Applied Mechanics and Materials Vol.

598 (2014) pp 38-42 Submitted: 2014-06-03

© (2014) Trans Tech Publications, Switzerland Accepted: 2014-06-03
doi:10.4028/www.scientific.net/AMM.598.38 Online: 2014-07-16

Iterative Taguchi Analysis: Optimizing the Grain Boundary Carbide in

Hadfield steel
Sh. Hosseini and M. B. Limooei
Department of Material Science and Engineering, Ayatollah Amoli Branch, Islamic Azad University,
Amol, Iran
sh_hosseini@yahoo.com

Keywords: Taguchi analysis, Manganese steel (Hadfield), Austenitizing temperature, Austenitizing

time, quenching solution, grain boundary carbide

Abstract. Manganese steel (Hadfield) is one of the important alloys in industry due to its special
properties. High work hardening ability with appropriate toughness and ductility are the properties
that caused this alloy to be used in wear resistance parts and in high strength condition. Heat
treatment is the main process through which the desired mechanical properties and microstructures
are obtained in Hadfield steel. Iterations of Taguchi designed experiments and analysis were used to
determine optimum heat treatment for minimizing grain boundary carbide content in Hadfield steel.
Experimental variable chosen for this study included austenitizing temperature and time and rate of
quenching. The austenitizing temperature and the cooling rate by changing in quenched solution
were seen to have the greatest influence on carbide content in Hadfield steel.

Introduction
The invention and production of Manganese Steels (so called Hadfield Steels) go back to 1883, and
such steels were registered as no. 200 in the name of Sir. Robert Hadfield (Sheffield, UK) [1,2].
Manganese steels are well-known in steel industries due to their particular characters such as their
resistance to wear as well as high impact strength, good ductility and also particular austenitic
structure enhancing after heat treatment [3]. It mainly contains 1.2% carbon, 12 – 14 % manganese.
Austenitic microstructure of steels is non magnetic, and its toughness and ductility due to FCC
matrix exhibit more advantages comparing to the ferritic phase (with BCC matrix). Within these
steels, manganese is playing main role as austenitic phase stabilizer to room temperature [1-9].
There are two significant stages during the production of desirable parts from Hadfield steel:
melting and developing of appropriate composition, and subsequently applying proper heat
treatment on casting parts. Microscopic structure of non-heat treated cast part includes austenite in
matrix, carbide phase and a small set of pearlite. Carbides generally develop on grain boundaries
and between dendrites. To obtain desired toughness, microstructure of Hadfield steels should be
thoroughly austenitic [2].
These kinds of steels are strictly sensitive to section thickness. Austenitic mono phase obtain
from heat treatment of austenitizing and rapid quenching. Nevertheless at the thicker sections due to
relatively low heat conductivity, the carbides precipitate as transgranular and grain boundaries and
consequently may result to mechanical properties reduction [3].
Heat treatment probably results in developing toughness in alloy; therefore solution temperature
would be too high so that carbide dissolves into austenite. In practice, the final structure will not be
formed as austenite and some of carbides will remain within grain boundaries, especially in thicker
section [2,10]. Formation of these carbides between austenite grain boundaries will minimize
impact strength of said steel [11].
Traditional heat treatment includes austenitizing and water quenching. Solution temperature
range of heat treatment is defined as 1010 to 1120˚C. Heating would be performed slowly to
prevent internal cracking or crack propagation. On the other hand quenching should be quick
enough to minimize the carbides. To obtain a fully austenite microstructure after heat treatment any

All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans
Tech Publications, www.ttp.net. (ID: 142.103.160.110, University of British Columbia, Kelowna, Canada-10/07/15,03:54:52)
 Applied Mechanics and Materials Vol. 598 39

transformation to austenite should not be allowed [1]. Since carbon solubility in austenite with 11%
Mn increases to remarkably higher values (around 1.4%) at temperatures higher than 1000˚C. the
austenitizing process is carried out at high temperature [3].
To obtain a fully austenite microstructure after heat treatment any transformation to austenite
should not be allowed [1]. The factors that affect on heat treatment are austenitizing temperature,
austenitizing time and quenching rate. Determination of appropriate times and temperatures for a
heat treating procedure that will achieve both low grains boundary carbide and austenite grain size
can appear initially to require extensive, Taguchi analysis provides an effective means of achieving
these goals. Taguchi method incorporate orthogonal arrays to minimize the number of experiments
required to achieve a given set of performance characteristics [12,13]. By applying the Taguchi
method, the time required for experimental investigations can be significantly reduced [14].
This research describes an application of a Taguchi analysis to obtain an optimum heat
treatment. L9(3)3 array, employing nine experiments, with three levels for each factor, was chosen
for this method. The heat treatment of Hadfield steel requiring minimum carbide content obtain
good wear resistance was chosen to demonstrate the approach.

