- Email: [email protected]
Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum
JMRTEC-1033; No. of Pages 9
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Available online at www.sciencedirect.com
www.jmrt.com.br
Original Article
Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum Xiaoxiao Huang, Shuchen Sun ∗ , Ganfeng Tu School of Metallurgy, Northeastern University, Shenyang 110819, China
a r t i c l e
i n f o
a b s t r a c t
Article history:
TiB2 anti-oxidation ceramic coatings were successfully synthesised on a Mo substrate by
Received 16 October 2019
using chemical vapor deposition (CVD). A scratch test was performed to measure the adhe-
Accepted 22 October 2019
sion strength of the TiB2 ceramic coatings to the Mo substrate. The results indicated good
Available online xxx
adhesion between the two materials; the adhesion strength was 125 mN. Furthermore, hightemperature oxidation tests indicated that the CVD TiB2 ceramic coating protected the Mo
Keywords:
substrate against oxidation up to 900 ◦ . Isothermal oxidation experiments revealed that the
Titanium diboride
Mo substrate covered with a 13-m-thick TiB2 ceramic coating was protected against oxida-
Coating
tion for a duration of 5 h. The oxidation kinetics followed a parabolic law for the first 2.5 h
Chemical vapor deposition
and a linear law in the next 2.5 h. © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the
Oxidation
CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1.
Introduction
Mo is a useful metal with many excellent physical and chemical properties, such as corrosion resistance, good electrical and thermal conductivities, and a high melting point (2622 ◦ C) [1,2]. Therefore, Mo and Mo alloys are widely used in the aerospace, metallurgy, nuclear, and energy industries, as well as in lighting technology, high-temperature furnace construction, sputtering targets, and high-performance electronics [3–5]. Nevertheless, the applicability of Mo is limited owing to its extremely poor resistance to oxidation at high temperatures. Mo is easily oxidised in air even at 427 ◦ C [5].
To improve the oxidation resistance of Mo, various methods have been employed, such as anti-oxidation coating and the addition of anti-oxidation elements (Si, B, etc.). However, the addition of B and Si is limited by the low solubility and poor workability [6]. The oxidation-resistance coating method is widely used for Mo and Mo-based alloys. In previous studies, two major types of coating materials were used for obtaining anti-oxidation protective layers on the surface of Mo: 1) MoSi2 [5] and MoSi2 -based composites, e.g., MoSi2 –SiO2 [2], (Mo, W)Si2 –Si3 N4 [7,8], MoSi2 + X SiC [9], and Mo-Si-B [10–17], and 2) metals (alloys), e.g., Ir [18], W [19–21], and Ni-20Cr [4]. Some ceramic materials also exhibit excellent properties, such as high-temperature oxidation resistance, a high melt-
∗
Corresponding author. E-mail: [email protected] (S. Sun). https://doi.org/10.1016/j.jmrt.2019.10.056 2238-7854/© 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
2
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Fig. 1 – Schematic design of the CVD system.
ing point, and high hardness. Titanium diboride (TiB2 ), which is a candidate material for ultrahigh-temperature ceramics, is attracting increasing attention. It has superior properties, such as a high melting point (approximately 3100 ◦ C [22]), high hardness (maximum 45 GPa [23]), good electrical and thermal conductivities, and chemical inertness [24]. Additionally, as a very important anti-oxidation ceramic material, TiB2 can resist oxidisation in air up to 1100 ◦ C [25]. Thus, TiB2 has been utilised for a protective coating for various applications. TiB2 as a protective coating has been used for cemented carbide cutting tools [26], moulds, and other tools [27,28]. However, there has been little research on the TiB2 ceramic as an anti-oxidation coating on the surface of Mo. In using a ceramic anti-oxidation coating, the major problem is poor adhesion between the ceramic coating and the metal substrate. However, the adhesion strength can be improved either via substrate surface treatment [29] or by increasing the substrate temperature [30]. Therefore, the TiB2 ceramic can be considered as a protective layer to improve the oxidation resistance of Mo metal. In discussions regarding the oxidation resistance of TiB2 coatings for Mo, Mo-B-Si coatings are also mentioned owing to their similar composition and oxidation performance. Mo-B-Si coatings have exhibited excellent high-temperature oxidation resistance for Mo and Mo alloys [11,12,14,15,31,32], particularly in the temperature range of 1200–1700 ◦ C [11,12,14]. However, the oxidation resistance of Mo-B-Si coatings is poor at temperatures below 1000 ◦ C [31–34]. At temperature of 650–750 ◦ C, pest oxidation of the coatings occurs [31,34]. Conversely, the TiB2 ceramic exhibits good and reliable oxidation resistance below 1000 ◦ C. Thus, it is of interest to examine the mechanical and oxidative properties of TiB2 ceramic coatings on Mo substrates. In our previous work [35], a high quality TiB2 ceramic coating was successfully synthesized on metal Mo by using a chemical vapor deposition (CVD) method, and the microstructure and morphology of the coatings on the Mo substrate have been examined. The objective of this study was to systematically investigate the effectiveness of a TiB2 ceramic coating for improving the oxidation resistance of Mo. Firstly, the adhesion strength and microhardness of the coating were measured.
Furthermore, the oxidation behaviour of the TiB2 ceramic coatings on the Mo substrate was investigated. Oxidation tests were performed on the coatings from room temperature to 900 ◦ C, which is the temperature required for Mo moulds in our future work.
2.
Experimental method
2.1.
Materials
The experimental materials are from our previous work. The TiB2 ceramic coatings were deposited on a sheet of metallic Mo using a gas mixture of H2 +TiCl4 + BCl3 under 1000 ◦ C by CVD. The overall chemical reaction is as follows. 2BCl3 (g) + 5H2 (g) + TiCl4 (g) = 10HCl (g) + TiB2 (s)
(1)
Within this work, the TiB2 ceramic coating was deposited on Mo substrate using a conventional vertical CVD reactor. The experimental equipment was self-made in this study. The schematic design of the experimental CVD system is illustrated in Fig. 1. The graphite deposition chamber was heated with a graphite resistance heating element. The temperature was measured by a platinum-rhodium thermocouple and with a temperature controller (CKW-3100, BCHY, Beijing, China) to ± 5 ◦ C. The TiCl4 was heated to boiling point of 135 ◦ C in a resistance furnace. TiCl4 , BCl3 and H2 were introduced into the bottom of the deposition chamber via stainless steel tubing, respectively. The TiCl4 gas path was heated by conventional heating tapes to o ensure that no condensation of the reagents occurred within the lines. It is well known that the adhesion between ceramic coatings and metal substrates is poor. Many factors affect the adhesion strength. Among the most important factors are the surface characteristics of the substrate. A roughened surface can improve the adhesiveness of ceramic coatings [36,37]. Hence, the surfaces of the Mo substrates were roughened by 600 grit silicon carbide sandpaper to improve the adhesiveness of the TiB2 ceramic coatings to the substrates. Then, the Mo substrates were ultrasonically cleaned with acetone for 30 min. Finally, the substrates were suspended in the centre
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
3
Fig. 2 – Appearance and cross-section images of experimental materials.
Table 1 – Deposition condition of TiB2 coating by CVD. substrate
molybdenum
Deposition temperature Deposition time Deposition pressure Pressure of H2 Flow rate of H2 Temperature of TiCl4 Flow rate of TiCl4 Flow rate of BCl3
1000 [◦ C] 2 [h] 40–50 [kPa] 0.15 [MPa] 635 [cm3 /min] 135 [◦ C] 130 [cm3 /min] 195 [cm3 /min]
of the deposition chamber by a Mo wire. A. J. Caputo et al. [38] reported that some properties of CVD TiB2 depend on the deposition temperature. They found that TiB2 coatings deposited at ≥900 ◦ C exhibited very low erosion rates and that the coating hardness increased with the deposition temperature. Hence, we used a high deposition temperature (1000 ◦ C) to obtain good coating properties. More detailed deposition conditions are presented in Table 1. The appearance and crosssection images of the materials we used in this work are shown in Fig. 2 [35].
2.2. Mechanical-property test of TiB2 ceramic coatings on Mo substrate Adhesion measurements were performed according to a standard test method for evaluating the adhesion strength between a vapor deposition film and a substrate (JB/T 8554, People’s Republic of China Machinery Industry Standard (PRCMIS)). The scratch method was employed for testing the adhesion of the TiB2 ceramic coatings to the Mo substrates. The adhesion strength was evaluated using a nanoscratch tester (TriboIndenter, Hysitron, Minneapolis, USA) with a diamond cube-corner stylus under a continuously increasing load. The load ranged from 0 to 250 mN, with a loading rate of 5 mN/s. Furthermore, a microhardness tester (FM-700, Future-tech, Kawasaki, Japan) was used to evaluate the microhardness of the TiB2 ceramic coating.
