Microstructural Evolution with Various Ti Contents in Fe ... - Springer Link

Met. Mater. Int., Vol. 20, No. 4 (2014), pp. 701~712 doi: 10.1007/s12540-014-4015-0

Microstructural Evolution with Various Ti Contents in Fe-Based Hardfacing Alloys Using a GTAW Technique Chih-Chun Hsieh, Yi-Chia Liu, Jia-Siang Wang, and Weite Wu* National Chung Hsing University, Department of Materials Science and Engineering, 250 Kuo-Kuang Rd., Taichung 402, Taiwan (received date: 3 May 2013 / accepted date: 26 November 2013) The aim of this study is to discuss the effect of microstructural development with different Ti contents in Fe-based hardfacing alloys. A series of Fe-Cr-C-Si-Mn-xTi alloy fillers was deposited on SS400 low carbon steel substrate using oscillating gas tungsten arc welding. The microstructure in the Fe-based hardfacing alloy without Ti content addition included: the primary , eutectic +(Fe,Cr)3C, eutectic +(Fe,Cr)2C and martensite. With increasing Ti contents, the microstructures showed the primary TiC carbide,  phase and eutectic +(Fe,Cr,Ti)3C. The amount and size of TiC carbide in the hardfacing layers increased as the Ti content increased. However, the eutectic +(Fe,Cr,Ti)3C content decreased as the Ti content increased. According to the results of the hardness test, the lowest hardness value (HRC 54.93) was found with 0% wt% Ti and the highest hardness (HRC 60.29) was observed with 4.87 wt% Ti. Keywords: alloys, welding, microstructure, mechanical properties, precipitation

1. INTRODUCTION The hardfacing alloy cladding technique melts and coats a matrix surface with an alloy filler by means of welding. The metallurgical junction between the cladding layer and the matrix surface is performed, and improves the wear and corrosion resistance of materials using this technique. The composition of the alloy filler can affect the microstructures and mechanical properties of the cladding layers. Generally, the hard microstructures include martensites, carbides (M23C6, M7C3, TiC, WC, VC and NbC) and oxides (Al2O3, ZrO2, and Y2O3). Damage due to wear and corrosion can be decreased when these hard phases form [1]. Therefore, hardfacing alloy cladding can increase both working life and working efficiency through the properties of the cladding layers [2,3]. Different hardfacing alloys are required for use in various wear environments. Many kinds of hardfacing alloys, including: Fe base, Ni base, Co base and Cr base [4-7], have been developed. Ferrous-based hardfacing alloys are widely used in many industries. Fe-Cr-C hardfacing alloys are cheaper than other hardfacing alloys. The properties of Fe-Cr-C hardfacing alloys are affected by chemical composition, dilution ratio, welding parameters and microstructure [8-11]. Some researchers [12-17] have pointed out that the amount of ele*Corresponding author: [email protected] KIM and Springer

ment additions can affect the microstructural morphologies of Fe-Cr-C hardfacing alloys, and that various morphologies have different mechanical strengths. In earlier investigations, scholars reported that the morphology and grain size of the primary carbides in the Fe-CrC alloy could be improved by controlled alloy design [18-20]. The strengthening phases in the Fe-Cr-C hardfacing alloys are the carbides of the M23C6 and M7C3. However, Ti has excellent mechanical strength and is suitable as a strengthening element in hardfacing alloys. Because of the ease with which Ti forms a carbide with C, TiC can be formed preferentially during solidification. Furthermore, the effect of Ti content on the microstructural development in the Fe-Cr-C hardfacing alloys using oscillating gas tungsten arc welding is unclear. Therefore, this study investigates the microstructures and mechanical properties of Fe-Cr-C-Si-Mn-xTi hardfacing alloys with different Ti contents.

2. EXPERIMENTAL PROCEDURES 2.1. Experimental Flow Fe-6Cr-1.9C-0.7Si-0.7Mn-xTi alloy fillers with various Ti contents were deposited on low carbon steel substrate using autogenous GTAW. A series of analyses of microstructures, chemical compositions, and mechanical properties was then performed using different equipment. In this study, the experimental flowchart is shown in Fig. 1.

702

Chih-Chun Hsieh et al.

Fig. 2. Schematic diagram of hardfacing cladding technique. Table 3. Welding parameters of hardfacing alloys Voltage Current

Fig. 1. Experimental flowchart.

