Effect of surface macroroughness on the microstructure and sliding wear properties of Al₂O₃ + 13 wt.% TiO₂ thick coatings

1. INTRODUCTION

Alumina (Al₂O₃) is a commercially available ceramic widely employed in industry to overcome wear and corrosion issues; coatings developed with Al₂O₃ offer a reasonable cost alternative to surface protection. In particular, Al₂O₃ + 13 wt.-% TiO₂ alloy coatings provide acceptable sliding, abrasion and erosion wear resistance owe to a combination of high hardness and reliable toughness. The addition of TiO₂ has several advantages; for instance, the melting temperature and hardness decrease as the content of TiO₂ increases; on the contrary, the toughness increase by increasing TiO₂ content, thus improving the overall wear properties of the coatings (Wang et al., 2000Wang, Y., Jiang, S., Wang, M., Wang, S., Xiao, T.D., Strutt, P.R. (2000). Abrasive wear characteristics of plasma sprayed nanostructured alumina/titania coatings. Wear 237 (2), 176-185. https://doi.org/10.1016/S0043-1648(99)00323-3.; Habib et al., 2006Habib, K.A., Saura, J.J., Ferrer, C., Damra, M.S., Giménez, E., Cabedo, L. (2006). Comparison of flame sprayed Al₂O₃/TiO₂ coatings: Their microstructure, mechanical properties and tribology behavior. Surf. Coat. Technol. 201 (3-4), 1436-1443. https://doi.org/10.1016/j.surfcoat.2006.02.011.; Yılmaz, et al., 2007Yılmaz, R., Kurt, A.O., Demir, A., Tatlı, Z. (2007). Effects of TiO₂ on the mechanical properties of the Al₂O₃ - TiO₂ plasma sprayed coating. J. Eur. Ceram. Soc. 27, 1319-1323. https://doi.org/10.1016/j.jeurceramsoc.2006.04.099.; Di Girolamo et al., 2014Di Girolamo, G., Brentari, A., Blasi, C., Serra, E. (2014). Microstructure and mechanical properties of plasma sprayed alumina-based coatings. Ceram. Int. 40 (8), 12861-12867. https://dx.doi.org/10.1016/j.ceramint.2014.04.143.; Matikainen et al., 2014Matikainen, V., Niemi, K., Koivuluoto, H., Vuoristo, P. (2014). Abrasion, erosion and cavitation erosion wear properties of thermally sprayed alumina based coatings. Coatings 4 (1), 18-36. https://doi.org/10.3390/coatings4010018.; Ghazali et al., 2016Ghazali, M.J., Forghani, S.M., Hassanuddin, N., Muchtar, A., Daud, A.R. (2016). Comparative wear study of plasma sprayed TiO₂ and Al₂O₃-TiO₂ on mild steels. Tribol. Int. 93, 681-686. https://doi.org/10.1016/j.triboint.2015.05.001.).

Flame thermal spraying is a cost-effective method for protecting or refurbishing components and structures because: a) it is easy to handle and quite adaptable to manufacturing processing and the automation is not needed; b) it is portable and easy to use as the operators do not require high-level of training; c) it produces lower dust and fume levels and the security equipment for operation is not highly sophisticated as compared to other thermal spraying processes; d) it is easy to coat parts or components with complex geometries due to the possibility of utilizing manual operation; e) it inputs lower distortion to substrate due low heat transfer; and f) it is suitable to be employed in environments which are less demanding. Flame spraying is often used to repair and/or protect industrial components subjected to sliding wear such as shafts, bearings, gears, or similar surfaces, by utilizing an ample variety of materials as feedstock (Oerlikon Metco, 1945Oerlikon Metco (1945). Metallizing Handbook. Long Island City, N.Y.; Romero, 1990Romero, R.R. (1990). Machinery Repairman. NAVEDTRA 12204-A. Naval Education and Training Program.; ASTM Committee, 1994ASTM Committee (1994). ASM Handbook “Surface Engineering”. Vol 5, ASM International, Materials Park, OH, USA.; Davis, 2004Davis, J.R. (2004). Handbook Thermal Spray Technology. ASM International, Materials Park, OH, USA.).

