Surface & Coatings Technology 269 (2015) 273–278
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Preparation of thermal control coatings on Ti alloy by plasma electrolytic oxidation in K2ZrF6 solution Zhongping Yao a,⁎, Peibo Su b, Qiaoxiang Shen a, Pengfei Ju b, Chao Wu b, Yunfei Zhai b, Zhaohua Jiang a a b
School of Chemical Engineering and Technology, State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150001, PR China Shanghai Aerospace Equipment Manufacture, Shanghai 200245, PR China
a r t i c l e
i n f o
Available online 23 January 2015 Keywords: Plasma electrolytic oxidation Ceramic coating Thermal control property Thermal shock resistance Zirconate Ti alloys
a b s t r a c t Ceramic coatings with high emission and low absorptance were prepared on Ti6Al4V alloy by plasma electrolytic oxidation (PEO) in zirconate electrolyte. The effects of current density and working frequency on the structure and thermal control properties of the coatings were investigated. The phase composition, microstructure, thickness and roughness of coatings were examined by XRD, SEM, EDS, thickness measurement gauge and roughness measuring instrument, respectively. Thermal control properties of the coatings were studied with a UV–VIS–NIR spectrophotometer and an infrared reflectometer. The results show that the coatings are porous and composed of a large amount of KZr2(PO4)3, and a little monoclinic ZrO2, tetragonal ZrO2 and ZrP2O7 as well. The thickness of the coatings increases with the increase of the current density or the decrease of the working frequency while the roughness of the coatings increases with the increase of the current density and the working frequency. The increase of current density reduces the absorptance, but improves the emissivity; the increase of working frequency improves the absorptance, but does not change emissivity apparently. The coating prepared at 10 A/dm2, 50 Hz and 50 min, has the lowest solar absorptance (0.34) and the highest infrared emissivity (0.9). © 2015 Elsevier B.V. All rights reserved.
1. Introduction A thermal control system ensures operational functionality of devices and instruments of the spacecraft when running in orbit with the harsh environment temperature changes [1]. A thermal control coating is a conventional method to balance the temperature of the surface by adjusting the absorptance–emissivity ratio of the coating surface. Currently, there are two kinds of thermal control coatings: (1) a low absorptance–emissivity ratio coating and (2) a high absorptance– emissivity ratio coating. The low absorptance–emissivity ratio coatings are generally white coatings, optical solar reflectors or ceramic coatings [1–4], which are usually prepared by chemical vapor deposition, electron beam physical vapor deposition, magnetron sputtering, ion implanting, plasma spray and sol–gel [5–8]. These methods are usually of high cost or confirm weak adhesion between the coating and the substrate. Plasma electrolytic oxidation (PEO), also called micro-arc oxidation (MAO), is becoming a universal coating technique to improve corrosion resistance and abrasion resistance for magnesium, aluminum and titanium and their alloys in recent years [9–12]. Furthermore, the prepared coatings exhibit good adhesion to the substrate, and the composition and structure and color of the coatings can be adjusted by the doping of the electrolytes or the changing of the process parameters. Therefore,
⁎ Corresponding author. E-mail address:
[email protected] (Z. Yao).
http://dx.doi.org/10.1016/j.surfcoat.2015.01.032 0257-8972/© 2015 Elsevier B.V. All rights reserved.
