Materials and Design 56 (2014) 345352

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matdes

Investigation on gas metal arc weldability of a high strength tool steel

C. Mandolno a,, E. Lertora a, L. Davini b, C. Gambaro a
a b

University of Genoa, Department of Mechanical Engineering, Via All Opera Pia 15, 16145 Genoa, Italy Ansaldo Energia S.p.a, Via Nicola Lorenzi 8, 16152 Genoa, Italy

a r t i c l e

i n f o

a b s t r a c t
In this paper, gas metal arc weldability results of a particular advanced tool steel are presented. Indeed, the study was focused on the weld prole, microhardness and microstructure of the joints. The aim was to identify an appropriate ller material and optimize the process parameter. The validation of results started with a careful metallographic analysis of the joints, in order to verify that the metallurgical properties of the material were not compromised by the welding process. In the following step, all the non-destructive and mechanical tests, imposed by procedure qualication, were performed in order to have a complete characterization of the joints. For all the wires used, hardness tests highlighted that the use of low heat input and a high number of beads causes an increase in the Heat Affected Zone (HAZ) hardness up to values equal to or exceeding the limits imposed by the European standard on the process qualication. To avoid this problem, it was therefore necessary to adopt high electric parameters and thus high heat inputs. The ller material that gave the best results, in terms of uniformity of mechanical properties, is the rutile ux wire. 2013 Elsevier Ltd. All rights reserved.

Article history: Received 8 July 2013 Accepted 19 November 2013 Available online 28 November 2013 Keywords: Tool steel Weldability Gas metal arc welding Welding procedure qualication

1. Introduction Tool steels are mainly used for machining and nishing materials, through operations such as turning and milling, and for mould realization. Because of the strict working conditions to which they are subjected, they must have excellent mechanical properties, such as hardness, toughness, wear and deformation resistance, and it is necessary to maintain these properties even at high temperatures [1]. Special high strength steels for tools are the Toolox series, produced by SSAB Oxelosund (Sweden). They are quenched and tempered steels, produced with an integrated process, able to guarantee a constant and accurate control of chemical composition. Toolox is based on the concept of metallurgical low carbon content and rapid cooling during quenching: the result is a steel characterized by a particular morphology and by a much greater toughness compared to steels of similar hardness. These characteristics ensure high wear resistance and an increased productivity rate for tools made with this material, even after long periods at high temperatures [1,2]. These kinds of steels are widely studied concerning their wear resistance and the mechanical characteristics in high temperature processes, including their microstructural modication [1,38].

Corresponding author. Tel.: +39 0103532679; fax: +39 010317750.

E-mail addresses: chiara.mandolno@unige.it (C. Mandolno), e.lertora@unige.it (E. Lertora), luca.davini@aen.ansaldo.it (L. Davini), gambaro@diptem.unige.it (C. Gambaro). 0261-3069/$ - see front matter 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.matdes.2013.11.042

Zhang et al., for example, investigated the microstructural evolutions of a martensitic hot-work tool steel during tempering and service, in order to control the tool lifetime. They also proposed a tempering ratio to describe the evolution of different types of hardness with temperature and time, during the tempering of martensitic steel [3]. Concerning the abrasive wear behavior, Colao and Vilar studied the relationship between the microstructure of a martensitic stainless tool steel and the abrasive wear coefcient. Their work shows that at lower loads, the material with a microstructure formed of martensite and carbide particles presents the higher wear resistance. On the contrary, at higher loads, a microstructure formed of martensite and 1525% of retained austenite presents a higher wear resistance [4]. Medvedeva et al. and Firrao et al. correlated the microstructure of tool steels for different applications with both their static and dynamic properties, especially in high temperature conditions [5,6]. Luo et al. in two of their works [7,8] investigated in depth the behavior of non-quenched prehardened steel for a large section plastic mould with particular attention to microstructure and hardness uniformity, machinability and few references to its Tungsten Inert Gas (TIG) weldability. From these studies it could be established that hardness is one of the main features to take into account while considering the machining, including welding, of these materials. Translating these excellent characteristics from machining tools to other kinds of tools, such as forks, knives and buckets for earth

