- Email: [email protected]
Substrate temperature dependence of splat morphology for plasma-sprayed cast iron on aluminum surface
Surface & Coatings Technology 283 (2015) 234–240
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Substrate temperature dependence of splat morphology for plasma-sprayed cast iron on aluminum surface Ya-Zhe Xing a,b,c,⁎, Xing-Hang Li a, Qiang Wang a, Yong Zhang a, Xu-Ding Song c a b c
School of Materials Science and Engineering, Chang'an University, Xi'an, Shaanxi 710061, China State Key Laboratory for Mechanical Behavior of Materials, Xi'an Jiaotong University, Xi'an, Shaanxi 710049, China School of Construction Machinery, Chang'an University, Xi'an, Shaanxi 710064, China
a r t i c l e
i n f o
Article history: Received 15 August 2015 Revised 28 October 2015 Accepted in revised form 29 October 2015 Available online 30 October 2015 Keywords: Plasma spraying Cast iron Splashing Substrate temperature
a b s t r a c t In the plasma-sprayed coatings, the morphology of the splats plays an important role in optimizing the microstructure and performance of the coating. Especially, the splashing of impinging droplets during deposition weakens the splat–substrate/intersplat bonding and increases the porosity of the coating. Consequently, the integration of the coating is deteriorated. In the present study, cast iron particles were plasma-sprayed on the surface of polished aluminum substrate to form a single splat. During spraying, the surface of aluminum substrate was preheated in a temperature range from 25 °C to 320 °C. The impact of the substrate preheating temperature on the morphology of the splats was studied using a field emission scanning electron microscope. Results showed that the substrate temperature had significant effects on the morphology of splats. At room temperature, the splats mainly exhibited a splash type with network or radial lines on the splat periphery. While, the splashed splats deposited onto a high temperature substrate showed a star shape on the splat periphery. When the substrate was preheated to 130 °C, the mean percentage of the splashed splats decreased to a minimum value of 18.4% and the disk-like splats prevailed. With the increase of the substrate temperature from 130 °C to 290 °C, the mean percentage of the splashed splats increased monotonically to 78.3%. When the substrate temperature reached to 320 °C, the mean percentage of the splashed splats slightly reduced to 76.6%. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Due to the superior wear resistance and low cost, cast iron has been widely used in many industrial fields. In automobile industry, to enlarge the weight reduction and improve the fuel efficiency for automobile engines, the application of aluminum alloy cylinder block without cast iron liner has been growing for years. Nevertheless, the poor antiwear ability of aluminum alloy makes the cylinder bores easy to be worn and broken down early. Therefore, many techniques have been employed to strengthen the aluminum alloy cylinder bores [1]. It was suggested that thermal spray technique is one of the most versatile processes of depositing coating to improve the wear and corrosion performance [2, 3]. Especially, a thermally sprayed cast iron coating was considered one possible candidate for surface coating to strengthen aluminum cylinder bores [4]. In order to obtain favorable service performance of the cylinder block, it is necessary to ensure no coating could peel off from the substrate (cylinder bores) during working. In other words, enhancing the adhesion strength between the coating and the substrate is expected. ⁎ Corresponding author at: School of Materials Science and Engineering, Chang'an University, Xi'an, Shaanxi 710061, China. E-mail address: [email protected] (Y.-Z. Xing).
http://dx.doi.org/10.1016/j.surfcoat.2015.10.073 0257-8972/© 2015 Elsevier B.V. All rights reserved.
Generally, it is thought that the coating is mechanically bonded together with the substrate. The adhesion strength is usually enhanced by roughing and cleaning the substrate surface prior to spraying coating. During the coating deposition process and following working process of coated aluminum alloy cylinder block, the thermal stress, derived from the large difference in thermal expansion coefficient between aluminum alloy and cast iron, would make against the enhancement of adhesion strength between the coating and the substrate. A plasma-sprayed coating is built up by a stream of molten droplets. The individual droplet forms a splat through the processes of impacting, flattening, rapid solidification and cooling. During spraying, the deposition behavior of single droplet determines the droplet–substrate interaction, the splat structure, and the coating properties. According to a previous report [5], the deposition behavior of a droplet is nearly independent of the impingement of forthcoming droplets. Therefore, the deposition behavior of single droplet is important in determining splat formation. As reported up to now, there are lots of factors, including spray conditions [6,7], spray materials [8,9], state of substrate surface such as oxidation state [6,8], surface roughness [10–12] and the wetting of droplet to substrate surface [6,8], that could influence the splat formation. Generally, the splat forms two different morphologies during spraying, one presents a disk-like shape and the other behaves as a splashed
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240 Table 1 Substrate preheating temperature for splat deposition. Sample
S1
S2
S3
S4
S5
S6
Ts (°C)
25
130
190
240
290
320
splat which may deteriorate the coating quality. Usually, splashing occurs during the flattening of a droplet even on a flat surface at ambient atmosphere, which leads to the formation of irregularly complicated splats [8]. The occurrence of these splats goes against the formation of strong splat-substrate or intersplat bonding, as well as the adhesion or cohesion strength [13]. Many studies have showed that splash-type splats formed at low substrate temperatures and disk splats appeared at high substrate temperature [14–17], although the transition temperature [18] on splat morphology varies for different splat–substrate combinations. Morks et al. [19] prepared the cast iron splats on preheated aluminum alloy substrate surface in low-pressure atmosphere. They found that splashtype splats appeared at low substrate temperatures and both disk-
235
and star-shaped splats appeared at high temperatures. Whereas, under atmospheric conditions, no literature has been found dealing with the influence of substrate preheating on formation of splat morphology for cast iron–aluminum combination. In the present work, polished aluminum substrates were preheated to different temperatures, then individual cast iron splat was plasma-sprayed onto the surface of preheated substrate. The influence of preheating temperature on splat morphology formation was investigated.
2. Materials and experimental procedure Commercially available Fe–4.1C–1.5Si–1.5B–36.6Cr (all compositions in weight percent) powder (DG.Fe-05, Chengdu Daguang Thermal Spraying Materials Co., Ltd., China) was used as the starting powder. The powder exhibits a spherical shape with the size ranged from 20 to 75 μm. A commercial plasma spray system (GP-80, Aerospace Research Institute of Materials & Processing Technology, China) was used to
Fig. 1. Top view of the splats for (a) S1, (b) S2, (c) S3, (d) S4, (e) S5, and (f) S6.