Adhesion is one of the most important characteristic of thermally sprayed coatings that strongly influence on the wear performance; thus, the surface roughness of the substrate is of critical importance since this guarantees suitable mechanical anchorage or mechanical interlocking (Davis, 2004Davis, J.R. (2004). Handbook Thermal Spray Technology. ASM International, Materials Park, OH, USA.; Wang et al., 2005Wang, Y.Y., Li, C.J., Ohmori, A. (2005). Influence of substrate roughness on the bonding mechanisms of high velocity oxy-fuel sprayed coatings. Thin Solid Films 485 (1-2), 141-147. https://doi.org/10.1016/j.tsf.2005.03.024.). In low velocity thermal spraying processes i.e. flame spraying and/or in processes where exist un-melted particles; mechanical interlocking of the splat within the substrate irregularities is the principal bonding mechanism influencing on the adhesion strength in coatings (Paredes et al., 2006Paredes, R.S.C., Amico, S.C., d’Oliveira, A.S.C.M. (2006). The effect of roughness and pre-heating of the substrate on the morphology of aluminium coatings deposited by thermal spraying. Surf. Coat. Technol. 200 (9), 3049-3055. https://doi.org/10.1016/j.surfcoat.2005.02.200.). Roughening is a critical step in preparing the surface before the coating application as the former ensures proper adhesion. Therefore, in order to obtain adequate roughening; the substrate surface must be correctly prepared and various common methods can be utilized viz. rough threading (used for cylindrical surfaces), grit blasting, and/or a combination of the previously mentioned (ASTM Committee, 1994ASTM Committee (1994). ASM Handbook “Surface Engineering”. Vol 5, ASM International, Materials Park, OH, USA.).

Thick coatings are found in various situations such as: built-up or reconditioning of machine parts, repairmen of components, thermal barrier, wear protection, corrosion protection, among others. Nevertheless, processing of thick coatings is not an easy task and several problems must be faced: i.e. weak mechanical anchoring (poor adhesion) (Berndt and McPherson, 1979Berndt, C.C., McPherson, R. (1979). The Adhesion of Flame and Plasma Sprayed Coatings - A Literature Review. Australas. Weld. Res. 6 (January), pp. 75-85.), presence of edge delamination failure (Bordeaux et al., 1991Bordeaux, F., Saint Jacques, R. G., Moreau, C. (1991). Study of surface preparation for enhanced resistance to thermal shocks of plasma-sprayed TiC coatings. Surf. Coat. Technol. 49 (1-3), 50-56. https://doi.org/10.1016/0257-8972(91)90030-Z.), increasing of residual stresses with thickness (Tucker, 1974Tucker, R.C. (1974). Structure property relationships in deposits produced by plasma spray and detonation gun techniques. J. Vac. Sci. Technol. 11 (4), 725-734. https://doi.org/10.1116/1.1312743.), generation of microcracks (Steffens et al., 1999Steffens, H.D., Babiak, Z., Gramlich, M. (1999). Some aspects of thick thermal barrier coating lifetime prolongation. J. Therm. Spray Tech. 8 (4), 517-522. https://doi.org/10.1361/105996399770350197.).

