Correlation between microstructural characteristics

Surface & Coatings Technology 200 (2005) 1310 – 1314 www.elsevier.com/locate/surfcoat

Correlation between microstructural characteristics and the abrasion wear resistance of sealed thermal-sprayed coatings Sugehis Liscano a,*, Linda Gil a, Mariana H. Staia b a

Departamento de Ingenierıá Metalu´rgica, Universidad Nacional Experimental Politećnica ‘‘Antonio Jose´ de Sucre’’, Urb. Villa Asia, Puerto Ordaz, Venezuela b Escuela de Ingenierıá Metalu´rgica y Ciencia de los Materiales, Facultad de Ingenierıá, Universidad Central de Venezuela, Apartado 49141, Caracas, Venezuela Available online 19 September 2005

Abstract Plasma-sprayed aluminum oxide coatings (Al2O3) are mainly used as a wear-resistant coating in mechanical applications. Previous studies have shown that both the abrasion wear resistance and the corrosion resistance of these coatings can be significantly improved by applying a sealing treatment. It was reported that this improvement is mainly due to the microstructural modifications which take place during the post treatment process. Therefore, this study was conducted in order to determine the influence of the microstructural characteristics on the wear resistance of these coatings, when different sealing treatments were applied to AISI 1020 thermal-sprayed specimens coated with alumina – 13% titania. Abrasion wear resistance of these coatings was evaluated employing the standard rubber wheel abrasion test and the ball-on-disc test. The coatings microstructure was characterized and their wear mechanisms were analyzed by using scanning electron microscopy. Porosity and microhardness values were also reported. It was shown that the abrasion wear resistance of both the phosphoric acid sealed coatings and the epoxy sealed coatings are significantly better than that of the unsealed coatings. D 2005 Elsevier B.V. All rights reserved. Keywords: Wear mechanisms; Sealants; Plasma-sprayed Al2O3 + TiO2 coatings

1. Introduction Given the importance of wear phenomena, it is natural that many technicians and scientists have worked on the problem. They have observed that with hard surfaces wear is considerably reduced. Generally ceramics (Al2O3, TiO2, Cr2O3, Z3O2) answer rather well to this criterion, but as it is often difficult to make ceramic bulk materials, one tends to make wear resistant coatings by detonation, flame spraying and plasma spraying [1,2]. Plasma-sprayed ceramic coatings are made up of layers that are formed when melted material droplets flatten and solidify on the surface of the substrate [3]. Because of their lamellar structure, the coatings have various amounts of pores. Pores and incomplete bonding between lamellae decrease the strength, wear resistance, and corrosion * Corresponding author. Tel.: +58 286 9626229; fax: +58 286 9626229. E-mail address: [email protected] (S. Liscano). 0257-8972/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2005.07.092

resistance of the coatings. The density of plasma-sprayed coatings can be increased by optimizing plasma parameters and spraying powders, or by using post-treatments. There are numerous sealants which can be used for thermally sprayed coatings [4]. For low temperature applications it is possible to use waxes and various organic sealants, e.g. epoxies and phenolics. These sealants are effective only to temperatures of about 100 -C. However, inorganic sealants can be utilized also at significantly higher temperatures. Other studies have shown that inorganic aluminum phosphates and phosphoric acid are very effective sealants, which also have a large improving effect on dry abrasion resistance of oxide coatings [5]. The tribological behavior of the coatings is closely linked to the microstructure [4 –6]. Accordingly, in this study the structure of plasma-sprayed Al2O3 + TiO2 coatings were modified by employing different sealants, with subsequent study of the correlation between the wear resistance and the microstructure.

