Direct Partial Remelting of XW-42 Steel in Semi-Solid Zone
Microstructure that consists of spheroid grains in liquid matrix is desirable for semi-solid metal forming. In this study, a direct partial remelting from as annealed condition was used to examine microstructural evolution of XW-42 steel prior to thixoforming process. It was found out that M7C3 carbides dissolve inside the semi solid zone (between 1250-1270°C) while M23C6 carbides dissolve completely before entering semi-solid zone. The dissolution of these grain boundary carbides during partial remelting helps the grain spheroidisation of the XW-42 steel samples. Partial remelting at 1300 and 1340°C resulted in the formation of relatively spherical grains with liquid percent around 17%. The results indicate the suitability of these steel as a potential material for thixoforming through direct partial remelting route.
Received: February 01, 2010;
Accepted: April 28, 2010;
Published: June 10, 2010
Semi solid forming (also called thixoforming) is a new technology in which
the metal is processed taking advantage of its thixotropic properties. In a
thixotropic condition, an alloy can experienced a fall in viscosity if it is
sheared but it will thicken again if it is allowed to stand. The first experiment
to demonstrate this thixotropic behavior was conducted by Flemings and his co-workers
in the early 1970s (Kirkwood, 1994). Since then, efforts
have been made to put the new method at industrial level. The efforts were partially
succeeded with utilization of thixoforming process to manufacture components
made of low melting point alloys such as aluminium and magnesium alloys. One
example of this component is space frame node for an Audi car structure made
from MHD cast billets of AlMg5Si2Mn aluminium alloy (Hirt
et al., 2009). Nevertheless, further development of Al and Mg alloys
are still ongoing (Bayoumi et al., 2009; Chen
et al., 2009).
On the other hand, development of high melting point material for thixoforming
process hasn`t yet reached its goal (i.e., commercialization of steel thixoforming).
However, efforts to realize the goal are steadily being pushed due to the benefits
of steel thixoforming. Early studies mentioned that the benefits are low forging
force compared to conventional forging process, possibility of forming components
with complex and intricate shapes, reduction of steps in forming process, less
air entrapment and also less shrinkage porosity compared to conventionally casting
process (Omar et al., 2004a). Several steels
have been investigated as potential thixoforming material candidates. The result
seems promising for some steels such as X210CrW12 (Bramann
et al., 2004; Seidl and Kopp, 2004; Uhlenhaut
et al. 2006), AISI M2 tool steel (Omar et
al., 2004b), 100Cr6 bearing steel (Bulte and Bleck.,
2004; Seidl and Kopp, 2004) and high chromium cast
iron (HCCI) (Huang et al., 2006). Other research
on thixoforming of steels are focused on tools (Behrens
et al., 2004; Kopp et al., 2004) and
also supporting facilities for the process (Kuthe et
al., 2004; Pierret et al., 2008).
Regarding these developments, however, research on thixoforming materials is still much needed to satisfy material selection process in order to bring steel thixoforming toward larger scale industrial application. XW-42 cold work tool steel is being studied in this research as a potential starting material for thixoforming process. The scope of this present study covers the microstructural evolution of XW-42 steel when the steel is directly heated to semi-solid zone from as annealed condition (Direct Partial Remelting)
Material: The material used in this work is an XW-42 cold work tool
steel supplied in as-annealed condition (heated to 850°C, then cooled in
the furnace at 10°C h-1 to 650°C, after that cooled freely
in air). Chemical composition for the material was acquired using Foundry-Master
Spark Spectrometer that is based on Optical Emission Spectroscopy (OES) method.
JMatPro simulation: JMatPro or Java-based Material Properties is software
that was developed to augment the thermodynamic calculation by incorporating
various theoretical material models and properties database that allow a quantitative
calculation for the requisite materials property to be made within a larger
software structure (Saunders et al., 2003). A
modified Scheil-Guliver model is employed by the software to perform the simulation.