Macro-roughening (an easier to handle technique, faster, and economical) is commonly applied for restoring greater dimensions (thick coatings) in damaged surfaces. Macro-roughening can be performed on shafts through a lathe machine operation by cutting a narrow, low-pitch, shallow groove or thread in the surface (Romero, 1990Romero, R.R. (1990). Machinery Repairman. NAVEDTRA 12204-A. Naval Education and Training Program.). It is worth mentioning that macro-roughening is an effective technique for residual stress reduction by inducing folds in the coatings thus promoting low shrinkage. In addition, macro-roughening can prevent shear or edge delamination, and further reduce cracking and/or spalling in coatings (Tucker, 1974Tucker, R.C. (1974). Structure property relationships in deposits produced by plasma spray and detonation gun techniques. J. Vac. Sci. Technol. 11 (4), 725-734. https://doi.org/10.1116/1.1312743.; James, 1984James, D.H. (1984). A review of Experimental Findings in surface preparation for thermal spraying. J. Mech. Work. Technol. 10 (2), 221-232. https://doi.org/10.1016/0378-3804(84)90069-X.; Bordeaux et al.,1992Bordeaux, F., Jacques, R.G.S., Moreau, C., Dallaire, S., Lu, J. (1992). Thermal shock resistance of TiC coatings plasma sprayed onto macroroughened substrates. Surf. Coat. Technol. 53 (1), 49-56. https://doi.org/10.1016/0257-8972(92)90102-G.; Matějíček et al., 2007Matějíček, J., Chráska, P., Linke, J. (2007). Thermal spray coatings for fusion applications - Review. J. Therm. Spray Tech. 16 (1), 64-83. https://doi.org/10.1007/s11666-006-9007-2.; Hollis et al., 2007Hollis, K.J., Bartram, B.D., Roedig, M., Youchison, D., Nygren, R. (2007). Plasma-sprayed beryllium on macro-roughened substrates for fusion reactor high heat flux applications. J. Therm. Spray Tech. 16 (1), 96-103. https://doi.org/10.1007/s11666-006-9011-6.). However, there is a lack of information in the literature relating the cylindrical surface macro-roughness viz. knurling and grooving, to the resultant microstructure and sliding wear performance upon flame spraying of thick coatings.

This study aims at evaluating the effect of the cylindrical surface macro-roughness on the overall microstructure and sliding wear behaviour of Al₂O₃ + 13 wt.-% TiO₂ thick coatings deposited on a AISI/SAE 1045 steel bar, using two different machining methods as anchorage viz. spiral grooving and knurling pattern.

2. MATERIALS AND METHODS

A commercial (Oerlikon Metco^TM), agglomerated and sintered Al₂O₃ + 13 wt.-% TiO₂ powder alloy with a particle size distribution of -45+15 µm (d₅₀=30.9 µm) was used as feedstock in this work (Fig. 1). An Oerlikon Metco^TM Ni-5Mo-5.5Al powder alloy with size distribution of -90+45 μm (d₅₀=59.6 μm) was additionally employed as bond-coat in order to improve the adhesion of alumina to the substrate.

An AISI/SAE 1045 cylindrical steel bar with diameter of 31.75 mm and 256 HV_0.5 in hardness was used as substrate. A drilled hole (15 mm in diameter) was performed in the center of the cylinder; afterwards, the steel bar was cross-sectioned with the purpose of forming a ring specimen of 5 mm in thickness.

Four different macro-roughened surfaces were prepared at the outer surface of the ring specimens: two diamond pattern knurling (DKA and DKB), and two spiral pattern grooving (SGA and SGB). Geometry and dimensions of the surface profile for both DK and SG patterns are indicated in the schematic of Fig. 2.

Before the coating deposition, all specimens were properly cleaned in an ultrasonic bath. The coating deposition was conducted towards the outer surface of the ring specimen, as it is shown in Fig. 3a, by using a CastoDyn DS8000™ thermal spray gun. Just after eight preliminary trials; suitable deposition parameters for Al₂O₃ + 13 wt.-% TiO₂ powders were obtained as it is listed in Table 1, the deposition parameters were constant for all conditions. Multi-pass torch (6 passes in total) was performed to obtain thick (1000-1200 µm) Al₂O₃ + 13 wt.-% TiO₂ coatings.