PEO technique presents a promising application prospect in the field of thermal control coatings [4,11–14]. Zirconate is usually regarded as a candidate material for high temperature applications due to its low thermal conductivity, large thermal expansion coefficient, high temperature resistant performance, strong sintering resistance and high stability at high temperature [15,16]. Presently, there are only a few instances of research work with the PEO technique in zirconate solution [17–20]. In this work, K2ZrF6 and NaH2PO4 solution was used as electrolytes for the PEO process on Ti6Al4V to prepare high emissivity and low absorption coatings. The composition, the structure and the thermal control properties of coatings were also investigated. 2. Experimental details 2.1. Preparation of PEO coatings Plate samples of Ti6Al4V (wt.%: 89.3 Ti, 6.26 Al, 4.01 V, 0. 03 Fe, 0.10 C, 0.03 N), with dimensions of 25 mm × 20 mm × 0.8 mm were prepared by a wire-electrode cutting machine. The samples were polished using SiC papers and washed in distilled water, served as the positive electrode. A water cooled electrochemical bath made of stainless steel served as the counter electrode. An in-house manufactured pulsed electrical source of 10 kW was employed for the plasma electrolytic oxidation process in a galvanostatic regime. The electrolyte solution contained 6 g/L K2ZrF6 and 5 g/L NaH2PO4. The electric parameters and the treatment time are
274
Z. Yao et al. / Surface & Coatings Technology 269 (2015) 273–278
3. Results and discussion
Table 1 The technique parameters of PEO process. Variables
Current density (A/dm2)
Frequency (Hz)
Time (min)
Current density
6 8 10 12 8 8 8
50 50 50 50 50 275 500
50 50 50 50 50 50 50
Frequency
shown in Table 1. During PEO treatment, the electrolyte temperature was maintained below 30 °C using a cooling water flow. After the treatment, the coatings were flushed with distilled water and dried in air. Three samples were made under each condition to ensure the reliability of the experiments. 2.2. Coating characterizations The surface and cross-sectional morphologies and elemental compositions of the coatings were examined by a JSM-6480 scanning electron microscopy (SEM, Japan's Hitachi LTD) equipped with energy dispersive spectroscopy (EDS, JED-2200, Japan). The phase composition of coatings was examined by TTR-III type X-ray diffraction (XRD, D/max-rB, Ricoh, Japan), using a Cu Kα radiation as the excitation source. The coating thickness was measured with an eddy current coating thickness measurement gauge (TT260, Time Group Inc., China). The thickness was the average of ten measurements made at different locations of the sample. The coating roughness was examined with a surface roughness measuring instrument (TR200, Time Group Inc., China). 2.3. Thermal control property evaluation of the coatings The emissivity of the coatings was measured by a TEMP 2000, a portable infrared emissivity spectrometer in the range 250 nm–2500 nm at room temperature. The Perkin Elmer Lambda 950 UV–Vis–NIR spectrophotometer was used to measure the absorptance of PEO coatings. Thermal shock resistance tests were carried out in a muffle furnace at 500 °C and in the water at room temperature. During the thermal shock test, coated samples were kept for 3 min to make the sample temperature uniform in the muffle. Then the coated samples were taken out from the furnace and quenched into the water. The tests were repeated for 50 cycles. After the thermal shock tests, the surface states of the coatings were observed and the thermal control properties of the coatings were measured.
3.1. Thickness and roughness of the coatings The coatings prepared on titanium alloy by the PEO technique demonstrate a roughness of Ra 8.1 to 11.6 μm and appear white in color. Fig. 1 shows the thickness of coatings produced under different current densities and working frequencies. The thickness and roughness of the coatings increase with an increase of the current density. During the experimental process, the number of micro-arc discharge locations was increased and the anode micro-arc discharge intensity was enhanced with the increasing current density, which would lead to the increase of the sintered particles on the coating surface in size and number. Thus the thickness and roughness of the coatings are increased. On the other side, the thickness decreases while the roughness increases with an increase of the working frequency. The increase of the frequency relates to a shorter duration of the anode process per pulse and the accumulative action time is effectively increased. The qualitative experimental observation is that the discharging sparks become more dispersed and smaller with the increase of the working frequency, which illustrates that the energy per pulse provided by the power source is decreased. Therefore, the amount of the sintered substance is comparatively decreased, which gives rise to the decrease of the thickness. Also, the decrease of the sintered substance may be liable for the formation of more micro-sized pores, which consequently results in the increase of the roughness. 3.2. Surface morphologies and elemental analysis of the coatings Fig. 2 shows the surface morphologies of the coating produced under different current densities and working frequencies. No matter what experimental conditions are, there are a great number of micron-sized pores distributing randomly on the surface of these coatings. Micro-sized pores are formed by molten oxide and gas bubbles during the discharge breakdown process. The electron avalanches are generally responsible for the pore formation, which takes place at the voltage higher than the breakdown potential of the oxygen gas layer or the surface oxide layer [21]. The pores are increased in size with the increase of the current density, the largest pores of the coating (in panel (c) produced under 12 A/dm2) are of around 75 μm. The increase of the current density leads to an enhanced discharging energy, which contributes to the enlarged pore sizes during the PEO process. Besides, as the discharge intensity increases continuously, the accumulated production mass of oxide increases, which induces a gradually increasing particle size and overlaps the pores nearby [22]. Panels (b), (d) and (e) present that
Fig. 1. Thickness and roughness of the coating prepared under different current densities and working frequencies.