By using this model, liquid fraction profile as well as phase stability diagram
can be created.
DTA measurement: DTA measurement was carried out to estimate solidus
and liquidus temperatures as well as liquid fraction profile within the semi-solid
zone and carbide stability zone of studied material. The measurement was performed
using Netzsch Simultaneous Thermal Analysis (STA) 409°C equipment. The sample
was cut into small pieces with total weight of 50-100 mg with alumina as reference
material. Heating rate of 5°C min-1 was employed for the test.
The solidus and liquidus temperatures as well as liquid fraction profile were
constructed using Proteus software (Puttgen and Bleck, 2004).
Direct partial remelting experiment: The direct partial remelting experiment was performed using vertical, high temperature carbolite furnace. The material was cut into coupons of 5x10x12 mm. For each test, a coupon was hung inside the furnace by using chromel wire. To ensure that coupon was heated into designated temperature, a K-type thermocouple was placed nearby the coupon without making any direct contact with the coupons to prevent damages to the thermocouple. The tests were repeated for various selected temperatures (based on liquid fraction profiles from DTA measurement) with fixed holding time of 2 min and subsequently quench into brine solution to freeze the structures. The heating was carried out in an argon atmosphere to reduce oxidation. Heating rate of the furnace was found to be around 20°C min-1.
Metallography and image analysis: Coupons from direct partial remelting experiments were ground and polished to get clean and mirror-like surface. Subsequently, the coupons were etched using Villela reagent (1 g picric acid, 5 mL hydrochloric acid and 95 mL ethyl alcohol) to reveal their microstructures. Microstructures observation and analysis were conducted by using BX-51 Olympus optical microscope as well as Hitachi S3400N scanning electron
microscope (SEM), equipped with Energy Dispersive Spectroscopy (EDX) to analyze
chemical composition of the selected area/spots. EDX identifies chemical composition
of the selected area/spots by recognizing x-ray in which its energy amount is
unique for every element (Goldstein et al., 1981).
RESULTS AND DISCUSSION
Starting material: Table 1 shows chemical analysis
result for XW-42 steel. Standard specification (Steiner and
American Society for Metals, 1990) confirmed that chemical analysis result
from the alloy fall within the required standard. Optical micrographs of as
received sample are shown in Fig. 1a and b. It shows arrays
of large carbides with smaller carbides distributed homogeneously. This structure
is known to be found in an annealed cold work tool steel (Roberts
et al., 1971) thus confirms that soft annealing treatment has been
done as described by the material supplier. According to Roberts
et al. (1971), the carbides in XW-42 steel (designated as D2 by AISI)
are mostly from M7C3 type carbide which is confirmed by
JMatPro simulation (Fig. 2). Beside the predominant M7C3
type carbide, JMatPro simulation also estimates the existence of MC and M23C6
types of carbide in this steel. However the amount of both types of carbides
after cooling to room temperature is negligible compare to M7C3
SEM observation using back scattered mode was used to analyze the carbides
of the starting material in detail (Fig. 1c). It was found
that both large carbides and small carbides are darkish in appearances, indicating
their similarity in major constituent elements. EDX analysis has been performed
to analyze the type of carbides based on their chemical composition (Table
2). It is found that the darkish carbides are rich in chromium and iron
with atomic proportion between carbon and other metallic elements (Fe, Cr, Mo
and V) reaches 3:7 ratio. Hence the darkish carbides analyzed thus far (Table
2) are most likely an M7C3-type carbide. M7C3
and M23C6 carbides cannot being distinguished based
on back scattered imaging mode because both are Cr-rich type carbides (Laigo
et al., 2008).
||Chemical composition of XW-42 Steel (wt. %)
||EDX analysis results of the darkish carbides
||Micrographs of As-Received XW-42 Steel from (a, b) Optical
Microscope and (c) SEM
||Phase equilibrium diagram of XW-42 steel obtained from JMatPro
Note that phases that are rich in elements with high atomic numbers such as
W and Ta (to name a few) will appear bright in Fe matrix, while those that are
rich in elements with low atomic numbers, such as Cr and V, will appear greyish
or darkish (Goldstein et al., 1981). Furthermore,
as will be later explained, the carbide of importance in this study will be
the M7C3-type carbide, as it has given a huge influenced
to the grain boundary liquation when heated into the semi-solid zones.