Type Size Type Flow Travel speed Oscillate speed

Electrode

2.2. Alloy design of hardfacing alloy and coated substrate material Low carbon steel (SS400) was selected as a coated substrate material; its chemical composition is listed in Table 1. The dimensions of the substrate material are 100 mm×100 mm×19 mm. Before welding, the oxidation and oil sludge on the surface was cleaned using a grinding wheel machine and acetone. A series of metal powders was used in this investigation. The purities of the C and Cr powders used were 100% and 99.2%, respectively, and those of the Ti and Mn powders used were both 99.9%. The compositional ratio of the Si powder used was 3:1 (Si:Fe). These compositional designs of alloy fillers are listed in Table 2. 2.3. Cladding process Various ratios of metal powders were mixed uniformly, and placed in a metal mold (dimensions: 60 mm×25 mm×3 mm). These metal powders were manufactured into block metals, and then placed on the SS400 low carbon steel substrate. Then, two block metals were coated onto the SS400 low carbon steel substrate using GTAW. During the welding, Table 1. Chemical composition of low carbon steel (SS400) substrate (wt%) C 0.11

Si 0.19

Mn 0.51

P 0.025

S 0.026

Fe 99.139

Protective gas Welding speed

15.5 (V) 230 (A) W-2% ThO 3.2 (mm) Argon 15 (l/mm) 25 (mm/min) 195 (mm/min)

coating was performed using an autogenous arc oscillator. The coating method and parameters are shown in Fig. 2 and Table 3. 2.4. Compositional analysis of hardfacing layer The coated samples were flattened using a grinding machine, and cut into sizes greater than 20 mm. The coated specimens were, then ground using #60 sandpaper, and the surfaces of the samples were cleaned with acetone. The compositions of the hardfacing layers were analyzed using a spark discharge spectrometer (ARC-MET8000). Argon was used as a protective 1 gas, and its flow rate was kept at 3 L·min . The coated specimens were placed on the electrode of the spark discharge spectrometer, and the compositions of the hardfacing layer were examined by a 5-point examination. Then, a maximum error value was deleted, and an average value of 4 points was obtained. 2.5. Microstructural analysis The coated specimens were ground using sand paper (from #100 to #1500), and then polished with Al2O3 powder (0.3

Table 2. Compositional design of alloy fillers (wt%) Specimen A B C D E F

C 3.8 3.8 3.8 3.8 3.8 3.8

Cr 7.7 7.7 7.7 7.7 7.7 7.7

Si(Si:Fe=3:1) 1.5 1.5 1.5 1.5 1.5 1.5

Mn 0.7 0.7 0.7 0.7 0.7 0.7

Ti 0 0.725 1.45 4.35 5.8 7.25

Fe Bal. Bal. Bal. Bal. Bal. Bal.

Microstructural Evolution with Various Ti Contents in Fe-Based Hardfacing Alloys using a GTAW Technique

703

Fig. 3. Schematic diagram of analytical position.

m). After polishing, the coated samples were etched using Beraha’s Etchant (1 g K2S2O5 + 0.5 ml HCl + 100 ml H2O). The microstructures of the hardfacing alloys were observed using an optical microscope (ZEISS Axioskop 2 MAT, OM), and the analytical position of the microstructures was determined, as shown in Fig. 3. 2.6. XRD analysis The XRD specimens were ground with sand paper (from #100 to #1500), and then polished with Al2O3 powder (0.3 m). An X-ray diffractometer (Siemens D5000, XRD) with Cu K radiation was utilized to analyze the constituent phases of the hardfacing layers at a scanning rate of 2°/min and 2 from 30° to 100°. The operating voltage and operating current were 40 kV and 30 mA. 2.7. FESEM, EDS and mapping analyses The coated samples were etched using Beraha’s color etchant, and the microstructures were observed using a field emission scanning electron microscope (JEOL JSM-6700F, FESEM). Morphological observations of the precipitates in the hardfacing layers were carried out using a backscattered electron image (BEI). The main purpose of this study is to discuss the microstructural evolution with different Ti contents. The composition and concentration profiles of the precipitates in the hardfacing layers were analyzed using the EDS and mapping techniques.

Fig. 4. Macrostructure of cross section for the B Group.

