Fabricating TiC particles reinforced Fe-based composite coatings produced by GTAW multi-layers melting process

Fabricating TiC particles reinforced Fe-based composite coatings produced by GTAW multi-layers melting process

Materials Science and Engineering A 441 (2006) 60–67 Fabricating TiC particles reinforced Fe-based composite coatings produced by GTAW multi-layers m...

2MB Sizes 0 Downloads 4 Views

Materials Science and Engineering A 441 (2006) 60–67

Fabricating TiC particles reinforced Fe-based composite coatings produced by GTAW multi-layers melting process X.H. Wang ∗ , S.L. Song, Z.D. Zou, S.Y. Qu School of Materials Science and Engineering, Shandong University, Jinan 250061, China Received 17 April 2006; received in revised form 7 June 2006; accepted 7 June 2006

Abstract The present study aims to analyze microstructure and properties of TiC particles reinforced Fe-based surface composite coatings produced by gas tungsten arc welding (GTAW) multi-layers melting process. The mixture powder of graphite and ferrotitanium (FeTi) was deposited evenly on an AISI 1045 steel substrate, which was then heated by GTAW heat source. The results showed that in situ synthesized TiC particle reinforced composite coatings can be achieved under suitable welding parameters. Cubic TiC carbides and fine needle-shape eutectic TiC carbides are formed by ternary eutectic reaction between FeTi and graphite powders. Together with these TiC carbides, radially grown dendrites of primary TiC particles are also found in the composite coatings. These TiC particles are evenly distributed in the composite coatings. Because of the generation of these carbide particles and their homogeneous distribution in the matrix, the composite coatings give very high hardness and excellent wear resistance. The wear resistance of multi-layers composite coatings is about three to four times higher than that of 1045 steel substrate. Moreover, the wear resistance of the composite coatings and the substrate increased with increasing wear sliding distance. © 2006 Elsevier B.V. All rights reserved. Keywords: Gas tungsten arc welding; Surface composite coating; TiC particles; Wear properties; Multi-layer melting process

1. Introduction It was reported that the surface layer of components is reinforced by ceramic particles to offer high wear resistance. In recent years, metal carbides have been widely used as reinforcement in metal matrix composites (MMCs) [1–3]. In all of ceramic particles, TiC particle crystallize in the cubic NaCl structure has lower density (4.90–4.93 g/cm3 ) and high hardness (3200 kg/mm2 ), as well as high thermal stability and more negative standard Gibbs formation free energy, which has been received interest worldwide as reinforcement in MMCs [4,5]. There are mainly two techniques available to incorporate reinforcement particles in the matrix of MMCs. One is mechanical mixing of the reinforcement externally [6–8]; the other is in situ formation of the reinforcement phase within the matrix [9–12]. The additional reinforcement ceramic particles



Corresponding author. Tel.: +86 531 88392208; fax: +86 531 82616431. E-mail address: [email protected] (X.H. Wang).

0921-5093/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2006.06.015

are directly added into the coating materials, however, the shape and chemical composition of additives hardly remain unchanged due to the dissolution into metal liquid or metallurgical reactions with the environment resulting in the deterioration of the toughness and crack resistance of MMCs. The latter is realized through creating conditions favorable for reaction of elements to form the particles. The eminent advantage of the in situ synthesis technology is that it eliminates interfacial incompatibility of matrices with reinforcements by creating more thermodynamically stable reinforcements based on their nucleation and growth from the matrix phase. In recent years, many researchers are focusing on in situ synthesis TiC particles reinforced Ni- or Fe-based surface composite coatings by using laser beam to enhance surface quality [13–15]. However, problems always exist owing to differences in the laser beam absorption rates of different cladded powders. Furthermore, compared to gas tungsten arc welding (GTAW) process, the components of laser beam is complex and expensive. Recently, new attempts have been made by using GTAW to achieve surface composites or surface alloying, in which an alloy powder of a desirable composition and a thin surface layer of