Estimation of semi-solid temperature interval: Liquid Fraction Profile
(LFP) of XW-42 steel from DTA measurement and also from JMatPro simulation is
shown in Fig. 3. The solidus line for this alloy is estimated
at 1200-1220°C while the liquidus temperature is at 1400°C.
||Liquid fraction profile of XW-42 steel obtained from DTA and
Plateau shape is found on both curves but at different temperature and liquid
fraction ranges. For LFP created from DTA measurement, the plateau shape is
found at 1300-1343°C that corresponds to liquid percent around 40%. Whereas
for LFP from JMatPro, the plateau shape is found rather earlier at 1250-1260°C
that corresponds to liquid percent around 15%. The plateau shape becomes an
advantage in semi solid metal forming due to its low fluctuation in liquid fraction
at given temperature range, thus makes the forming process easier. Previous
work discovered that such plateau shape is caused by dissolution of certain
carbide(s) inside semi-solid zone (Puttgen and Bleck, 2004;
Omar et al., 2009; Meuser
and Bleck, 2001). By using the phase equilibrium diagram in Fig.
2 and also explanation about types of carbides in as received sample above,
it is strongly believed that the carbides that dissolved at this plateau region
are M7C3 carbides.
Temperatures for partial re-melting experiment were selected based on the liquid
fraction profile. The experiment were performed at sub solidus (1190°C)
and at semi-solid zone (1220, 1250, 1270, 1300 and 1340°C) The selected
temperatures are up to 50% liquid percent that was said to be the upper limit
of liquid percent for thixoforming process (Omar et al.,
2004a). Above this limit, some alloys exhibits fall in viscosity value which
could be detrimental to its thixotropic behaviour (Kirkwood,
Microstructures in sub-solidus and semi-solid zone: Microstructures of XW-42 Steel at 1190°C and 1220°C are shown in Fig. 4 a-c. By comparing Fig. 1c and 4c, it could be seen that at 1190°C some of the darkish carbides have already dissolved. It was believed that the dissolved carbides are all M23C6 carbides along with some M7C3 carbides. Thus, the remaining darkish carbide could be positively confirmed as M7C3 carbides due to its high dissolution temperature. EDX examination results also support the claim by obtaining the similar carbide chemical composition as in Table 2. Other than carbides, equiaxed austenite grains are also observed, especially at the SEM micrograph (Fig. 4c). However, it is only partly visible due to etchant attacking preference effect.
At 1250°C (Fig. 5 a) the M7C3 carbides
start to dissolve more rapidly and the equiaxed austenite grains in background
are now more visible. Also at this temperature, liquid phase should be appearing
at around 10% volume fraction according to DTA profile at Fig.
3. However, due its low quantity of liquid phase (<25% volume fraction),
partial remelting experiment failed to reveal the presence of liquid phase other
than thin film of dark phase at the grain boundaries (Fig. 5a).
This is attributed to insufficient cooling rates during quenching that enables
the liquid phase at the vicinity of grains to solidify epitaxially (Tzimas
and Zavaliangos, 2000a).
An increase in partial remelting temperature to 1270°C will dissolve all remaining M7C3 carbides. Furthermore, dark eutectics formed from quenched liquid phase start to appear at this temperature, indicating the much reduced epitaxial effect due to significant increase in liquid phase (up to 17% according to Fig. 3). Calculation on the dark eutectics phase (LFP from image analysis) observed at Fig. 7 shows that its average volume fraction reaches 12%. However, the dark eutectics seem to be grouped along rolling direction and not dispersed around the grains.