The coated specimens were sectioned in the transverse direction. After performing the usual metallographic procedures all micrographs were taken in a Tescan Mira3™ scanning electron microscope. By following the ASTM E384 (2017)ASTM E384 (2017). Standard Test Method for Microindentation Hardness of Materials. ASTM International, West Conshohocken, USA. hardness standard and employing a Shimadzu tester, Vickers hardness was measured with 500 g (4.905 N) of applied load and a dwell time of 15 s. Three linear paths of indentations parallel to each other with a minimum number of 8 indentations (located 200 µm apart as a minimum to each other) were obtained at the substrate and throughout the coating thickness in all specimens, it is important to declare that porosity was avoided when placing the hardness indentations. The coatings were evaluated by X-ray diffraction (XRD) using Cu-Kα radiation on a D8Advance Bruker diffractometer by setting up a scan step of 0.002° (2q) and a time per scan of 1 s. The standard test method ASTM E2109−01 (2021)ASTM E2109−01 (2021). Standard Test Methods for Determining Area Percentage Porosity in Thermal Sprayed Coatings. ASTM International, West Conshohocken, USA. was used for determining the area percentage porosity in the thermal sprayed coatings using image analysis.

Sliding wear testing was conducted as per the configuration illustrated in Fig. 3b. An Al₂O₃ disk with an initial roughness of Ra=18 µm and hardness within a range of 2110-2450 HV was utilized as counterface. Sliding wear testing was conducted by setting up a constant load of 16 N, a rotational speed of 450 rpm, and a fixed sliding distance of 1979 m. All samples were cleaned with an ultrasonic acetone bath and weighed before and after the trial in ±0.01 mg AUW220D Shimadzu™ precision balance with the purpose of evaluating mass loss of the specimen. Four replicas per condition were subjected to sliding wear testing at room temperature.

3. RESULTS

The corresponding cross-sectioned microstructure of the Al₂O₃ + 13 wt.-% TiO₂ coatings deposited onto four different macro-roughened surfaces is shown in Fig. 4. The Ni-5Mo-5.5Al bond coat (hereafter named as “Ni bonding layer”) with an average layer thickness of 177 µm was properly anchored to the steel substrate. The subsequent flame sprayed region is composed of various layers of Al₂O₃ + 13 wt.-% TiO₂ that form the final thick coating structure.

The porosity (as pointed by black arrows) was revealed throughout the thickness of the coating in both spiral grooving surface conditions (SGA and SGB); it is worth mentioning that higher density of porosity is located about the tip of the threads (Fig. 4a-b). Porosity is clearly correlated to the sharp tip and the slope of the groove. Due to the grooving surface is not parallel to the particles stream; there is poor adhesion along the slope of the grooves, and additionally, few particles are trapped around the sharp tip, thus promoting porosity. On the other hand, porosity is also observed upon the knurling conditions (DKA and DKB) as shown in Fig. 4c-d; however, there is a clear reduction in porosity content as compared to the SGA and SGB conditions. The absence of a sharp tips as well as the reduced slope of the grooves over the substrate actually creates an extensive active zone, leaving a larger amount of deposited particles onto the surface with reduced porosity. The porosity percentage is clearly reduced from 8.2% in spiral grooving pattern up to 3.5% upon knurling pattern due to the macro-roughness effect.

Based on the abovementioned results, the surface topography viz. macro-roughness has a clear effect on the distribution and density of porosity; for instance, the absence of sharp tips and the reduced slope in the knurling conditions, leaded to a reduction in porosity; an indirect evidence of acceptable adhesion conditions. Indeed, the top surface of the coating takes the original shape of the initial substrate macro-roughness, for this reason a smoother surface can be obtained when employing the knurling pattern.

A laminar structure formed by the particle impact and the subsequent flattening of splats commonly found in flame spraying processes is observed in all Al₂O₃ + 13 wt.-% TiO₂ coatings (Fig. 5). A relatively dense microstructure with distinguishable microcracks is revealed; however, the extension of cracking depended on macro-roughness condition, hence, visibly dense and coarser microcracks are observed upon the SG pattern (Fig. 5a-b); whereas, a reduced fraction of thinner microcracks are revealed upon the DK condition as shown in Fig. 5c-d. It is clear that microcracks are minimized upon the pattern knurling DKA and DKB.