Z. Yao et al. / Surface & Coatings Technology 269 (2015) 273–278
275
Fig. 2. The surface morphologies of the coatings for different current densities and working frequencies. a) 6 A/dm2, 50 Hz; b) 8 A/dm2, 50 Hz; c) 12 A/dm2, 50 Hz; d) 8 A/dm2, 275 Hz; e) 8 A/dm2, 500 Hz.
the increase of the working frequency do not increase the particle size apparently, but increase the size of the micro-sized pores, which is consistent with the measured results of the roughness in Fig. 1(b). The elemental composition on the surface of the coating was analyzed by EDS with the results showed in Table 2. The coating is mainly composed of P, Zr, O and K. As we can see, Ti is not incorporated into
Table 2 The relative contents of main elements on the surface of the coatings by EDS analysis (at.%).
8 A/dm2, 50 Hz 12 A/dm2, 50 Hz 8 A/dm2, 500 Hz
O
P
Zr
F
Ti
Na
K
25 25 25
48 45 48
21 25 19
0 2 1
0 0 0
1 1 1
5 4 5
the surface of the coatings. Also, it can be noted from Table 2 that the changes of the current density and the frequency do not influence the relative contents of the main elements of the coatings under experimental conditions. 3.3. Section morphologies and elemental distribution of the coating Fig. 3 shows the section morphology of PEO coating produced under 8 A/dm2 and 50 Hz. The coating is porous, which is consistent with the surface images. These micro-sized pores are distributed on the section image and some of them nearly perforate through the entire oxide film. The thickness of the coatings is approximately more than 100 μm, which corroborates the data obtained from the thickness measure in Fig. 1. The two-layer structure of the coating is not very obvious, but it seems that the inner layer is much thinner while the outer layer is
276
Z. Yao et al. / Surface & Coatings Technology 269 (2015) 273–278
Fig. 3. The cross-sectional morphology (a) and the elemental distribution (b) of coatings prepared under 8 A/dm2 and 50 Hz.
much larger and part of the outer layer is spalled off and embodied by the resin. The reason for this two-layer structure can be explained as different transient temperature fields in the outer layer and in the inner layer [23]. Due to periodical plasma discharge leading to periodical temperature change of the coating surface, melt and solidification occur in the outer layer repeatedly, leading to a porous structure. While the transient temperature of the inner layer is lower than that of the outer layer, thus the inner layer grows slowly, resulting to compact structure. The elemental analysis along the inset line in Fig. 3(a) is shown in Fig. 3(b). From the substrate to the outer layer, Ti content decreases sharply and then remains nearly constant at a very low value, which shows that there is only a very small amount of Ti in the coating, which matches the EDS results in Table 2 well. The contents of other elements like O and P present obvious increasing trends. The elements Zr and K keep almost constant through the coating. Therefore it can be confirmed that the main mode of spark discharge in this PEO process could be the breakdown of the vapor envelopes instead of the breakdown of the oxide films, which facilitates the elements from the electrolyte into the coating [24].
3.4. Phase composition of the coatings Fig. 4 shows the XRD patterns of the coatings under different current densities and working frequencies. The coatings are crystalized well and consist mainly of KZr2(PO4)3, and there is a little t-ZrO2, m-ZrO2 and ZrP2O7 as well. Combined with the results of EDS analysis, the current density and the working frequency do not change the phase composition of the coatings clearly. Under the experimental conditions, the coatings have the similar compositions.
3.5. Emissivity and absorptance of the coatings Table 3 shows the absorptance and the emissivity of the coatings prepared under different current densities and working frequencies. As for the substrate, the absorptance is 0.47 and the emissivity is 0.15, which means that Ti alloy presents the weak thermal control property. After PEO process, the prepared coatings present different absorptance and emissivity. It is clear that the current density and the working frequency have a greater influence on the absorptance than on the emissivity of the coating. The absorptance decreases with the increase of the current density or the decrease of the working frequency. On the other hand, there is only a slight increase of the emissivity of the coating with the increase of the current density and the working frequency. The thermal control properties are related to the composition and structure of the coatings. According to EDS and XRD analyses, the coatings have similar phase composition and constant elemental distribution along the coating section. Therefore, the thermal control properties are mainly determined by their structure. In general, both thickness and roughness influence the thermal control properties of the coatings. Table 3 Absorptance and emissivity of coatings prepared under different current densities and working frequencies. Conditions Current density
Frequency Fig. 4. XRD patterns of the coatings prepared at different current densities and working frequencies.