Further heating to 1300 and 1340°C (Fig. 6a, b)
increases the dark eutectics volume fractions to around 17% due to increase
in liquid phase up to 40% according LFP in Fig. 3. The increase
of dark eutectics volume fraction, which represents the increase of liquid phase,
is thought to give a significant surface tension that affect grain size and
spheroidity level. The grain spheroidity level is calculated by using shape
factor equation (Bergsma et al., 1997):
||Relation between liquid fraction profile, average grain diameter
and shape factor of XW-42 steel at semi-solid zone
where, F is the shape factor, U is the grain circumference and A is the grain area. In the case of F =1, the grains are fully spherical, while for more complex shape, the shape factor is >1.
Figure 7 shows the effect of partial remelting temperature
at 1270, 1300 and 1340°C to the average grain sizes and shape factor of
the austenite grains. The average grain diameters at these temperatures are
found to be around 40-60 μm. It is found that increase in partial remelting
temperature will reduce the grain size due to melting at the sharp asperities
of the equiaxed grains that is preceded by liquid penetration into the grain
boundaries. Then the melting is followed by solidification in regions of negative
curvature due to diffusion (Omar et al., 2009).
Hence, the reduction of the grain size is resulted in the decline of shape factor
value, which means better grain spheroidity level.
||Grain agglomeration at 1340°C that leads to grain boundary
elimination. Small unattached isolated grains will be slowly liquidated
due to Otswald Ripening
However, it also can be proven that for other alloys (such as AISI M2), grain
growth can overcome the reduction process and will result in bigger grain at
higher partial remelting temperatures without sacrificing its spheroidity (Omar
et al., 2009).
Grain agglomeration occurs for some grains at semi-solid zone as depicted at
Fig. 8. Agglomerates could have a beneficial effect for thixoforming
process by creating solid skeleton structure that will support the alloy in
high liquid content condition (Tzimas and Zavaliangos, 2000b).
Previous works show that the agglomerate consists of grains connected with low
angle grain boundary (i.e., with misorientation less than 10°) (Liu
et al., 2005; Arnberg et al., 1999).
Subsequent process of grain rotation and/or grain boundary migration minimizes
the misorientation and then eliminates the grain boundary line. On the other
hand, isolated unattached grains will be liquidated and become smaller due to
Otswald ripening process (Tzimas and Zavaliangos, 2000b).
In order to examine the suitability of an alloy for thixoforming based on its
microstructure prior to forming process, specific criterions for average grain
size and shape factor is used. Uggowitzer et al. (2000)
specified that the average grain size of an alloy must not exceed 100 μm,
while its shape factor must be below 2 in order to have good mould filling behavior
(Bünck et al., 2009). By using this criterions,
it is apparent that the average grain sizes and shape factor for XW-42 steel
at 1270, 1300 and 1340°C are suitable for thixoforming process because they
are in the range of 50-60 μm and 1.2-1.4, respectively (which is well below
the limit value of grain size and shape factor at 100 μm and 2, respectively).
However, unbalanced liquid phase distribution found at 1270°C is thought
to be detrimental for thixoforming process owing to its lack of wetting effect
during forming process. Hence, higher temperature (such as at 1300 and 1340°C)
is more recommended for forming process.
The microstructural evolution of XW-42 steel has been examined using direct partial remelting experiment. The as-received material shows arrays of large carbides with smaller carbides distributed homogeneously. The carbides were identified as M7C3 and M23C6 carbides. Further heating to sub solidus zone at 1190°C reveals the dissolution of M23C6 carbides while the M7C3 carbides are still observed after entering semi solid zone and finally dissolve between 1250-1270°C. Epitaxial effect during quenching reduces significantly the volume fraction of dark eutectics that represents the liquid phase. Subsequent calculation after partial remelting shows that the liquid fraction at 1250°C is 12% and increase to 17% at 1300-1340°C. Equiaxed grains found at 1250°C have transformed into more spherical grains at higher partial remelting temperatures (i.e., 1270, 1300 and finally 1340°C). Based on criterions for average grain size and shape factor and also acknowledging the distribution of the liquid phase around grains, it could be said that structures at 1300 and 1340°C are thought to be suitable for thixoforming process.
