Materials Science and Engineering A332 (2002) 1 10

www.elsevier.com/locate/msea

Phase transformation in AISI 410 stainless steel

M.C. Tsai *, C.S. Chiou 1, J.S. Du 2, J.R. Yang
Institute of Materials Science and Engineering, National Taiwan Uni6ersity, 1 Roose6elt Rd., Sec. 4, Taipei, Taiwan, ROC
Received 9 April 2001

Abstract
AISI 410 stainless steel can virtually fully transform to dislocated lath martensite at a very low cooling rate by air cooling. It
is found that at the same cooling rate the specimen austenitized at the higher temperature has a lower martensite start temperature
(Ms); besides, the difference of Ms becomes much larger at the higher cooling rate. In the continuously cooled specimens
investigated, significant amounts of inter-martensite retained austenite film may be imaged, where the martensite laths tend to be
in the same crystallographic orientation. On the other hand, micro-twinning is generally detected in martensite laths when the
adjacent variants of martensite form in a twin-related manner. Another important result indicates that no bainitic structure can
be obtained in this steel for the isothermal transformation just above the martensitic start temperature (Ms), and suggests that the
single C curve in the timetemperaturetransformation (TTT) diagram of AISI 410 is for a diffusional transformation. The prior
heavy deformation of austenite also gives strong evidence to suggest that severe deformation causes mechanical stabilization of
austenite against martensitic transformation. 2002 Elsevier Science B.V. All rights reserved.
Keywords: Martensitic stainless steel; Continuous cooling transformation; Ms temperature; Retained austenite; Micro-twinning; Isothermal
transformation; Mechanical stabilization

1. Introduction
In low alloy steels, the pearlitic and bainitic temperature ranges overlap each other to a considerable extent,
and this make the interpretation of microstructure and
kinetics very difficult. However, in medium alloy steels,
the
isothermal
time temperature transformation
(TTT) diagram possesses two C curves [1,2]. In such a
diagram, the upper C curve represents the time taken
for the initiation of diffusional transformation such as
allotriomorphic ferrite and pearlite, whereas the lower
C curve represents the time taken for the initiation of
displacive transformation such as bainite [3,4]. On the
other hand, in commercial martensitic stainless steels,
the TTT diagram possesses single C curve only. In this
respect, Bhadeshia had used the concepts of nucleation
of ferrite (which exhibited an invariant plane strain)
* Corresponding author.
E-mail address: jryang@ccms.ntu.edu.tw (M.C. Tsai).
1
Present address: Department of Mechanical Engineering, YuanZe University, Chung-Li, Taoyuan, Taiwan, ROC.
2
Present address: Applied Materials Twain Co., Hsin-Chu, Taiwan, ROC.

and stored energy due to shape change, and proposed a

thermodynamic model to explain why the bainitic
transformation does not occur in these high alloy steels
[3]. However, there was no experimental work on the
isothermal transformation just above the Ms temperature for these high alloy steels to elucidate whether the
bainitic transformation can occur or not. An attempt in
this work is therefore to justify the above outstanding
claim.
AISI 410 stainless flsteel is the progenitor in the
commercial low-carbon martensitic stainless steels, and
is widely used in many industries for the essential parts
(such as parts for HDD, plastic moulds, screws for
extruder, valves, shafts and bearings) due to its corrosion-resistance, high strength and high toughness. The
study of retained austenite and martensitic structure in
this type of martensitic steels has assumed great significance, due to their influence on the mechanical properties. There was extensive research work [511] in the
1970s on retained austenite, twinned plate martensite
and lath martensite in alloy steels. In order to fully
understand the martensitic transformation and existence of retained austenite in this steel, a series of
continuous cooling treatment, isothermal treatment and

0921-5093/02/$ - see front matter 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 1 - 5 0 9 3 ( 0 1 ) 0 1 7 1 0 - 5

M.C. Tsai et al. / Materials Science and Engineering A332 (2002) 110

Table 1
Chemical composition of the steel (wt.%)
Fe

Cr

Mo

Si

Cu

Sn

Mn

Ni

Al

Balance

0.12

12.80

0.006

0.02

0.33

0.04

0.01

0.44

0.12

0.01

0.031

0.023

thermomechanical treatment have been carried out on a

Dilatronic III dilatometer of Theta Industries, Inc. The
dilatometric curves, optical metallographs and transmission electron micrographs have been examined to
elucidate the microstructural development. Furthermore, the effects of austenitizing temperature and cooling rate on Ms temperature have been investigated. The
interpretations for the observed phenomena have been
proposed.