346

C. Mandolno et al. / Materials and Design 56 (2014) 345352

moving machines, its metallurgical, mechanical and wear-resistant characteristics make Toolox an excellent choice for heavy duty applications, such as high temperature environments. For all these reasons, its use in the manufacturing of these and other welded structures makes for a very interesting possibility. Furthermore, other high strength steels are effectively welded through gas metal arc welding techniques, and there are several studies in the literature that show the effect of process parameter changing on the quality of the welded joint [915]. et al. in three works emphasize the importance of the Lazic selection of the optimal procedure and technology for welding high strength steels, with particular reference to Weldox 700, developed by the same producer of the Toolox series. After a detailed analysis of the properties of the base metal and the evaluation of the main aspects related to its weldability, they selected the optimal combination of ller materials, methods and technologies of welding, as well as the conduction of a model and other standard tests, establishing the optimal technology of welding, which was then applied to a very secure welded structure [911]. The welding techniques are also largely and effectively employed to functionalised steels, for example realizing tool steel hardfacing deposits, as studied by Gualco et al. and by Coronado et al. [12,13]. In this case too, the investigation into welding parameters, heat input and shielding gas is essential to realize a durable deposit. Magudeeswaran et al. analyzed in detail the effect of welding consumables and processes on tensile, impact and fatigue properties of armour grade quenched and tempered joints fabricated by ux cored arc welding (FCAW) processes. In particular they stated that the use of low hydrogen ferritic steel consumables is found to be benecial to enhance the tensile properties and fatigue resistance of these steel joints, also compared to the joints fabricated by conventional austenitic stainless steel consumables, which are much more expensive [14,15]. On the contrary, only little information on the weldability of tool steel is reported, because welding processes occur mainly in repair operations [16]. In order to nd detailed information about industrial applications of welded components with this particular material, using Gas Metal Arc (GMA) process, a careful experimental campaign was carried out. In particular, the aim was the identication of an appropriate ller material and the process parameters ensuring the best quality of the joints. The validation of results started with a careful metallographic analysis of the joints, in order to conrm that the metallurgical properties of the material were not compromised by the welding process. In the following step, all the non-destructive and mechanical tests, imposed by procedure qualication, were performed to have a complete characterization of the joints and verify industrial applicability. 2. Materials and methods 2.1. Base metal and ller material Toolox 33 is a quenched and tempered steel, marked by very low residual stress and good dimensional stability. It offers low carbon content and its production process is marked by a rapid cooling: this leads to a particular carbide morphology which ensures limited wear and high production rate, even after regular use at elevated temperatures. In spite of the elevated hardness, Toolox 33 preserves very high toughness values, especially compared to steels with the same hardness, making it an effective and better choice [1,2].

Fig. 1 shows Toolox 33 martensitic microstructure and Fig. 2 the ne carbide dispersion in its matrix. Due to the high content of alloying elements, this steel has a carbon equivalent, evaluated with the CEV (Carbon Equivalent Value) index [18,19], which is quite high. The specic casting used for the tests (Table. 1), for example, has a CEV index equal to 0.62, for which some precautions during welding were suggested [2,17,18]:      preheat temperature of about 170 C; low-hydrogen ller materials (max 5 ml/100 g); heat input in order to have Dt8/5 between 10 and 20 s; minimum interpass temperature of 170 C; post-heating treatment of about 200 C maintained for 120 min.

For the welding tests, 14-mm-thick sheets of Toolox 33 were used for the fabrication of single V butt joint conguration. Table 2 reports the specimen dimensions and the indication of bevel angle and the side edge of root face. Three types of wire were used as ller material:  metal cored wire, a tubular electrode that consists of a metal sheath and a core of various powdered materials, primarily iron.  rutile ux cored wire, which gives a remarkable uidity to the weld pool and a good nishing of the bead. Its arc stability is quite high and usually the joints are free of spatter and its slag is easily removable. This ux, however, does not provide any purifying action of the weld metal.  basic ux cored wire, which is lled with iron oxides, ferroalloys of manganese and silicon, silicates and carbonates, especially calcium and magnesium. Fluorite (calcium uoride) is usually added to make the arc ignition easier. Indeed, calcium and magnesium carbonates are used as puriers but their melting temperature is quite high [1719]. As a general statement, ller material with yield strength (Rp0,2) up to 700 MPa has a CEV index lower than base material while very high yield strength ller materials have CEV index higher than base material. In the latter case, it is necessary to pay attention to the thermal cycle, since this kind of material is sensitive to high interpass temperatures [1719]. In Table 3, the mechanical properties of the specic ller materials selected are reported. All the data are referred to supplier Inspection Certicate, class 3.1.

Fig. 1. Toolox 33 martensitic microstructure, 100.