Microcracks are the product of the shrinkage due to rapid solidification of the deposited particles on the surface. Both spiral grooving conditions resulted in coarser and larger microcracks due to abrupt surface geometry changes (i.e. sharp tips, with high slope within the groove shape), thus causing distortion during the splat disposition affecting the shrinkage pattern, and hence, the inter-splat adhesion. On the other hand, a properly defined pattern such as knurling, ensures uniform surface geometry changes, free of sharp-tips, and reduced slope, the splats are properly trapped to the substrate with smooth geometry changes, having an effective deposition, thus facilitating the stacking (coating build up), and promoting good inter-splat adhesion. Therefore, it is again inferred that the surface topography greatly affects the characteristics of microcracks present in the coatings. Predominantly melted regions are encounter in all coatings; however, partially melted and/or un-melted round particles are also revealed.

XRD analysis of the feedstock powder provided in Fig. 6a, shows the presence of stable α- Al₂O₃ and TiO₂ (rutile) phases along with a reduced fraction of Al₂TiO₅ which can be ascribed to the sintering reaction of Al₂O₃ + 13 wt.-% TiO₂ powder mixture during the manufacturing process of the powder.

No differences in terms of phase transformation were found among SGA, SGB, DKA and DKB conditions as shown in Fig. 6b. Transformation from α-Al₂O₃ phase to the metastable γ-Al₂O₃ phase during flame spraying processing resulted in all specimens as it is seen in Fig. 6b, not evidence neither of TiO₂ nor Al₂TiO₅ was recorded. The transformation from α-Al₂O₃ phase to γ-Al₂O₃ phase during solidification is attributed to its lower critical free energy for nucleation; on this basis, γ-Al₂O₃ would be nucleated rather than α-Al₂O₃ at low temperatures (McPherson, 1973McPherson, R. (1973). Formation of metastable phases in flame- and plasma-prepared alumina. J. Mater.s Sci. 8, 851-858. https://doi.org/10.1007/BF02397914.; R. McPherson, 1980McPherson, R. (1980). On the formation of thermally sprayed alumina coatings. J. Mater. Sci. 15, 3141-3149. https://doi.org/10.1007/BF00550387.; Fervel et al., 1999Fervel, V., Normand, B., Coddet, C. (1999). Tribological behavior of plasma sprayed Al₂O₃ -based cermet coatings. Wear 230 (1), 70-77. https://doi.org/10.1016/S0043-1648(99)00096-4.; Normand et al., 2000Normand, B., Fervel, V., Coddet, C., Nikitine, V. (2000). Tribological properties of plasma sprayed alumina-titania coatings: role and control of the microstructure. Surf. Coat. Technol. 123 (2-3), 278-287. https://doi.org/10.1016/S0257-8972(99)00532-0.; Yılmaz et al., 2007Yılmaz, R., Kurt, A.O., Demir, A., Tatlı, Z. (2007). Effects of TiO₂ on the mechanical properties of the Al₂O₃ - TiO₂ plasma sprayed coating. J. Eur. Ceram. Soc. 27, 1319-1323. https://doi.org/10.1016/j.jeurceramsoc.2006.04.099.; Di Girolamo et al., 2014Di Girolamo, G., Brentari, A., Blasi, C., Serra, E. (2014). Microstructure and mechanical properties of plasma sprayed alumina-based coatings. Ceram. Int. 40 (8), 12861-12867. https://dx.doi.org/10.1016/j.ceramint.2014.04.143.).