S. Liscano et al. / Surface & Coatings Technology 200 (2005) 1310 – 1314

1311

2. Experimental details 2.1. Coatings Alumina– 13% titanium oxides of the type FST-335.23 (Flame Spray Technology, Italy) with a nominal particle size range of 45 + 15 Am was deposited onto substrates of AISI 1020. The coatings were commercially deposited (Thermacoat, Venezuela) by atmospheric plasma spraying, using APS 45 Flame Spray Technology [7]. The coating thickness in all cases was between 300 and 400 Am. The dimensions of the specimens for the rubber wheel abrasion test were 19.05 mm square bars and 50 mm in length, and in ball-on-disc test, 5 mm thickness and 25.4 mm diameter cylinders were used. The substrates were grit blasted just before spraying. 2.2. Sealing treatment Three different sealants were used to seal the plasmasprayed coatings: phosphoric acid, phenol and epoxy based sealants. Some characteristics of these sealants are listed in Table 1. Previous to the sealing treatment, dry grinding by using 600-grit SiC paper was applied to all samples in order to decrease their roughness below Ra = 1 Am. Subsequently, the samples were cleaned in ethanol by using an ultrasonic bath. All sealants were applied by brushing, with a nylon brush, and were allowed to impregnate into the coatings at room temperature, in air, and at atmospheric pressure for the following time periods: 12 h for the phenol sealant, 4 h for the epoxy sealant and 12 h for the phosphoric acid sealant. The characteristics of the cure treatments and the adequate concentration of the sealants, which were indicated by the suppliers, are shown in Table 1. The sealant excess was ground off using 120 SiC grit. 2.3. Coating microstructure Microstructure coatings cross-sections and the wear tracks on the surface of coatings were studied using an Olympus BH2-UMA optical microscope and a Phillips XL30 scanning electron microscope (SEM) with X-ray energy dispersive system (EDAX-DXA). Vickers microhardness scan on the cross-section of coatings was made using a Micromet Tester Series 2100 of Buehler Ltd. The magnification used was 400 for the measurements. The load was 300 g (HV0.3) and the testing time was 10 s. Ten measurements were made for each sample. Porosity

Fig. 1. Scanning electron microscope (SEM) micrographs of the crosssection of the following: (a) unsealed Al2O3 – 13%TiO2 plasma-sprayed coating; (b) epoxy sealed coating.

measurements were calculated by image analysis, analyzing 10 fields at magnification of 200 on the same crosssection of each coating. 2.4. Wear test Wear resistance of the sealed and unsealed coatings was studied by a rubber wheel abrasion test and the ball-on-disc test. The rubber wheel abrasion test equipment is a modified version of the equipment described in standard ASTM G6500 [6,8]. The samples were pressed against the rubber wheel with a force of 16 N. Dry quartz sand with a grain size range of 0.1 to 0.6 mm was used as an abrasive, which was fed between the rotating rubber wheel and the specimen. Test duration was 2160 s, which is equivalent to a wear length of 5904 m. Ball-on-disc tests were performed on a CSEM Tribometer Model Pin on Disc. The sliding counter body was a 6 mm sphere of alumina covering a radius of 5 mm with normal load of 5 N. The sliding length was 1000 m and the

Table 1 Characteristics and curing conditions of the sealant studied Sealant

Curing treatment

Optimum impregnation time (h) Sealant concentration

Phosphoric acid Temperature: 100, 200, 400 -C; time: 2 h for each temperature 12 Phenol resin Temperature: room temperature; time: 2 h 12 Epoxy resin Temperature: 120 -C; time: 1/2 h 4

Acid: 97.5%; solvent: 2.5% Resin: 10%, volatile vehicle: 90% Resin: 21.1%, volatile vehicle: 78.9%

1312

S. Liscano et al. / Surface & Coatings Technology 200 (2005) 1310 – 1314 45

Table 2 Microhardness and porosity of the coatings

40

Porosity (%)

Microhardness (HV0.3)

As-sprayed Phenol sealed Phosphoric sealed Epoxy sealed

23.1 T 3.8 14.3 T 4.4 18.5 T 2.5 14.8 T 2.6

848 T 40 953 T 22 1379 T 49 1021 T 32

sliding speed was fixed at 0.1 m/s. Both the sphere and the coated specimens were ultrasonically cleaned and degreased prior to wear testing. After ball-on-disc test, the resultant roughness of sliding surfaces was measured by Zygo 200 profilometer, and examined by scanning electron microscopy to assess the operating wear mechanism.