The authors would like to thank to Universiti Kebangsaan Malaysia (a.k.a. National University of Malaysia) and Ministry of Science, Technology and Innovation (MOSTI) Malaysia for sponsoring this work.
1: Arnberg, L., A. Bardal and H. Sund, 1999. Agglomeration in two semisolid type 6082 aluminium alloys. Mater. Sci. Eng., A262: 300-303.
2: Bayoumi, M.A., M.I. Negm and A.M. El-Gohry, 2009. Microstructure and mechanical properties of extruded Al-Si alloy (A356) in the semi-solid state. Mater. Des., 30: 4469-4477.
3: Behrens, B., B. Haller and D. Fischer, 2004. Thixoforging of steel using ceramic tool materials. Steel Res. Int., 75: 561-568.
4: Bergsma, S.C., M.C. Tolle, M.E. Kassner, X. Li and E. Evangelista, 1997. Semi-solid thermal transformations of Al-Si alloys and the resulting mechanical properties. Mater. Sci. Eng., 237: 24-34.
5: Bramann, H., C. Afrath, T. Grimmig, M. Fehlbier and A. Buhrig-Polaczek, 2004. Casting of high alloy steels in the mushy state. Steel Res. Int., 75: 537-544.
Direct Link |
6: Bulte, R. and W. Bleck, 2004. Effects of pre-processing on thixoformability of steel grade 100Cr6. Steel Res. Int., 75: 588-592.
7: Bunck, M., F. Kuthe and A. Buhrig-Polaczek, 2009. Rheocasting of Aluminium and Thixocasting of Steel. In: Thixoforming-Semi-Solid Metal Processing, Hirt, G. and R. Kopp (Eds.). Willey-VCH Verlag GmbH, USA., pp: 311-368.
8: Chen, T.J., Y. Ma, B. Li, Y.D. Li and Y. Hao, 2009. Effects of processing parameters on wear behaviors of thixoformed AZ91D magnesium alloys. Mater. Des., 30: 235-244.
9: Goldstein, J.I., D.E. Newbury, P. Echlin, D.C. Joy, C. Fiori and E. Lifshin, 1981. Scanning Electron Microscopy and X-Ray Microanalysis, A Text for Biologists, Materials Scientists and Geologists. 1st Edn., Plenum Press, New York.
10: Hirt, G., L. Khizhnyakova, R. Baadjou, F. Knauf and R. Kopp, 2009. Semi-Solid Forming of Aluminium and Steel-Introduction and Overview. In: Thixoforming-Semi-Solid Metal Processing, Hirt, G. and R. Kopp (Eds.). Willey-VCH Verlag GmbH, USA., pp: 1-27.
11: Huang, Z., J. Xing and A. Zhang, 2006. Investigation of microstructure and impact toughness of semisolid hypereutectic high chromium cast iron prepared by slope cooling body method. J. Applied Sci., 6: 1635-1640.
CrossRef | Direct Link |
12: Kirkwood, D.H., 1994. Semisolid metal processing. Int. Mater. Rev., 39: 173-189.
CrossRef | Direct Link |
13: Kopp, R., H. Shimahara, J.M. Schneider, D. Kurapov, R. Telle and S. Munstermann, 2004. Characterization of steel thixoforming tool materials by high temperature compression tests. Steel Res. Int., 75: 569-576.