X.H. Wang et al. / Materials Science and Engineering A 441 (2006) 60–67

61

the substrate material are simultaneously melted and then rapidly solidified to form a dense coating, metallurgical bonded to the base material [16–21]. Wang et al. [16] have produced the wear resistant clad layers on medium carbon steel by GTAW, where WC and TiC particles were directly added into the specified metal powders. The results showed that the TiC with W clad layer had superior wear performance under low sliding speed condition. Ero˘glu et al. [17] have investigated the tungsten-inert gas surface alloying with pre-placed graphite, chromium and high-carbon-ferro-chromium powders on SAE1020 low carbon steel. Buytoz [18,19] has studied the effect of GTAW parameters on the microstructure properties of SiC-based hardfacing on low alloy steel. It was found that the microstructure of the cladding layer is M7 C3 primary carbides, Fe3 Si, SiC, and the graphitic carbon precipitates. All of these indicated that GTAW cladding coatings provided remarkable enhancement on the corrosion resistance, wear resistance, and thermal conductivity without impairing the bulk properties; and it has been demonstrated for Fe-, Co-, and Ni-based alloy coatings synthesized on various traditional substrate materials. However, a limited application of this process is updated in the literature on the formation of TiC particles via liquid reaction. In the present study, an attempt has been made to prepare TiC reinforced Fe-based composite coatings by direct melting of the mixture of graphite and ferrotitanium powders on an AISI 1045 steel substrate during GTAW process under a non-oxidizing atmosphere, rather than the TiC particles being directly added into the GTAW weld pool. 2. Experimental procedures In this study, AISI 1045 steel in a quenched and tempered condition with hardness HRC 32–35 was used for the substrate material; it contains (wt.%) 0.45C, 0.25Si, 0.66Mn, and balance Fe. The dimensions of the substrate are 100 mm × 25 mm × 10 mm. A powder mixture of ferrotitanium (FeTi) alloy and crystalline graphite (99.5% purity) was used as the raw coating alloy. The chemical composition of FeTi is (wt.%): 41.5Ti, 0.08C, 0.035P, 0.025S, and balance Fe. The ratio of FeTi alloy to graphite powder corresponds to that of stoichiometric TiC, thus, the weight ratio wFeTi : wC was 9.5:1. The average size of the FeTi and graphite particles was less than10 ␮m. In order to obtain homogeneous distribution, the combined powders attrition-milled for 1 h using agate ball mill with an agate container and balls operated at 300 rpm. The blended powders were mixed with a small amount of sodium silicate to keep the powders on the surface under the flow of argon during GTAW process. Then, the blended powders with sodium silicate were dried in hot air. Finally, the blended powders were pre-placed on the surface of the substrate, which were thoroughly cleaned, dried and finally rinsed by acetone, to give a thickness of about 1.5 mm for a single pass. To decrease the effect of dilution of the substrate, some specimens were cladded two or three pass under the same welding parameters. The thickness of second and third pass is also about 1.5 mm. For convenience, specimen clad single pass, two pass and three pass are referred to as “S1”, “S2”, and “S3”, respectively.

Fig. 1. A schematic of the GTAW process: (a) photo of apparatus and (b) schematic welding area.

Cladding was carried out by GTAW process to produce a series of clad tracks, which is presented in Fig. 1. Table 1 lists the parameters of the cladding process used in this work. GTAW torch was held stationary above the moving specimens, while a shielding gas of pure argon was supplied. Tracks were produced along the length of the specimens. After surface alloying, samples were cut from the alloyed specimens for microstructural examination and hardness measurement. The samples were prepared for metallographic exam-

Table 1 Welding parameters of GTAW processing Welding current (A) Welding speed (cm/min) Arc voltage (V) Arc gap (mm) Electrode Electrode polarity Argon flow rate (L/min)