236
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
deposit cast iron splats. Pure aluminum disk (20 mm in diameter, 4 mm in thickness) was employed as substrate. Before spraying, all aluminum disks were mirror-polished until their surface roughness (Ra) reached about 0.03 μm, which was measured by a surface roughmeter (TR-240, Beijing Time Group, China). Argon, hydrogen, and nitrogen were employed as primary gas, auxiliary gas and carrier gas, respectively. During spraying, the arc power and the spray distance were set at 30 kW and 120 mm, respectively. The pressures of Ar and H2 were fixed at 0.8 and 0.36 MPa, respectively. The flow rates of Ar and H2 were 32 and 6.3 L/min, respectively. The traverse speed of plasma torch was fixed at 1200 mm/s. The flow rate of the powders was fixed at 4 L/min. Table 1 lists substrate preheating temperature (Ts) for deposition of cast iron splats. After spraying, the surface morphology of the collected splats was characterized by a field emission scanning electron microscope (FESEM, S-4800, Hitachi, Japan) equipped with energy dispersive X-ray spectrometry (EDX, EX-250, Horiba, Japan). To study the effect of substrate preheating on the surface geometry of aluminum substrate during spraying, the surface roughness of post-sprayed substrates was measured by the roughmeter. This measurement in surface roughness was repeated for six times at the locations near the deposited splats, and the average value with standard deviation of each specimen was calculated. 3. Experimental results Fig. 1 shows the evolution of the splat morphology with the substrate preheating temperature ranged from 25 to 320 °C (S1–S6). It was found that the splat morphology changed greatly with the increase of the substrate temperature. This variation demonstrates that the splat morphology is greatly affected by the substrate preheating temperature. Generally, there are two typical splat morphologies named splash shape and disk shape. Interestingly, in the present work, many splats appear to present a pit at the center of each splat surface, named center-depressed splats. Clearly, no central depression was found in the splats deposited at room temperature, while the vast majority of the splats exhibited a center-depressed shape when the substrate temperature reached around 300 °C, as shown in Fig. 2. When the substrate temperature reached 130 °C, a small amount of minute splats exhibited a centerdepressed morphology (see Figs. 1b and 2). With the substrate temperature increased to 190 °C, the number of the center-depressed splats increased significantly. However, with the substrate temperature increased from 190 °C to 290 °C, a sharply rise in the number of the center-depressed splats appeared. With further increasing the substrate temperature to 320 °C, a slightly decrease in the number of the center-
Fig. 2. Percentage of center-depressed splats with increasing substrate temperature.
Fig. 3. Change in the size of central depression in the splats deposited at different substrate preheat temperatures.
depressed splats could be found. To further study the effect of the substrate temperature on the central depression of the deposited splats, the size ratio d/D (d is diameter of the center-depressed part, D is the splat diameter regardless of splashed part) was counted, as shown in Fig. 3. It was found that the substrate temperature had a significant effect on the scale of the center-depressed area. As the substrate temperature increased from 130 °C to 240 °C, the mean value of d/D reduced linearly to a minimum value of 0.67. With further increasing the substrate temperature to 320 °C, a significant increase in d/D could be observed in Fig. 3. When the substrate temperature was kept at room temperature, all deposited splats performed a typical splash-shape with a center splat surrounded by a ring of fragments, named annular-shaped splash splats (see Figs. 1a and 4a). While, as the substrate temperature was kept at above 130 °C, the deposited splashed splats showed a star-shaped splash on the splat periphery, named star-shaped splash splats (see Figs. 1b–f and 4b–d). In order to build the relationship between the substrate temperature and splat morphology, the statistical percentage of splashed splats was plotted in Fig. 5. It was found that a minimum value (18.4%) in mean percentage of the splashed splats appeared at 130 °C. As shown in Figs. 1b and 4b, there are many approximately disk-shaped splats. This indicates that the splash of the splats is suppressed by impeding the formation of the star-shaped splash on the splat periphery. However, when the substrate temperature increased from 130 °C to 290 °C, a notable increase in the number of the splashed splats occurred. With the substrate temperature increased to above 290 °C, the percentage of the splashed splats remained a stable value and was close to 80%, indicating that most splats exhibit a splash-type at high substrate temperature (above 290 °C). Moreover, from Fig. 1b–f, two interesting phenomena can be observed. One is the star-shaped splash for every splat accompanies the central depression, the other is the size of the star-shaped splash splats presents a temperature dependence at low temperature range. Clearly, all star-shaped splash splats showed a center-depressed morphology, and in turn overwhelming majority of the center-depressed splats displayed a distinct star-shaped splash morphology except for minimal center-depressed splats. The magnified photograph (Fig. 4c) confirms that these center-depressed splats are still presenting a splashed morphology, but the splash is not obvious. Therefore, SEM observations of low magnification photographs inevitably produce a deviation in the statistical results, i.e., statistical data for splashed splats are lower than that for center-depressed splats. As shown in Fig. 1b, only some small splats splashed for the specimen deposited at 130 °C. When the substrate temperature increased to 190 °C, both the small splats and the
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
237
Fig. 4. Classic top-view of single splat for (a) S1, (b) S3, (c) S4, and (d) S5.
medium size splats splashed (see Fig. 1c). However, with further increase of the substrate temperature, all splats (the splats of various sizes) splashed. That is to say, the high substrate temperature cannot produce any effect on the size of the splashed splats. 4. Discussion Previous reports [20,21] suggested that the initial substrate temperature has a profound effect on the splat cooling and the temperature history of the substrate. In the flattening process of a molten particle on a substrate surface, preheating substrate promotes the heat transfer from flattening particle to the substrate. As a result, the melting of the
Fig. 5. Change in the percentage of the splashed splats deposited at different substrate temperatures.
