses on martensite
transformation. The relationship between the stress
and the TRIP effects will be shown. Though the
maximum value of applied stress 6 is only more than
half of the yield strength of austenite 6, ( T , ) , it is
shown evidently that M sis affected slightly because
of 6.
1 Experimental and Calculation Methods
1. 1 Test material
The composition of t h e experimental steel AI-
SI4340 (mass percent, %> is: C 0. 39, M n 0. 65, Ni
1 . 6 0 , Cr 0. 67, Mo 0.15, A1 0. 03, N 0.010, and
Fe balance. Cylindrical specimens (10 mm in diame-
ter, 15 mm in length) were machined for the pur-
pose of the dilatometric experiments. In all experi-
ments, the specimens were austenitized a t 900 "C for
10 min.
1.2 Thermomechanical testing
The thermomechanical tests were carried out in
a Gleeble 3500 thermomechanical simulator, which
typically has a high speed heating system, a servo
hydraulic system, and a computer control and data
acquisition system. The thermal cycle is shown in
Fig. 1. The sample was heated in a high vacuum cham-
ber. T h e loading mode is compression. T h e loading
stress levels were 0 , 2 0 , 4 0 , 6 0 , and 80 MPa; the

Foundation Item: Item Sponsored by Hi-Tech Research and Development Program of China (2001AA332020)
Biography:ZHAO Hong-zhuang(l966-), Male, Doctor, Lectureship; E-mail: zhzl966@sohu. corn; Revised Date: November 24, 2006
 64 * Journal of Iron and Steel Research, International
(a) Thermal
-
7
t (b) Stress holding test
1
2 - Fittedline
b
a=O,20,40,60,80 MPa
I Fig. 2
0 200 400 600 800
Time/s
Fig. 1 Thermomeehanical loading path
loading temperature was 320 "C,and the loading
speed was 80 MPa s-'. T h e heating and cooling
rates were 10 "C s-' and 11 "C s - l , respectively.
Thermocouple was attached centrally to the
specimen's surface. Uniaxial loads were applied by
using the servo-hydraulic system which allowed ei-
ther load, displacement, or stress control of the ram
position. The specimen temperature, the uniaxial
load, and both diametral and longitudinal displace-
ments were measured concurrently.
1 . 3 Austenite yield strength
Since the value of the austenite yield strength is
difficult to measure in the range of martensite trans-
formation, its evaluation method is used. T h e yield
strength is expressed using the empirical model by
Young and BhadeshiaC7'as follows :
U: =(1-0. 26X10-'Tr+0. 4 7 X 1 0 P 5 T - 0 . 3 2 6 X
lO-'T?) X 15.4[4.4+23w(C)+l. 3w(Si)+
0.24w(Cr) +0.94w(M0)+32w(N)] (2)
where, T , = T - 2 5 , T is the temperature ("C),and
w represents the concentration of the element (mass
percent, %). T h e yield strength of the austenite is
given in unit of MPa. T h e relation between U, and
temperature is shown in Fig. 2. U, can be briefly fit-
ted to a function of temperature (TI in the temperature
range of 50-350 "C.
a,(T>=O. 000 6 T 2 - 0 . 538T+226. 52 (3)
The maximum value of the external applied
stress is 80 MPa, amounting to about 67% of a,(T,)
Vol. 1 4
200
5
2 150
9
.-x
s
'2
9
J
4
I
50 150
Temperature/ "c
250 360
Austenite yield strength as a function of temperature
at 320 "C (120 MPa). T h u s , there will be no plastic
deformation because of the applied stresses before
martensitic transformation.
1.4 Martensite transformation start temperature
T h e calculated M , was found to be 280 "C.T h e
effect of stress on the M, was examined, with M,
measured as the temperature at which the slope of
the radial strain vs. temperature graph began to
change during quenching. Since the load will affect
the M , , the loads were applied at 320 "Cat the same
loading rate of 80 MPa s P 1 , just before the mar-
tensite transformation start temperature.
T h e M , for steel AISI4340 can be estimated using
the following equationc8' :
MS=561-474w(C) -33w(Mn) -17wCNi) -
17w(Cr) - 2 l w ( M o ) (4)
For this steel, M , =280 "C, and Mf(90% mar-
tensite transferred) = 210 "C.This indicates that the
temperature at the end of martensite transformation
is well above the room temperature ( R T ) , and
therefore, a quench directly to R T transforms al-
most all the austenite to martensite. Martensite can
be furnished without any additional cooling.
1.5 Analysis of strain in martensitic continuous
transformation
Specimens were cooled to the test temperature
in the same way. T h e load was applied t o the speci-
men at 320 "C; since the cooling rate is 11 "C sP1 ,
the full load could be applied to the specimen above
310 "C. Radial and longitudinal strains ( E and
~ E ~ . )
were measured simultaneously. T h e cylindrical
No. 6 Effects of Applied Stresses on Martensite Transformation in AISI4340 Steel * 65 a