150 5.5 15–17 2 W-2% thorium DCSP 8

62

X.H. Wang et al. / Materials Science and Engineering A 441 (2006) 60–67

Fig. 2. A schematic diagram of wear testing: (a) scheme of MM200 model friction and wear tester and (b) worn surface of specimen (1, specimen; 2, cermet ring; 3, support axes; 4, support; 5, lever; 6, spring; 7, weight).

ination by grinding on SiC wheels followed by polishing. The composite coatings were etched with a solution of 3% nital. The microstructure and compositions were analyzed by using a JXA840 scanning electron microscopy (SEM) and a JXA-8800R electron microprobe microanalysis (EPMA), respectively. A type of D/Max-Rc X-ray diffraction with Cu K␣ radiation operated at 60 kV and 40 mA was used to analyze the coating phase structure. Micro-hardness along the depth of the cross-section was measured by using a Shimadzu HMV-2000 type micro Vickers. The load used was 200 g and loading time was set at 15 s. The macro-hardness was determined by using an HR-150A type of Rockwell hardness tester. An average value of hardness was taken from five measurements. The block-on-ring wear testing was carried out without lubrication at room temperature using a friction and wear tester, which is presented in Fig. 2. The ring material of the wear couple was a cermet containing 92 wt.% WC and 8 wt.% Co. Its hardness is 90 HRA. The outer radius of the circular test ring is 20 mm, and its width is 10 mm. The test specimens were machined to block with size of 30 mm × 6 mm × 8 mm. The wear conditions were, a normal load of 49 N, a sliding speed of 0.84 m/s, and a sliding distance of 1008 m. The average width of the wear track was measured with the help of a microscope, and the wear volume was calculated using the following formula [22]:      2 π 2 −1 b b b V =w r2 − r sin − (1) 180 2r 2 4 where w is the width of the specimens (mm), b the width of the wear track (mm), and r is the out radius of wear ring (mm). The wear volume loss was calculated after a certain time interval.

Fig. 3. Gibbs free energy possible reactions within Fe–Ti–C system as a function of temperature.

such as temperature and the activities of the element. Therefore, for the Fe–Ti–C system, possible reactions between Ti–C, Fe–Ti, and Fe–C should be considered. The possible products could be TiC, Fe3 C, Fe2 Ti, and FeTi. These products may be formed via the following reactions: FeTi + C = TiC + Fe

(2)

3Fe + C = Fe3 C

(3)

2Fe + C = Fe2 C

(4)

2Fe + Ti = Fe2 Ti

(5)

Fe + Ti = FeTi

(6)

Fig. 3 shows the change in Gibbs free energy for each reaction as a function of temperature, in which the thermodynamic data are from Refs. [25–29]. It can be seen that the Gibbs free energy of formation of TiC is always negative, and the free energy values of TiC is much lower than Fe3 C, FeTi, and Fe2 Ti. Namely, Ti has stronger carbide forming tendency than Fe, and TiC has better stability than Fe–C and Fe–Ti. Meanwhile, it is also indicated from Fig. 3 that some possibility of formation of Fe3 C exists at the temperature over 1100 K. Apart from the temperature, activity of carbon also plays an important role. In the present Fe–Ti–C system, TiC formation occurred with Ti depletion in the system that could lead to the tendency of Fe to form Fe3 C. In addition, it also shows that formation of Fe2 Ti exists at temperature lower than 1000 K. While

3. Results and discussion 3.1. Thermodynamic calculations The thermodynamics of Fe–Ti–C system had been studied by several researchers [23,24]. It has been concluded that Fe–Ti–C is a very active and complex system due to several factors. This is mainly due to the presence of carbon, which easily migrates from TiC to Fe and vice-versa depending upon reaction conditions,

Fig. 4. Melted track produced on the surface of S1-specimen (welding current = 150 A; welding speed = 5.5 cm/min; arc voltage = 15–17 V).