substrate surface may be possible. When the aluminum substrate is preheated, a reduction in the substrate hardness is expected. In other words, with increasing the substrate temperature, the substrate hardness decreases and the ductility of the substrate enhances accordingly. Therefore, during the impact of the liquid droplets, the substrate surface is prone to deformation. Furthermore, according to Li's reports [22,23], the droplet transient impact pressure mainly appears in the center area of the splat, which indicates a maximum transient impact pressure is imposed on the position in the substrate just below the center of the splat. Moreover, in consideration of fluid dynamics of melt during splat flattening, a splat has the tendency to pit in the center zone under the action of inertial force. Consequently, under the action of the kinetic energy of an impact droplet, a central depression with the spherical crown shape forms naturally. According to above results, the central depression and star-type splash of a splat always occur simultaneously when the substrate temperature is above 130 °C. Nevertheless, for the splats deposited at room temperature, no depression can be observed. From Fig. 1b, when the substrate temperature is 130 °C, there are less than 38% centerdepressed splats, and all of these splats are small. When a small powder particle is injected into plasma jet, a high temperature and a high impact velocity of the in-flight particle are expected. With a low temperature, it is difficult for the substrate to deform due to its high hardness. Therefore, the central depression of the splats is mainly dependent on the characteristics of the in-flight particles. High temperature improves the flowability of the melt by lowering its dynamic viscosity, while high speed enhances the kinetic energy by a square multiple. Consequently, both the high temperature and the high speed resulted from small powder particles promote the central depression of the splats. Moreover, the excellent flowability and high kinetic energy of the melt in small powder particles lead to a strong central depression. As a result, a high d/D appears for the small splat deposited at low temperature (see Fig. 3). With the increase of the substrate temperature, the center-depressed splats become bigger than those at low temperatures,
238
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
which results in a decrease in d/D and a minimum d/D for the splats collected at 240 °C. With the further increase of the substrate temperature, the deformation of the substrate becomes another factor besides the flowability of the melt for the depression of the flattening particles. Therefore, when the substrate temperature increased from 240 °C to 320 °C, an increase in d/D can be expected, as shown in Fig. 3. From above observation, when the substrate temperature increased from 25 °C to 130 °C, a significant change in splat morphology from annular-type to disk-type occurred. This can be reasonably explained by an evaporated-gas-induced splashing model [24]. When the substrate temperature remained at room temperature, the adsorbates (water) covered on the substrate surface. During the deposition process of the droplet, the substrate surface was heated rapidly by impact droplets. As a result, the adsorbates (water) gasified and evaporated from the substrate surface. Then, the air flow generated by the evaporation gas induced the splashing. Nevertheless, when the substrate temperature increased to 130 °C, the adsorbates (water) on the substrate surface were removed by high temperature (above boiling point of water). The annular-type splashing induced by the evaporation gas could not appear. Therefore, the diskshaped splats prevailed except for minute star-shaped splash splats due to the high in-flight speed and impact temperature of small droplets. In addition, the adsorbates on the substrate surface can reduce the wettability between the splats and the substrate surface. When the substrate temperature reached 130 °C, most adsorbents were evaporated and resulted in an improvement in the splat–substrate wettability, which would promote the formation of the disk-shaped splats [25]. Notably, with increase of the substrate temperature to above 130 °C, the splashed splats became a star-type morphology. At room temperature, due to the lowest flowing obstruction from the smooth substrate surface, the in-plane flow of the melt during the splashing induced by evaporated-gas exhibited an annular shape. While, the adsorbates (water) were removed prior to deposition of the droplet when the substrate surface was preheated to above 130 °C, the oxidation of the substrate surface became more and more serious and led to an increase in the roughness of the substrate surface (see Fig. 6), which resulted in an uneven radial flowing obstruction of the melt. Consequently, a starshaped splash splat was formed. When the substrate temperature increased from 130 °C to 290 °C, the number of the star-shaped splash splats increased dramatically. Clearly, this kind of splashing cannot be suppressed through substrate preheating. When the substrate temperature increases, the substrate becomes softening and even melting accordingly. As reported by Li [26], the melt of the substrate surface can induce the splashing of a flattening droplet. However, the substrate surface melting-induced splashing splats showed a split blocks shape rather than a star shape.
Fig. 6. Surface roughness of the substrates for S1, S2, S3, S4, S5 and S6.
This indicates that the present splashing cannot be explained by the substrate surface melting-induced splashing model. In other words, the substrate surface may not melt, but only soften in the present work. From above analyses, with increase of the substrate temperature, the substrate deformation during the impact of the droplets becomes more and more serious. The deformed substrate surface alters the flow direction of the droplet fluid and tends to form a free liquid jet that detaches from the substrate surface, which promotes the formation of the starshaped splash splats with the central depression. On the other hand, the oxide film generated on the substrate surface reduces the heat exchange between the flattening droplet and the substrate, which leads to a reduction in the cooling rate of the flattening droplet. As a result, the bottom of the splats maybe still remain liquid or semi-liquid with good mobility, which is easy to splash during flattening. Furthermore, according to previous reports [27,28], with the increase in the substrate surface roughness, the restriction to the flattening of the splats strengthens, and nonuniform flow resistance along different radial directions results in nonuniform flow of flowing fluid, which leads to the formation of star-shaped splash splats. Therefore, the number of disk-shaped or approximate disk-shaped splats reduces significantly, and while the number of the star-splashed splats increases with the increase of the substrate temperature. Specially, when the substrate temperature reached above 290 °C, as shown in Fig. 4d, a layered splash was observed, i.e., both the top and bottom of the splat splashed. Generally, it is difficult for the bottom of the splat to splash due to early solidification of the bottom of the
Fig. 7. EDS analysis of the splashed splat. (a) SEM photograph, (b) element distribution analyzed by EDS line scan of (a).
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
spreading droplet. In order to illuminate the formation of the splashing in Fig. 4d, an EDS analysis (15 kV accelerating voltage, 15.2 mm working distance) with line scan mode from splat center was conducted, as shown in Fig. 7. The EDS results suggested the presence of Fe (31.17%), C (10.5%), Cr (21.22%), Si (0.91%), O (2.99%), Al (32.48%) and Mg (0.73%). It is believed that Fe, Cr, C, Si, and O are from the splat, while Al and Mg (as impurity element in pure aluminum) are from the substrate. Importantly, strong Fe and Cr peaks in center of line scan patterns from 27 to 45 μm confirmed that the bottom of the spreading droplet splashed. A schematic model was proposed based on Li's models [24,26] to describe the formation of this splashing morphology, as shown in Fig. 8. When a molten cast iron droplet impacts on a polished aluminum substrate surface, the droplet deforms dramatically and starts to spread along the substrate surface. Meanwhile, deformation occurs at the substrate immediately beneath the droplet (Fig. 8a). As the flattening proceeds, the deformation of the substrate becomes more and more severe (Fig. 8b). As a consequence, pits form at the center of the droplet due to the deformation of the substrate and inertia force derived from the kinetic energy of the flowing fluid. As a result, a center-depressed morphology forms (Fig. 8c). Moreover, the deformed substrate surface alters the flow direction of the droplet fluid to form a free liquid jet that detaches from the substrate surface, which leads to the formation of the star-shaped splash splats with the central depression. As analyzed above, oxide film formed on the substrate surface reduces the heat transfer efficiency between the flattening droplet and the substrate. Consequently, the solidification and
239
cooling of the droplet are postponed. This keeps the flattening droplet in the liquid or semi-liquid at the bottom. Therefore, the drastic flowing of the droplet fluid prompts the splashing of the droplet at the bottom and top, as shown in Fig. 8d.