2.3.
Oxidation test
To study the oxidation process of the specimens, oxidation experiments were performed in a tube furnace with a temperature accuracy of ±5 ◦ C. These experiments were conducted in dry air at 300, 450, 600, 750, and 900 ◦ C. Additionally, isothermal-oxidation tests were performed at 900 ◦ C for 1, 2, 3, 4, and 5 h. Thermal analysis, thermogravimetry (TG), and differential scanning calorimetry (DSC) were performed to investigate the oxidation mechanisms. The tests were performed using a Netzsch STA449F3 thermoanalytical system (TG measurement error was 10−4 mg, temperature measurement error was <1 ◦ C) at a constant heating rate of 10 ◦ C/min in flowing air at a flow rate of 150 mL/min in the temperature range of 20–900 ◦ C. The duration of the isothermal was 5 h at 900 ◦ C. For comparison, thermogravimetric tests of uncoated Mo were also performed under the same oxidation conditions.
2.4. Characterisations of coating before and after oxidation The phases of the coatings formed on the Mo substrate were examined using X-ray diffraction (XRD, X’ Pert Pro, PANalytical B.V., Almelo, Netherlands) at 40 mA and 40 kV with Cu K␣ radiation ( = 1.541 Å, step size = 0.0334◦ , integration time of 19.69 s/step, scan rate of 0.216◦ s−1 from 5◦ to 90◦ ). Scanning electron microscopy (SEM, Ultra Plus, Zeiss, Heidenheim, Germany) and energy-dispersive X-ray spectroscopy (EDS, X-Max 50, Oxford Instruments, Oxford, UK) were performed to investigate the microstructure, morphology, and composition of the coatings. XRD was also used to determine the phase composition of the oxidation products on the surface of the specimens after the oxidation experiments. The oxidation products were imaged and analysed using SEM, and their compositions were analysed using EDS.
3.
Results and discussion
3.1.
Mechanical properties of TiB2 coatings
According to a standard method for testing the adhesion strength between a vapor deposition film and a substrate (JB/T
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
4
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Fig. 3 – Coefficient of friction plots for the scratch testing of the TiB2 coating on the Mo substrate.
8554 from PRCMIS), scratch adhesion tests were performed by applying continuous loading from 0 to 250 mN with a 5-mN/s increase rate. The adhesive strength was assessed according to the critical load, which was identified by an abrupt change in the coefficient of friction. To reduce the influence of the surface roughness in the evaluation of the coefficient of friction, the surface roughness must be <0.32 m according to the test standard JB/T 8554. Hence, the coating was polished by 2000 grit silicon carbide sandpaper before the scratch tests. The surface roughness of the coating was measured using a three-dimensional laser scanning microscope (LEXT OLS4100, Olympus, Tokyo, Japan). The results of the roughness test indicated that after polishing, the average surface roughness was 0.26, which is 0.32 lower than the value specified by the test standard JB/T 8554. Fig. 3 shows the coefficient of friction plot for the scratch testing of the TiB2 ceramic coating on a pure Mo substrate. Two turning points are observed in the coefficient of friction plot: the lower critical load and upper critical load, corresponding to the critical loads for failure of the coating. Before the lower critical load was reached, the coefficient of friction increased slowly and fluctuated slightly with the increasing load. At this stage, the coating exhibited good adhesion to the substrate and was polished by the indenter. Then, we observed an abrupt increase in the coefficient of friction over a short period of time, indicating that the coating began to crack and peel off. Therefore, the lower critical load was defined as the adhesion strength of the coating to the substrate. As the load continued to increase, an upper critical load was observed. Subsequently, the coating was removed from the substrate. Hence, the friction coefficient increased slowly with the increasing load. According to the results, as shown in Fig. 3, the adhesion strength of the TiB2 ceramic coatings to the Mo substrate was approximately 125 mN. Q. L. Lv and J. Gao [39] reported that good adhesion of a Hf metal coating to a Mo substrate was achieved using ion-beam sputtering; the adhesion strength was approximately 120 mN. The adhesion of the as-prepared CVD TiB2 coatings to the Mo substrate in the present study was as good as that of the previously reported
Fig. 4 – TG weight-gain data under continuous oxidation for the Mo substrate covered with the TiB2 coating and the pure Mo specimen in air up to 900 ◦ C with a heating rate of 10 ◦ C/min.
Hf metal coating. Additionally, the TiB2 coating exhibited high microhardness relatively to MoSi2 coating on Mo substrate. The mean average microhardness value of the coating we obtained was 28 GPa and the maximum microhardness value was 30 GPa. But the MoSi2 coating microhardness only have a mean value of 13 GPa [40]. Generally, the hardness values for TiB2 films can vary between 25 and 36 GPa [27]. In our work, hardness of as-preparation TiB2 coatings on Mo substrate is in accordance with the consistently range of TiB2 hardness value [23,41–43].
3.2.
TG and DSC of TiB2 coated on Mo
Fig. 4 presents the weight-gain data obtained during continuous oxidation of Mo covered with the TiB2 coating. For comparison, the data for a pure Mo specimen are also shown. The possible oxidation reactions are as follows. 2TiB2 (s) +5O2 (g) = 2TiO2 (s) +2B2 O3 (l)
(2)
B2 O3 (l) = B2 O3 (g)
(3)
Mo (s)+3O2 (g) = 2MoO3 (s)
(4)
3MoO3 (s) = Mo3 O9 (g)
(5)
The weight gain of the pure Mo and the Mo covered with the TiB2 coating started at 550 and 720 ◦ C, respectively, in agreement with previously reported data for the oxidation of TiB2 [44]. The rate of weight gain for the pure Mo spec˜ ◦ C; subsequently, the weight-gain imen increased up to 790 rate remained high up to 850 ◦ C, and then rapid weight loss occurred. These results can be explained by the fact that Mo is very easily oxidised at low temperatures in air, and its oxide (MoO3 ) is easily volatilised according to reaction (5) [45]. However, the Mo covered with the TiB2 coating exhibited entirely different weight-gain trends. Above 720 ◦ C, the sample exhib-
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
5
Fig. 5 – TG and DSC curves for the oxidation of TiB2 coated on Mo in air, with the heating segment at a heating rate of 10 ◦ C/min up to 900 ◦ C and isothermal oxidation at 900 ◦ C up to 5 h. Fig. 6 – XRD patterns of Mo coated with TiB2 oxidised in air at different temperatures for 5 h. ited very slow weight gain up to the final temperature. This indicates that the TiB2 functioned as a protective layer on the surface of the pure Mo metal and improved the oxidation resistance of the Mo. The thermal-analysis results for the TiB2 coated on Mo in air are presented in Fig. 5. The heating rate was 10 ◦ C/min in the heating segment up to 900 ◦ C, and the duration of the isothermal oxidation was 5 h at 900 ◦ C. According to the TG curve in Fig. 5, a very small weight gain (0.03%) occurred in the initial oxidation period up to 120 ◦ C. Although this weight gain was negligible, it indicates the start of the oxidation, as supported by the DSC curve. The oxidation reaction was followed by reaction (2). Subsequently, the weight did not change until
the temperature reached 400 ◦ C, owing to the formation of low-viscosity B2 O3 glass that functioned as an O barrier. When the oxidation temperature increased to 450 ◦ C, the vaporisation of B2 O3 (liquid) (see reaction (3)) occurred more vigorously, resulting in weight loss. When the temperature increased to 720 ◦ C, slight weight gain occurred, because a large amount of B2 O3 fluid covered the surface of the TiB2 coating, preventing its further oxidation. Above this temperature, rapid weight gain occurred up to 900 ◦ C because of the rapid vaporisation consumption of the B2 O3 glass, causing part of the TiB2 to be exposed in air, which led to further oxidation. Thereafter,
Fig. 7 – SEM images and EDS mapping of Mo coated with TiB2 oxidised in air at 300–900 ◦ C for 5 h. Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
6
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
isothermal oxidation occurred for 5 h. The weight-change rate of the specimen was dependent on the oxidation time. For the first 2.5 h, the weight-gain rate decreased with the oxidation time, and the oxidation kinetics of the specimen essentially followed a parabolic law. Then, the weight-gain rate increased, and the oxidation kinetics followed a linear law.
3.3.