X.H. Wang et al. / Materials Science and Engineering A 441 (2006) 60–67

63

3.2. Microstructure of coating

Fig. 5. XRD spectrum of (a) GTAW composite coating of S1-specimen and (b) raw materials (welding current = 150 A; welding speed = 5.5 cm/min; arc voltage = 15–17 V).

the C/Ti ratio is too low, carbon content is not enough to react with titanium to form TiC, thus excess titanium reacts with iron to form Fe2 Ti phase. Therefore, carbon content should be appropriately raised to restrain formation of Fe2 Ti compound.

Fig. 4 shows the macro-morphology of the coating. It can be seen that melted track gave a smooth rippled surface topography and was found to be free from gas porosity and cracks. The X-ray diffraction pattern of the composite coating and the raw materials of the mixture powders are shown in Fig. 5(a and b), respectively. It can be seen that the phases of composite coating are mainly TiC, Fe, and Fe3 C. It clearly confirms that TiC particles can be synthesized by direct reaction between the ferrotitanium and graphite. Additionally, from Fig. 5, it also indicates that the reaction seemed to be complete since FeTi phase was not observed. Fig. 6 shows the EPMA area scanning of the elements in the top surface of S2-specimen. It can be identified that the reinforcement of second phases in composite coating are the precipitate of TiC particles. Although some areas show more TiC particles than others, the distribution of TiC particles is, in general, uniform in a matrix. These fine TiC particles act as reinforcement and are expected to improve the wear properties of the composite coatings significantly. Additionally, these carbides embedded in the clad layer also imply that the distribution of reinforcement phase in the clad layer can cause precipitates strengthening. Fig. 7(a and b) show the representative microstructures of the composite coatings produced by GTAW process for S1-

Fig. 6. EPMA area scanning of elements in the cladding coating for two tracks: (a) EPMA morphology; (b) Ti; (c) C; (d) EDS of point A in (a) (welding current = 150 A; welding speed = 5.5 cm/min; arc voltage = 15–17 V).

64

X.H. Wang et al. / Materials Science and Engineering A 441 (2006) 60–67

tion is completed. Thus, a lot of primary TiC dendrites and cubic TiC carbides are formed, together with fine eutectic TiC particles. Due to the effect of the dilution of the substrate, the content of Ti and C of S1-specimen is lower than that of S3-specimen in the weld pool. TiC carbides do not grow much, but mainly form in a cubic shape. In addition, among the composite coatings in this study, S3-specimen possesses a largest volume fraction of TiC particles in the top surface of coatings. 3.3. Hardness of the coating Micro-hardness of TiC particles reinforced Fe-based composite coatings were measured along the coating depth from the surface, and results are shown in Fig. 9(a–c). It can be seen that the micro-hardness of the coating gradually increased with increase of distance from bottom of the coatings. This may be attributed to the different-density-driven force. Compared with the Fe-based alloy, the relative lower density of TiC particles (4.90–4.93 g/cm3 ) tended to segregate to the upper regions in the coating, which causes a gradient distribution on a macroscale. The gradient distribution of TiC particles led to a gradual hardness distribution of the coating. Fig. 10 shows the macro-hardness of the top surface of the composite coatings. It can be seen that multiple claddings possesses a higher hardness than that of single pass and substrate. This may be attributed to the increasing of the fraction of TiC carbides. 3.4. Wear properties of the coatings

Fig. 7. Morphology of the top surface of composite coating produced by GTAW process: (a) S1-specimen and (b) S3-specimen (welding current = 150 A; welding speed = 5.5 cm/min; arc voltage = 15–17 V).