350 C for tempering the existing martensite. The

schematic diagram for thermomechanical treatment
is shown in Fig. 1c.
The transmission electron microscopy samples were
prepared from 0.25 mm thick discs cut from the
dilatometer specimens. The discs were thinned to 0.05
mm by abrasion on silicon carbide papers and then
twin-jet electropolished using a mixture of 5% perchloric acid, 25% glycerol and 70% ethanol at ambient

2. Experimental procedure
A commercial AISI 410 steel bar with 16 mm diameter in the annealed state was used as the starting
material. The steel was produced by Gloria Material
Technology Corporation through the process of electric
arc furnace melting, vacuum ladle furnace treatment,
ingot casting, forging, rolling and annealing. The chemical composition of the steel studied is listed in Table 1.
In this work, all the thermal treatments have been
carried out on a Dilatrometric III RDP deformation
dilatometer of Theta Industries, Inc. Before preparation
for dilatrometric specimens, the pieces of steel plate
were homogenized at 1200 C for 3 days, then
quenched into water, and finally removed the decarburization layer. The specimens were machined in the form
of 3 mm diameter cylindrical rods with 6 mm length.
The experiments included three parts as described in the
following.
1. Continuous cooling treatment. The processing
scheme performed in the dilatometer is presented in
Fig. 1a; the dilatometer specimens were austenitized
at 900, 1000, 1100, 1200 C for 10 min, and then
cooled at different cooling rates (0.5 100 C s 1)
to room temperature.
2. Isothermal treatment. The specimens were heated to
1000 C for 10 min, and then rapidly cooled to the
temperature range 370 600 C for isothermal transformation. The schematic diagram for isothermal
treatment is shown in Fig. 1b.
3. Thermomechanical treatment. After austenitization
at 1000 C for 10 min, the specimens were cooled to
750 C at a cooling rate of 60 C s 1, immediately,
different amounts (0, 10, 20 and 40%) of compressive deformation were executed within 3 s. After
deformation, the specimens were held at 290 C for
isothermal treatment until the transformation was
completed, and immediately rapidly up-quenched to

Fig. 1. Schematic diagrams showing the processing performed in the

dilatometer, (a) continuous cooling treatment; (b) isothermal treatment; (c) thermomechanical treatment.

M.C. Tsai et al. / Materials Science and Engineering A332 (2002) 110

3. Results and discussion

In order to investigate the effects of cooling rate on
Ms temperature in AISI 410 stainless steel, a series of
continuous cooling treatment has been done. Fig. 2
shows the dilatometric curves for the specimens austenitized at 900 C for 10 min and followed by the cooling
at different rates (0.5100 C s 1). It is obvious that

Fig. 2. The dilatometric curves for the specimen austenitized at

900 C for 10 min followed by cooling at 0.5 100 C s 1.

Fig. 5. The dilatometric curves for the specimen austenitized at

1200 C for 10 min followed by cooling at 0.5 100 C s 1.
Table 2
The Ms temperature measured from dilatometric curves
Cooling rate (C s1)

Fig. 3. The dilatometric curves for the specimen austenitized at

1000 C for 10 min followed by cooling at 0.5 100 C s 1.

0.5
5
10
20
40
60
80
100

Austenitization temperature
900 C

1000 C 1100 C 1200 C

375
360
350
345
340
335
330
320

360
340
335
320
300
295
290
285

355
340
335
320
300
290
285
280

340
335
330
315
290
270
250
245

Fig. 4. The dilatometric curves for the specimen austenitized at

1100 C for 10 min followed by cooling at 0.5 100 C s 1.

temperature and 60 V polishing potential. They were

examined using a JEOL 100CX transmission electron
microscope operating at 100 kV.