Phase and microstructure of oxide scales
To further examine the oxidation process of the TiB2 ceramic coatings formed on Mo, samples prepared under identical conditions were oxidised at 300, 450, 600, 750, and 900 ◦ C for 5 h. XRD was employed to identify the phases of the oxide scales, and their morphology was examined using SEM. The results are shown in Figs. 6 and 7, respectively. When the samples were exposed to air at temperatures of 300 and 450 ◦ C for 5 h, only the peaks of TiB2 were detected, as shown in Fig. 6. However, as shown in Fig. 5, the TG and DSC curves indicated that oxidation (reaction (2)) occurred at these two temperatures. It is considered that the phases of the reaction products were undetected because of their small amount. Indeed, the weight gain was only 0.03%, according to the foregoing TG data. At 600 ◦ C, the TiO2 peaks were observed, but the B2 O3 peaks were absent. Additionally, the B2 O3 peaks were absent at 750 and 900 ◦ C, as shown in Fig. 6. This is because the B2 O3 evaporated with the increasing temperature, as confirmed by the EDS mapping in Fig. 7. With the increasing temperature, the B content gradually decreased, and the O content increased. When the temperature increased to 900 ◦ C, the TiB2 peaks completely disappeared, and only TiO2 peaks remained. After the total consumption of TiB2 (T = 900 ◦ C and 5-h oxidation time), important changes were observed, as shown in Fig. 7(e): high-crystallinity TiO2 (crystallites) was formed on the surface. The XRD patterns (Fig. 6) and EDS mapping (Fig. 7(e)) indicated that Mo was present on the surface of the coating, suggesting that the TiB2 coating (13 m thick) lost its ability to protect the Mo substrate after 5 h of oxidation at 900 ◦ C in air. Furthermore, isothermal oxidation studies for the Mo substrate coated with the TiB2 ceramic coating were performed in air at 900 ◦ C for 1, 2, 3, 4, and 5 h. Fig. 8 presents the XRD patterns of Mo coated with TiB2 oxidised in air at 900 ◦ C for different durations. As shown, TiO2 and B2 O3 peaks were detected, as well as TiB2 peaks, after 1 h of oxidation. Thus,
Fig. 8 – XRD patterns of Mo coated with TiB2 oxidised in air at 900 ◦ C for different durations.
both TiO2 and B2 O3 were formed on the surface after 1 h of oxidation. The intensity of the B2 O3 peak was reduced after 2 h, and the B2 O3 peak disappeared after 3 h. As described previously, the volatilisation of B2 O3 increased with time, causing the disappearance of B2 O3 . After 2 h of oxidation, only the TiB2 peak remained. After 5 h of oxidation, the TiB2 peak disappeared, and a Mo peak was observed. Hence, when the sample was exposed to air for 5 h at 900 ◦ C, the only phase on the surface was TiO2 . Fig. 9 shows SEM images of the specimens after oxidation at 900 ◦ C for various durations. As indicated by the SEM images of the as-prepared TiB2 coating surface in our previous work [35], the coating typically exhibited a rod-like particle microstructure with a small gap. As shown in Fig. 9(a), a liquid phase filled the small gap. According to the results of the foregoing analysis, the liquid phase was B2 O3 . This is consistent with the EDS mapping results in Fig. 9(a). Comparing Fig. 9(a)–(c) reveals that the B2 O3 content decreased with the exposure time. Fig. 9(d) and (e) indicate that crystalline TiO2 was formed on the surface of the Mo substrate, supporting the foregoing XRD results and TG data.
Fig. 9 – SEM images and EDS mapping of Mo coated with TiB2 oxidised in air at 900 ◦ C for different durations. Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
7
Fig. 10 – Cross-sectional SEM images and element line scanning EDS analyses of the TiB2 coating after oxidation at 900 ◦ C for different durations in air.
Fig. 10 presents cross-sectional SEM images and element line scanning EDS analysis results for the oxidation of the TiB2 coating at 900 ◦ C for different durations. Oxide scales were formed after oxidation for 1, 2, 3, 4, and 5 h at 900 ◦ C. As shown in Fig. 10(a)–(e), the TiB2 coatings were generally oxidised with the increasing oxidation time, and there were obvious boundaries between the oxide layer and the unoxidised layer (TiB2 ). For the five aforementioned oxidation durations, the thicknesses of the oxide scales were approximately 3, 6, 9, 10, and 13 m, respectively. These results indicate that the oxidation rate was higher in the initial oxidation period than in the later stage, in accordance with the TG results shown in Fig. 5. The results of element line scanning EDS analysis indicated that the oxide scales mainly comprised O and Ti. The B content was lower in the oxide scales than in the TiB2 coatings. This suggests that the B2 O3 evaporated during the oxidation at 900 ◦ C. The jagged concentration line profiles of B, Ti, and O for the oxide scales resulted from pores in the oxide scales. The pores were due to the evaporation of B2 O3 formed during the oxidation of the dense TiB2 coating. Conversely, the smooth concentration line profiles indicate that the TiB2 coatings were dense.
3.4.
the oxidation of the coating depended on the formation of B2 O3 . The oxidation process of the TiB2 coating started at a low temperature (100 ◦ C), as indicated by the TG data (Fig. 5). At this temperature, the surface of the coating with the rod-like TiB2 phase was surrounded by low-viscosity B2 O3 glass resulting from the slight oxidation of the TiB2 coating (see Figs. 8(b) and 9 (a)). The B2 O3 , as a protective layer, prevented the further access of O ions. Thus, the formation of the B2 O3 layer prevented the further oxidation of the TiB2 coating. However, the volatilisation of B2 O3 became noticeable above 720 ◦ C, resulting in rapid oxidation of the TiB2 coating. A high oxidation rate was detected up to 900 ◦ C through TG (Fig. 5). Isothermal oxidation studies (at 900 ◦ C) revealed that the oxidation rate decreased with the oxidation time, owing to the formation of a new B2 O3 layer. However, the oxidation kinetics of the sample exhibited different trends at isothermal temperature 900 ◦ C duration of 5 h. For the first 2.5 h, the oxidation kinetics of the sample followed a parabolic law. For the next 2.5 h, the oxidation kinetics essentially followed a linear law. After oxidation for 5 h at 900 ◦ C, XRD data (Fig. 8) and SEM (Fig. 9) images indicated the complete consumption of TiB2 and the volatilisation of B2 O3 , and only a porous and non-protective TiO2 crystal was observed on the Mo substrate.
Oxidation mechanism
The foregoing results indicate that the CVD TiB2 ceramic coating formed on the surface of the Mo metal protected the Mo substrate against oxidation up to 900 ◦ C in air. B2 O3 was formed in the process of oxidation, and the protection against
4.
Conclusions
To improve the oxidation resistance of Mo metal, a TiB2 coating was synthesised on the surface of pure metallic Mo using
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
8
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
CVD. According to the experimental results, the following conclusions are drawn. (1) The TiB2 ceramic coating, as an anti-oxidation protective layer on the Mo metal, was successfully obtained on the pure Mo substrate via CVD. The as-deposited TiB2 ceramic coating exhibited good adhesion to the Mo substrate and a high microhardness. These good mechanical properties of the coating were due to the roughened surface of the substrate and the high deposition temperature. (2) The weight gain of only 0.03% indicates that the TiB2 coating exhibited high oxidation resistance during oxidation below 720 ◦ C. The TiB2 coatings also exhibited good oxidation resistance in the range of 720–900 ◦ C; the weight gain increased to 0.4% but was far smaller than the 2.3% weight gain of pure Mo. In this temperature range, the oxidation resistance depended on the coating-layer thickness and oxidation time. At 900 ◦ C, the 13-m-thick TiB2 layer protected the Mo substrate against oxidation for up to 5 h. (3) The oxidation mechanism of the TiB2 coating was examined from room temperature to 900 ◦ C. The oxidation process of the TiB2 coating started at a low temperature (100 ◦ C). The oxidation resistance of the TiB2 coating depended on the state of the reaction product B2 O3 . Below 720 ◦ C, B2 O3 functioned as a low-viscosity diffusion barrier that covered the surface of the coating, preventing further ingress of O ions and thereby preventing further oxidation of the TiB2 . At 720–900 ◦ C, the B2 O3 glass was consumed by vaporisation, which slightly reduced the protective ability. (4) The oxidation kinetics of the sample exhibited different trends during oxidation for 5 h at 900 ◦ C. For the first 2.5 h, the oxidation kinetics essentially followed a parabolic law. For the next 2.5 h, they followed a linear law.
Conflict of interest statement The authors declared that they have no conflicts of interest to this work.
Acknowledgments The authors thank Prof. Puqi Jia (Lanzhou University) for technically reviewing the manuscript. We gratefully acknowledge Prof. Bart Blanpain (KU Leuven) for his careful reading of and linguistic assistance with the manuscript.
references
[1] Zhang H, Gu S. Preparation and oxidation behavior of MoSi2 –CrSi2 –Si3 N4 composite coating on Mo substrate. Int J Refract Met Hard Mater 2013;41:128–32. [2] Nyutu EK, Kmetz MA, Suib SL. Formation of MoSi2 –SiO2 coatings on molybdenum substrates by CVD/MOCVD. Surf Coat Technol 2006;200(12–13):3980–6. [3] Babinsky K, Weidow J, Knabl W. Atom probe study of grain boundary segregation in technically pure molybdenum. Mater Charact 2014;87:95–103.