specimen and S3-specimen, respectively. It can be seen that cubic TiC carbides and fine needle-shaped eutectic carbides are formed in the composite coatings. Together with these TiC carbides, radially grown dendrites of primary TiC carbides are found in the specimens. This precipitation of TiC carbides in the GTAW cladding coatings can be explained from a Fe–Ti–C ternary phase diagram. Fig. 8 shows a basal projection of surfaces of the iron-rich corner [25]. During GTAW process, as the arc is started between tungsten electrode and powder coated surface, the temperature rises to the melting point of mixture powders, the powder layer melts with the substrate together forming a weld pool, and then weld pool cools down below the liquidus temperature, primary starts to nucleate and grow by diffusion of nearby Ti and B. When the temperature decreases to the point “A”, the ␣-ferrite dendrites also to form and continue growing with TiC until the point “B” (the ternary peritectic temperature). At the ternary peritectic temperature, liquid interacts with ferrite dendrites, and then partially or totally formed to ␥-dendrites. Subsequently, liquid cools to the ternary peritectic temperature, solidifica-

Fig. 11 shows the wear volume loss of composite coatings and 1045 steel substrate at the same processing parameters. It can be seen that the composite coatings are more effective in improving wear resistance with increasing sliding distance than that of AISI 1045 steel. It indicates that a significantly enhanced wear resistance is caused by the in situ formed TiC particles in the composite coatings. Additionally, it also shows that multilayers cladding coatings give a higher wear resistant than that of single pass and substrate.

Fig. 8. Iron-rich corner of the Fe–Ti–C phase diagram.

X.H. Wang et al. / Materials Science and Engineering A 441 (2006) 60–67

65

Fig. 9. Micro-hardness of the composite coatings vs. layer depth: (a) S1-specimen; (b) S2-specimen; (c) S3-specimen (welding current = 150 A; welding speed = 5.5 cm/min; arc voltage = 15–17 V).

Fig. 12(a–d) show the wear scar of the composite coatings for S3-specimen, S2-specimen, S1-specimen, and substrate, respectively. It indicates that there is a mild wear with fine scratches for the multi-layers composite coatings, and there is no indication of brittle failure or loose debris formation of TiC ceramic phase. As mentioned earlier, TiC particles in the Fe-based matrix are formed in an in situ manner, which improves the bonding

strength between the TiC particles and Fe-based matrix interface. As a result the hard TiC particles do not easily pulled out from the matrix during wear sliding. Therefore, the composite coating is found to possess a much higher resistance to plastic deformation and scoring, which increases the resistance to plastic erasing and removal of the edges of grooves during subsequent passes.

Fig. 10. Macro-hardness of the top surface of composite coatings and substrate (welding current = 150 A; welding speed = 5.5 cm/min; arc voltage = 15–17 V).

Fig. 11. The wear volume loss vs. sliding distance under an normal load of 49 N and a sliding speed of 0.84 m/s.

66

X.H. Wang et al. / Materials Science and Engineering A 441 (2006) 60–67

Fig. 12. Wear scar of the composite coatings under a normal load of 49 N, a sliding speed of 0.84 m/s and sliding distance of 1008 m: (a) S3-specimen; (b) S2-specimen; (c) S1-specimen; (d) 1045 steel.

4. Conclusions

References

1. GTAW multi-layers melting process has been used to produce Fe-based composite coatings reinforced by in situ TiC particles. The melted tracks gave a smooth rippled surface topography and were found to be free from gas porosity and cracks. 2. Cubic TiC carbides and fine needle-shape eutectic TiC carbides are formed by ternary eutectic reaction between FeTi and graphite powders. Together with these TiC carbides, radially grown dendrites of primary TiC particles are also found in the composite coatings. TiC particle formed from graphite and ferrotitanium, which distribute in the matrix dispersedly on the surface of the coatings. 3. The composite coatings give very high hardness and excellent wear resistance. The wear resistance of the multi-layers composite coatings is three to four times higher than that of 1045 steel substrate. Moreover, the wear resistance of the composite coatings and the substrate increased with increasing wear sliding distance.