Tester (SHIMADZU, HMV-2). The base position was selected as a fusion line. The starting examination position was 0.4 mm below the fusion line, and every examination point was separated by 0.1 mm. The final measured point was at the upper 2 mm of the fusion zone. The test load was 1.96 N, and load time was kept at 10 s.

3. RESULTS AND DISCUSSION 3.1. Macrostructural observation of hardfacing layers Figure 4 shows the macrostructures of the cross section of the hardfacing layers in Group B. There was a better metallurgical reaction between the hardfacing layer and substrate, and no cracks were formed in the interfacial junction site. Figures 5(a)-(f) show the macrostructural morphologies of the hardfacing layers with various Ti contents. The width of every group is about 3.1 cm, and the hardfacing layers indicate a smooth surface. The different Ti alloy fillers do not affect

2.8. Rockwell hardness test The mechanical property of the top surface of the hardfacing layer was examined using the Rockwell Hardness Tester. The test load was 150 kgw, and a diamond indenter was selected as the test indenter. The measured method was performed by 9 points, and the center of the sample was selected as the base point. Then, 8 points were taken surrounding the base point. Every point was separated by a distance of 5 mm. Maximum and minimum error values were deleted for the final data, and 7 points were taken as an average value. The purpose of this experiment is to discuss the effect of Ti content on the hardness values of hardfacing layers. 2.9. Vickers hardness test Profiles of hardness values for the heat-affected zone and the hardfacing layer were analyzed using the Vickers Hardness

Fig. 5. Macrostructures of the hardfacing layers with different Ti contents (a) A Group, (b) B Group, (c) C Group, (d) D Group, (e) E Group, and (f) F Group.

704


the flow ability under high temperatures during welding. On the other hand, there were no cracks in the surface of the hardfacing layers because of the lower heat input during GTAW. Therefore, no cracks were found under lower strain relief in the surface of the hardfacing layers. 3.2. Compositional analysis of hardfacing layers Regarding the design of alloy fillers in this study, the C (2.0 wt%), Si (0.7 wt%), Mn (0.7 wt%), and Cr (6.0 wt%) contents were fixed with different Ti contents (Group A: 0.0 wt%, Group B: 0.5 wt%, Group C: 1.0 wt%, Group D: 3.0 wt%, Group E: 4.0 wt%, and Group F: 5.0 wt%). The alloy fillers were then coated onto the substrate materials. Elemental contents on the surface of the hardfacing layers were analyzed using a Spark. The analysis results are shown in Table 4, where it can be seen that the C, Mn and Cr contents approximatd the original design content, and the Si content was higher than the original design content. For the Ti element, the contents of Groups A, B, C, E and F were lower than the original design content except for Group D. Experimental results showed the difference between the analyzed composition and the original composition; the composition between the hardfacing layer and the substrate indicated a certain difference, and then caused elemental diffusion. For the above reasons, the dilution ratio is an important factor. Soda et al. [21] reported that C easily increases the dilution ratio when the carbon and other metal powders are mixed for welding. On the other hand, increments of C in the hardfacing layer also decreased the formation of carbides. However, there has, until now, been no suitable equation for predicting the dilution ratio in hardfacing welding. Because the dilution ratio is widely applied in the butt welding of dissimilar metals welding, the compositional variation was calculated in this study, and it had a similar effect to the calculation of the dilution ratio. The equation is shown as Eq. (1): Wfiller – Whardfacing W = ---------------------------------------- 100% Wfiller

(1)

where W: The variation of composition (%) Whardfacing: Weight percent of the hardfacing layer (wt%) Wfiller: Weight percent of the designed filler alloy (wt%) Table 5 shows the compositional variations of the filler alloy design and the hardfacing layer. The compositional variations

Table 5. Compositional variation of Groups A~F Compositional variation (%) C Cr Ti 49.5 23.0 0.00 42.5 17.9 65.0 46.6 19.5 64.3 42.4 5.10 26.2 42.1 12.0 38.8 46.3 26.4 32.8