Fig. 6. Plot of Ms vs. cooling rate for four different austenitization

temperatures.

M.C. Tsai et al. / Materials Science and Engineering A332 (2002) 110

the Ms temperature decreases with increasing cooling

rate. The similar phenomena were also found for the
specimens austenitized at 1000, 1100 and 1200 C as
shown in Figs. 35.
The Ms temperatures have been measured (as presented in Table 2) from the dilatometric curves in Figs.
25 for different austenitization temperature conditions
with a wide range of cooling rate. The Ms temperature
versus cooling rate is plotted in Fig. 6. It indicates that
at the same cooling rate the specimen with the higher

Fig. 7. TEM obtained from the specimens austenitized at 1000 C

followed by continuous cooling at (a) 0.5 and (b) 100 C s 1.

Fig. 8. Bright and dark field images of TEM for retained austenite
films obtained from the specimen continuously cooled at 0.5 C s 1
after austenitization at 1100 C for 10 min.

Fig. 9. Bright and dark field images of TEM for retained austenite
films obtained from the specimen continuously cooled at 100 C s 1
after austenitization at 1100 C for 10 min.

Fig. 10. Transmission electron micrograph obtained from the specimen continuously cooled at 0.5 C s 1 after austenitization at
1100 C for 10 min.

M.C. Tsai et al. / Materials Science and Engineering A332 (2002) 110

Fig. 11. Micro-twins in AISI 410 steels obtained from specimens cooling at 0.5 C s 1, (a) bright field micrograph; (b) dark field of left side
martensitic lath; (c) dark field of right side martensitic lath.

Fig. 12. Micro-twins in AISI 410 steels obtained from specimens cooling at 100 C s 1, (a) bright field image; (b) dark field image using (2( 1( 1)
reflection; (c) dark field image using a1% (01( 1) reflection; (d) dark field image using a2% (011( ) reflection; (e) corresponding diffraction pattern; (f)
interpretation of (e).

M.C. Tsai et al. / Materials Science and Engineering A332 (2002) 110

Table 3
Twenty-four equivalent axisangle pairs relating twinning within the
martensitic laths
Number

The 23 equivalent axisangle pairs

Axis

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

Angle

0
0.4082
0.7071
0.5774
0.9045
0.3015
0.7071
0.5774
0.3015
0.4082
0.8944
0.4472
0.7071
0.9045
0.3015
0.7071
0
0.4472
0.3015
0
0.5774
0
0.8944
0.8165

0.7071
0.8165
0.7071
0.5774
0.3015
0.3015
0.7071
0.5773
0.9045
0.4082
0.4472
0
0
0.3015
0.3015
0
0.8944
0.8944
0.9045
0.7071
0.5774
0.4472
0
0.4082

0.7071
0.4082
0
0.5774
0.3015
0.9045
0
0.5774
0.3015
0.8165
0
0.8944
0.7071
0.3015
0.9045
0.7071
0.4472
0
0.3015
0.7071
0.5773
0.8944
0.4472
0.4082

70.5
180
70.5
180
146.4
146.4
109.5
60
146.4
180
131.8
131.8
109.5
146.4
146.4
70.5
131.8
131.8
146.4
109.5
60
131.8
131.8
180

Table 4
Twenty-four independent variants of the KS orientation relationships
Variant number Description
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

(011)a%//(111)g
(011)a%//(111)g
(011)a%//(111)g
(011)a%//(111)g
(011)a%//(111)g
(011)a%//(111)g
(011)a%//(1( 1( 1)g
(011)a%//(1( 1( 1)g
(011)a%//(1( 1( 1)g
(011)a%//(1( 1( 1)g
(011)a%//(1( 1( 1)g
(011)a%//(1( 1( 1)g
(011)a%//(1( 11( )g
(011)a%//(1( 11( )g
(011)a%//(1( 11( )g
(011)a%//(1( 11( )g
(011)a%//(1( 11( )g
(011)a%//(1( 11( )g
(011)a%//(11( 1( )g
(011)a%//(11( 1( )g
(011)a%//(11( 1( )g
(011)a%//(11( 1( )g
(011)a%//(11( 1( )g
(011)a%//(11( 1( )g