[4] Huang C, Zhang Y, Vilar R. Microstructure and anti-oxidation behavior of laser clad Ni–20Cr coating on molybdenum surface. Surf Coat Technol 2010;205(3):835–40. [5] Suzuki RO, Ishikawa M, Ono K. MoSi2 coating on molybdenum using molten salt. J Alloys Compd 2000;306(1-2):285–91. [6] Kruzic JJ, Schneibel JH, Ritchie RO. Fracture and fatigue resistance of Mo–Si–B alloys for ultrahigh-temperature structural applications. Scripta Mater 2004;50(4):459–64. [7] Zhang H, Lv J, Wu Y. Oxidation behavior of (Mo,W)Si2 –Si3 N4 composite coating on molybdenum substrate at 1600◦ C. Ceram Int 2015;41(10):14890–5. [8] Yoon J-K, Kim G-H, Han J-H. Low-temperature cyclic oxidation behavior of MoSi2 /Si3 N4 nanocomposite coating formed on Mo substrate at 773 K. Surf Coat Technol 2005;200(7):2537–46. [9] Govindarajan S, Moore JJ, Ohno TR. Interfacial characterization of MoSi2 + x SiC composite thin films on molybdenum substrates. Mrs Proc 2011;458. [10] Taylor M, Perepezko JH. Hot corrosion of Mo–Si–B coatings. Oxid Met 2017;87(5-6):705–15. [11] Kiryukhantsev-Korneev PV, Iatsyuk IV, Shvindina NV. Comparative investigation of structure, mechanical properties, and oxidation resistance of Mo-Si-B and Mo-Al-Si-B coatings. Corros Sci 2017;123:319–27. [12] Su L, Lu-Steffes O, Zhang H. An ultra-high temperature Mo–Si–B based coating for oxidation protection of NbSS /Nb5 Si3 composites. Appl Surf Sci 2015;337:38–44. [13] Lu-Steffes OJ, Su L, Jackson DM. Mo-Si-B coating for improved oxidation resistance of niobium. Adv Eng Mater 2015;17(7):1068–75. [14] Wu J, Wang W, Zhou C. Microstructure and oxidation resistance of Mo–Si–B coating on Nb based in situ composites. Corros Sci 2014;87:421–6. [15] Zhang Y-L, Li H-J, Hu Z-X. Microstructure and oxidation resistance of Si–Mo–B coating for C/SiC coated carbon/carbon composites. Corros Sci 2013;72:150–5. [16] Ritt P, Sakidja R, Perepezko JH. Mo–Si–B based coating for oxidation protection of SiC–C composites. Surf Coat Technol 2012;206(19–20):4166–72. [17] Perepezko JH, Rioult F, Sakidja R. Oxidation performance of high temperature Mo-Si-B alloys and coatings. Mater Sci Forum 2008;595–598:1065–74. [18] Wang L, Chen Z, Zhang P. Ir coating prepared on Mo substrate by double glow plasma. J Coat Technol Res 2008;6(4):517–22. [19] Lian Y, Liu X, Wang J. Influence of surface morphology and microstructure on performance of CVD tungsten coating under fusion transient thermal loads. Appl Surf Sci 2016;390:167–74. [20] Jiang F, Zhang Y, Sun N. Tungsten coating prepared on molybdenum substrate by electrodeposition from molten salt in air atmosphere. Appl Surf Sci 2015;327:432–6. [21] Jihong D, Zhengxiang L, Gaojian L. Surface characterization of CVD tungsten coating on molybdenum substrate. Surf Coat Technol 2005;198(1–3):169–72. [22] Shimada S, Takahashi M, Kiyono H. Coatings and microstructures of monolithic TiB2 films and double layer and composite TiCN/TiB2 films from alkoxide solutions by thermal plasma CVD. Thin Solid Films 2008;516(19):6616–21. [23] Tkadletz M, Schalk N, Mitterer C. Cross-sectional characterization techniques as the basis for knowledge-based design of graded CVD TiN-TiB2 coatings. Int J Refract Met Hard Mater 2018;71:280–4. [24] Tang W-M, Zheng Z-X, Wu Y-C. Synthesis of TiB2 nanocrystalline powder by mechanical alloying. Trans Nonferrous Met Soc China 2006;16(3):613–7. [25] Raju GB, Basu B. Development of high temperature TiB2 -based ceramics. Key Eng Mater 2009;395:89–124.
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
[26] Schalk N, Keckes J, Czettl C. Investigation of the origin of compressive residual stress in CVD TiB2 hard coatings using synchrotron X-ray nanodiffraction. Surf Coat Technol 2014;258:121–6. [27] Silva FJG, Casais RCB, Martinho RP. Mechanical and tribological characterization of TiB2 thin films. J Nanosci Nanotechnol 2012;12(12):9187–94. [28] Silva FJG, Casais RB, Martinho RP. Role of abrasive material on micro-abrasion wear tests. Wear 2011;271(9–10):2632–9. [29] Alves SM, Albano W, De Oliveira AJ. Improvement of coating adhesion on cemented carbide tools by plasma etching. J Braz Soc Mech Sci Eng 2016;39(3):845–56. [30] Pershin V, Lufitha M, Chandra S, Mostaghimi J. Effect of substrate temperature on adhesion strength of plasma-sprayed nickel coatings. J Therm Spray Technol 2003;12(3):370–6. [31] Parthasarathy TA, Mendiratta MG, Dimiduk DM. Oxidation mechanisms in Mo-reinforced Mo5 SiB2 (T2)–Mo3 Si alloys. Acta Mater 2002;50(7):1857–68. [32] Mendiratta MG, Parthasarathy TA, Dimiduk DM. Oxidation behavior of ␣Mo–Mo3Si–Mo5 SiB2 (T2) three phase system. Intermetallics 2002;10(3):225–32. [33] Lange A, Braun R, Heilmaier M. Magnetron sputtered Mo(Six ,Al1−x )2 oxidation protection coatings for Mo–Si–B alloys. Oxid Met 2015;84(1–2):91–104. [34] Perepezko JH, Sakidja R. Oxidation-resistant coatings for ultra-high-temperature refractory Mo-based alloys. JOM 2010;62(10):13–9. [35] Huang X, Sun S, Lu S. Synthesis and characterization of oxidation-resistant TiB2 coating on molybdenum substrate by chemical vapor deposition. Mater Lett 2018;228:53–6.
9
[36] Amada S, Hirose T. Planar fractal characteristics of blasted surfaces and its relation with adhesion strength of coatings. Surf Coat Tech 2000;130(2–3):158–63. [37] Mellali M, Fauchais P, Grimaud A. Influence of substrate roughness and temperature on the adhesion/cohesion of alumina coatings. Surf Coat Tech 1996;81(2–3):275–86. [38] Caputo AJ. Chemical vapor deposition of erosion-resistant TiB2 coatings. J Electrochem Soc 1985;132(9). [39] Lv QL, Gao J, Li T. Structure, composition and adhesion strength of Hf coatings deposited on Mo substrate by ion beam sputtering. Mater Sci Forum 2016;852:1108–11. [40] Wang Y, Wang D, Yan J. Preparation and characterization of molybdenum disilicide coating on molybdenum substrate by air plasma spraying. Appl Surf Sci 2013;284:881–8. [41] Kullmer R, Lugmair C, Figueras A. Microstructure, mechanical and tribological properties of PACVD Ti(B,N) and TiB2 coatings. Surf Coat Technol 2003;174–175:1229–33. ¨ H. Preparation of TiB2 and TiBx Ny [42] Karner H, Laimer J, StoRi coatings by PACVD. Surf Coat Technol 1989;39–40:293–300. [43] Martinho R, Silva FJ, Alexandre R. TiB2 Nanostructured coating for GFRP injection moulds. J Nanosci Nanotechnol 2011;11(6):5374–82. [44] Murthy TSRC, Sonber JK, Vishwanadh B. Densification, characterization and oxidation studies of novel TiB2 +EuB6 compounds. J Alloys Compd 2016;670:85–95. [45] Roy B, Das J, Mitra R. Transient stage oxidation behavior of Mo76 Si14 B10 alloy at 1150 ◦ C. Corros Sci 2013;68:231–7.