Specimen Group A Group B Group C Group D Group E Group F

of C, Cr, and Ti were 42.1%~39.5%, 12%~26.4% and 26.2% ~49.5%, respectively. The compositional variation of C indicated an obvious increment because the graphite powder had a light specific weight, and was easily sprayed and lost during welding. On the other hand, the compositional variation of Cr was lower than that of C and Ti. Cr has a high specific weight, and is not easily sprayed and lost during welding. 3.3. Solidification path of hardfacing layers Figure 6(a)[22] shows the Fe-Cr-C liquidus projection, and the solidification path of the hardfacing layer can be explained using this diagram. The compositional regions of Groups A~F (from Table 4) can be noted on Fig. 6. The black circle point is a compositional location of the hardfacing layers. This diagram can be used to predict the primary phases of the hardfacing layers at a high temperature. The U3U4 line is a eutectic reaction line of LFe+M3C. The composition of the hardfacing layer of Group A located in the austenite phase () region at high temperatures and the primary austenite phase will be formed during cooling. With further cooling, a eutectic reaction of LFe+M3C occurs, and the austenite and M3C can be formed. Figure 6(b)[22] shows the Fe-Cr-C phase diagram at 870 °C, and the black circle point is the composition of the hardfacing layer; this indicates the hardfacing layer of Group A located in the Fe+M3C phase region at 870 °C. It can be proven that the hardfacing layer has two phases of the austenite and M3C at a temperature of 870 °C. Figures 7(a)-(d)[23] show the Fe-Ti-C ternary phase diagram at various temperatures of 1500 °C, 1400 °C and 1300 °C, and Fig. 7(e)[20] shows the Fe-Ti-C ternary phase diagram with 0.3 wt% Ti. The dotted line in Fig. 7(e) shows the diffusion path for Group B of the hardfacing layer. However, Group B of the hardfacing layer can preferentially form a pri-

Table 4. Chemical composition on the top surface of hardfacing layer (wt%) Specimen Group A Group B Group C Group D Group E Group F

C 1.92 2.19 2.03 2.19 2.20 2.04

Si 0.80 1.02 0.86 0.83 0.93 0.94

Mn 0.70 0.72 0.74 0.74 0.67 0.70

Cr 5.93 6.32 6.20 6.23 6.78 5.67

Ti 0 0.25 0.52 3.21 3.55 4.87

Fe Bal. Bal. Bal. Bal. Bal. Bal.


705

Fig. 6. The Fe-Cr-C ternary phase diagram (a) Liquidus projection and (b) at 870 °C.

Fig. 7. The Fe-Ti-C ternary phase diagram (a) 1500 °C, (b) 1400 °C, (c) 1300 °C, (d) 1200 °C, and (e) 0.3 wt%Ti (Symbol meanings: ▲ Group B, ◆ Group C, ▼ Group D, ■ Group E, ★ Group F).

706

Chih-Chun Hsieh et al. Table 6. Solidification path of Groups A~F Specimen Group A Group B Group C Group D Group E Group F

Microstructure /M3C Hypoeutectic TiC + /M3C Hypoeutectic TiC + /M3C Hypoeutectic TiC + /M3C Hypoeutectic TiC + /M3C Hypoeutectic TiC + /M3C Hypoeutectic

mary austenite () at high temperatures, and react as Fe+TiC phases during cooling. With further cooling, three phases can be formed (Fe+TiC+Fe3C). A comparison of the marked point in Fig. 6(b) shows that Fe+TiC+Fe3C existed in the hardfacing layer of Group B at 870 °C. In Figs. 7(a)-(d), the primary TiC can be precipitated at high temperatures for the solidification path of Groups C~F. During cooling, two phases (Fe+TiC) can be formed. With further cooling, three phases (Fe+TiC+M3C) can be precipitated. Therefore, the diffusion paths for Groups A~F can be predicted according to the FeCr-C and Fe-Ti-C phase diagrams, as shown in Table 6. A schematic diagram was used to explain the microstructural evolution with Ti addition during the solidification, as shown in Fig. 8. The primary TiC can be preferentially precipitated with the addition of Ti from the liquid during solidification, as shown in Fig. 8(a). The primary TiC can then grow gradually, as indicated in Fig. 8(b). During the cooling progress, M3C carbides were formed surrounding the TiC phases, as shown in Fig. 8(c). Finally, the eutectic structures (+M3C), and a small amount of martensite was precipitated in surrounding TiC phases, and in the TiC grains, as exhibited in Fig. 8(d). 3.4. XRD analysis of hardfacing layers Figure 9 shows the XRD results for the hardfacing layers of Groups A~F. The microstructures of Groups A, B, and C

Fig. 8. Microstructural evolution in hardfacing layer with Ti addition.