C. Mandolno et al. / Materials and Design 56 (2014) 345352 Table 2 Specimen dimensions. Length (mm) 300 Width (mm) 55 Thickness (mm) 14 Bevel angle () 60

347

Side edge of root face (mm) 3

Table 3 Geometrical and mechanical characteristics of ller materials used for welding tests. Type of ller material Metal cored Diameter (mm) Rp0.2 (N/mm2) Rm (N/mm2) Strain (%) kV (J) 20 C 30 C 40 C 1.2 495 570 26 90 60 Rutile 1.2 786 806 17 Basic 1.2 562 602 26

Fig. 2. Carbide dispersion in Toolox 33, 1000.

84 84

118

2.2. Welding tests and joint characterization 3. Results and discussion The joints were realized using the ller materials selected, with gas metal arc welding technique, using an ArCO2 blend as shielding gas and spray-arc as metal transfer mode. The experimental campaign was carried out in two phases. The rst focused on research and optimization of welding parameters, both electrical (voltage and current) and thermal (preheating, interpass and post heating temperatures). This investigation made it possible to obtain the information necessary for the choice of size and number of beads required to ll the welding gap, and identication of the ller materials most suitable for the application. Visual examination, macrographic and micrographic analyses and hardness tests were conducted to assess absence of macroscopic imperfections, degree of surface nish and mechanical characteristics, especially in the heat affected zone. For each ller material, the joints were realized in two different ways using, when possible:  low heat input and therefore greater number of beads;  high heat input and thus fewer beads. Table 4 shows some details of the welding performed in the rst phase, reporting the welding sequence, the number of beads, the welding parameters and the values of average heat input used for each joint. All joints were realized with a preheating to 175 C and, after welding, a post-heating for 2 h at 200 C, followed by slow cooling in the furnace to room temperature, in order to facilitate hydrogen emission and avoid crack occurrence. The maximum interpass temperature was maintained at 225 C [2]. In the second phase, a joint with parameters, ller material and thermal cycle that had guaranteed the best results in the rst stage was realized. This joint was completely investigated through non-destructive testing, mechanical and structural characterization, obtaining a qualied procedure, according to UNI EN ISO 15614-1 [20]. 3.1. First phase joints After welding, the specimens were rst examined by visual inspection, all resulting free from imperfections. In order to investigate the inuence of heat input and type of wire on the microstructure, they were cut from transverse sections and the cut surfaces prepared for metallographic inspection by polishing and etching using a 2% Nital solution, to display weld shape and microstructure. Microhardness was evaluated using a Vickers hardness tester at a load of 10 N. The macrographic analysis of the specimens revealed perfect fusion, full penetration and complete absence of porosity and inclusions. Fig. 3 shows the macrographic investigation of the samples, in which the various zones of the joint and the bead distribution are clearly visible. The type of wire used inuenced the possibility of using both high and low heat input approach, or only one of them. In fact, calcium carbonates and uorite in the basic ux wire, make arc stability at low parameters really difcult [17,19]. Indeed, with this kind of wire it was not possible to realize the comparison between the two approaches and the parameters adopted were the highest among all wires, to ensure good arc stability and a correct realization of the joint. Always referring to Fig. 3, the HAZ is generally greater and more evident for higher heat input values adopted, as could be imagined [1719]. Fig. 4 highlights a considerable difference of the HAZ microstructure in Toolox 33 (close to the nishing bead, both taken from the right side) between two specimens: the rst underwent a heat input of 1 kJ/mm (A) and the second experienced a double value (B). A ner grain structure of low heat input samples allows us to predict higher mechanical properties in this area of the joint [19,21]. From the operative point of view, metal cored wire is very easy to manage. Rutile cored wire, however, does not need to enlarge

Table 1 Chemical composition [% weight] of the specic casting of Toolox 33. C 0.22 Si 1.07 Mn 0.79 P 0.008 S 0.001 Cr 1.04 Ni 0.06 Mo 0.19 V 0.098 Ti 0.014 Cu 0.01 Al 0.013 Nb 0.017 B 0.002 N 0.003

348

C. Mandolno et al. / Materials and Design 56 (2014) 345352

Table 4 Number of beads and heat input values used for the rst phase of test campaign. Welding sequence Filler material Metal cored Bead number 18 910 Current (A) 203 210 Voltage (V) 27.0 27.7 Average heat input (KJ/mm) 0.64