[1] D.B. Miracle, Comp. Sci. Technol. 65 (2005) 2526. [2] Q.C. Jiang, B.X. Ma, H.Y. Wang, Y. Wang, Y.P. Dong, Composites Part A: Appl. Sci. Manuf. 37 (2006) 133. [3] L. Dubourg, D. Ursescu, F. Hlawka, A. Cornet, Wear 258 (2005) 1745. [4] H.C. Man, Y.Q. Yang, W.B. Lee, Surf. Coat. Technol. 185 (2004) 74. [5] X.L. Wu, Y.S. Hong, Mater. Sci. Eng., A 318 (2001) 15. [6] T.C. Lei, J.H. Ouyang, Y.T. Pei, Y. Zhou, Mater. Sci. Technol. 11 (5) (1995) 520. [7] X.H. Wang, M. Zhang, Z.D. Zou, S.Y. Qu, Surf. Coat. Technol. 161 (2002) 195. [8] K.V. Acker, D. Vanhoyweghen, R. Persoons, J. Vangrunderbeek, Wear 258 (2005) 194. [9] S. Yang, M.L. Zhong, W.J. Liu, Mater. Sci. Eng., A 343 (2003) 57. [10] L. Lu, J.Y.H. Fuh, Z.D. Chen, C.C. Leong, Y.S. Wong, Mater. Res. Bull. 35 (2000) 1555–1561. [11] H. Berns, B. Wewers, Wear 251 (2001) 1386. [12] Y.J. Kim, H. Chung, S.L. Kang, Mater. Sci. Eng., A 333 (2002) 343. [13] A. Singh, W.D. Porter, N.B. Dahotre, Mater. Sci. Eng., A 399 (2005) 318. [14] Y. Chen, H.M. Wang, Intermetallics 14 (2006) 325. [15] S. Zhang, W.T. Wu, M.C. Wang, H.C. Man, Surf. Coat. Technol. 138 (2001) 95. [16] S.W. Wang, Y.C. Lin, Y.Y. Tsai, J. Mater. Process. Technol. 140 (2003) 682. ¨ [17] M. Ero˘glu, N. Ozdemir, Surf. Coat. Technol. 154 (2002) 209. [18] S. Buytoz, M. Ulutan, Surf. Coat. Technol. 200 (2006) 3698. [19] S. Buytoz, Surf. Coat. Technol. 200 (2006) 3734. [20] S. Mridha, H.S. Ong, L.S. Poh, P. Cheang, J. Mater. Proc. Technol. 113 (2001) 516.

Acknowledgements This research was supported by, the Specialized Research Fund for the Doctoral Program of Higher Education (No. 20020422032) and Doctoral Program of Shandong Province (2004BS04004).

X.H. Wang et al. / Materials Science and Engineering A 441 (2006) 60–67 [21] [22] [23] [24]

M.H. Korkut, O. Yilmaz, S. Buytoz, Surf. Coat. Technol. 157 (2002) 5. H.L. Wang, H.H. Li, F.Y. Yan, Wear 258 (2004) 1562. S. Jonsson, Metall. Mater. Trans., B 29 (1998) 371. C. Raghunath, M.S. Bhat, P.K. Rohatgi, Comp. Scr. Metall. 32 (1995) 577. [25] Y. Murakami, H. Kimura, Y. Nishimura, Trans. Natl. Res. Inst. Met. 1 (1959) 7.

67

[26] K.O. Knacke, K. Hesselmann, Thermo-Chemical Properties of Inorganic Substances, 2nd ed., Springer-Verlag, New York, 1991. [27] W.J. Lu, X.N. Zhang, D. Zhang, R.J. Wu, Y.J. Bian, P.W. Fang, Acta Metall. Sinica 35 (1999) 536. [28] A. Agarwal, N.B. Dahotre, J. Mater. Eng. Perform. 8 (1999) 479. [29] Y.J. Liang, Y.C. Che, Handbook of Data on Abiothermodynamics, Northeast University Press, Shenyang, 1991 (in Chinese).