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Available online at www.sciencedirect.com
www.jmrt.com.br
Original Article
Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum Xiaoxiao Huang, Shuchen Sun ∗ , Ganfeng Tu School of Metallurgy, Northeastern University, Shenyang 110819, China
a r t i c l e
i n f o
a b s t r a c t
Article history:
TiB2 anti-oxidation ceramic coatings were successfully synthesised on a Mo substrate by
Received 16 October 2019
using chemical vapor deposition (CVD). A scratch test was performed to measure the adhe-
Accepted 22 October 2019
sion strength of the TiB2 ceramic coatings to the Mo substrate. The results indicated good
Available online xxx
adhesion between the two materials; the adhesion strength was 125 mN. Furthermore, hightemperature oxidation tests indicated that the CVD TiB2 ceramic coating protected the Mo
Keywords:
substrate against oxidation up to 900 ◦ . Isothermal oxidation experiments revealed that the
Titanium diboride
Mo substrate covered with a 13-m-thick TiB2 ceramic coating was protected against oxida-
Coating
tion for a duration of 5 h. The oxidation kinetics followed a parabolic law for the first 2.5 h
Chemical vapor deposition
and a linear law in the next 2.5 h. © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the
Oxidation
CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1.
Introduction
Mo is a useful metal with many excellent physical and chemical properties, such as corrosion resistance, good electrical and thermal conductivities, and a high melting point (2622 ◦ C) [1,2]. Therefore, Mo and Mo alloys are widely used in the aerospace, metallurgy, nuclear, and energy industries, as well as in lighting technology, high-temperature furnace construction, sputtering targets, and high-performance electronics [3–5]. Nevertheless, the applicability of Mo is limited owing to its extremely poor resistance to oxidation at high temperatures. Mo is easily oxidised in air even at 427 ◦ C [5].
To improve the oxidation resistance of Mo, various methods have been employed, such as anti-oxidation coating and the addition of anti-oxidation elements (Si, B, etc.). However, the addition of B and Si is limited by the low solubility and poor workability [6]. The oxidation-resistance coating method is widely used for Mo and Mo-based alloys. In previous studies, two major types of coating materials were used for obtaining anti-oxidation protective layers on the surface of Mo: 1) MoSi2 [5] and MoSi2 -based composites, e.g., MoSi2 –SiO2 [2], (Mo, W)Si2 –Si3 N4 [7,8], MoSi2 + X SiC [9], and Mo-Si-B [10–17], and 2) metals (alloys), e.g., Ir [18], W [19–21], and Ni-20Cr [4]. Some ceramic materials also exhibit excellent properties, such as high-temperature oxidation resistance, a high melt-
∗
Corresponding author. E-mail: [email protected] (S. Sun). https://doi.org/10.1016/j.jmrt.2019.10.056 2238-7854/© 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
2
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Fig. 1 – Schematic design of the CVD system.
ing point, and high hardness. Titanium diboride (TiB2 ), which is a candidate material for ultrahigh-temperature ceramics, is attracting increasing attention. It has superior properties, such as a high melting point (approximately 3100 ◦ C [22]), high hardness (maximum 45 GPa [23]), good electrical and thermal conductivities, and chemical inertness [24]. Additionally, as a very important anti-oxidation ceramic material, TiB2 can resist oxidisation in air up to 1100 ◦ C [25]. Thus, TiB2 has been utilised for a protective coating for various applications. TiB2 as a protective coating has been used for cemented carbide cutting tools [26], moulds, and other tools [27,28]. However, there has been little research on the TiB2 ceramic as an anti-oxidation coating on the surface of Mo. In using a ceramic anti-oxidation coating, the major problem is poor adhesion between the ceramic coating and the metal substrate. However, the adhesion strength can be improved either via substrate surface treatment [29] or by increasing the substrate temperature [30]. Therefore, the TiB2 ceramic can be considered as a protective layer to improve the oxidation resistance of Mo metal. In discussions regarding the oxidation resistance of TiB2 coatings for Mo, Mo-B-Si coatings are also mentioned owing to their similar composition and oxidation performance. Mo-B-Si coatings have exhibited excellent high-temperature oxidation resistance for Mo and Mo alloys [11,12,14,15,31,32], particularly in the temperature range of 1200–1700 ◦ C [11,12,14]. However, the oxidation resistance of Mo-B-Si coatings is poor at temperatures below 1000 ◦ C [31–34]. At temperature of 650–750 ◦ C, pest oxidation of the coatings occurs [31,34]. Conversely, the TiB2 ceramic exhibits good and reliable oxidation resistance below 1000 ◦ C. Thus, it is of interest to examine the mechanical and oxidative properties of TiB2 ceramic coatings on Mo substrates. In our previous work [35], a high quality TiB2 ceramic coating was successfully synthesized on metal Mo by using a chemical vapor deposition (CVD) method, and the microstructure and morphology of the coatings on the Mo substrate have been examined. The objective of this study was to systematically investigate the effectiveness of a TiB2 ceramic coating for improving the oxidation resistance of Mo. Firstly, the adhesion strength and microhardness of the coating were measured.
Furthermore, the oxidation behaviour of the TiB2 ceramic coatings on the Mo substrate was investigated. Oxidation tests were performed on the coatings from room temperature to 900 ◦ C, which is the temperature required for Mo moulds in our future work.
2.
Experimental method
2.1.
Materials
The experimental materials are from our previous work. The TiB2 ceramic coatings were deposited on a sheet of metallic Mo using a gas mixture of H2 +TiCl4 + BCl3 under 1000 ◦ C by CVD. The overall chemical reaction is as follows. 2BCl3 (g) + 5H2 (g) + TiCl4 (g) = 10HCl (g) + TiB2 (s)
(1)
Within this work, the TiB2 ceramic coating was deposited on Mo substrate using a conventional vertical CVD reactor. The experimental equipment was self-made in this study. The schematic design of the experimental CVD system is illustrated in Fig. 1. The graphite deposition chamber was heated with a graphite resistance heating element. The temperature was measured by a platinum-rhodium thermocouple and with a temperature controller (CKW-3100, BCHY, Beijing, China) to ± 5 ◦ C. The TiCl4 was heated to boiling point of 135 ◦ C in a resistance furnace. TiCl4 , BCl3 and H2 were introduced into the bottom of the deposition chamber via stainless steel tubing, respectively. The TiCl4 gas path was heated by conventional heating tapes to o ensure that no condensation of the reagents occurred within the lines. It is well known that the adhesion between ceramic coatings and metal substrates is poor. Many factors affect the adhesion strength. Among the most important factors are the surface characteristics of the substrate. A roughened surface can improve the adhesiveness of ceramic coatings [36,37]. Hence, the surfaces of the Mo substrates were roughened by 600 grit silicon carbide sandpaper to improve the adhesiveness of the TiB2 ceramic coatings to the substrates. Then, the Mo substrates were ultrasonically cleaned with acetone for 30 min. Finally, the substrates were suspended in the centre
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
3
Fig. 2 – Appearance and cross-section images of experimental materials.
Table 1 – Deposition condition of TiB2 coating by CVD. substrate
molybdenum
Deposition temperature Deposition time Deposition pressure Pressure of H2 Flow rate of H2 Temperature of TiCl4 Flow rate of TiCl4 Flow rate of BCl3
1000 [◦ C] 2 [h] 40–50 [kPa] 0.15 [MPa] 635 [cm3 /min] 135 [◦ C] 130 [cm3 /min] 195 [cm3 /min]
of the deposition chamber by a Mo wire. A. J. Caputo et al. [38] reported that some properties of CVD TiB2 depend on the deposition temperature. They found that TiB2 coatings deposited at ≥900 ◦ C exhibited very low erosion rates and that the coating hardness increased with the deposition temperature. Hence, we used a high deposition temperature (1000 ◦ C) to obtain good coating properties. More detailed deposition conditions are presented in Table 1. The appearance and crosssection images of the materials we used in this work are shown in Fig. 2 [35].
2.2. Mechanical-property test of TiB2 ceramic coatings on Mo substrate Adhesion measurements were performed according to a standard test method for evaluating the adhesion strength between a vapor deposition film and a substrate (JB/T 8554, People’s Republic of China Machinery Industry Standard (PRCMIS)). The scratch method was employed for testing the adhesion of the TiB2 ceramic coatings to the Mo substrates. The adhesion strength was evaluated using a nanoscratch tester (TriboIndenter, Hysitron, Minneapolis, USA) with a diamond cube-corner stylus under a continuously increasing load. The load ranged from 0 to 250 mN, with a loading rate of 5 mN/s. Furthermore, a microhardness tester (FM-700, Future-tech, Kawasaki, Japan) was used to evaluate the microhardness of the TiB2 ceramic coating.
2.3.