5. Conclusions The morphology of cast iron splats was significantly influenced by the substrate temperature. When the substrate was kept at room temperature, all deposited splats exhibited an annular-type splash morphology. While, the splashed splats deposited onto a high temperature substrate showed a star shape on the splat periphery. When the substrate was preheated to 130 °C, the mean percentage of the splashed splats decreased to a minimum value of 18.4% and the disk-like splats prevailed, indicating an annular to disk shape transition of the splats. However, when the substrate temperature was preheated to above 130 °C, the number of the disk-shaped splats decreases and the number of star-shaped splash splats with the central depression increases. With further increasing substrate temperature to above 290 °C, the mean percentage of the star-shaped splash splats increased to nearly 80%. Moreover, the bottom of the splats splashed during spreading, indicating the retardation in solidification and cooling of the spreading droplets. The proposed model could illustrate reasonably the formation of the starshaped splash splats collected on a high temperature substrate with central depression.
Fig. 8. Schematic diagram showing the flattening behavior of a cast iron droplet impacting a high temperature aluminum substrate. (a) Early stage of the droplet flattening, (b) further flattening of the droplet, (c) formation of the distinct central depression, (d) formation of the droplet splashing. Fk: flow direction of the spreading fluid driven by kinetic energy; Ff: flow trend inside the droplet; St: splashing at the top of the splat; Sb: splashing at the bottom of the splat.
240
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
Acknowledgments This work was supported by State Key Laboratory for Mechanical Behavior of Materials (No. 20131312), National Natural Science Foundation of China (No. 51301023) and the Fundamental Research Funds for the Central Universities, Chang'an University (No. 310831151079). References [1] G. Barbezat, Surf. Coat. Technol. 200 (2005) 1990–1993. [2] J. Rodríguez, A. Martín, R. Fernández, J.E. Fernández, Wear 255 (2003) 950–955. [3] W.J. Kim, S.H. Ahn, H.G. Kim, J.G. Kim, I. Ozdemir, Y. Tsunekawa, Surf. Coat. Technol. 200 (2005) 1162–1167. [4] M.F. Morks, Y. Tsunekawa, N.F. Fahim, M. Okumiya, Mater. Chem. Phys. 96 (2006) 170–175. [5] S. Kuroda, T.W. Clyne, Thin Solid Films 200 (1991) 49–66. [6] L. Bianchi, A. Grimaud, F. Blein, P. Lucchèse, P. Fauchais, J. Therm, Spray Technol. 4 (1995) 59–66. [7] G. Montavon, S. Sampath, C.C. Berndt, H. Herman, C. Coddet, J. Therm, Spray Technol. 4 (1995) 67–74. [8] M. Vardelle, A. Vardelle, A.C. Leger, P. Fauchais, D. Gobin, Influence of particle parameters at impact on splat formation and solidification in plasma spraying processes, J. Therm. Spray Technol. 4 (1995) 50–58. [9] C.-J. Li, J.-L. Li, W.-B. Wang, A. Ohmori, K. Tani, in: C. Coddet (Ed.), Thermal Spray: Meeting the Challenges of the 21st Century, ASM International, Nice, France 1998, pp. 481–487. [10] C. Moreau, P. Gougeon, M. Lamontagne, J. Therm. Spray Technol. 4 (1995) 25–33.
[11] S. Amada, K. Imagawa, S. Aoki, Surf. Coat. Technol. 154 (2002) 27–33. [12] M. Raessi, J. Mostaghimi, M. Bussmann, Thin Solid Films 506-507 (2006) 133–135. [13] M. Fukumoto, T. Yokoyama, K. Oku, Y. Tanaka, J. High Temp. Soc. 23 (1997) 240–246. [14] A. Hasui, S. Kitahara, T. Fukushima, Trans. Natl. Res. Inst. Met. 12 (1970) 9–20. [15] V.V. Sobolev, Morphology of splats of thermally sprayed coatings, in: C. Coddet (Ed.), Thermal Spray Meeting the Challenges of the 21st Century, 1, ASM International, Metals Park, OH 1998, pp. 507–510. [16] M. Fukumoto, Y. Huang, J. Therm. Spray Technol. 8 (1999) 427–432. [17] C.-J. Li, Y.-Z. Xing, Y. Li, C.-X. Li, G.-J. Yang, Proceedings of 2008 International Thermal Spray Conference, Maastricht, The Netherlands, 2008 1423–1428. [18] M. Fukumoto, S. Katoh, I. Okane, J. High Temp Soc. 1 (1995) 353–358. [19] M.F. Morks, Y. Tsunekawa, M. Okumiya, M.A. Shoeib, J. Therm. Spray Technol. 11 (2002) 226–232. [20] Y.-Z. Xing, C.-J. Li, J.-H. Qiao, G.-X. Wang, Proceedings of the ASME International Mechanical Engineering Congress and Exposition 2007, ASME, Seattle, 2008 1837–1844. [21] Y.-Z. Xing, Q.-L. Wei, J.-M. Hao, Adv. Mater. Res. 1129-1132 (2009) 78–82. [22] J.-L. Li, C.-J. Li, J. Mech. Sci. Technol. 19 (2000) 88–91. [23] J.-L. Li, C.-J. Li, J. Therm. Spray Technol. 13 (2) (2004) 229–238. [24] C.-J. Li, J.-L. Li, Surf. Coat. Technol. 184 (2004) 13–23. [25] K. Yang, M. Fukumoto, T. Yasui, M. Yamada, J. Thermal Spray Technol. 19 (2010) 1195–1205. [26] C.-J. Li, C.-X. Li, G.-J. Yang, Y.-Y. Wang, J. Thermal Spray Technol. 15 (2006) 717–724. [27] Z. Ma, S. Cao, F. Wang, Y. Liu, X. Xiao, Trans. Beijing Ins. Technol. 26 (4) (2006) 373–376 (in Chinese). [28] S. Shakeri, S. Chandra, Int. J. Heat Mass Transf. 45 (2002) 4561–4575.
