Surface and Coatings Technology 151 – 152 (2002) 483–489 Some properties of atmospheric air and inert gas high-pressure plasma sprayed ZrB2 coatings M. Tuluia, F. Ruffinia, F. Arezzoa, S. Lasiszb, Z. Znamirowskib, L. Pawlowskic,* a b CSM, via di Castel Romano 100, 00128 Rome, Italy IMT, Wroclaw University of Technology, ul. Janiszewskigo 11y17, 50372 Wroclaw, Poland c ¨ BP 108, 59652 Villeneuve d’Ascq, France ENSCL, 8, Avenue Mendeleıev, Abstract This work was aimed at developing a plasma spraying process for the deposition of ZrB2 coatings. The ZrB2 powder was prepared by a spray drying technique and characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), electron microprobe analysis (EMPA) and laser granulometry. XRD revealed the presence of ZrO2 in the coatings deposited by atmospheric plasma spray (APS), while ZrB2 was only present in the coatings deposited by inert plasma spray (IPS). The microhardness increased from 900 to over 1500 HV by increasing the spraying pressure over the atmospheric one. The coatings were submitted to pin-on-disc testing and their behavior was compared to the WC – Co and TiO2 qAl2 O3 coatings, sprayed by high velocity oxy fuel (HVOF) and APS, respectively. The characterization results showed that the ZrB2 coatings, deposited by IPS, were good electrical conductors and they compared favorably with the reference coatings, in terms of tribological performance. 䊚 2002 Elsevier Science B.V. All rights reserved. Keywords: Plasma spraying; ZrB2; Electrical conductivity; Tribology; Spray dryer; Coatings 1. Introduction ZrB2 is a very interesting material, especially for its high hardness, associated with high electrical and thermal conductivities w1x. The very high melting point (Tms3313 K) allows its utilization for very high () 1500 K) temperature applications, providing it is combined with other materials, such as SiC, for preventing oxidation. Furthermore, ZrB2 alone has been used successfully in many tribological applications (e.g. wear protection). ZrB2 components are currently produced by hot isostatic pressing (HIP) technology w2x. Protective ZrB2 coatings are obtained by chemical vapor deposition (CVD) w3x. The present work aims at developing a thermal spraying process, alternative to the CVD. Since ZrB2 powder, having shape and dimensions suitable for thermal spraying technology, is not commercially available, it was necessary to start the investigation with the preparation of the powder. Then, a simplified model of * Corresponding author. Tel.: q33-3-20336165; fax: q33-320336165. E-mail address: lech.pawlowski@ensc-lille.fr (L. Pawlowski). the plasma spray process was applied to optimize the spray parameters. Finally, the tribological and electrical properties of sprayed coatings were tested. 2. Experimental methods 2.1. Powder preparation The powder was prepared by agglomerating fine (2mm average size) particles of ZrB2. The agglomeration was carried out from slurry by means of a spray dryer. The slurry composition, per 1 l of water, was composed of polyvinyl alcohol binder (40 g), dispersing agent (30 g), and ZrB2 (1000 g). The spray dryer inlet and outlet temperatures were 523 and 413 K, respectively. 2.2. Thermal spraying ZrB2 powder was sprayed using a commercial plasma spray equipment called controlled atmosphere plasma spray (CAPS) by Sulzer-Metco (Wohlen, Switzerland). This apparatus w4x includes a 80-kW plasma torch installed in a vessel that can work in air, argon or nitrogen atmospheres, with a pressure ranging from 10 0257-8972/02/$ - see front matter 䊚 2002 Elsevier Science B.V. All rights reserved. PII: S 0 2 5 7 - 8 9 7 2 Ž 0 1 . 0 1 5 7 2 - 9 484 M. Tului et al. / Surface and Coatings Technology 151 – 152 (2002) 483–489 Table 1 Plasma spray parameters relative to coatings of the first and second group Sample Atmosphere Gas flow Power Group No. Composition Pressure (hPa) Ar (slpm)a H2 (slpm) Input (kW) Outputb (kW) 1st I-1 I-2 I-3 Air Ar 1000 600 1500 30 40 16 12 13 57 54 54 28 20 23 2nd II-1 II-2 Ar 400 2000 40 12 54 27 43 14 a b Standard liters per second. Effective power is calculated subtracting from input power the heat dispersed in cooling water. to 4000 hPa. Before starting the spraying runs, a simplified theoretical approach was carried out to select the process parameters. Such an approach, detailed in Pawlowski w5x, can be briefly described as a model based on two fundamental plasma-spraying parameters: ● AHF (ability of heating factor), depending on the plasma properties; and ● DMF (difficulty of melting factor), depending on the powder properties. These parameters are defined as follows: B Tg D 300 | LC AHFs DMFs E2 lgdTF h gu G wJ2 y(kg=m)x H2md2prp 2 wJ y(kg=m)x 16 (1) (2) Where L is the length of the plasma stream, Tg, lg, hg, and u are average values of temperature, thermal conductivity, viscosity and velocity, respectively, of the plasma; Hm is the enthalpy of one melted particle, dp is the mean diameter of the powder particles and rp is their density. It is assumed that melting of the powder particles in the plasma takes place when AHFGDMF (3) The terms on the right of Eq. (1) were obtained by following the procedures given in Pawlowski w5x. The AHF was derived from these terms for various combinations of the following process parameters: ● power of the torch; ● pressure in the deposition vessel; and ● flow of the gases generating the plasma. The DMF was calculated from literature data, considering different diameters of ZrB2 particles. The spraying experiments can be divided into five groups: 1. The first group includes the spraying, under air and inert gas (Ar), of ZrB2 powder into water. Purpose of these experiments was to investigate the effects of spraying atmosphere on the characteristics, including composition, of the deposited particles upon cooling without deformation. 