Methods and Materials

In the present study, cylindrical specimens of Hadfield steel were used. Chemical composition of
this alloy which analyzed by Spectrolab model quantometer, is shown in table 1.

Table (1) Weight percent of chemical composition

C Si Mn P S Cr Mo Ni Al Co Cu Nb V Sn
1.32 0.49 13.32 0.041 0.00004 1.91 0.009 0.039 0.014 0.016 0.041 0.005 0.02 0.005

Three parameters identified as affecting the carbide content and ASTM grain size are the
austenitizing temperature, austenitizing time and cooling rate. These factors are normally specified
in heat treating references as being the most important. The austenitizing temperature and time must
be enough to transform to homogeneous FCC austenite. The cooling rate was varied by changing
the salt amount of quenching solution. Heat treatment has been performed within electrical furnace.
Table 2 shows these factors and their level. L9(3)3 array, employing nine experiments, with three
levels for each factor, was chosen for DOE (table 3).

Table (2) Factor and level descriptions for Taguchi DOE

Factors Level 1 Level 2 Level 3
A Quenching rate 0% salt 1.5% salt 3% salt
B Austenitizing temperature (˚C) 1050 1100 1150
C Austenitizing time (hour) 1.5 2 3

Table (3) L9(3)3 array for Taguchi DOE

3
L9(3) A B C
Factors
Austenitizing Austenitizing
Exp. A B C Quenching rate
temperature(˚C) time(hour)
1 1 1 1 0% salt 1050 1.5
2 1 2 2 0% salt 1100 2
3 1 3 3 0% salt 1150 3
4 2 1 2 1.5% salt 1050 2
5 2 2 3 1.5% salt 1100 3
6 2 3 1 1.5% salt 1150 1.5
7 3 1 3 3% salt 1050 3
40 Advanced Materials, Mechanics and Industrial Engineering

8 3 2 1 3% salt 1100 1.5

9 3 3 2 3% salt 1150 2

Metallographic examination has been done in according to ASTM E3-01 (metallographic sample
preparation), ASTM E407-99 (metal micro-etching) and ASTM 883-02(optical microscopic
images). Specimens microstructures have been investigated by optical microscope (Olympus,
PMG3 model) and austenite grain size determined in according to with the ASTM Code E112-
06.

Results and discussion

According to table 3 the experiments were carried out and its results are shown in table 4. Figure 1
is illustrated heat treated microstructures. Within mentioned figure, carbide volume difference is
well pointed out.
Table (4) Response table for Taguchi DOE
Carbide content
Experiments Carbide type
(Volume Percent)
1 42% Fiber, disperse in grain boundary, inter granular
2 30% Fiber, disperse in grain boundary, inter granular
3 13% Very fine and disperse
4 24% Fiber, disperse in grain boundary
5 15% Fiber, disperse in grain boundary
6 8% Very fine and disperse
7 18% Fiber, disperse in grain boundary
8 12% Fiber, disperse in grain boundary
9 5% Fiber, disperse in grain boundary

Figure (1) The microstructure of 9 experiments

Figure 2 shows the linear graphs of the main effects and their variation between levels of
parameters on the grain boundary carbide content. As it shown in the picture, because of dissolving
the carbide in austenite matrix, by increasing the austenitizing temperature, the grain boundary
 Applied Mechanics and Materials Vol. 598 41

carbides that are produced in the casting process, decrease. This effect is seen in austenitizing time,
too. On the other hand, the carbide volume is decreased with increasing the cooling rate. Salt weight
increasing in quenching solution has significantly decreased carbide content in matrix structure and
grain boundaries. Considering undesirable carbides in final structure, it could be concluded that,
with increasing NaCl content, may produce non carbide structure.
The relative slope of the linear graphs indicates significance of the parameters. Furthermore the
result of analysis of variance (ANOVA) for the response function is given in table 5. Hence from
figure 2 and table 5, it can be concluded that the austenitizing temperature is the most significant
parameter on carbide content. Two factors of austenitizing temperature and quenching rate have the
most influence and austenitizing time has the least influence on the carbide content.
Since the lower is better, the minimum carbide volume is in the level 3 of each parameter.