Solidification path LL +  + M3C LL +   TiC + TiC + M3C LL + TiCL +  + TiC + TiC + M3C LL + TiCL +  + TiC + TiC + M3C LL + TiCL +  + TiC + TiC + M3C LL + TiCL +  + TiC + TiC + M3C

Fig. 9. XRD diffraction patterns in Groups A~E.

include austenite, martensite, M3C and M2C. On the other hand, the diffraction peak of TiC was detected in Groups D, E and F with increased Ti content. Therefore, the main microstructures of the hardfacing layers were austenite, M3C and TiC. In a comparison of the XRD result and the phase region of the phase diagram, Group A had austenite and M3C, while Groups B~C had austenite, M3C and TiC. Furthermore, there was no diffraction peak of TiC for Groups B~C because a small amount of TiC formed in the hardfacing layers of Groups B~C. 3.5. Microstructural analysis of hardfacing layers Figures 10 (a)-(c) show the OM and SEM microstructures of Group A. The results indicated that the matrix was the primary  phase without Ti addition. The microstructures included the eutectic structures of +(Fe,Cr)3C, +(Fe,Cr)2C, and a small amount of martensite (from the transformation of the austenite). The morphology of the primary austenite was the dendritic structure, and the white lamellar region was a eutectic structure of +(Fe,Cr)3C. The gray and black parts surrounding the +(Fe,Cr)3C were eutectic structures of +(Fe,Cr)2C. The martensite phases were formed at the interface between the +(Fe,Cr)3C and the +(Fe,Cr)2C. The morphologies of the martensite were lathy structures. Magnifying the image revealed that the lamellar eutectic structures were linked together, and


707

Fig. 10. Microstructures of hardfacing layer of Group A (0 wt%Ti) (a) OM, (b) SEM, and (c) Magnification: 20000X.

Fig. 11. Microstructures of hardfacing layer of Group B (0.25 wt%Ti) (a) OM, (b) SEM, and (c) Magnification: 20000X.

formed networked structures, as shown in Fig. 10(c). Finer lamellar structures of eutectic +(Fe,Cr)2C were also observed. An EDS elemental analysis was used to examine the elemental segregation in Groups A~F, as shown in Table 7. The C was detected from the austenite because the austenite was a solid solution phase. Higher C contents existed in the +(Fe, Cr)3C and +(Fe,Cr)2C due to the formation of carbide. PeroSanz et al.[16] reported that the lamellar structure was the phase and (Fe,Cr)3C when the Cr content was less than 10 wt%.

Jacuinde et al. [24] noted that carbides were formed along the austenite grain boundaries. The C content from the interface can then be completely consumed at the same time. Because of the above results, the martensitic formation temperature (Ms) will be increased, and the austenite can be transformed into the martensite. The microstructures of Group A contained martensite at the interface between the +(Fe,Cr)3C and the +(Fe,Cr)2C. Figures 11(a)-(b) and 12(a)-(b) show the microstructures

708


Fig. 12. Microstructures of hardfacing layer of Group C (0.52 wt%Ti) (a) OM and (b) SEM.

Fig. 13. Microstructures of hardfacing layer of Group D (3.21 wt%Ti) (a) OM and (b) SEM.

of the hardfacing layers of Groups B (0.25 wt% Ti) and C (0.52 wt% Ti) using the OM and SEM. The microstructures of the hardfacing layers in Groups B and C were similar to those of the hardfacing layers in Group A. Tiny precipitates were found in the hardfacing layers of Groups B and C. The tiny precipitates had higher Ti and C contents, and were identified as the carbide of TiC. The eutectic structures of the + (Fe,Cr)3C and the +(Fe,Cr)2C indicated a higher C content, and their grain size decreased when the Ti content increased to 0.52 wt%. The morphologies of +(Fe,Cr)2C exhibited fiber lamellar structures, as shown in Fig. 12. The microstructures of the hardfacing layers in Group D included the primary TiC, austenite and eutectic +(Fe,Cr,Ti)3C, as shown in Figs. 13(a)-(b). The primary TiC was precipitated with 3.21 wt%Ti, and their morphologies showed small (pellet and lathy) and large structures (irregular island). The maximum TiC size was about 8 m. The Cr content of +(Fe,Cr,Ti)3C was higher than that of other groups, as shown in Table 7. The microstructures of the hardfacing layers in Group E showed the TiC, austenite and +(Fe,Cr,Ti)3, with 3.55 wt%Ti, as shown in Figs. 14(a)-(b). The TiC for the most part had large dendrite structures, except for some small pellets. The grain size of the small pellets was about 1~5 m, and that of the large dendrite was about 20 m.