Group number
[1( 1( 1]a%//[1( 01]g
[11( 1]a%//[1( 01]g
[1( 1( 1]a%//[01( 1]g
[11( 1]a%//[01( 1]g
[1( 1( 1]a%//[11( 0]g
[11( 1]a%//[11( 0]g
[1( 1( 1]a%//[11( 0]g
[11( 1]a%//[11( 0]g
[1( 1( 1]a%//[101]g
[11( 1]a%//[101]g
[1( 1( 1]a%//[011]g
[11( 1]a%//[011]g
[1( 1( 1]a%//[110]g
[11( 1]a%//[110]g
[1( 1( 1]a%//[011]g
[11( 1]a%//[011]g
[1( 1( 1]a%//[1( 10]g
[11( 1]a%//[1( 01]g
[1( 1( 1]a%//[110]g
[11( 1]a%//[110]g
[1( 1( 1]a%//[101]g
[11( 1]a%//[101]g
[1( 1( 1]a%//[01( 1]g
[11( 1]a%//[01( 1]g

1
1
1
1
1
1
2
2
2
2
2
2
3
3
3
3
3
3
4
4
4
4
4
4

austenitization temperature has a lower martensite start

temperature (Ms); besides, the difference of Ms becomes much larger at the higher cooling rate. The Ms
temperatures for the specimens austenitized at 900,
1000, 1100 and 1200 C separately and followed by
cooling at 100 C s 1 are determined to be 320, 285,
280 and 245 C, respectively. It is noticed that at the
cooling rate of 100 C s 1, Ms in the case with austenitization at 1200 C is lower than that in the case with
austenitization at 900 by 75 C. The result provides
strong evidence showing that the higher austenitization
temperature with the higher cooling rate depresses the
martensitic transformation.
Some related research work has been done by Hsu et
al. [12] who claimed that values of Ms decreased with
the increase of the quenching temperature above
900 C in Fe29Ni (wt.%), Fe30Ni (wt.%) and Fe
31Ni (wt.%) alloys. They suggested that the pinning of
clustered vacancies to the partial dislocations hindered
the nucleation of martensite and consequently depressed the Ms temperature. The detailed research
about interaction between vacancies and the edge dislocation in body centered cubic (b.c.c.) metal had been
studied by Ingle et al. [13] The variation of Ms in
Fe C, Fe Mn, Fe Si, Fe Ni C and Fe Cr C [14]
alloys have also been investigated through considering
the strength of matrices. Because carbon atom has most
powerful solid-solution strengthening effect on austenite than other alloying atoms (such as Cr, Ni and Si), it
has been supposed to be the main element in steels to
influence the Ms temperature. When the chromium
carbide precipitates in matrix, the lower amount of
carbon retained in solid-solution austenite would lead
to a higher Ms temperature.
The possible factors to affect the Ms temperature can
be listed as follows, (1) chemical composition in matrix;
(2) grain size of austenite [15]; (3) cooling rate [16]; (4)
crystal defect and inclusion [17,18]; (5) occurrence of
other pre-transformations [19]; (6) stress and strain [20];
(7) magnetization [21]; (8) hydrostatic pressure. In this
work, the factors (6)(8) could be neglected. Generally,
the main factor to influence the Ms temperature is the
strength of parent austenite structure. If the strength of
prior austenite is high, the Ms temperature will be
depressed to a lower temperature. Factors (1) and (2)
would be related with this main factor. For example,
the increase in cooling rate could induce the larger
quantities of vacancy in matrix to strength the austenite, and leads to a lower Ms temperature. On the other
hand, the variation of Ms could be due to the difference
in the grain size of austenite. The explanation for the
grain size effect has been well known; the increased
strength of austenite by grain refining makes it more
difficult to accommodate plastically the shape strain of
the transforming martensite plate, thus depressing the
Ms temperature in fine-grained specimens. However,