Table (5) Analysis of variance for grain boundary carbide content

DOF Sum of square Variance F-ratio Pure sum Percent

Factors
(f) (S) (V) (F) (S') P(%)
Quenching rate 2 454.222 227.111 16.094 426 40.003
Austenitizing temperature 2 561.555 280.777 19.897 533.333 50.83
Autenitizing time 2 48.222 24.111 1.708 19.999 2.83

30

25
grain boundary
carbide(%)

20

15

10
quenching rate
5 austenizing temperature
austenizing time
0
1 2 3
levels of design parameters
Figure (2) Plot of response data for main factor of Taguchi DOE, Effect of main factor on grain
boundary carbide content

Conclusion
The experiments conducted show that austenitizing temperature has the most influence on the grain
boundary carbide content in the heat treatment of Hadfield steel. The highest austenitizing
temperature gave the lowest carbide content. The optimum carbide content condition is said to be at
austenitizing temperature= 1150˚C, austenitizing time= 3 hour and quenching solution with 3% salt.

Reference:
[1] D .K. Subramanya , A.E. Swansiger, H.S. Avery, Austenitic Manganese Steels, 10th Edition,
ASM Metals Handbook, Vol.1, 1991, pp.822-840.
[2] D.K. Subramanyam, G.W. Grube, H.J. Chapin, “Austenitic Manganese Steel Castings”, 9th
Edition, ASM Metals Handbook, Vol. 9, 1985, pp.251-256.
[3] S.Chakrabartti, Cast Austenitic Manganese Steels Some Practical Notes ,
www.cbgfoundry.com.
[4] I.Karaman, Ph.D. Thesis, University of Illinois, Chicago, 2000.
42 Advanced Materials, Mechanics and Industrial Engineering

[5] E.A. Aleshina , O.V. Sizova , “Formation of Structural-Phase States of the Surface of
Hadfield Steel”, Steel in Translation, 2007, Vol.12, pp.989-990.
[6] E.G. Zakharova, I.V. Kireeva, Yu.I. Chumlyakov, “ Strain Hardening and Fracture of
Austenitic Steel Single Crystals with High Concentration of Interstitial Atoms”, Russian Physics
J, 2002, Vol. 45, pp. 274-284.
[7] T.Sasaki, “Physical and Mechanical Properties of High Manganese Non-Magnetic Steel and
its Application toVarious Products for Cammerical Use”, Transactions ISU, 1982, Vol. 22,
pp.184-192.
[8] A.K. Srivastava, K.Das, “Microstructural Characterization of Hadfield Austenitic Manganese
Steel”, Materials Sciences, 2008, Vol. 43, pp.5654-5658.
[9] P.H. Adler, G.B. Olson, and W.S. Owen, “Strain Hardening of Hadfield Manganese Steel”,
Metall Trans, 1986, Vol.17A, pp. 1725-1737.
[10] M.Ravandoost, M.Naghavi, M.H.Shaeri, “Effect of Direct Quenching on Hadfield Steel
Micro Structure”11 th annual conference of Iranian Metallurgical Engineers Society,1368.
[11] G.F. Liang, and C.J. Song, “Eutectic Decomposition in Ca-Si Modified Austenitic Medium
Mn Steel after Solidification”, Materials Sciences, 2005, Vol. 40, pp. 2081-2084.
[12] P.W.Mason and P.S.Prevey, “Iterative Taguchi Analysis: optimizing the austenite content
and hardness in 52100 steel”, Journal of Materials Engineering and performance, Vol. 10(1),
February 2001,14-21.
[13]M. Joseph Davidson, K. Balasubramanian, G.R.N. Tagore, “Experimental investigation on
flow-forming of AA6061 alloy-A Taguchi approach”, journal of materials processing technology ,
2008(200): 283–287
[14] Adem Cicek, Turgay Kivak, Gurcan Samtas, “Application of Taguchi Method for Surface
Roughness and Roundness Error in Drilling of AISI 316 Stainless Steel”,Journal of Mechanical
Engineering, 58(2012)3, 165-174.
Advanced Materials, Mechanics and Industrial Engineering
10.4028/www.scientific.net/AMM.598

Iterative Taguchi Analysis: Optimizing the Grain Boundary Carbide in Hadfield Steel
10.4028/www.scientific.net/AMM.598.38