Metal cored

1, 2, 4 3, 5

235 203

28.0 27.0

Rutile

18

206

25.0

0.91

Rutile

1, 2, 4, 5, 6 3

235 206

25.5 25.0

1.3

Basic

1, 4 2, 3, 5, 6, 7

220 240

27.0 28.0

2.05

weld pool, to avoid slag cooling and its solidication at bead sides, increasing the risk of inclusions. This problem is offset by the fact that joints realized with this type of wire, unlike other joints, do not show dendritic structure in high heat input beads: micrographic analysis conrms that the structure is very ne in every part of the fusion zone (Fig. 5), promising higher strength values compared to other llers. The image was taken from the middle of the fusion zone of the joint realized with rutile wire and high heat input. Indeed it is possible to observe the effect of grain renement of the last beads on the rst. To conrm the microstructure remarks, microhardness tests Vickers HV1 [22] were carried out, collecting three lines of sampling for each joint. Standard sampling method is reported in Fig. 6. In Fig. 7, the microhardness values of joint nishing beads (Pos. 1) are compared to the average hardness value of the base material. The realization of the two types of passes (fewer and greater number of beads) has allowed us to compare the different approaches, in order to verify how a large heat input affects the nal characteristics of the joint and whether there are obvious reasons to prefer one method over the other. Unlike what commonly happens to quenched and tempered steel [8,13,14], the most marked result is that all joints

experimented the same tendency: hardness, compared to the base material, increases in the HAZ. This is even clearer on the right side of the joint (indentation 101112), in which the low heat input samples experimented the highest values. This is due to the fact that the heat affected zones corresponding to these indentations underwent a heat treatment produced by the realization of nal beads on the left side, which causes the effect of grain rening already reported in Fig. 4A. This renement is connected to an increase of mechanical properties and therefore of the zone hardness [17,19]. This might become a problem for the process qualication: in fact, it is necessary to keep hardness increasing in HAZ under control, in order not to reach maximum levels permitted by UNI EN ISO 15614-1 standard, which limits the maximum hardness to 450 HV10 for this class of material, not heat treated after welding. On the other hand, the fusion zone experimented a softening, since, thanks to the extraordinary Toolox 33 characteristics, it is very difcult to guarantee performance of the same level using conventional ller materials, without employing super alloys that would increase production costs excessively [17,18,20].

C. Mandolno et al. / Materials and Design 56 (2014) 345352

349

Fig. 3. Macrographic investigation on Toolox 33 joints, etching Nital 2%.

Fig. 4. Microstructures in the Toolox 33 HAZ produced with different parameters and types of wire, etching Nital 2% (50).

Fig. 5. Fusion zone (FZ) microstructure of Toolox 33 using high strength rutile wire, etching Nital 2%.

In any case, the best results were given by rutile ux wire and no marked difference between the two approaches occurred. For the joint realized with this wire, the performance gap between base material, HAZ and the weld metal remains quite small, especially for the high heat input one, giving a more regular behavior to the material. Considering the discrete user friendliness and the high performance revealed, the rutile ux cored wire was considered

the best choice for the implementation of the second test campaign. 3.2. Welding procedure qualication In this second phase, a joint realized with the rutile-ux wire, using high heat input-fewer beads approach and all the parameters that guaranteed the best results in the previous stage, was realized.

350

C. Mandolno et al. / Materials and Design 56 (2014) 345352

Fig. 6. Indentation position for hardness test.

Fig. 7. Hardness trend of joint nishing beads and average HV1 value of base material.

This joint was completely investigated through non-destructive testing, mechanical and structural characterization, obtaining a qualied procedure, according to UNI EN ISO 15614-1 [20]. Visual and penetrant test [23,24], showed no evidence of surface imperfections, good nishing and bead regularity. The same good result was conrmed by the third non-destructive test, the X-ray examination [25], by which the joint is considered acceptable. In the same way, macrographic investigation [26] shows a very symmetrical and regular transverse section, with very well distributed beads, absence of lack of penetration and slag inclusions, and a very at nishing bead (Fig 8). The nishing pass was realized in a unique bead, to avoid the irregular hardness behavior registered in the rst part of the test campaign. The absence of internal imperfections is also conrmed by the successful results of bending test [27], performed reaching an angle slightly smaller than 180. No principle of failure occurred, also thanks to the excellent behavior of both base and ller material (Fig. 9).

Fig. 8. Macrograph of welded joint, obtained using rutile-ux wire as ller material, Nital 5% etching.