Figure 15 shows the mapping analysis of the TiC at a high magnification in order to observe the concentration profile of Ti and C. The result indicated that the TiC had a segregation region of C and Ti. The C and Ti were then concentrated on the dendrites, and were identified as the TiC carbide. From Table 7, the +(Fe,Cr,Ti)3C had a higher C and Cr content than other groups did. Figures 16(a)-(b) show the microstructures of the hardfacing layer in Group F. Groups E and F had the same microstructures. The amount of primary TiC in Group F was higher than in Group E. The grain size of the TiC was about 5~15 m. The TiC amounts increased when the Ti contents in the hardfacing layer were increased to 4.87 wt% and the amounts of +(Fe,Cr,Ti)3C were decreased. However, +(Fe,Cr,Ti)3C can be greatly decreased when the Ti content is above 3.21 wt% because the Ti has a high affinity, and the TiC was formed by Ti and C during solidification. The C content will be consumed largely as primary TiC carbide. The Fe, C and a residual C were then reacted as the (Fe,Cr)3C. Furthermore, the Cr content of +(Fe,Cr,Ti)3C increased when the Ti content was above 3.21 wt%, as shown in Table 7. In this study, the affinities between Ti and C were higher than those between Cr and C, and it was therefore easy to form the (Fe,Cr,Ti)3C when the amounts of eutectic +(Fe,Cr,Ti)3C were reduced. Consequently, the Cr

709

Microstructural Evolution with Various Ti Contents in Fe-Based Hardfacing Alloys using a GTAW Technique Table 7. EDS analysis of hardfacing layer of Groups A~F (wt%) Specimen Group A (0 wt%Ti) Group B (0.25 wt%Ti)

Group C (0.52 wt%Ti) Group D (3.21 wt%Ti) Group E (3.55 wt%Ti) Group F (4.87 wt%Ti)

Phase Primary   + (Fe,Cr)3C  + (Fe,Cr)2C Primary   + (Fe,Cr)3C  + (Fe,Cr)2C TiC   + (Fe,Cr)3C  + (Fe,Cr)2C Primary TiC   + (Fe,Cr,Ti)3C Primary TiC   + (Fe,Cr,Ti)3C Primary TiC   + (Fe,Cr,Ti)3C Primary TiC

C 2.57 7.52 8.38 3.80 7.44 9.71 17.39 3.85 6.16 8.74 15.61 2.22 8.23 17.06 2.37 8.20 20.70 3.14 6.71 16.37

Cr 4.24 15.53 9.46 3.88 18.26 8.02 11.88 4.26 18.21 11.31 6.91 5.00 33.30 0.66 6.42 33.83 7.26 37.81 0.70

Si 0.82 0.72 0.59 0.81 0.27 0.32 0.62 1.29 0.84 1.09 1.00 0.10 1.19 1.67 1.01 0.16

Mn 0.64 2.42 2.72 0.82 2.26 1.84 0.21 1.42 2.51 1.48 0.31 1.09 2.23 1.11 0.92 1.14 1.53 0.13

Ti 0.08 0.36 0.05 43.89 1.36 0.38 56.32 1.32 80.48 0.23 2.22 79.30 1.22 1.54 80.60

Fe 91.74 73.81 78.79 90.61 71.41 80.05 26.00 87.81 72.73 77.64 19.76 90.69 54.83 1.8 88.68 54.82 85.57 51.40 2.04

content of eutectic +(Fe,Cr,Ti)3C was higher in the hardfacing layers of Groups D, E and F than that of other groups.

Fig. 14. Microstructures of hardfacing layer of Group E (3.55 wt%Ti) (a) OM and (b) SEM.