Ti alloy substrate
6 A/dm2 8 A/dm2 10 A/dm2 12 A/dm2 50 Hz 275 Hz 500 Hz
Absorptance
Emissivity
0.39 0.36 0.33 0.33 0.36 0.40 0.48 0.47
0.88 0.88 0.89 0.90 0.88 0.89 0.89 0.15
Z. Yao et al. / Surface & Coatings Technology 269 (2015) 273–278
277
phase composition of the coatings after the thermal shock tests does not change, but the thickness decreases by 5% or so and the roughness seems to increase a little which might be due to the partial spalloff of the coating surface. Table 4 shows the absorptance and emissivity of the coatings before and after the thermal shock tests, which also indicates that absorptance and emissivity of the coatings are mainly influenced by the thickness and the roughness, respectively. By comparison, it seems that the higher current density is liable for the stability of the thermal control properties of the coatings. 4. Conclusions
Fig. 5. Absorptance and emissivity of the coating at the different thickness produced under 8 A/dm2 and 50 Hz.
Since current density and frequency influence the thickness and the roughness, they are further bound to change the absorptance and the emissivity of the coatings. The thickness and roughness are increased with the current density, which are conducive to reduce the absorptance and improve the emissivity but the former changes greater than the latter. However, the thickness and roughness present the opposite changing trends with the increase of the working frequency. As for the absorptance, the influence of the thickness is larger than that of the roughness, which means that the absorptance is mainly reduced with the increase of the thickness. As for the emissivity, the influence of the roughness is a little larger than that of the thickness, which means that the emissivity is mainly improved with the increase of the roughness. In order to further investigate the effects of thickness and roughness of the coating on the absorptance and emissivity. The coating was polished to different thickness and meantime the absorptance and the emissivity were measured with the results shown in Fig. 5. After polishing treatment, the thickness is decreased quickly and the roughness is also reduced slowly. Therefore, it is assumed that the roughness is comparatively constant. The absorptance and the emissivity change gradually at the different positions within the coating. From the surface to the inner layer, the absorptance increases quickly whereas the emissivity decreases slowly. This further confirms that the thickness is the more important factor for the change of the absorptance than the roughness and the roughness influences the emissivity more than the thickness. Interestingly, when the thickness is less than 60 μm within the coating, there is a different slope for the change of the absorptance and the emissivity, i.e. the absorptance increases more quickly while the emissivity decreases more slowly, compared with the outer layer of the coating. This also proves that the coating has a double layer structure. 3.6. Thermal shock resistance of the coatings The thermal shock resistance is another important aspect for the application of the thermal control coating, which was evaluated by the thermal shock tests under 20–500 °C for 50 times. After the thermal shock tests, the coating surface is almost the same. The coating is stable and there is a good adhesion between the substrate and the coating. The Table 4 Absorptance and emissivity of the coatings before and after thermal shock tests. Current density
6 A/dm2 12 A/dm2
Before
After
Absorptance
Emissivity
Absorptance
Emissivity
0.39 0.34
0.88 0.90
0.43 0.35
0.86 0.88
The high emissivity and low absorptance ceramic coating is prepared successfully on Ti6Al4V alloy in zirconate electrolyte by PEO technique. Based on the characterization and the investigation of the thermal control properties of the coatings, the following conclusions can be drawn. (1) The porous ceramic coating is composed of a large amount of KZr2(PO4)3, and a little t-ZrO2, m-ZrO2 and ZrP2O7 as well. The increase of the current density increases the thickness and the roughness of the coatings. And the increase of frequency decreases the thickness, but increases the roughness. The changes of the current density and the working frequency do not influence the phase composition of the coating apparently. (2) The coatings present high emissivity and low absorptance compared with the untreated substrate. The absorptance decreases with the increase of the current density or the decrease of the working frequency. The emissivity of the coating slightly increases with the increase of the current density and the working frequency. From the surface to the inner layer of the coating, the absorptance increases gradually whereas the emissivity decreases slowly.