Tensile tests required for process qualication [28] were conducted on two specimens, with a deformation rate of 0.007 1/s. They gave excellent results, as expected, even though the failure always occurred in the fusion zone, which represents the weakest area of the whole joint. As already mentioned, in the case of very high-strength base material, it is especially difcult for the fusion zone to have the same performance as the base material using conventional ller materials. This issue is not always negative, et al. in several studies rebut has some advantages. Indeed, Lazic ported that by application of ller materials with lower strength compared to the base metal, it is possible to achieve the higher toughness of the welded joint, a higher resistance to cold cracks and lower residual stresses in the welded joint, especially in HAZ [10,11]. In any case, the value of the welded joint tensile strength is quite high, considering that the certied value by the supplier reported a Rm equal to 806 MPa, in tests carried out on the ller material only. The efciency of the joint was also calculated, comparing the value to which the products break with the base material Rm. As shown in Table 5, the joint efciency reaches good levels, between 80% and 90%. The joint realized showed, in addition to a good ductility, also a reasonable toughness, evaluated by impact tests [29], performed at 20 C, providing high energy values summarized in table 6. As a useful comparison, it has to be reported that the Charpy energy of base material is 96 J [2]. The samples are identied by the terms VWT, if notched in the melted zone, and VHT, if notched in the heat affected zone. Finally, Fig. 10 reports Vickers test results. They were performed with a 100 N load, according to the standard [30]. The sampling method was the same as reported in Fig. 6.

C. Mandolno et al. / Materials and Design 56 (2014) 345352

351

Fig. 9. Results of bending tests on the specimens. Detail of the plastic deformation of the molten zone (B C).

Table 5 Tensile test report. Sample F1 F2 Rm base material (MPa) 962 Rm welded joint (MPa) 837 841 Failure zone Fusion zone Fusion zone Efciency 87% 87%

4. Conclusions The weldability of a particular tool steel, Toolox 33, using GMA process was investigated, carrying out a careful experimental campaign. Homogeneous joints using three kinds of ller materials and two different heat input approaches were realized. The development of the optimal parameters for each ller wire heat input couple was the rst step. The results of macrographic and micrographic analysis and hardness test on the rst samples allowed the scheduling of a second test campaign, using the ller material that gave the best results. From the nal outcomes, the following conclusions can be drawn. (1) Regarding the operational characteristics of ller materials, the metal cored was the easiest to manage in a wide range of parameters. On the contrary, calcium carbonates and uorite, in basic wire, make arc stability at low parameters really difcult. In fact, to ensure a good arc stability and a correct

Table 6 Results of Charpy test at 20 C. Charpy energy (J) VWT VHT 65.5 168.1 61.5 161.0 61.4 150.5

The trend of the curves is quite irregular but it conrms what has been seen previously. From Fig. 10, it is possible to see how the fusion zone settles the expected high values and how the presence of only one nishing bead contributes to the limited hardness increasing in HAZ on both sides.

Fig. 10. Hardness HV10 values on the rutile-ux wire specimen.