2. The second group includes spraying runs, carried out under inert gas at low and high pressure, onto stainless steel substrates. A very high relative motion of the torch with respect to the substrate was used, to enhance the impact of single particles and make easier the investigation of their morphology. Table 1 shows a detailed description of process parameters relative to coatings of the first and second group. 3. Then, several stainless steel substrates were coated, using different IPS parameters for each sample. This change of parameters, shown in Table 2, allowed to investigate the effects of different temperature and density of the plasma on coating microstructure and hardness. 4. The fourth group includes disc shaped stainless steel specimens coated with IPS ZrB2 for tribological investigation. It also includes two sets of the same specimens, coated with Al2O3q13%TiO2 by plasma spray and WCq12%Co by HVOF, respectively. Table 2 Plasma spray parameters relative to coatings of the third group No. III-1 III-2 III-3 III-4 III-5 III-6 III-7 III-8 III-9 Spray atmosphere Gas flow Power Composition Pressure (hPa) Ar (slpm) H2 (slpm) Input (kW) Output (kW) Ar 900 900 1250 1500 1500 1600 2000 2020 2070 40 46 38 40 46 32 46 47 35 12 6 10 12 6 10 6 10 10 56 46 43 58 49 41 52 48 55 29 22 18 25 20 18 19 20 20 M. Tului et al. / Surface and Coatings Technology 151 – 152 (2002) 483–489 485 Fig. 1. XRD spectrum of ZrB2 powder, after spraying it into water by IPS. These additional coatings are widely used in tribological applications and, thus, they were used as references. 5. The fifth group includes IPS ZrB2 coatings on Al2O3 substrates having dimensions of 50=30=2 mm. Purpose of these samples was the investigation of d.c. electrical conductivity. 2.3. Powders and coatings characterization ZrB2 powder samples were characterized before and after spraying them into water. They were analyzed by XRD Model D500, by Siemens (Germany) and by SEM, model 505, by Philips (Netherlands). Powder size distribution was measured by means of a laser granulo- Fig. 2. XRD spectrum of ZrB2 powder, after spraying it into water by APS. 486 M. Tului et al. / Surface and Coatings Technology 151 – 152 (2002) 483–489 Fig. 3. SEM micrograph showing morphology of ZrB2 powder after spraying it into water by APS. Fig. 4. SEM micrograph showing morphology of ZrB2 powder after spraying it into water by IPS. meter ‘Master sizer’, by Malvern (UK). The single splats (samples II-1 and II-2) were observed by SEM. Vickers microhardness measurements, at 50- and 200-g weights, were carried out on the samples belonging to the third group. The tribological characterization of samples belonging to the fourth group was carried out by a pin-on-disc apparatus, by RTM (Italy). Coated discs were polished before testing. During the tests, the friction coefficient was continuously monitored. The d.c. electrical resistance of the coatings belonging to the fifth group was measured by a standard four-point probe method. In total, 25 measurements on each sample were carried out. The actual thickness of the tested samples was determined by SEM. 3.3. Coating characterization Fig. 7 shows an example of the correlation between plasma characteristics (temperature and density) and microhardness of the third group coatings. Fig. 8 shows microhardness, in the range of 1000–1600 HV, plotted vs. AHF. The data were fitted by a linear regression. Fig. 9 shows the friction coefficients measured on ZrB2, obtained by IPS at different chamber pressures, and reference samples. Table 3 reports the d.c. electrical resistance values measured on the surface of the samples belonging to the fifth group. The average thickness of 3. Results 3.1. Powder characterization Figs. 1 and 2 show the XRD spectra of the ZrB2 powder, after spraying into water under inert gas and air, respectively. Figs. 3 and 4 show the morphologies of these powder samples by SEM. 3.2. Optimization of spray parameters Fig. 5 gives an example of the AHF parameter behavior as a function of the pressure in deposition chamber. In this case, AHF was determined by varying the flow of the plasma gases, while maintaining constant the input power. Fig. 6 shows, for one combination of gas flows, the maximum diameter in microns at which a particle could melt in the plasma stream, depending on power and pressure. Fig. 5. AHF parameters vs. chamber pressure at different compositions and flow rates of the plasma gases. M. Tului et al. / Surface and Coatings Technology 151 – 152 (2002) 483–489 487 Fig. 6. Power and pressure parameters