Acknowledgment This work was supported by Shanghai Aerospace Science and Technology Innovation Fund Projects (Grant No. SAST201431) and State Key Laboratory of Urban Water Resource and Environment (Harbin Institute of Technology) (No. 2015DX07). References [1] A.K. Sharma, N. Sridhara, Adv. Space Res. 50 (2012) 1411–1424. [2] H.Y. Xiao, C.D. Li, D.Z. Yang, S.Y. He, Nucl. Inst. Methods Phys. Res. B Beam Interact. Mater. Atoms 266 (2008) 3375–3380. [3] N. Kayhan, R.S. Razavi, S. Choopani, Prog. Org. Coat. 74 (2012) 603–607. [4] S.T. Lu, W. Qin, X.H. Wu, X.D. Wang, G.M. Zhao, Mater. Chem. Phys. 135 (2012) 58–62. [5] M. Touzin, F. Beclin, J. Eur. Ceram. Soc. 31 (2011) 1661–1667. [6] H.K. Moon, S.J. Lee, J. Yoon, H. Kim, N.E. Lee, Mater. Res. Bull. 47 (2012) 2961–2965. [7] C.W. Hsu, T.C. Cheng, W.H. Huang, J.S. Wu, C.C. Cheng, K.W. Cheng, S.C. Huang, Thin Solid Films 518 (2010) 1953–1957. [8] R.G. Turri, R.M. Santos, E.C. Rangel, N.C. da Cruz, J.R.R. Bortoleto, J.H. Dias da Silva, C.A. Antonio, S.F. Durrant, Appl. Surf. Sci. 280 (2013) 474–481. [9] Y.H. Gu, S. Bandopadhyay, C.F. Chen, C.Y. Ning, Y.J. Guo, Mater. Des. 46 (2013) 66–75. [10] T. Lampke, D. Meyer, G. Alisch, B. Wielage, H. Pokhmurska, M. Klapkiv, M. Student, Mater. Sci. 46 (2011) 591–598. [11] S. Durdu, S. Bayramoglu, A. Demirtas, M. Usta, A.H. Ucisik, Vacuum 88 (2013) 130–133. [12] Y.J. Xu, Z.P. Yao, F.Z. Jia, Y.L. Wang, Z.H. Jiang, H.T. Bu, Curr. Appl. Phys. 10 (2010) 698–702. [13] Y. Goueffon, L. Arurault, C. Mabru, C. Tonon, P. Guigue, J. Mater. Process. Technol. 209 (2009) 5145–5151. [14] Y. Wang, L. Wang, China Surf. Eng. 22 (6) (2009) 8–18 (in Chinese). [15] L. Zhao, Y. Wang, X. Li, A.J. Wang, C.S. Song, Y.K. Hu, Int. J. Hydrog. Energy 38 (2013) 14415–14423. [16] L. Wang, L. Chen, Z.C. Yan, H.L. Wang, J.Z. Peng, J. Alloys Compd. 493 (2010) 445–452. [17] Z.P. Yao, Z.H. Jiang, X.L. Zhang, J. Am. Ceram. Soc. 89 (2006) 2929–2932. [18] Y. Han, J.F. Song, J. Am. Ceram. Soc. 92 (2009) 1813–1816.
278
Z. Yao et al. / Surface & Coatings Technology 269 (2015) 273–278
[19] V.S. Rudnev, K.N. Kilin, I.V. Malyshev, T.P. Yarovaya, P.M. Nedozorov, A.A. Popovich, Prot. Met. Phys. Chem. Surf. 46 (2010) 704–709. [20] T. Akatsu, T. Kato, Y. Shinoda, F. Wakai, Surf. Coat. Technol. 223 (2013) 47–51. [21] V. Shoaei-Rad, M.R. Bayati, F. Golestani-Fard, H.R. Zargar, J. Javadpour, Mater. Lett. 65 (2011) 1835–1838.
[22] X. Wu, P. Su, Z. Jiang, S. Meng, ACS Appl. Mater. Interfaces 2 (2010) 808–812. [23] Y.J. Guan, Y. Xia, G. Li, Surf. Coat. Technol. 202 (2008) 4602–4612. [24] Z.P. Yao, Y.L. Jiang, F.Z. Jia, Z.H. Jiang, F.P. Wang, Appl. Surf. Sci. 254 (2008) 4084–4091.