Interestingly, a semi-quantification analysis made to the multilayered coating (obtained with more than one torch pass) resulted in an average fraction of approximately 25% of α-Al₂O₃ and 755 of γ-Al₂O₃. In further analysis, XRD was executed in a single layer coating (obtained with only one torch pass) thus resulting in 5 to 10% content of α-Al₂O₃; hence, the presence of α-Al₂O₃ is correlated to the partially melted and/or the un-melted particles in single-layered coatings (Normand et al., 2000Normand, B., Fervel, V., Coddet, C., Nikitine, V. (2000). Tribological properties of plasma sprayed alumina-titania coatings: role and control of the microstructure. Surf. Coat. Technol. 123 (2-3), 278-287. https://doi.org/10.1016/S0257-8972(99)00532-0.; Lou et al., 2003Lou, H., Goberman, D., Shaw, L., Gell, M. (2003). Identation fracture behavior of plasma-sprayed nanostructured Al₂O₃-13wt.%TiO₂ coatings. Mater. Sci. Eng. A 346 (1-2), 237-245. https://doi.org/10.1016/S0921-5093(02)00523-3.; Yılmaz et al., 2007Yılmaz, R., Kurt, A.O., Demir, A., Tatlı, Z. (2007). Effects of TiO₂ on the mechanical properties of the Al₂O₃ - TiO₂ plasma sprayed coating. J. Eur. Ceram. Soc. 27, 1319-1323. https://doi.org/10.1016/j.jeurceramsoc.2006.04.099.; Di Girolamo, et al., 2014Di Girolamo, G., Brentari, A., Blasi, C., Serra, E. (2014). Microstructure and mechanical properties of plasma sprayed alumina-based coatings. Ceram. Int. 40 (8), 12861-12867. https://dx.doi.org/10.1016/j.ceramint.2014.04.143.; Yang et al., 2015Yang, Y., Wang, Y., Tian, W., Yan, D-ran, Zhang, J.-xin, Wang, L. (2015). Influence of composite powders’ microstructure on the microstructure and properties of Al₂O₃-TiO₂ coatings fabricated by plasma spraying. Mater. Des. 65, 814-822. https://doi.org/10.1016/j.matdes.2014.09.078.). However, the remained fraction of α-Al₂O₃ phase in the multilayered specimens, is absolutely related to re-transformation of γ-Al₂O₃ → α-Al₂O₃ during the multi-pass torch spray processing (reheating). For instance, during the deposition process; the melted particles resulted in full transformation from α-Al₂O₃ to γ-Al₂O₃ phase; nevertheless, with the subsequent passing of the torch on the already deposited layer, a localized heat treatment is developed with enough temperature for causing γ-Al₂O₃ → α-Al₂O₃ reversed transformation. It is well known that transformation from γ-Al₂O₃ to α-Al₂O₃ phase is possible when heating the substrate approximately about 1100 °C (McPherson, 1980McPherson, R. (1980). On the formation of thermally sprayed alumina coatings. J. Mater. Sci. 15, 3141-3149. https://doi.org/10.1007/BF00550387.; Di Girolamo, et al., 2014Di Girolamo, G., Brentari, A., Blasi, C., Serra, E. (2014). Microstructure and mechanical properties of plasma sprayed alumina-based coatings. Ceram. Int. 40 (8), 12861-12867. https://dx.doi.org/10.1016/j.ceramint.2014.04.143.). It is important to mention that approximately 15 to 20% of the preliminary layer is re-transformed to α-Al₂O₃ due to the fact that the maximum temperature for transformation penetrates few microns within the previous layer. The subsequent torch passes created a layering which combines the presence of thinner α-Al₂O₃ phase and thicker γ-Al₂O₃ phase layers as schematically depicted in Fig. 7a. It is worth mentioning that spraying parameters (nozzle, flame type, spraying distance, etc.) were constant for all specimens; then it is deduced that the surface geometry has no effect on the phase transformation of the coatings as no differences are observed according to XRD.