M.C. Tsai et al. / Materials Science and Engineering A332 (2002) 110

Fig. 13. The explanation of twin relative using stereograph, (a) KS orientation relationship for a1% with g; (b) KS orientation relationship for h %2
with g; (c) superimposition of (a) and (b); (d) h %1 and h %2 being twin-related.

the plausible explanation in this work is suggested that

the higher austenitization temperature with the higher
cooling rate gives the austenite matrix with a higher
concentration of vacancies and without any precipitation of chromium carbides. The increased strength of
austenite by the interaction between dislocations and
vacancies, and by the solid solution due to carbon and
chromium elements preserved in the matrix leads to the
lower Ms temperature.
The study of retained austenite and martensitic structure in this type of martensitic steels has assumed great
significant, primarily due to their influence on mechanical properties of commercially used quenched and tempered ultra-high strength steels. However, the detailed
microstructure development is still need to elucidate.
Transmission electron micrographs with a low magnification in Fig. 7 were obtained from the specimens
austenitized at 1000 C and followed by continuous
cooling at 0.5 and 100 C s 1, respectively, to room
temperature. For the slow cooling rate condition (Fig.
7a), large laths of martensite partitioned the prior
austenite grain and fine laths of martensite swamped
the partitioned areas. On the other hand, for the high
cooling rate condition (Fig. 7b), much more uniform
fine laths of martensite could be obtained. The result

implies that at a slow cooling rate, the martensite lath

forms at the high temperature has enough time to grow
without impingement with other martensite variants.
Besides, auto-tempered martensite (with several variants of carbides) could be identified in the specimens
treated by the slow continuous cooling. Bright and dark
field images of transmission electron micrographs for
retained austenite in the martensitic matrices were
shown in Figs. 8 and 9, obtained from the specimen
austenitized at 1100 C and cooled at 0.5 and 100 C
s 1, respectively. Significant amounts of inter-martensite retained austenite films could be detected and it was
also found that the martensite laths tended to be in the
same crystallographic orientation in space (over a distance of approximately six martensite laths at least). It
was noted that the wider the lath of martensite, the
thicker the film of retained austenite (as shown in Figs.
8 and 9). When the laths constituting a packet were in
the same orientation, it was presumed that the shape
deformation of each unit within the packet was identical. It is therefore suggested that mechanical stabilization, which essentially involves the jamming of the
austenitemartensite interface by accommodation defects could be responsible for the stability of the austenite films.

M.C. Tsai et al. / Materials Science and Engineering A332 (2002) 110

Fig. 16. The optical microscope microstructure by austenitizing to

1000 C and given 0 40% deformation at 750 C, then hold temperature at 290 C, (a) 0%; (b) 10%; (c) 20%; (d) 40%.
Fig. 14. Optical metallographs showing the specimens isothermally
treated at 600 C for various periods of time, (a) 1 h; (b) 4 h; (c) 8 h;
(d) 16 h.

Fig. 15. (a) Optical micrographs and (b) transmission electron micrograph obtained from the specimen tempered at 370 C for 11
days.

Martensitic structure is normally considered to give

dislocated laths if carbon content of the steel is less
than 0.4 wt.%. However, it has been frequently found
that micro-twins exist within the martensite laths (as
shown in Fig. 10) in this steel. To resolve the mechanism of twin formation, before attempting to study the
effect of twinning on the mechanical properties, is
important. Although AISI 410 steel has a high Ms over
room temperature, micro-twinning could often be observed in martensite laths. These twins extended partially cross the martensite units and the inter-twin
spacing were generally large; these features are different
from those in twinned plate martensite. Examination of
bright field, diffraction patterns and dark field images
(Figs. 11 and 12) found that micro-twinning was extensive when adjacent martensite laths were twin related,
and that these interpenetrating twins had the same
orientation with the adjacent martensite lath. It was
apparent that the formation of these kinds of twins
resulted from the high degree of mutually compensating
accommodation between twin related martensite variants. Such micro-twins were not intrinsic transformation features. Table 3 shows the 24 equivalent
axisangle pairs for twins calculated from electron
diffraction patterns of Fig. 12f.
The prevalent orientation relationships describing the
austenite and martensite disposition after transformation can be listed as follows [22].