specimens used were of axial symmetry, and there-

0
fore, assuming that the strains in radial direction are
identical. In this study, only weak loads, which
were less than 6 7 % of the yield stress of austenite,
were applied to the specimens. T h u s , the strain due -0.005
to the loads was small. When all strains are small,
the volume strain (nV/V) in the specimens is given G
.d

approximately by the following equationcg1: 4

t:
c;l
-0.010
N
V
= (I+EL) ( I + ~ R > z - I%&L +2ER (5)
T h e total strain E is decomposed into:
E = E ~ +E' +EP+~' (6)
where E~ is the elastic strain, E' is the thermal strain, -0.015 Y
0.006
and eP is the plastic strain, which is composed of the
"classical" transformation plastic strain E~~ ( stress-
free condition) and the T R I P strain E" ( d u e t o ap- 0.004

r
plied stress), i. e.
EP = E c P fE'P

The T R I P strain component etP depends on the

(7)
0.002 fl
Stress free
constant load stress and the varying martensitic vol-
ume fraction e,i. e. , etP= E ' ~ ( C J , E ) . T h e volume frac-
tion of martensite can be measured directly, using 0
the dilatometric curves of the radius direction. -0.001
The parameter K in Eqn. ( 1 ) comprises two 50 100 150 200 250 300
Temperature/9:
contributions, one due to plastic accommodation and
the other due to the orientation effect. Both contri- Fig. 4 Strain evaluation along longitudinal (a> and
butions are important in the TRIP study. diametrical directions (b)

2 Results and Discussion of the specimens in longitudinal and diametrical di-

T h e specimens were austenitized at 900 "C for
10 min in vacuum and cooled at 11 "C spl to RT.
Fig. 3 shows an optical micrograph of the specimen
after transformation. The fully martensitic trans-
formed microstructure appears in the final product
because of the sufficient hardenability of the alloy at
this cooling rate.
Fig. 4 shows the experimental dilatometric curves
Fig. 3 Optical micrograph of specimen obtained
after martensite transformation
rections during cooling at l l "C s-' under the vari-
ous uniaxial compressive stresses. It can be seen in
Fig. 4 ( a ) that in the longitudinal direction, the lon-
gitudinal strain (L-strain) due t o the transformation
lessens with the increase of the applied compressive
stress. Fig. 4 ( b ) shows that in the diametrical di-
rection, the larger the value of the applied stress,
the larger is the diametrical strain (D-strain) due to
transformation.
The final differences between the maximum
T R I P strain along the longitudinal direction obtained
under the various compressive stress conditions and
the stress-free one after transformation are shown in
Fig. 5. Under the stress-free condition, isotropic ex-
pansion occurs in the material with random orienta-
tion during martensitic transformation. With an in-
crease of the applied compressive stress, the expan-
sion in the direction of applied stress decreases.
Therefore, the total volume expansion after trans-
formation must be the same. It is obvious that the
absolute value of the maximum TRIP strain along the
 * 66 Journal of Iron and Steel Research, International Vol. 14

Table 1 Measured and calculated values of M.