Direct Link |
14: Kuthe, F., A. Schonbohm, D. Abel and R. Kopp, 2004. An automated thixo-forging plant for steel parts. Steel Res. Int., 75: 593-600.
Direct Link |
15: Laigo, J., F. Christien, R. Le Gal, F. Tancret and J. Furtado, 2008. SEM, EDS, EPMA-WDS and EBSD characterization of carbides in HP type heat resistant alloys. Mater. Characterization, 59: 1580-1586.
16: Liu, D., H.V. Atkinson and R.L. Higginson, 2005. Disagglomeration in thixoformed wrought aluminium alloy 2014. Mater. Sci. Eng., 392: 73-80.
17: Meuser, H and W. Bleck, 2001. Microstructural investigations in the semi-solid state of the steel X210CrW12. Steel Res. Int., 72: 271-276.
18: Omar, M.Z., H.V. Atkinson, E.J. Palmiere, A.A. Howe and P. Kapranos, 2004. Microstructural development of HP9/4/30 steel during partial remelting. Steel Res. Int., 75: 552-560.
19: Omar, M.Z., E.J. Palmiere, A.A. Howe, H.V. Atkinson and P. Kapranos, 2004. Thixoforming of GFM M2 tool steel from the as-annealed condition. Proceeding of the 8th International Conference on Semi-Solid Processings of Alloys and Composites, (ICSSPAC`04), Limassol, Cyprus, pp: 521-531.
20: Omar, M.Z., H.V. Atkinson, A.A. Howe, E.J. Palmiere, P. Kapranos and M.J. Ghazali, 2009. Solid-liquid structural break-up in M2 tool steel for semi-solid metal processing. J. Mater. Sci., 44: 869-874.
CrossRef | Direct Link |
21: Pierret, J.C., A. Rassili, G. Vaneetveld, J. Lecomte-Beckers and R. Bigot, 2008. Robustness of a thixoforming production line. Solid State Phenomena, 141-143: 207-212.
Direct Link |
22: Puttgen, W. and W. Bleck, 2004. DTA-Measurements to determine the thixoformability of steels. Steel Res. Int., 75: 531-536.
Direct Link |
23: Roberts, G.A., J.C. Hamaker and A.R. Johnson, 1971. Tool Steels. 2nd Edn., ASM, USA.
24: Saunders, N., Z. Guo, X. Li, A.P. Miodownik and J. Schille, 2003. Using JmatPro to model materials properties and behavior. JOM J. Minerals Metals Mater. Soc., 12: 60-65.
CrossRef | Direct Link |
25: Seidl, I. and R. Kopp, 2004. Semi-solid rheoforging of steel. Steel Res. Int., 75: 545-551.
Direct Link |
26: Tzimas, E. and A. Zavaliangos, 2000. Evaluation of volume fraction of solid in alloys formed by semi-solid processing. J. Mater. Sci., 35: 5319-5330.
CrossRef | Direct Link |
27: Tzimas, E. and A. Zavaliangos, 2000. Evolution of near-equiaxed microstructure in the semi-solid state. Mater. Sci. Eng., 289: 228-240.
28: Uggowitzer, P.J., G. Gullo and A. Wahlen, 2000. Metallkundliche aspekte der semi-solid formgebung von leichtmetallen. First Ranshofener Leichtmetalltage. http://e-collection.ethbib.ethz.ch/eserv.php?pid=eth:25645&dsID=eth-25645-01.pdf.
29: Uhlenhaut, D.I., J. Kradolfer, W. Puttgen, J.F. Loffler and P.J. Uggowitzer, 2006. Structure and properties of a hypoeutectic chromium steel processed in the semi-solid state. Acta Materialia, 54: 2727-2734.
Direct Link |
30: Steiner, R. and American Society for Metals, 1990. ASM Handbook Volume 1: Properties and Selection: Irons, Steels and High-Performance Alloys. 10th Edn., ASM International, USA., ISBN-10: 0871703777.