M.C. Tsai et al. / Materials Science and Engineering A332 (2002) 110

Kurdjumov Sachs (KS) (111)g//(011)a% [1( 01]g//[1( 1( 1]a%.

Greninger Troiano (GT) (111)g 0.2 from (011)a%
[1( 01]g 2.7 from [1( 1( 1]a%.
NishiyamaWasserman (NW) (111)g//(011)a% [1( 01]g//
[100]a%.
Considering the KS orientation relationship, 24 independent crystallographic variants of martensite can be
obtained within a single crystal of austenite as shown in
Table 4. Four groups can be assigned to the four
independent {111}g planes in a given austenite crystal. In
each group, six independent variants can be obtained by
ration about the pole of (011)a%//{111}g. Hence, for four
groups 24 independent variants exist. The 24 KS relationships have the special property that variants of martensie
can be twin-related in pairs (such as variants 1 2, 34,
5 6, ). It can be illustrated as follows. Fig. 13a and
b show variants 1 and 2 of martensite with their
respective KS orientation relationship. It is clear that
variants 12 have the twin-related orientation (Fig. 13d).
Similarly, variants 3 4, 5 6, 7 8, and so on are twin-related. It is found that twin related variants cannot be
obtained by choosing variants from different groups. The
twin-related orientation happens in only 12 combinations
out of 276. Bhadeshia examined the crystal orientations
for NW and GT relationships, and claimed that for NW
and GT relationships, the adjacent laths of martensite
can not have twin-related orientation [23]. It is suggested
that micro-twins in lath martensite structure depends on
the local crystallography.
It had been well know that the AISI 410 stainless steel
only has single C curve in the TTT diagram [24].
Bhadeshia [3] has proposed a thermodynamic mode to
explain the bainitic transformation does not occur in this
steel. Fig. 14 illustrates the microstructures obtained
from the specimens after completion of isothermal transformation at 600 C for 1, 4, 8 and 16 h. The optical
micrographs show a thin diffusion layer of carbide
decorated at prior austenite grain boundaries. Examination of a specimen isothermally transformed slightly
above the Ms temperature, at 370 C, for 11 days (before
finally quenching to ambient temperature) revealed that
alloy-pearlite formed along martensite grain boundaries
(Fig. 15). No bainitic structure could be identified and
there was no evidence for bainitic transformation in this
alloy steel. The result is consistent with that proposed by
Bhadeshia, and strongly indicates that the single C curve
in the TTT diagram of AISI 410 is for a diffusional
transformation.
Optical metallographs in Fig. 16 were obtained from
the specimens after austenitization at 1000 C for 10
min, followed by different amounts of prior deformation
(0, 10, 20 and 40%) at 750 C, followed by isothermal
transformation at 290 C for 30 min (i.e. the time is
longer enough to complete the isothermal transformation), and then up-quenched to 350 C in order to temper
the existing martensite. The metallographs in Fig. 16 are

illustrated as follows. The grey etching regions represented the martensite, which formed and was immediately tempered after isothermal treatment (grey etching
structure was due to the existence of interlath carbides
within martensite). The white etching regions represented
the martensite, which formed during cooling to room
temperature after up-quench treatment. They show an
increase in difficulty of martensite lath propagation into
the austenite phase due to work hardening of austenite.
This phenomenon elucidates severe deformation leads to
mechanical stabilization of austenite against martensitic
transformation.