determining, during plasma spraying, maximum sizes of meltable ZrB2 particles. the samples was 43 mm and the mean resistivity, calculated from this value and from the results reported in Table 3, was approximately rs3.2 mV-m. 4. Discussion XRD analysis of samples sprayed into water under different environments gave quite different results. In air the ZrB2 was partially oxidized to ZrO2 and particles were not completely melted. The oxidation was probably initiated by a partial evaporation of boron w6x, as oxide, during the plasma deposition. On the contrary, under inert gas, the powder particles maintained their initial composition and they were completely melted. Furthermore, the powder size distribution seemed to be shifted to lower values after spraying under argon atmosphere. This could be explained by a shrinkage of the particles resulting from their sintering in the liquid phase and disappearing of the internal porosity of spray dried particles w7x. On the basis of these results it was decided to continue the spraying under inert gas atmosphere. The different morphologies that were observed for the splats sprayed at low and high pressures could be attributed to different velocities of the molten particles at impact with the substrate. The particle sprayed at low pressure gave a splat having a flower shape, indicative Fig. 7. Correlation between plasma characteristics (temperature and density) and microhardness (measured under a load of 0.5 N) of ZrB2 coatings. 488 M. Tului et al. / Surface and Coatings Technology 151 – 152 (2002) 483–489 Table 3 Sheet resistances (mVyh), measured in different areas, relative to coatings of the fifth group V-1 sample V-2 sample 76.59 73.42 78.41 95.63 123.28 91.10 58.92 60.73 50.31 66.62 64.36 68.44 80.22 114.67 65.72 61.19 66.17 80.22 119.20 §50 87.93 66.17 57.11 53.94 mm™ 87.47 70.70 60.73 61.19 69.35 62.09 68.44 84.30 114.67 60.28 61.19 66.63 80.68 142.32 87.47 72.52 60.73 55.29 111.04 78.41 63.45 66.63 ≠ 30 mm x ≠ 30 mm x melting of the powder particles. Further remarks are as follows: Fig. 8. ZrB2 coating microhardness as a function of AHF parameter. of its high velocity at impact. This was lower at high pressure and the splat had a pancake form w5x. Regarding the process parameters, there was a strong correlation between AHF and coating microhardness. This confirms well the usefulness of the proposed simplified model. In fact, the coating microhardness grows as coating cohesion increases, which is favored by a complete ● High input of electrical power to the plasma is needed in order to melt the ZrB2 particles. This was expected taking into account the high melting point of this compound. ● The power required to melt particles of a given diameter decreases by increasing the chamber pressure. This is due to the increase of the plasma density which enhances the heat transfer among the powder particles. Moreover, the high chamber pressure results in a lower particle speed within the plasma, thus increasing the dwell time. ● Similar AHF parameters can be obtained using different combinations of process parameters (gas flow rate, power input, etc.). It gives a possibility to choose the parameters in a way more convenient for the spraying torch, e.g. the lifetime of torch electrodes increases at smaller flow rate of H2. Fig. 9. Friction coefficients of ZrB2, Al2O3 – TiO2, and WC – Co coatings obtained by IPS, APS, and HVOF, respectively. M. Tului et al. / Surface and Coatings Technology 151 – 152 (2002) 483–489 The ZrB2 coatings sprayed using IPS optimized parameters showed good tribological characteristics. The ZrB2 friction coefficient was lower than the friction coefficients measured for the reference materials (Al2O3qTiO2 and WCqCo). Finally, the measured ZrB2 d.c. resistivity (rs3.2 mV-m) was similar to that of metals. 5. Conclusions A powder suitable for thermal spraying was prepared by means of spray dryer agglomeration. The usefulness of the simplified model to optimize the spray parameters was verified. The experimental results proved that thermal spraying can be a valid alternative to CVD technology. ZrB2 coatings obtained by plasma spraying compared favorably with the standard materials, in terms of tribological performance. The d.c. resistivity indicated also that the coatings were good electrical conductors. 489 Acknowledgements Authors are grateful to Prof T. Valente from University of Rome 1 ‘La Sapienza’ for his assistance and useful discussion. References w1x R. Steinitz, Borides: Fabrication, Properties, Applications, Academic Press, New York, 1960. w2x H.D. Hanes, D.A. Seifert, C.R. Watts, Hot Isostatic Processing, Battelle Press, Columbus, Ohio, 1992. w3x US patent n804446357, Resistance-heated Boat for Metal Vaporization, 1984. w4x T. Valente, L. Bertamini, M. Tului, Effects of Pressure Deposition on Plasma Jet and Coating Microstructure, CSM — Centro Sviluppo Materiali rapport, Rome, Italy, 1996. w5x L. Pawlowski, The Science and Engineering of Thermal Spray Coatings, Wiley, Chichester, 1995. w6x W.C. Tripp, H.C. Graham, J. Electrochem. Soc.: Sol. State Sci. 118 (7) (1971) 1195–1199. w7x K. Masters, Spray Drying Handbook, Longman Scientific & Technical, London, 1985.