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Substrate temperature dependence of splat morphology for plasma-sprayed cast iron on aluminum surface Ya-Zhe Xing a,b,c,⁎, Xing-Hang Li a, Qiang Wang a, Yong Zhang a, Xu-Ding Song c a b c
School of Materials Science and Engineering, Chang'an University, Xi'an, Shaanxi 710061, China State Key Laboratory for Mechanical Behavior of Materials, Xi'an Jiaotong University, Xi'an, Shaanxi 710049, China School of Construction Machinery, Chang'an University, Xi'an, Shaanxi 710064, China
a r t i c l e
i n f o
Article history: Received 15 August 2015 Revised 28 October 2015 Accepted in revised form 29 October 2015 Available online 30 October 2015 Keywords: Plasma spraying Cast iron Splashing Substrate temperature
a b s t r a c t In the plasma-sprayed coatings, the morphology of the splats plays an important role in optimizing the microstructure and performance of the coating. Especially, the splashing of impinging droplets during deposition weakens the splat–substrate/intersplat bonding and increases the porosity of the coating. Consequently, the integration of the coating is deteriorated. In the present study, cast iron particles were plasma-sprayed on the surface of polished aluminum substrate to form a single splat. During spraying, the surface of aluminum substrate was preheated in a temperature range from 25 °C to 320 °C. The impact of the substrate preheating temperature on the morphology of the splats was studied using a field emission scanning electron microscope. Results showed that the substrate temperature had significant effects on the morphology of splats. At room temperature, the splats mainly exhibited a splash type with network or radial lines on the splat periphery. While, the splashed splats deposited onto a high temperature substrate showed a star shape on the splat periphery. When the substrate was preheated to 130 °C, the mean percentage of the splashed splats decreased to a minimum value of 18.4% and the disk-like splats prevailed. With the increase of the substrate temperature from 130 °C to 290 °C, the mean percentage of the splashed splats increased monotonically to 78.3%. When the substrate temperature reached to 320 °C, the mean percentage of the splashed splats slightly reduced to 76.6%. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Due to the superior wear resistance and low cost, cast iron has been widely used in many industrial fields. In automobile industry, to enlarge the weight reduction and improve the fuel efficiency for automobile engines, the application of aluminum alloy cylinder block without cast iron liner has been growing for years. Nevertheless, the poor antiwear ability of aluminum alloy makes the cylinder bores easy to be worn and broken down early. Therefore, many techniques have been employed to strengthen the aluminum alloy cylinder bores [1]. It was suggested that thermal spray technique is one of the most versatile processes of depositing coating to improve the wear and corrosion performance [2, 3]. Especially, a thermally sprayed cast iron coating was considered one possible candidate for surface coating to strengthen aluminum cylinder bores [4]. In order to obtain favorable service performance of the cylinder block, it is necessary to ensure no coating could peel off from the substrate (cylinder bores) during working. In other words, enhancing the adhesion strength between the coating and the substrate is expected. ⁎ Corresponding author at: School of Materials Science and Engineering, Chang'an University, Xi'an, Shaanxi 710061, China. E-mail address: [email protected] (Y.-Z. Xing).
http://dx.doi.org/10.1016/j.surfcoat.2015.10.073 0257-8972/© 2015 Elsevier B.V. All rights reserved.
Generally, it is thought that the coating is mechanically bonded together with the substrate. The adhesion strength is usually enhanced by roughing and cleaning the substrate surface prior to spraying coating. During the coating deposition process and following working process of coated aluminum alloy cylinder block, the thermal stress, derived from the large difference in thermal expansion coefficient between aluminum alloy and cast iron, would make against the enhancement of adhesion strength between the coating and the substrate. A plasma-sprayed coating is built up by a stream of molten droplets. The individual droplet forms a splat through the processes of impacting, flattening, rapid solidification and cooling. During spraying, the deposition behavior of single droplet determines the droplet–substrate interaction, the splat structure, and the coating properties. According to a previous report [5], the deposition behavior of a droplet is nearly independent of the impingement of forthcoming droplets. Therefore, the deposition behavior of single droplet is important in determining splat formation. As reported up to now, there are lots of factors, including spray conditions [6,7], spray materials [8,9], state of substrate surface such as oxidation state [6,8], surface roughness [10–12] and the wetting of droplet to substrate surface [6,8], that could influence the splat formation. Generally, the splat forms two different morphologies during spraying, one presents a disk-like shape and the other behaves as a splashed
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240 Table 1 Substrate preheating temperature for splat deposition. Sample
S1
S2
S3
S4
S5
S6
Ts (°C)
25
130
190
240
290
320
splat which may deteriorate the coating quality. Usually, splashing occurs during the flattening of a droplet even on a flat surface at ambient atmosphere, which leads to the formation of irregularly complicated splats [8]. The occurrence of these splats goes against the formation of strong splat-substrate or intersplat bonding, as well as the adhesion or cohesion strength [13]. Many studies have showed that splash-type splats formed at low substrate temperatures and disk splats appeared at high substrate temperature [14–17], although the transition temperature [18] on splat morphology varies for different splat–substrate combinations. Morks et al. [19] prepared the cast iron splats on preheated aluminum alloy substrate surface in low-pressure atmosphere. They found that splashtype splats appeared at low substrate temperatures and both disk-
235
and star-shaped splats appeared at high temperatures. Whereas, under atmospheric conditions, no literature has been found dealing with the influence of substrate preheating on formation of splat morphology for cast iron–aluminum combination. In the present work, polished aluminum substrates were preheated to different temperatures, then individual cast iron splat was plasma-sprayed onto the surface of preheated substrate. The influence of preheating temperature on splat morphology formation was investigated.
2. Materials and experimental procedure Commercially available Fe–4.1C–1.5Si–1.5B–36.6Cr (all compositions in weight percent) powder (DG.Fe-05, Chengdu Daguang Thermal Spraying Materials Co., Ltd., China) was used as the starting powder. The powder exhibits a spherical shape with the size ranged from 20 to 75 μm. A commercial plasma spray system (GP-80, Aerospace Research Institute of Materials & Processing Technology, China) was used to
Fig. 1. Top view of the splats for (a) S1, (b) S2, (c) S3, (d) S4, (e) S5, and (f) S6.