Oxidation test
To study the oxidation process of the specimens, oxidation experiments were performed in a tube furnace with a temperature accuracy of ±5 ◦ C. These experiments were conducted in dry air at 300, 450, 600, 750, and 900 ◦ C. Additionally, isothermal-oxidation tests were performed at 900 ◦ C for 1, 2, 3, 4, and 5 h. Thermal analysis, thermogravimetry (TG), and differential scanning calorimetry (DSC) were performed to investigate the oxidation mechanisms. The tests were performed using a Netzsch STA449F3 thermoanalytical system (TG measurement error was 10−4 mg, temperature measurement error was <1 ◦ C) at a constant heating rate of 10 ◦ C/min in flowing air at a flow rate of 150 mL/min in the temperature range of 20–900 ◦ C. The duration of the isothermal was 5 h at 900 ◦ C. For comparison, thermogravimetric tests of uncoated Mo were also performed under the same oxidation conditions.
2.4. Characterisations of coating before and after oxidation The phases of the coatings formed on the Mo substrate were examined using X-ray diffraction (XRD, X’ Pert Pro, PANalytical B.V., Almelo, Netherlands) at 40 mA and 40 kV with Cu K␣ radiation ( = 1.541 Å, step size = 0.0334◦ , integration time of 19.69 s/step, scan rate of 0.216◦ s−1 from 5◦ to 90◦ ). Scanning electron microscopy (SEM, Ultra Plus, Zeiss, Heidenheim, Germany) and energy-dispersive X-ray spectroscopy (EDS, X-Max 50, Oxford Instruments, Oxford, UK) were performed to investigate the microstructure, morphology, and composition of the coatings. XRD was also used to determine the phase composition of the oxidation products on the surface of the specimens after the oxidation experiments. The oxidation products were imaged and analysed using SEM, and their compositions were analysed using EDS.
3.
Results and discussion
3.1.
Mechanical properties of TiB2 coatings
According to a standard method for testing the adhesion strength between a vapor deposition film and a substrate (JB/T
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
4
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
Fig. 3 – Coefficient of friction plots for the scratch testing of the TiB2 coating on the Mo substrate.
8554 from PRCMIS), scratch adhesion tests were performed by applying continuous loading from 0 to 250 mN with a 5-mN/s increase rate. The adhesive strength was assessed according to the critical load, which was identified by an abrupt change in the coefficient of friction. To reduce the influence of the surface roughness in the evaluation of the coefficient of friction, the surface roughness must be <0.32 m according to the test standard JB/T 8554. Hence, the coating was polished by 2000 grit silicon carbide sandpaper before the scratch tests. The surface roughness of the coating was measured using a three-dimensional laser scanning microscope (LEXT OLS4100, Olympus, Tokyo, Japan). The results of the roughness test indicated that after polishing, the average surface roughness was 0.26, which is 0.32 lower than the value specified by the test standard JB/T 8554. Fig. 3 shows the coefficient of friction plot for the scratch testing of the TiB2 ceramic coating on a pure Mo substrate. Two turning points are observed in the coefficient of friction plot: the lower critical load and upper critical load, corresponding to the critical loads for failure of the coating. Before the lower critical load was reached, the coefficient of friction increased slowly and fluctuated slightly with the increasing load. At this stage, the coating exhibited good adhesion to the substrate and was polished by the indenter. Then, we observed an abrupt increase in the coefficient of friction over a short period of time, indicating that the coating began to crack and peel off. Therefore, the lower critical load was defined as the adhesion strength of the coating to the substrate. As the load continued to increase, an upper critical load was observed. Subsequently, the coating was removed from the substrate. Hence, the friction coefficient increased slowly with the increasing load. According to the results, as shown in Fig. 3, the adhesion strength of the TiB2 ceramic coatings to the Mo substrate was approximately 125 mN. Q. L. Lv and J. Gao [39] reported that good adhesion of a Hf metal coating to a Mo substrate was achieved using ion-beam sputtering; the adhesion strength was approximately 120 mN. The adhesion of the as-prepared CVD TiB2 coatings to the Mo substrate in the present study was as good as that of the previously reported
Fig. 4 – TG weight-gain data under continuous oxidation for the Mo substrate covered with the TiB2 coating and the pure Mo specimen in air up to 900 ◦ C with a heating rate of 10 ◦ C/min.
Hf metal coating. Additionally, the TiB2 coating exhibited high microhardness relatively to MoSi2 coating on Mo substrate. The mean average microhardness value of the coating we obtained was 28 GPa and the maximum microhardness value was 30 GPa. But the MoSi2 coating microhardness only have a mean value of 13 GPa [40]. Generally, the hardness values for TiB2 films can vary between 25 and 36 GPa [27]. In our work, hardness of as-preparation TiB2 coatings on Mo substrate is in accordance with the consistently range of TiB2 hardness value [23,41–43].
3.2.
TG and DSC of TiB2 coated on Mo
Fig. 4 presents the weight-gain data obtained during continuous oxidation of Mo covered with the TiB2 coating. For comparison, the data for a pure Mo specimen are also shown. The possible oxidation reactions are as follows. 2TiB2 (s) +5O2 (g) = 2TiO2 (s) +2B2 O3 (l)
(2)
B2 O3 (l) = B2 O3 (g)
(3)
Mo (s)+3O2 (g) = 2MoO3 (s)
(4)
3MoO3 (s) = Mo3 O9 (g)
(5)
The weight gain of the pure Mo and the Mo covered with the TiB2 coating started at 550 and 720 ◦ C, respectively, in agreement with previously reported data for the oxidation of TiB2 [44]. The rate of weight gain for the pure Mo spec˜ ◦ C; subsequently, the weight-gain imen increased up to 790 rate remained high up to 850 ◦ C, and then rapid weight loss occurred. These results can be explained by the fact that Mo is very easily oxidised at low temperatures in air, and its oxide (MoO3 ) is easily volatilised according to reaction (5) [45]. However, the Mo covered with the TiB2 coating exhibited entirely different weight-gain trends. Above 720 ◦ C, the sample exhib-
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
5
Fig. 5 – TG and DSC curves for the oxidation of TiB2 coated on Mo in air, with the heating segment at a heating rate of 10 ◦ C/min up to 900 ◦ C and isothermal oxidation at 900 ◦ C up to 5 h. Fig. 6 – XRD patterns of Mo coated with TiB2 oxidised in air at different temperatures for 5 h. ited very slow weight gain up to the final temperature. This indicates that the TiB2 functioned as a protective layer on the surface of the pure Mo metal and improved the oxidation resistance of the Mo. The thermal-analysis results for the TiB2 coated on Mo in air are presented in Fig. 5. The heating rate was 10 ◦ C/min in the heating segment up to 900 ◦ C, and the duration of the isothermal oxidation was 5 h at 900 ◦ C. According to the TG curve in Fig. 5, a very small weight gain (0.03%) occurred in the initial oxidation period up to 120 ◦ C. Although this weight gain was negligible, it indicates the start of the oxidation, as supported by the DSC curve. The oxidation reaction was followed by reaction (2). Subsequently, the weight did not change until
the temperature reached 400 ◦ C, owing to the formation of low-viscosity B2 O3 glass that functioned as an O barrier. When the oxidation temperature increased to 450 ◦ C, the vaporisation of B2 O3 (liquid) (see reaction (3)) occurred more vigorously, resulting in weight loss. When the temperature increased to 720 ◦ C, slight weight gain occurred, because a large amount of B2 O3 fluid covered the surface of the TiB2 coating, preventing its further oxidation. Above this temperature, rapid weight gain occurred up to 900 ◦ C because of the rapid vaporisation consumption of the B2 O3 glass, causing part of the TiB2 to be exposed in air, which led to further oxidation. Thereafter,
Fig. 7 – SEM images and EDS mapping of Mo coated with TiB2 oxidised in air at 300–900 ◦ C for 5 h. Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
6
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
isothermal oxidation occurred for 5 h. The weight-change rate of the specimen was dependent on the oxidation time. For the first 2.5 h, the weight-gain rate decreased with the oxidation time, and the oxidation kinetics of the specimen essentially followed a parabolic law. Then, the weight-gain rate increased, and the oxidation kinetics followed a linear law.
3.3.