0
Uniaxial stress/MPa M./% Source
-0.0005 0 280
0 280 Measured

$
20 281 Measured
-0.001 5 40 283 Measured
4 60 285 Measured

2 80 284 Measured

8 -O.OO2
Some acceleration of transformation because of the
-0.003 5 increase of applied stress can be observed especially
0 20 40 60 80 in the stress range of 60-80 MPa in Fig. 6.
Applied stress/MPa
Fig. 7 shows the difference between the calcu-
Fig. 5 Maximum TRIP strain in longitudinal direction as lated and the measured longitudinal strains during
a function of applied compressive stress transformation under various applied stresses. It can
be found that the calculated results agree well with
longitudinal direction E:: decreases linearly with the the experimental data and that the absolute value of
applied stress u: strain of TRIP along the longitudinal direction is
E%:=-3. 54X10-5~ (8) proportional t o the transformed fraction.
In fact, during the cooling t o R T , the material
cannot be transformed to martensite completely.
There is a small amount of retained austenite ( R A )
that can be neglected in the final specimen. It must
be assumed that the martensitic transformation is
complete when cooled to RT. T h e volume fraction
of martensite can then be measured directly by the
linear mixture rule method, using the dilatometric
curves of the radius direction, as shown in Fig. 6. It
can be seen in Fig. 6 that when stress is applied,
martensite transformation starts slightly earlier than
that for the stress-free one. T h e relationship be-
tween the values of applied stress and the calculated
and measured M sis shown in Table 1. M , increases Fig. 7 Comparison between calculated (lines)
with the applied stress in a micro-scale, and this strains and measured (symbols) ones
tendency is similar t o t h e results of t h e Ref. [S].
3 Conclusions
1.0 Transformation plasticity for the martensitic
transformation in low-alloy steel is considered under
5 0.8
the application of stress.
2
c (1) A relatively simple model has been adopt-
6
I o.6 ed to depict the transformation plasticity under dif-
5 0.4
ferent applied compressive stress. From this model,
the TRIP effect along the longitudinal direction
9E 0.2
caused by the martensitic transformation can be cal-
culated. T h e calculated results fit the experimental
i:
0 data well.
150 200 250 30 (2) T h e linear relationship between the plastic
Temperature&
strain along the longitudinal direction and the trans-
Fig. 6 Martensite transformation kinetics obtained by formed martensite fraction is obviously observed.
measured dilatometric curves As the applied stress 5 is increased, the maximum
No. 6 Effects of Applied Stresses on Martensite Transformation in AIS14340 Steel
magnitude of the TRIP effect along the longitudinal
direction EZ$decreases: E%$ = -3. 54 X 10-5a.
( 3 ) Martensite transformation kinetics is af-
fected slightly by applied stress.
References:
Lakhdar Taleb, Francois Sidoroff. A Micromechanical Model-
ing of the Greenwood-Johnson Mechanism in Transformation
Induced Plasticity [J]. International Journal of Plasticity,
2003, 19(10): 1821-1842.
Heung Nam Han, Dong-Woo Suh. A Model for Transforma-
tion Plasticity During Bainite Transformation of Steel Under