236
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
deposit cast iron splats. Pure aluminum disk (20 mm in diameter, 4 mm in thickness) was employed as substrate. Before spraying, all aluminum disks were mirror-polished until their surface roughness (Ra) reached about 0.03 μm, which was measured by a surface roughmeter (TR-240, Beijing Time Group, China). Argon, hydrogen, and nitrogen were employed as primary gas, auxiliary gas and carrier gas, respectively. During spraying, the arc power and the spray distance were set at 30 kW and 120 mm, respectively. The pressures of Ar and H2 were fixed at 0.8 and 0.36 MPa, respectively. The flow rates of Ar and H2 were 32 and 6.3 L/min, respectively. The traverse speed of plasma torch was fixed at 1200 mm/s. The flow rate of the powders was fixed at 4 L/min. Table 1 lists substrate preheating temperature (Ts) for deposition of cast iron splats. After spraying, the surface morphology of the collected splats was characterized by a field emission scanning electron microscope (FESEM, S-4800, Hitachi, Japan) equipped with energy dispersive X-ray spectrometry (EDX, EX-250, Horiba, Japan). To study the effect of substrate preheating on the surface geometry of aluminum substrate during spraying, the surface roughness of post-sprayed substrates was measured by the roughmeter. This measurement in surface roughness was repeated for six times at the locations near the deposited splats, and the average value with standard deviation of each specimen was calculated. 3. Experimental results Fig. 1 shows the evolution of the splat morphology with the substrate preheating temperature ranged from 25 to 320 °C (S1–S6). It was found that the splat morphology changed greatly with the increase of the substrate temperature. This variation demonstrates that the splat morphology is greatly affected by the substrate preheating temperature. Generally, there are two typical splat morphologies named splash shape and disk shape. Interestingly, in the present work, many splats appear to present a pit at the center of each splat surface, named center-depressed splats. Clearly, no central depression was found in the splats deposited at room temperature, while the vast majority of the splats exhibited a center-depressed shape when the substrate temperature reached around 300 °C, as shown in Fig. 2. When the substrate temperature reached 130 °C, a small amount of minute splats exhibited a centerdepressed morphology (see Figs. 1b and 2). With the substrate temperature increased to 190 °C, the number of the center-depressed splats increased significantly. However, with the substrate temperature increased from 190 °C to 290 °C, a sharply rise in the number of the center-depressed splats appeared. With further increasing the substrate temperature to 320 °C, a slightly decrease in the number of the center-
Fig. 2. Percentage of center-depressed splats with increasing substrate temperature.
Fig. 3. Change in the size of central depression in the splats deposited at different substrate preheat temperatures.
depressed splats could be found. To further study the effect of the substrate temperature on the central depression of the deposited splats, the size ratio d/D (d is diameter of the center-depressed part, D is the splat diameter regardless of splashed part) was counted, as shown in Fig. 3. It was found that the substrate temperature had a significant effect on the scale of the center-depressed area. As the substrate temperature increased from 130 °C to 240 °C, the mean value of d/D reduced linearly to a minimum value of 0.67. With further increasing the substrate temperature to 320 °C, a significant increase in d/D could be observed in Fig. 3. When the substrate temperature was kept at room temperature, all deposited splats performed a typical splash-shape with a center splat surrounded by a ring of fragments, named annular-shaped splash splats (see Figs. 1a and 4a). While, as the substrate temperature was kept at above 130 °C, the deposited splashed splats showed a star-shaped splash on the splat periphery, named star-shaped splash splats (see Figs. 1b–f and 4b–d). In order to build the relationship between the substrate temperature and splat morphology, the statistical percentage of splashed splats was plotted in Fig. 5. It was found that a minimum value (18.4%) in mean percentage of the splashed splats appeared at 130 °C. As shown in Figs. 1b and 4b, there are many approximately disk-shaped splats. This indicates that the splash of the splats is suppressed by impeding the formation of the star-shaped splash on the splat periphery. However, when the substrate temperature increased from 130 °C to 290 °C, a notable increase in the number of the splashed splats occurred. With the substrate temperature increased to above 290 °C, the percentage of the splashed splats remained a stable value and was close to 80%, indicating that most splats exhibit a splash-type at high substrate temperature (above 290 °C). Moreover, from Fig. 1b–f, two interesting phenomena can be observed. One is the star-shaped splash for every splat accompanies the central depression, the other is the size of the star-shaped splash splats presents a temperature dependence at low temperature range. Clearly, all star-shaped splash splats showed a center-depressed morphology, and in turn overwhelming majority of the center-depressed splats displayed a distinct star-shaped splash morphology except for minimal center-depressed splats. The magnified photograph (Fig. 4c) confirms that these center-depressed splats are still presenting a splashed morphology, but the splash is not obvious. Therefore, SEM observations of low magnification photographs inevitably produce a deviation in the statistical results, i.e., statistical data for splashed splats are lower than that for center-depressed splats. As shown in Fig. 1b, only some small splats splashed for the specimen deposited at 130 °C. When the substrate temperature increased to 190 °C, both the small splats and the
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
237
Fig. 4. Classic top-view of single splat for (a) S1, (b) S3, (c) S4, and (d) S5.
medium size splats splashed (see Fig. 1c). However, with further increase of the substrate temperature, all splats (the splats of various sizes) splashed. That is to say, the high substrate temperature cannot produce any effect on the size of the splashed splats. 4. Discussion Previous reports [20,21] suggested that the initial substrate temperature has a profound effect on the splat cooling and the temperature history of the substrate. In the flattening process of a molten particle on a substrate surface, preheating substrate promotes the heat transfer from flattening particle to the substrate. As a result, the melting of the
Fig. 5. Change in the percentage of the splashed splats deposited at different substrate temperatures.