Phase and microstructure of oxide scales
To further examine the oxidation process of the TiB2 ceramic coatings formed on Mo, samples prepared under identical conditions were oxidised at 300, 450, 600, 750, and 900 ◦ C for 5 h. XRD was employed to identify the phases of the oxide scales, and their morphology was examined using SEM. The results are shown in Figs. 6 and 7, respectively. When the samples were exposed to air at temperatures of 300 and 450 ◦ C for 5 h, only the peaks of TiB2 were detected, as shown in Fig. 6. However, as shown in Fig. 5, the TG and DSC curves indicated that oxidation (reaction (2)) occurred at these two temperatures. It is considered that the phases of the reaction products were undetected because of their small amount. Indeed, the weight gain was only 0.03%, according to the foregoing TG data. At 600 ◦ C, the TiO2 peaks were observed, but the B2 O3 peaks were absent. Additionally, the B2 O3 peaks were absent at 750 and 900 ◦ C, as shown in Fig. 6. This is because the B2 O3 evaporated with the increasing temperature, as confirmed by the EDS mapping in Fig. 7. With the increasing temperature, the B content gradually decreased, and the O content increased. When the temperature increased to 900 ◦ C, the TiB2 peaks completely disappeared, and only TiO2 peaks remained. After the total consumption of TiB2 (T = 900 ◦ C and 5-h oxidation time), important changes were observed, as shown in Fig. 7(e): high-crystallinity TiO2 (crystallites) was formed on the surface. The XRD patterns (Fig. 6) and EDS mapping (Fig. 7(e)) indicated that Mo was present on the surface of the coating, suggesting that the TiB2 coating (13 m thick) lost its ability to protect the Mo substrate after 5 h of oxidation at 900 ◦ C in air. Furthermore, isothermal oxidation studies for the Mo substrate coated with the TiB2 ceramic coating were performed in air at 900 ◦ C for 1, 2, 3, 4, and 5 h. Fig. 8 presents the XRD patterns of Mo coated with TiB2 oxidised in air at 900 ◦ C for different durations. As shown, TiO2 and B2 O3 peaks were detected, as well as TiB2 peaks, after 1 h of oxidation. Thus,
Fig. 8 – XRD patterns of Mo coated with TiB2 oxidised in air at 900 ◦ C for different durations.
both TiO2 and B2 O3 were formed on the surface after 1 h of oxidation. The intensity of the B2 O3 peak was reduced after 2 h, and the B2 O3 peak disappeared after 3 h. As described previously, the volatilisation of B2 O3 increased with time, causing the disappearance of B2 O3 . After 2 h of oxidation, only the TiB2 peak remained. After 5 h of oxidation, the TiB2 peak disappeared, and a Mo peak was observed. Hence, when the sample was exposed to air for 5 h at 900 ◦ C, the only phase on the surface was TiO2 . Fig. 9 shows SEM images of the specimens after oxidation at 900 ◦ C for various durations. As indicated by the SEM images of the as-prepared TiB2 coating surface in our previous work [35], the coating typically exhibited a rod-like particle microstructure with a small gap. As shown in Fig. 9(a), a liquid phase filled the small gap. According to the results of the foregoing analysis, the liquid phase was B2 O3 . This is consistent with the EDS mapping results in Fig. 9(a). Comparing Fig. 9(a)–(c) reveals that the B2 O3 content decreased with the exposure time. Fig. 9(d) and (e) indicate that crystalline TiO2 was formed on the surface of the Mo substrate, supporting the foregoing XRD results and TG data.
Fig. 9 – SEM images and EDS mapping of Mo coated with TiB2 oxidised in air at 900 ◦ C for different durations. Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
7
Fig. 10 – Cross-sectional SEM images and element line scanning EDS analyses of the TiB2 coating after oxidation at 900 ◦ C for different durations in air.
Fig. 10 presents cross-sectional SEM images and element line scanning EDS analysis results for the oxidation of the TiB2 coating at 900 ◦ C for different durations. Oxide scales were formed after oxidation for 1, 2, 3, 4, and 5 h at 900 ◦ C. As shown in Fig. 10(a)–(e), the TiB2 coatings were generally oxidised with the increasing oxidation time, and there were obvious boundaries between the oxide layer and the unoxidised layer (TiB2 ). For the five aforementioned oxidation durations, the thicknesses of the oxide scales were approximately 3, 6, 9, 10, and 13 m, respectively. These results indicate that the oxidation rate was higher in the initial oxidation period than in the later stage, in accordance with the TG results shown in Fig. 5. The results of element line scanning EDS analysis indicated that the oxide scales mainly comprised O and Ti. The B content was lower in the oxide scales than in the TiB2 coatings. This suggests that the B2 O3 evaporated during the oxidation at 900 ◦ C. The jagged concentration line profiles of B, Ti, and O for the oxide scales resulted from pores in the oxide scales. The pores were due to the evaporation of B2 O3 formed during the oxidation of the dense TiB2 coating. Conversely, the smooth concentration line profiles indicate that the TiB2 coatings were dense.
3.4.
the oxidation of the coating depended on the formation of B2 O3 . The oxidation process of the TiB2 coating started at a low temperature (100 ◦ C), as indicated by the TG data (Fig. 5). At this temperature, the surface of the coating with the rod-like TiB2 phase was surrounded by low-viscosity B2 O3 glass resulting from the slight oxidation of the TiB2 coating (see Figs. 8(b) and 9 (a)). The B2 O3 , as a protective layer, prevented the further access of O ions. Thus, the formation of the B2 O3 layer prevented the further oxidation of the TiB2 coating. However, the volatilisation of B2 O3 became noticeable above 720 ◦ C, resulting in rapid oxidation of the TiB2 coating. A high oxidation rate was detected up to 900 ◦ C through TG (Fig. 5). Isothermal oxidation studies (at 900 ◦ C) revealed that the oxidation rate decreased with the oxidation time, owing to the formation of a new B2 O3 layer. However, the oxidation kinetics of the sample exhibited different trends at isothermal temperature 900 ◦ C duration of 5 h. For the first 2.5 h, the oxidation kinetics of the sample followed a parabolic law. For the next 2.5 h, the oxidation kinetics essentially followed a linear law. After oxidation for 5 h at 900 ◦ C, XRD data (Fig. 8) and SEM (Fig. 9) images indicated the complete consumption of TiB2 and the volatilisation of B2 O3 , and only a porous and non-protective TiO2 crystal was observed on the Mo substrate.
Oxidation mechanism
The foregoing results indicate that the CVD TiB2 ceramic coating formed on the surface of the Mo metal protected the Mo substrate against oxidation up to 900 ◦ C in air. B2 O3 was formed in the process of oxidation, and the protection against
4.
Conclusions
To improve the oxidation resistance of Mo metal, a TiB2 coating was synthesised on the surface of pure metallic Mo using
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
8
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
CVD. According to the experimental results, the following conclusions are drawn. (1) The TiB2 ceramic coating, as an anti-oxidation protective layer on the Mo metal, was successfully obtained on the pure Mo substrate via CVD. The as-deposited TiB2 ceramic coating exhibited good adhesion to the Mo substrate and a high microhardness. These good mechanical properties of the coating were due to the roughened surface of the substrate and the high deposition temperature. (2) The weight gain of only 0.03% indicates that the TiB2 coating exhibited high oxidation resistance during oxidation below 720 ◦ C. The TiB2 coatings also exhibited good oxidation resistance in the range of 720–900 ◦ C; the weight gain increased to 0.4% but was far smaller than the 2.3% weight gain of pure Mo. In this temperature range, the oxidation resistance depended on the coating-layer thickness and oxidation time. At 900 ◦ C, the 13-m-thick TiB2 layer protected the Mo substrate against oxidation for up to 5 h. (3) The oxidation mechanism of the TiB2 coating was examined from room temperature to 900 ◦ C. The oxidation process of the TiB2 coating started at a low temperature (100 ◦ C). The oxidation resistance of the TiB2 coating depended on the state of the reaction product B2 O3 . Below 720 ◦ C, B2 O3 functioned as a low-viscosity diffusion barrier that covered the surface of the coating, preventing further ingress of O ions and thereby preventing further oxidation of the TiB2 . At 720–900 ◦ C, the B2 O3 glass was consumed by vaporisation, which slightly reduced the protective ability. (4) The oxidation kinetics of the sample exhibited different trends during oxidation for 5 h at 900 ◦ C. For the first 2.5 h, the oxidation kinetics essentially followed a parabolic law. For the next 2.5 h, they followed a linear law.
Conflict of interest statement The authors declared that they have no conflicts of interest to this work.
Acknowledgments The authors thank Prof. Puqi Jia (Lanzhou University) for technically reviewing the manuscript. We gratefully acknowledge Prof. Bart Blanpain (KU Leuven) for his careful reading of and linguistic assistance with the manuscript.
references
[1] Zhang H, Gu S. Preparation and oxidation behavior of MoSi2 –CrSi2 –Si3 N4 composite coating on Mo substrate. Int J Refract Met Hard Mater 2013;41:128–32. [2] Nyutu EK, Kmetz MA, Suib SL. Formation of MoSi2 –SiO2 coatings on molybdenum substrates by CVD/MOCVD. Surf Coat Technol 2006;200(12–13):3980–6. [3] Babinsky K, Weidow J, Knabl W. Atom probe study of grain boundary segregation in technically pure molybdenum. Mater Charact 2014;87:95–103.