352

C. Mandolno et al. / Materials and Design 56 (2014) 345352 [8] Luo Yi, Wu Xiao-chun, Wang Hong-bin, Min Yong-an. A comparative study on non-quenched and quenched prehardened steel for large section plastic mould. J Mater Process Technol 2009;209:543742. M, Lazic V, Marinkovic P, Nedeljkovic B, Aleksandrovic S, [9] Jovanovic Milosavljevic S. Estimates of weldability and selection of the optimal procedure and technology of welding of high strength steel WELDOX 700. 26th Meeting Welding 2010, Tara, Serbia; 2010. p. 82. V, Aleksandrovic S, Nikolic R, Prokic -Cvetkovic R, Popovic O, [10] Lazic D, et al. Estimates of weldability and selection of the optimal Milosavljevic procedure and technology for welding of high strength steels. Procedia Eng 2012;40:3105. V, Adamovic D, Josifovic D, Jovanovic M. Contribution to dening the [11] Lazic limiting parameters of the structural high strength steels welding. Fac Mech Eng Kragujevac 1995; 133140. [12] Gualco A, Svoboda HG, Surian ES, De Vedia LA. Effect of welding procedure on wear behaviour of a modied martensitic tool steel hardfacing deposit. Mater Des 2010;31:416573. [13] Coronado JJ, Caicedo HF, Gomez AL. The effects of welding processes on abrasive wear resistance for hardfacing deposits. Tribol Int 2009;42:7459. [14] Magudeeswaran G, Balasubramanian V, Madhusudhan Reddy G. Effect of welding processes and consumables on high cycle fatigue life of high strength, quenched and tempered steel joints. Mater Des 2008;29:18217. [15] Magudeeswaran G, Balasubramanian V, Madhusudhan Reddy G, Balasubrarnaniarn TS. Effect of welding processes and consumables on tensile and impact properties of high strength quenched and tempered steel joints. J Iron Steel Res Int 2008;15(6):8794. [16] Vedani M, Previtali B, Vimercati GM, Sanvito A, Somaschini G. Problems in laser repair-welding a surface-treated tool steel. Surf Coat Technol 2007;201:451825. [17] Saldatura per fusione vol. II Metallurgia e saldabilit dei materiali metallici Qualit e controllo dei giunti saldati. Hoepli editore; 2006. [18] Saldatura per fusione vol. I Procedimenti di saldatura economia sicurezza e salute. Hoepli editore; 2006. [19] ASM Metals Handbook vol. 6 Welding, brazing and soldering; ASM International; 2005. [20] UNI EN ISO 15614-1: Specication and qualication of welding procedures for metallic materials-Welding procedure test Part 1: Arc and gas welding of steels and arc welding of nickel and nickel alloys. [21] ASM Metals Handbook vol. 9 Metallography and Microstructures; ASM International; 2005. [22] UNI EN ISO 9015-2: Destructive tests on welds in metallic materials Hardness testing Part 2: Microhardness testing on welded joints. [23] UNI EN 970: Non-destructive testing of welds Visual testing of fusion welded joints. [24] UNI EN 571-1: Non-destructive testing Penetrant testing Part 1: General principles. [25] UNI EN 1435: Non-destructive testing of welds Radiographic examination of fusion-welded joints. [26] UNI EN 1321: Destructive tests on welds in metallic materials Macroscopic and microscopic examination of welds. [27] UNI EN ISO 5173: Destructive tests on welds in metallic materials Bend tests. [28] UNI EN ISO 4136: Destructive tests on welds in metallic materials Transverse tensile test. [29] UNI EN ISO 9016: Destructive tests on welds in metallic materials Impact tests Test specimen location, notch orientation and examination. [30] UNI EN ISO 9015-1: Destructive tests on welds in metallic materials Hardness testing Part 1: Hardness test on arc welded joints.

realization of joints with this wire, the parameters had to be maintained the highest among all wires. Using rutile wire, however, is better not to enlarge weld pool, to avoid slag cooling and its solidication at bead sides, increasing the risk of inclusions. (2) For all the wires used, hardness tests highlight that the use of low heat input and a high number of beads causes an increase in the HAZ hardness up to values equal to or exceeding the limits imposed by the European standard on the process qualication. To avoid this problem, it is therefore necessary to adopt high electric parameters and thus high heat inputs. (3) The ller material that gave the best results, in terms of uniformity of mechanical properties, is the rutile ux wire. In fact, considering also the good user friendliness, it was considered the best choice for the implementation of the second test campaign. (4) The second stage of this study established that GMA process associated with a suitable ller material can be successfully used in Toolox 33 welding, with high productivity rate and user-friendliness, ensuring process quality imposed by recent standards.

Acknowledgements The authors wish to thank Lameter S.r.l and Ansaldo Energia for their technical support, the material supply and all their assistance in carrying out this investigation. References
[1] Davis & Associates. ASM specialty handbook tool materials. ASM International; 1995. [2] Available from <www.ssab.com>, accessed on May 2013. [3] Zhang Z, Delagnes D, Bernhart G. Microstructure evolution of hot-work tool steels during tempering and denition of a kinetic law based on hardness measurements. Mater Sci Eng A 2004;380:22230. [4] Colao R, Vilar R. On the inuence of retained austenite in the abrasive wear behaviour of a laser surface melted tool steel. Wear 2005;258:22531. [5] Medvedeva A, Bergstrm J, Gunnarsson S, Andersson J. High-temperature properties and microstructural stability of hot-work tool steels. Mater Sci Eng A 2009;523:3946. [6] Firrao D, Matteis P, Scavino G, Ubertalli G, Ienco MG, Pinasco MR, et al. Relationships between tensile and fracture mechanics properties and fatigue properties of large plastic mould steel blocks. Mater Sci Eng A 2007;468:193200. [7] Luo Yi, Wu Xiao-chun, Wang Hong-bin, Min Yong-an. Study on microstructure and hardness uniformity of non-quenched prehardened steel for large section plastic mould. Mater Sci Eng A 2008;492:20510.