External Stress [J]. Acta Materialia, 2003, 51(16) : 4907-
4917.
Fischer F D, Reisner G , Tanaka K , et al. A New View on
Transformation Induced Plasticity (TRIP) [J]. International
Journal of Plasticity, 2000, 51(16) : 723-748.
(Continued From Page 57)
References:
ill H U I Wei-jun, DONG Han, WENG Yu-qing. Development
Trend of High Strength Steels Used for Automotive Suspen-
sion Coil Springs [J]. Journal of Iron and Steel Research,
2001, 13(2) : 67-72 (in Chinese).
c21 Nam W J , Choi H C. High Strength Steel Wire Rod for Sus-
pension Coil Spring [J]. Wire J Inter, 1996, ( 5 ) : 94-97.
C31 Matsuyama S. Delayed Fracture of High Strength Steels [J].
Tetsu-to-Hagene, 1994, 80(9) : 679-684 (in Japanese).
c41 H U I Wei-jun, DONG Han, WENG Yu-qing. Development of
High Strength Bolt Steels With High Delayed Fracture Resist-
ance [J]. Iron and Steel, 2001, 36(3): 69-73 (in Chinese).
c51 CHEN J M, W U J K. Hydrogen Diffusion Through Copper-
Plated AISI 4140 Steels [J]. Corrosion Science, 1992, 33(5):
6 5 7-6 6 6.
C6l Nakasato F, Takahashi M. Delayed-Failure Analysis of High-
Strength Structural Bolts [J]. Metals Technology, 1980,
(11) : 449-455.
c71 Inoue T , Namba Y. T h e Effect of Phosphorus Segregation at
Grain Boundaries and Boron Addition on the Impact Toughness
- .
67
[4] Heung Nam Han, Jae Kon Lee, Hong Joon Kim, et al. A
Model for Deformation, Temperature and Phase Transforma-
tion Behavior of Steels on Run-Out Table in Hot Strip Mill [J].
Journal of Materials Processing Technology, 2002, 128( 1-3) :
216-225.
C5l LIU Zhuang, W U Zhao-ji, WU Jing-zhi, et al. Numerical Sim-
ulation on Heat Treatment [MI. Beijing: Science Press, 1996
(in Chinese).
[6] Lakhdar Taleb, Nathalie Cavallo, Francois Waeckel. Experi-
mental Analysis of Transformation Plasticity [J]. International
Journal of Plasticity, 2000, 1 7 ( 1 ) : 1-20.
c71 Singh S B, Bhadeshia H K B H. Estimation of Bainite Plate-
Thickness in Low-Alloy Steels [J]. Materials Science and
Engineering, 1998, 245Ac1) : 72-79.
C8l Paul G Shewmon. Transformation in Metals [MI. New York:
McGraw-Hill Book Company, 1969.
191 Shipway P H , Bhadeshia H K D H. T h e Effect of Small Stres-
ses on the Kinetics of the Bainite Transformation [J]. Materi-
als Science and Engineering, 1995, 201A(1-2) : 143.149.
of Medium Carbon Tempered Martensite Steels With 150 kg/
mm2 Strength Level [J]. Tetsu-to-Hagane, 1991, 77: 2155-
2162 (in Japanese).
Liu C M , Nagoya T , Abiko K , et al. Effect of Boron on the
Grain Boundary Segregation of Phosphorus and Intergranular
Fracture in High-Purity Fe-0. 2 %P-B Alloys [J]. Metallurgical
Transactions A , 1992, 23A: 263-269.
Komazazki, Watanabe, Misawa. Influence of Phosphours and
Boron on Hydrogen Embrittlement Susceptibility of High
Strength Low Alloy Steel [J]. ISIJ International, 2003, 43
(11) :1851-1857.
C H U Wu-yang. Hydrogen Damage and Delayed Fracture
[MI. Beijing : Chinese Metallurgical Industry Press, 1988
(in Chinese).
Yamsaki S, Takahashi T. Delayed Fracture Mechanism in
High Strength Steels by Acoustic Emission Source Wave
Analysis [J]. Testu-to-Hagane, 1997, 83(7) : 460-465 (in
Japanese).
He H L , Chu Y Y , Chang H L. Distribution of Boron in
Steel [J]. Acta Metallurgica Sinica, 1977, 1 3 ( 4 ) : 235-245
(in Chinese).