3.6. Phase fraction and hardness analyses of hardfacing layer Figure 17 shows the relationship between the phase fraction and hardness value of the hardfacing layers in Groups A~F. The phase fractions of +(Fe,Cr)3C and +(Fe,Cr)2C decreased when the Ti content was increased because of the formation and growth of TiC carbide. The phase fractions of the austenitic phase were then increased. The hardness value of the hardfacing layer in Group A was 54.9 HRC, indicating a slight increment (56.71 HRC) for Groups B and C. The average hardness value of the hardfacing layers had an obvious increment when the Ti content was increased to above 3.21 wt%. In this study, the eutectic structures of +(Fe,Cr)3C and +(Fe,Cr)2C were about 1013 HV, and the austenitic phases were about 390 HV. Therefore, the main microstructures of the hardfacing layers without Ti addition were softer austenite phases, and those with Ti addition were harder +(Fe,Cr)3C and +(Fe,Cr)2C. On the other hand, a strengthening effect resulted when the (Fe,Cr)3C precipitated. This demonstrates that (Fe,Cr)3C is an important factor in increasing hardness value. However, TiC can be formed by Ti and C, so the amount of TiC in austenite can be increased when the Ti content is increased. Therefore, an increment of TiC can increase the hardness value. Wang et al. [25] noted that TiC has higher hardness and thermal stability than chromium carbide, and can be used to reinforce Fe-based alloys. This meant that the strengthening effect of TiC

710


Fig. 16. Microstructures of hardfacing layer of Group F (4.87 wt%Ti) (a) OM and (b) SEM.

Fig. 15. Mapping analysis of TiC carbide.

carbide in Fe-Cr-C-Si-Mn-xTi hardfacing alloys is better than that of chromium carbide in Fe-Cr-C hardfacing alloys. On the other hand, the phase fractions of +(Fe,Cr,Ti)3C and (Fe,Cr)3C decreased when the Ti content increased. The hardness values can be increased when the TiC amount is increased. The phase fraction of the eutectic structures decreased as the Ti content increased. However, TiC has a dispersion strengthening effect in the austenite matrix. Therefore, TiC can efficiently increase the hardness value. The hardness value profiles of the hardfacing layer cross sections for Groups A~F were examined using the Vickers Hardness Tester, as shown in Fig. 18. The purpose of this experiment was to evaluate the mechanical properties between the

Fig. 17. Relationships between the hardness values and phase fractions of TiC and eutectic structures in hardfacing layers of Groups A~F (Eutectic: +M3C and +M2C).

hardfacing layers and the substrate materials. The hardness indicated the highest values from the top of the hardfacing layers in Groups A~F. There was a larger difference in composition between the hardfacing layer and the substrate material, so that the compositional variation near the fusion line was higher; therefore, the hardness values decreased significantly.


711

ness showed a slight variation from the middle to the top of the hardfacing layer.

4. CONCLUSIONS

Fig. 18. The hardness distribution from the substrate to the top of the hardfacing layers. Table 8. Composition and hardness in various regions of Group F Regions Top Medium Bottom Oscillating direction

Hardfacing layer

Cr 6.21 5.67 2.44 6.15

Element (wt%) C Ti 2.03 4.83 1.97 4.54 0.58 2.06 1.95 4.37

Hardness (HRC) 65.5 63.3 51.1 60.4

However, the compositional variation gradually decreased further from the fusion line, and the hardness value increased. Furthermore, the compositional variation near the surface of the hardfacing layer exhibited the lowest value, and the hardness indicated the highest value at the same time. In order to further understand the relationship between chemical composition and hardness value, the average values with 20 measured points were taken, from the top to the bottom of the hardfacing layer, and in an oscillating direction for Group F. These data are listed in Table 8. The results showed that the Cr, C and Ti contents showed a slight increment from the middle to the top of the hardfacing layers, and in the oscillating direction. The hardness value can maintain a certain degree of strength in these regions attributed to some precipitates, (TiC, +(Fe,Cr)3C and +(Fe,Cr)2C). However, the Cr, C and Ti contents obviously decreased from the bottom of the hardfacing layer. When these strengthening elements were reduced, the hardness values also decreased in the bottom of the hardfacing layer. The lower Ti content in the bottom of the hardfacing layer caused an unobvious precipitation of TiC, + (Fe,Cr)3C and +(Fe,Cr)2C). Therefore, Ti, C and Cr were determinant factors for the variation of composition and hardness. Furthermore, the TiC had a larger grain size than in previous microstructural observation of other eutectic phases, so the composition and hardness of the TiC could be more easily identified. Consequently, the composition and hard-