Figure 7b, shows the micro-hardness profile taken from the substrate throughout the coating thickness. A schematic depicting the paths of hardness indentations on the various regions viz. substrate and the Al₂O₃ layering is provided in Fig. 7a. The hardness of the substrate averaged 256 HV_0.5, whereas the mean value hardness of the alumina layering was 972 HV_0.5. A clear fluctuation of the hardness values is observed along the coating in all specimens which is markedly related to the distinct α- and γ-Al₂O₃ layers throughout the coating. Therefore, as some indentations lay into α-Al₂O₃ and others into γ-Al₂O₃ along the distance from the substrate/coating line over the coating; the higher hardness values are correlated to α-Al₂O₃ because of its highest hardness as compared with γ-Al₂O₃ phase (Yilmaz, 2009Yilmaz, Ş. (2009). An evaluation of plasma-sprayed coatings based on Al₂O₃ and Al₂O₃-13 wt.% TiO₂ with bond coat on pure titanium substrate. Ceramics International 35 (5), 2017-2022. https://doi.org/10.1016/j.ceramint.2008.11.017.; Singh et al., 2011Singh, V.P., Sil, A., Jayaganthan, R. (2011). A study on sliding and erosive wear behaviour of atmospheric plasma sprayed conventional and nanostructured alumina coatings. Mater. Des. 32 (2), 584-591. https://doi.org/10.1016/j.matdes.2010.08.019.; Islak et al., 2012Islak, S., Buytoz, S., Orhan, N., Stokes, J. (2012). Effect on microstructure of TiO₂ rate in Al₂O₃-TiO₂ composite coating produced using plasma spray method. Optoelectron. Adv. Mater. Rapid Commun. 6 (9), 844-849.).

Wear track of multi-layered Al₂O₃ + 13 wt.-% TiO₂ coatings are shown in Fig. 8. Scratched surfaces are revealed for all specimens. Severe wear damage is clearly observed upon SGB condition followed by SGA. Coating defects (i.e. cracks) located parallel to the sliding wear track are clearly distinguishable in Fig. 8b. On the contrary, coating defects in SGA, DKA and DKB conditions are perpendicular to the wear tracks. The initial contact area between the counter disk (alumina) and the coating is minimum; in fact, the actual contact area is supported by the coating tips inherited by the substrate geometry. Once the sliding friction displacement has begun; coalescence of existing cracks as well as nucleation of new ones are generated, following this, particles are detached out from coatings (debris), it is worth mentioning that the presence of porosity and microcracks also generate splats exfoliation (as shown in the SEM micrograph in Fig. 9) (Normand et al., 2000Normand, B., Fervel, V., Coddet, C., Nikitine, V. (2000). Tribological properties of plasma sprayed alumina-titania coatings: role and control of the microstructure. Surf. Coat. Technol. 123 (2-3), 278-287. https://doi.org/10.1016/S0257-8972(99)00532-0.; Psyllaki et al., 2001Psyllaki, P.P., Jeandin, M., Pantelis, D.I. (2001). Microstructure and wear mechanisms of thermal-sprayed alumina coatings. Mater. Lett. 47 (1-2), 77-82. https://doi.org/10.1016/S0167-577X(00)00215-9.; Michalak et al., 2021Michalak, M., Sokołowski, P., Szala, M., Walczak, M., Łatka, L., Toma, F.-L., Björklund, S. (2021). Wear Behavior Analysis of Al₂O₃ Coatings Manufactured by APS and HVOF Spraying Processes Using Powder and Suspension Feedstocks. Coatings 11 (8), 879. https://doi.org/10.3390/coatings11080879.), which is conducted by the weak splat unions and microcracks, the above mentioned produce a progressive increment in contact area by the constant material removing from the tips. Additionally, the hard debris trapped at the inter-face between the disk and the coatings promotes the abrasion mechanism. Actually, debris is generated from both: the detached splats coming from coatings and the separated material from the counter disk. Scratching scar demonstrates an abrasion mechanism involved in the wear process.

Figure 10, shows the relationship between the wear ratio (coating mass loss/ substrate Mass loss) and porosity. It is seen that the specimens with the knurling pattern possess a higher wear resistance as compared to specimens with the spiral grooving pattern. It is clear that the wear ratio is higher when the porosity increases; for instance, SGB specimen having the largest amount of porosity and microcracks (Fig. 8b) had a higher wear ratio. Therefore, the porosity influenced on the wear resistance by weakening the splats joining; if porosity the increases, splats adhesion is logically reduced by promoting splats detachment, which clearly affects the wear resistance. Furthermore, microcracks limit splats contact, influencing on splats adhesion, and consequently on the coating toughness. If splats joint is deficient, then the toughness is reduced thus facilitating splats decohesion during wear.

4. CONCLUSIONS