[4] Huang C, Zhang Y, Vilar R. Microstructure and anti-oxidation behavior of laser clad Ni–20Cr coating on molybdenum surface. Surf Coat Technol 2010;205(3):835–40. [5] Suzuki RO, Ishikawa M, Ono K. MoSi2 coating on molybdenum using molten salt. J Alloys Compd 2000;306(1-2):285–91. [6] Kruzic JJ, Schneibel JH, Ritchie RO. Fracture and fatigue resistance of Mo–Si–B alloys for ultrahigh-temperature structural applications. Scripta Mater 2004;50(4):459–64. [7] Zhang H, Lv J, Wu Y. Oxidation behavior of (Mo,W)Si2 –Si3 N4 composite coating on molybdenum substrate at 1600◦ C. Ceram Int 2015;41(10):14890–5. [8] Yoon J-K, Kim G-H, Han J-H. Low-temperature cyclic oxidation behavior of MoSi2 /Si3 N4 nanocomposite coating formed on Mo substrate at 773 K. Surf Coat Technol 2005;200(7):2537–46. [9] Govindarajan S, Moore JJ, Ohno TR. Interfacial characterization of MoSi2 + x SiC composite thin films on molybdenum substrates. Mrs Proc 2011;458. [10] Taylor M, Perepezko JH. Hot corrosion of Mo–Si–B coatings. Oxid Met 2017;87(5-6):705–15. [11] Kiryukhantsev-Korneev PV, Iatsyuk IV, Shvindina NV. Comparative investigation of structure, mechanical properties, and oxidation resistance of Mo-Si-B and Mo-Al-Si-B coatings. Corros Sci 2017;123:319–27. [12] Su L, Lu-Steffes O, Zhang H. An ultra-high temperature Mo–Si–B based coating for oxidation protection of NbSS /Nb5 Si3 composites. Appl Surf Sci 2015;337:38–44. [13] Lu-Steffes OJ, Su L, Jackson DM. Mo-Si-B coating for improved oxidation resistance of niobium. Adv Eng Mater 2015;17(7):1068–75. [14] Wu J, Wang W, Zhou C. Microstructure and oxidation resistance of Mo–Si–B coating on Nb based in situ composites. Corros Sci 2014;87:421–6. [15] Zhang Y-L, Li H-J, Hu Z-X. Microstructure and oxidation resistance of Si–Mo–B coating for C/SiC coated carbon/carbon composites. Corros Sci 2013;72:150–5. [16] Ritt P, Sakidja R, Perepezko JH. Mo–Si–B based coating for oxidation protection of SiC–C composites. Surf Coat Technol 2012;206(19–20):4166–72. [17] Perepezko JH, Rioult F, Sakidja R. Oxidation performance of high temperature Mo-Si-B alloys and coatings. Mater Sci Forum 2008;595–598:1065–74. [18] Wang L, Chen Z, Zhang P. Ir coating prepared on Mo substrate by double glow plasma. J Coat Technol Res 2008;6(4):517–22. [19] Lian Y, Liu X, Wang J. Influence of surface morphology and microstructure on performance of CVD tungsten coating under fusion transient thermal loads. Appl Surf Sci 2016;390:167–74. [20] Jiang F, Zhang Y, Sun N. Tungsten coating prepared on molybdenum substrate by electrodeposition from molten salt in air atmosphere. Appl Surf Sci 2015;327:432–6. [21] Jihong D, Zhengxiang L, Gaojian L. Surface characterization of CVD tungsten coating on molybdenum substrate. Surf Coat Technol 2005;198(1–3):169–72. [22] Shimada S, Takahashi M, Kiyono H. Coatings and microstructures of monolithic TiB2 films and double layer and composite TiCN/TiB2 films from alkoxide solutions by thermal plasma CVD. Thin Solid Films 2008;516(19):6616–21. [23] Tkadletz M, Schalk N, Mitterer C. Cross-sectional characterization techniques as the basis for knowledge-based design of graded CVD TiN-TiB2 coatings. Int J Refract Met Hard Mater 2018;71:280–4. [24] Tang W-M, Zheng Z-X, Wu Y-C. Synthesis of TiB2 nanocrystalline powder by mechanical alloying. Trans Nonferrous Met Soc China 2006;16(3):613–7. [25] Raju GB, Basu B. Development of high temperature TiB2 -based ceramics. Key Eng Mater 2009;395:89–124.
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056
JMRTEC-1033; No. of Pages 9
ARTICLE IN PRESS j m a t e r r e s t e c h n o l . 2 0 1 9;x x x(x x):xxx–xxx
[26] Schalk N, Keckes J, Czettl C. Investigation of the origin of compressive residual stress in CVD TiB2 hard coatings using synchrotron X-ray nanodiffraction. Surf Coat Technol 2014;258:121–6. [27] Silva FJG, Casais RCB, Martinho RP. Mechanical and tribological characterization of TiB2 thin films. J Nanosci Nanotechnol 2012;12(12):9187–94. [28] Silva FJG, Casais RB, Martinho RP. Role of abrasive material on micro-abrasion wear tests. Wear 2011;271(9–10):2632–9. [29] Alves SM, Albano W, De Oliveira AJ. Improvement of coating adhesion on cemented carbide tools by plasma etching. J Braz Soc Mech Sci Eng 2016;39(3):845–56. [30] Pershin V, Lufitha M, Chandra S, Mostaghimi J. Effect of substrate temperature on adhesion strength of plasma-sprayed nickel coatings. J Therm Spray Technol 2003;12(3):370–6. [31] Parthasarathy TA, Mendiratta MG, Dimiduk DM. Oxidation mechanisms in Mo-reinforced Mo5 SiB2 (T2)–Mo3 Si alloys. Acta Mater 2002;50(7):1857–68. [32] Mendiratta MG, Parthasarathy TA, Dimiduk DM. Oxidation behavior of ␣Mo–Mo3Si–Mo5 SiB2 (T2) three phase system. Intermetallics 2002;10(3):225–32. [33] Lange A, Braun R, Heilmaier M. Magnetron sputtered Mo(Six ,Al1−x )2 oxidation protection coatings for Mo–Si–B alloys. Oxid Met 2015;84(1–2):91–104. [34] Perepezko JH, Sakidja R. Oxidation-resistant coatings for ultra-high-temperature refractory Mo-based alloys. JOM 2010;62(10):13–9. [35] Huang X, Sun S, Lu S. Synthesis and characterization of oxidation-resistant TiB2 coating on molybdenum substrate by chemical vapor deposition. Mater Lett 2018;228:53–6.
9
[36] Amada S, Hirose T. Planar fractal characteristics of blasted surfaces and its relation with adhesion strength of coatings. Surf Coat Tech 2000;130(2–3):158–63. [37] Mellali M, Fauchais P, Grimaud A. Influence of substrate roughness and temperature on the adhesion/cohesion of alumina coatings. Surf Coat Tech 1996;81(2–3):275–86. [38] Caputo AJ. Chemical vapor deposition of erosion-resistant TiB2 coatings. J Electrochem Soc 1985;132(9). [39] Lv QL, Gao J, Li T. Structure, composition and adhesion strength of Hf coatings deposited on Mo substrate by ion beam sputtering. Mater Sci Forum 2016;852:1108–11. [40] Wang Y, Wang D, Yan J. Preparation and characterization of molybdenum disilicide coating on molybdenum substrate by air plasma spraying. Appl Surf Sci 2013;284:881–8. [41] Kullmer R, Lugmair C, Figueras A. Microstructure, mechanical and tribological properties of PACVD Ti(B,N) and TiB2 coatings. Surf Coat Technol 2003;174–175:1229–33. ¨ H. Preparation of TiB2 and TiBx Ny [42] Karner H, Laimer J, StoRi coatings by PACVD. Surf Coat Technol 1989;39–40:293–300. [43] Martinho R, Silva FJ, Alexandre R. TiB2 Nanostructured coating for GFRP injection moulds. J Nanosci Nanotechnol 2011;11(6):5374–82. [44] Murthy TSRC, Sonber JK, Vishwanadh B. Densification, characterization and oxidation studies of novel TiB2 +EuB6 compounds. J Alloys Compd 2016;670:85–95. [45] Roy B, Das J, Mitra R. Transient stage oxidation behavior of Mo76 Si14 B10 alloy at 1150 ◦ C. Corros Sci 2013;68:231–7.
Please cite this article in press as: Huang X, et al. Investigation of mechanical properties and oxidation resistance of CVD TiB2 ceramic coating on molybdenum. J Mater Res Technol. 2019. https://doi.org/10.1016/j.jmrt.2019.10.056