This study discusses the effect of Ti contents on the microstructural evolution of Fe-based hard-facing alloys. A series of Fe-Cr-C-Si-Mn-xTi alloy fillers were deposited on SS400 low carbon steels using oscillating GTAW. Some significant results can be summarized as follows: (1) A primary  phase, eutectic structures of +(Fe,Cr)3C and +(Fe,Cr)2C, and martensite were observed without Ti addition in the hardfacing layers. (2) Microstructures of the  phase, TiC carbide, eutectic structures of +(Fe,Cr)3C and +(Fe,Cr)2C, and martensite were found with Ti content between 0.25 wt% and 0.52 wt% in the hardfacing layers. (3) The primary phases of the hardfacing layers with 0.25 wt% Ti and 0.52 wt% Ti were the  phase and the TiC carbide, respectively. (4) When Ti content was over 3.21 wt%, the primary TiC,  phase, and +(Fe,Cr,Ti)3C were observed in the hardfacing layers. (5) The hardness value in the hardfacing layer increased as Ti content increased, due to an increment of TiC carbide. The highest hardness value (HRC 60.29) was found when the Ti content was 4.87 wt%.

ACKNOWLEDGEMENT The authors would like to thank the National Science Council of the Taiwan for financial support under projects numbered NSC 101-2623-E-005-002-ET and NSC 101-2811-E005-001.

REFERENCES 1. E. Pagounis, M. Talvite, and V. K. Lindroos, Compos. Sci. Technol. 56, 1329 (1996). 2. P. Crook and H. N. Farmer, Metals Handbook, 18th ed., p. 758, American Society for Metals, Ohio (1992). 3. M. B. Peterson and W. O. Winer, Wear Control Handbook, 1st ed., p.373, ASME, New York (1980). 4. K. Gurumoorthy, Mater. Sci. Eng. A 456, 11 (2007). 5. C. A. Mayer, Weld. Des. Fab. 82, 61 (1982). 6. R. Menon, Weld. J. 75, 43 (1996). 7. M. Zucchetti, J. Nucl. Mater. 233, 1486 (1996). 8. D. J. Kotecki and J. S. Ogborn, Weld. J. 74, 269 (1995). 9. C. K. Kim, S. Lee, J. Y. Jung, and S. Ahm, Mater. Sci. Eng. A 349, 1 (2003). 10. K. Y. Lee, S. H. Lee, Y. Kim, H. S. Hong, Y. M. Oh, and S. J. Kim, Wear 255, 81 (2003). 11. T. Sharma, S. Maria, and D. Dwivedi, ISIJ Int. 45, 1322

712


(2005). 12. S. Chatterjee and T. K. Pal, Wear 255, 417 (2003). 13. S. Atamert and H. K. D. H. Bhadeshia, Mater. Sci. Eng. A 130, 101 (1990). 14. A. Wiengmoon, T. Chairuangsri, A. Brown, R. Brydson, D. V. Edmonds, and J. T. H. Pearce, Acta Mater. 53, 4143 (2005). 15. L. E. Svensson, B. Gretoft, B. Ulander, and H. K. D. H. Bhadeshia, J. Mater. Sci. 21, 1015 (1986). 16. J. A. Pero-Sanz, D. Plaza, J. I. Verdeja, and J. Asensio, Mater. Charact. 43, 33 (1999). 17. M. F. Buchely, J. C. Gutierrez, L. M. Leon, and A. Toro, Wear 259, 52 (2005). 18. C. M Chang, C. M. Lin, C. C. Hsieh, J. H. Chen, C. M. Fan, and W. Wu, Mater. Chem. Phys. 117, 257 (2009).

19. C. M Chang, C. C. Hsieh, C. M. Lin, J. H. Chen, C. M. Fan, and W. Wu, Mater. Chem. Phys. 123, 241 (2009). 20. C. M. Lin, C. M Chang, J. H. Chen, C. C. Hsieh, and W. Wu, Metall. Mater. Trans. A 40, 1031 (2009). 21. L. Lu, H. Soda, and A. Mclean, Mater. Sci. Eng. A 347, 214 (2003). 22. G. V. Raynor and V. G. Rivlin, Int. Metals Rev. 29, 299 (1984). 23. H. Ohtani, T. Tanaka, M. Hasebe, and T. Nishizawa, Calphad 12, 225 (1988). 24. A. B. Jacuinde, R. Correa, J. G. Quezada, and C. Maldonado, Mater. Sci. Eng. A 398, 297 (2005). 25. X. H. Wang, Z. D. Zou, S. Y. Qu, and S. L. Sou, J. Mater. Process. Technol. 168, 89 (2004).