4. Conclusion
An investigation has been made of the phase transformations in AISI 410 stainless steel. Based on the study
of dilatometry, optical metallography and transmission
electron microscopy, the following conclusion were
drawn.
The results show that the Ms temperature decreases
with the increase of cooling rate, and that at the same
cooling rate the specimen austenitized at the higher
temperature has a lower Ms temperature. The plausible
explanation is suggested that the austenite matrix with
a higher concentration of vacancies and a lack of
chromium carbide precipitation leads to a lower Ms
temperature.
The existence of retained austenite films and microtwins in lath martensite structure depends on the local
crystallography, which results in mechanical stabilization
of austenite films or mechanical accommodation of
interpenetrating twins.
Examination of isothermal transformation just above
Ms temperature confirms that the single C curve in the
TTT diagram of AISI 410 stainless steel is for a diffusional transformation.
Heavy deformation of prior austenite causes mechanical stabilization of the austenite and leads to the
difficulty of martensite lath propagation into the austenite.

Acknowledgements
This work was carried out with financial support from
the National Science Council of the Republic of China,
Taiwan, under Contract NSC 88-2216-E-002-037.

References
[1] H.K.D.H. Bhadeshia, Acta Metall. 29 (1981) 1117 1130.
[2] A. Ali, H.K.D.H. Bhadeshia, Mater. Sci. Technol. 6 (1990)
781 784.

10

M.C. Tsai et al. / Materials Science and Engineering A332 (2002) 110

[3] H.K.D.H. Bhadeshia, Bainite in Steels, The Institute of Materials, London, 1992, pp. 126 133.
[4] A. Ali, H.K.D.H. Bhadeshia, Mater. Sci. Technol. 7 (1991)
895 903.
[5] S.K. Das, G. Thomas, Metall. Trans. 1 (1970) 325 327.
[6] D.H. Huang, G. Thomas, Metall. Trans. 2 (1971) 1587 1598.
[7] M. Raghavan, G. Thomas, Metall. Trans. 2 (1971) 3433 3439.
[8] G.Y. Lai, W.E. Wood, R.A. Clark, V.F. Zackay, E.R. Parker,
Metall. Trans. 5 (1974) 1663 1670.
[9] E.R. Parker, Metall. Trans. A 8A (1977) 1025 1042.
[10] G. Thomas, Metall. Trans. A 9A (1978) 439 450.
[11] K.H. Khan, W.E. Wood, Metall. Trans. A 9A (1978) 899 907.
[12] T.Y. Hsu, Y. Linfah, J. Mater. Sci. 18 (1983) 3213 3218.
[13] K.W. Ingle, A.G. Crocker, Acta Metall. 26 (1978) 1461 1469.
[14] T.Y. Hsu, J. Mater. Sci. 20 (1985) 23 31.
[15] S. Kajiwara, Metall. Trans. A 17A (1986) 1693 1702.
[16] J.C. Brachet, L. Gavard, C. Boussidan, C. Lepoittevin, S. Denis,

C. Servant, J. Nucl. Mater. 258 263 (1998) 1307 1311.

[17] J.W. Brooks, M.H. Loretto, R.E. Smallman, Acta Metall. 27
(1979) 1829 1838.
[18] J.W. Brooks, M.H. Loretto, R.E. Smallman, Acta Metall. 27
(1979) 1839 1847.
[19] A. Alamo, J.C. Brachet, A. Castaing, C. Lepoittevin, F. Barcelo,
J. Nucl. Mater. 258 263 (1998) 1228 1235.
[20] S. Kajiwara, Mater. Trans. JIM 33 (1992) 1027 1034.
[21] T. Kakeshita, T. Saburi, K. Shimizu, Mater. Sci. Eng. A273
A275 (1999) 21 39.
[22] H.K.D.H. Bhadeshia, Ph.D. thesis, University of Cambridge,
1979.
[23] H.K.D.H. Bhadeshia, Worked Examples in the Geometry of
Crystals, The Institute of Metals, London, 1987, pp. 16 18.
[24] P.M. Unterweiser, Heat Treaters Guide Standard Practices and
Procedures for Steel, American Society for Metals, USA, 1982,
pp. 423 426.