substrate surface may be possible. When the aluminum substrate is preheated, a reduction in the substrate hardness is expected. In other words, with increasing the substrate temperature, the substrate hardness decreases and the ductility of the substrate enhances accordingly. Therefore, during the impact of the liquid droplets, the substrate surface is prone to deformation. Furthermore, according to Li's reports [22,23], the droplet transient impact pressure mainly appears in the center area of the splat, which indicates a maximum transient impact pressure is imposed on the position in the substrate just below the center of the splat. Moreover, in consideration of fluid dynamics of melt during splat flattening, a splat has the tendency to pit in the center zone under the action of inertial force. Consequently, under the action of the kinetic energy of an impact droplet, a central depression with the spherical crown shape forms naturally. According to above results, the central depression and star-type splash of a splat always occur simultaneously when the substrate temperature is above 130 °C. Nevertheless, for the splats deposited at room temperature, no depression can be observed. From Fig. 1b, when the substrate temperature is 130 °C, there are less than 38% centerdepressed splats, and all of these splats are small. When a small powder particle is injected into plasma jet, a high temperature and a high impact velocity of the in-flight particle are expected. With a low temperature, it is difficult for the substrate to deform due to its high hardness. Therefore, the central depression of the splats is mainly dependent on the characteristics of the in-flight particles. High temperature improves the flowability of the melt by lowering its dynamic viscosity, while high speed enhances the kinetic energy by a square multiple. Consequently, both the high temperature and the high speed resulted from small powder particles promote the central depression of the splats. Moreover, the excellent flowability and high kinetic energy of the melt in small powder particles lead to a strong central depression. As a result, a high d/D appears for the small splat deposited at low temperature (see Fig. 3). With the increase of the substrate temperature, the center-depressed splats become bigger than those at low temperatures,
238
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
which results in a decrease in d/D and a minimum d/D for the splats collected at 240 °C. With the further increase of the substrate temperature, the deformation of the substrate becomes another factor besides the flowability of the melt for the depression of the flattening particles. Therefore, when the substrate temperature increased from 240 °C to 320 °C, an increase in d/D can be expected, as shown in Fig. 3. From above observation, when the substrate temperature increased from 25 °C to 130 °C, a significant change in splat morphology from annular-type to disk-type occurred. This can be reasonably explained by an evaporated-gas-induced splashing model [24]. When the substrate temperature remained at room temperature, the adsorbates (water) covered on the substrate surface. During the deposition process of the droplet, the substrate surface was heated rapidly by impact droplets. As a result, the adsorbates (water) gasified and evaporated from the substrate surface. Then, the air flow generated by the evaporation gas induced the splashing. Nevertheless, when the substrate temperature increased to 130 °C, the adsorbates (water) on the substrate surface were removed by high temperature (above boiling point of water). The annular-type splashing induced by the evaporation gas could not appear. Therefore, the diskshaped splats prevailed except for minute star-shaped splash splats due to the high in-flight speed and impact temperature of small droplets. In addition, the adsorbates on the substrate surface can reduce the wettability between the splats and the substrate surface. When the substrate temperature reached 130 °C, most adsorbents were evaporated and resulted in an improvement in the splat–substrate wettability, which would promote the formation of the disk-shaped splats [25]. Notably, with increase of the substrate temperature to above 130 °C, the splashed splats became a star-type morphology. At room temperature, due to the lowest flowing obstruction from the smooth substrate surface, the in-plane flow of the melt during the splashing induced by evaporated-gas exhibited an annular shape. While, the adsorbates (water) were removed prior to deposition of the droplet when the substrate surface was preheated to above 130 °C, the oxidation of the substrate surface became more and more serious and led to an increase in the roughness of the substrate surface (see Fig. 6), which resulted in an uneven radial flowing obstruction of the melt. Consequently, a starshaped splash splat was formed. When the substrate temperature increased from 130 °C to 290 °C, the number of the star-shaped splash splats increased dramatically. Clearly, this kind of splashing cannot be suppressed through substrate preheating. When the substrate temperature increases, the substrate becomes softening and even melting accordingly. As reported by Li [26], the melt of the substrate surface can induce the splashing of a flattening droplet. However, the substrate surface melting-induced splashing splats showed a split blocks shape rather than a star shape.
Fig. 6. Surface roughness of the substrates for S1, S2, S3, S4, S5 and S6.
This indicates that the present splashing cannot be explained by the substrate surface melting-induced splashing model. In other words, the substrate surface may not melt, but only soften in the present work. From above analyses, with increase of the substrate temperature, the substrate deformation during the impact of the droplets becomes more and more serious. The deformed substrate surface alters the flow direction of the droplet fluid and tends to form a free liquid jet that detaches from the substrate surface, which promotes the formation of the starshaped splash splats with the central depression. On the other hand, the oxide film generated on the substrate surface reduces the heat exchange between the flattening droplet and the substrate, which leads to a reduction in the cooling rate of the flattening droplet. As a result, the bottom of the splats maybe still remain liquid or semi-liquid with good mobility, which is easy to splash during flattening. Furthermore, according to previous reports [27,28], with the increase in the substrate surface roughness, the restriction to the flattening of the splats strengthens, and nonuniform flow resistance along different radial directions results in nonuniform flow of flowing fluid, which leads to the formation of star-shaped splash splats. Therefore, the number of disk-shaped or approximate disk-shaped splats reduces significantly, and while the number of the star-splashed splats increases with the increase of the substrate temperature. Specially, when the substrate temperature reached above 290 °C, as shown in Fig. 4d, a layered splash was observed, i.e., both the top and bottom of the splat splashed. Generally, it is difficult for the bottom of the splat to splash due to early solidification of the bottom of the
Fig. 7. EDS analysis of the splashed splat. (a) SEM photograph, (b) element distribution analyzed by EDS line scan of (a).
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
spreading droplet. In order to illuminate the formation of the splashing in Fig. 4d, an EDS analysis (15 kV accelerating voltage, 15.2 mm working distance) with line scan mode from splat center was conducted, as shown in Fig. 7. The EDS results suggested the presence of Fe (31.17%), C (10.5%), Cr (21.22%), Si (0.91%), O (2.99%), Al (32.48%) and Mg (0.73%). It is believed that Fe, Cr, C, Si, and O are from the splat, while Al and Mg (as impurity element in pure aluminum) are from the substrate. Importantly, strong Fe and Cr peaks in center of line scan patterns from 27 to 45 μm confirmed that the bottom of the spreading droplet splashed. A schematic model was proposed based on Li's models [24,26] to describe the formation of this splashing morphology, as shown in Fig. 8. When a molten cast iron droplet impacts on a polished aluminum substrate surface, the droplet deforms dramatically and starts to spread along the substrate surface. Meanwhile, deformation occurs at the substrate immediately beneath the droplet (Fig. 8a). As the flattening proceeds, the deformation of the substrate becomes more and more severe (Fig. 8b). As a consequence, pits form at the center of the droplet due to the deformation of the substrate and inertia force derived from the kinetic energy of the flowing fluid. As a result, a center-depressed morphology forms (Fig. 8c). Moreover, the deformed substrate surface alters the flow direction of the droplet fluid to form a free liquid jet that detaches from the substrate surface, which leads to the formation of the star-shaped splash splats with the central depression. As analyzed above, oxide film formed on the substrate surface reduces the heat transfer efficiency between the flattening droplet and the substrate. Consequently, the solidification and
239
cooling of the droplet are postponed. This keeps the flattening droplet in the liquid or semi-liquid at the bottom. Therefore, the drastic flowing of the droplet fluid prompts the splashing of the droplet at the bottom and top, as shown in Fig. 8d.
5. Conclusions The morphology of cast iron splats was significantly influenced by the substrate temperature. When the substrate was kept at room temperature, all deposited splats exhibited an annular-type splash morphology. While, the splashed splats deposited onto a high temperature substrate showed a star shape on the splat periphery. When the substrate was preheated to 130 °C, the mean percentage of the splashed splats decreased to a minimum value of 18.4% and the disk-like splats prevailed, indicating an annular to disk shape transition of the splats. However, when the substrate temperature was preheated to above 130 °C, the number of the disk-shaped splats decreases and the number of star-shaped splash splats with the central depression increases. With further increasing substrate temperature to above 290 °C, the mean percentage of the star-shaped splash splats increased to nearly 80%. Moreover, the bottom of the splats splashed during spreading, indicating the retardation in solidification and cooling of the spreading droplets. The proposed model could illustrate reasonably the formation of the starshaped splash splats collected on a high temperature substrate with central depression.
Fig. 8. Schematic diagram showing the flattening behavior of a cast iron droplet impacting a high temperature aluminum substrate. (a) Early stage of the droplet flattening, (b) further flattening of the droplet, (c) formation of the distinct central depression, (d) formation of the droplet splashing. Fk: flow direction of the spreading fluid driven by kinetic energy; Ff: flow trend inside the droplet; St: splashing at the top of the splat; Sb: splashing at the bottom of the splat.
240
Y.-Z. Xing et al. / Surface & Coatings Technology 283 (2015) 234–240
Acknowledgments This work was supported by State Key Laboratory for Mechanical Behavior of Materials (No. 20131312), National Natural Science Foundation of China (No. 51301023) and the Fundamental Research Funds for the Central Universities, Chang'an University (No. 310831151079). References [1] G. Barbezat, Surf. Coat. Technol. 200 (2005) 1990–1993. [2] J. Rodríguez, A. Martín, R. Fernández, J.E. Fernández, Wear 255 (2003) 950–955. [3] W.J. Kim, S.H. Ahn, H.G. Kim, J.G. Kim, I. Ozdemir, Y. Tsunekawa, Surf. Coat. Technol. 200 (2005) 1162–1167. [4] M.F. Morks, Y. Tsunekawa, N.F. Fahim, M. Okumiya, Mater. Chem. Phys. 96 (2006) 170–175. [5] S. Kuroda, T.W. Clyne, Thin Solid Films 200 (1991) 49–66. [6] L. Bianchi, A. Grimaud, F. Blein, P. Lucchèse, P. Fauchais, J. Therm, Spray Technol. 4 (1995) 59–66. [7] G. Montavon, S. Sampath, C.C. Berndt, H. Herman, C. Coddet, J. Therm, Spray Technol. 4 (1995) 67–74. [8] M. Vardelle, A. Vardelle, A.C. Leger, P. Fauchais, D. Gobin, Influence of particle parameters at impact on splat formation and solidification in plasma spraying processes, J. Therm. Spray Technol. 4 (1995) 50–58. [9] C.-J. Li, J.-L. Li, W.-B. Wang, A. Ohmori, K. Tani, in: C. Coddet (Ed.), Thermal Spray: Meeting the Challenges of the 21st Century, ASM International, Nice, France 1998, pp. 481–487. [10] C. Moreau, P. Gougeon, M. Lamontagne, J. Therm. Spray Technol. 4 (1995) 25–33.
[11] S. Amada, K. Imagawa, S. Aoki, Surf. Coat. Technol. 154 (2002) 27–33. [12] M. Raessi, J. Mostaghimi, M. Bussmann, Thin Solid Films 506-507 (2006) 133–135. [13] M. Fukumoto, T. Yokoyama, K. Oku, Y. Tanaka, J. High Temp. Soc. 23 (1997) 240–246. [14] A. Hasui, S. Kitahara, T. Fukushima, Trans. Natl. Res. Inst. Met. 12 (1970) 9–20. [15] V.V. Sobolev, Morphology of splats of thermally sprayed coatings, in: C. Coddet (Ed.), Thermal Spray Meeting the Challenges of the 21st Century, 1, ASM International, Metals Park, OH 1998, pp. 507–510. [16] M. Fukumoto, Y. Huang, J. Therm. Spray Technol. 8 (1999) 427–432. [17] C.-J. Li, Y.-Z. Xing, Y. Li, C.-X. Li, G.-J. Yang, Proceedings of 2008 International Thermal Spray Conference, Maastricht, The Netherlands, 2008 1423–1428. [18] M. Fukumoto, S. Katoh, I. Okane, J. High Temp Soc. 1 (1995) 353–358. [19] M.F. Morks, Y. Tsunekawa, M. Okumiya, M.A. Shoeib, J. Therm. Spray Technol. 11 (2002) 226–232. [20] Y.-Z. Xing, C.-J. Li, J.-H. Qiao, G.-X. Wang, Proceedings of the ASME International Mechanical Engineering Congress and Exposition 2007, ASME, Seattle, 2008 1837–1844. [21] Y.-Z. Xing, Q.-L. Wei, J.-M. Hao, Adv. Mater. Res. 1129-1132 (2009) 78–82. [22] J.-L. Li, C.-J. Li, J. Mech. Sci. Technol. 19 (2000) 88–91. [23] J.-L. Li, C.-J. Li, J. Therm. Spray Technol. 13 (2) (2004) 229–238. [24] C.-J. Li, J.-L. Li, Surf. Coat. Technol. 184 (2004) 13–23. [25] K. Yang, M. Fukumoto, T. Yasui, M. Yamada, J. Thermal Spray Technol. 19 (2010) 1195–1205. [26] C.-J. Li, C.-X. Li, G.-J. Yang, Y.-Y. Wang, J. Thermal Spray Technol. 15 (2006) 717–724. [27] Z. Ma, S. Cao, F. Wang, Y. Liu, X. Xiao, Trans. Beijing Ins. Technol. 26 (4) (2006) 373–376 (in Chinese). [28] S. Shakeri, S. Chandra, Int. J. Heat Mass Transf. 45 (2002) 4561–4575.