3. Results and discussion 3.1. Microstructure and microhardness Fig. 1 shows the cross-sectional scanning electron microscope (SEM) micrographs of the as-sprayed and epoxy sealed coatings. In the latter, the decrease in porosity throughout all the thickness of specimen is obvious. The porosity and average microhardness values of each condition are presented in Table 2. Sealing treatment increases the hardness of alumina coatings in the case when the epoxy and phosphoric sealants were used but had practically no effect on the hardness when the phenol sealant was used. Because the sealants are transparent colorless films, it is difficult to measure their penetration depth into ceramic coatings directly from the scanning electron microscope micrographs [9]. In order to recognize the penetration depth of the sealants in the ceramic coating, scan microhardness was made 50 Am away from the coating surface and every 100 Am. Fig. 2 shows the results of these scan microhardness measurements. Note that the penetration depth of

35

Weight loss (mg)

Condition

30 25 20 15 10 5 0 As-sprayed

Epoxy Sealed Phenolic Sealed Phosphoric Sealed

Fig. 3. Rubber wheel abrasion test results of sealed and unsealed Al2O3 – 13%TiO2 plasma-sprayed coating.

phosphoric and epoxy sealed coating was almost 400 Am. In the case of the phenol sealed coating, the penetration depth is almost 150 Am and from here on, coatings microhardness was the same as that of as-sprayed condition. Porosity values, calculated by image analysis, are a clear evidence of the densification effect of sealants; however, they are too high. This may be attributed to a change in residual stress state of coating due to the sealing treatment, which introduces new pullouts during metallographic preparation and, thus, increases apparent porosity [6].

1500

HV0.3

1300

1100

900

700

500 0

10

100

150

200

250

300

350

400

450

Distance from the top of the coating, µm As-sprayed

epoxy sealed

phenol sealed

phosphoric sealed

Fig. 2. Effect sealing treatment on the microhardness of the Al2O3 – 13% TiO2 coating.

Fig. 4. Scanning electron microscope (SEM) micrographs of the wear traces of abrasive wear tested: (a) unsealed Al2O3 – 13%TiO2 plasma-sprayed coating; (b) epoxy sealed coating.


3.2. Wear resistance Rubber wheel abrasion wear test results of the sealed and unsealed coatings are presented in Fig. 3. As it can be seen from the results, the epoxy and phosphoric sealants decreased wear, but the latter showed more accentuated effect. In contrast, with the previously mentioned conditions, the phenol sealed coating showed higher wear than the unsealed ones. A similar behavior was reported previously in the literature [10]. The results obtained showed a direct correlation with microhardness behavior mentioned previously, which corroborates the fact that effective strengthening method also improves the dry abrasion resistance [10]. Abrasive wear traces of the as-sprayed coatings and phenol sealed coatings were very rough and observable with the bare eye (Fig. 4a), which indicates that the material was removed in large pieces with the size of a lamella [11]. Although the phenol sealant had penetrated some microns in depth of coatings, it is possible that it may have suffered a high shrinkage in curing; it could create local tensile stress fields, causing easier and large wear-particle detachment [12]. The epoxy and phosphoric sealed coating behavior was different, they wore smoothly (Fig. 4b). The epoxy sealant shows a better penetration and adhesion to pore walls, and for the phosphoric sealant it is possible that it reacts during cure treatment with coating [6,11]; in both cases, the result is a better adhesion between lamellae. Wear traces were featureless and smooth, so the particles removed were very fine. In the graphic of coefficient of friction shown in Fig. 5, it can be observed that the epoxy sealant modified the coefficient of friction of the system, decreasing it by 0.3. In contrast, the behavior of coating in phenol sealed condition does not show any difference with respect to the unsealed condition. The alterations in the coefficient of friction curve of epoxy sealed condition can be a conse-

1313

quence of the increase in the friction force necessary to overcome polymer particles that accumulate in the ball front. Likewise, the roughness values of the sliding surfaces after ball-on-disc test are also presented in Fig. 5. Only the epoxy sealed condition maintained the initial roughness which is indicative of a good protective behavior of the sealant. In the other cases, the high roughness values are evidence of material detaching from coating surface during wear. Sliding surfaces of sealed coatings show two different wear mechanisms (see the micrographs presented in Fig. 6). The first is shown in unsealed and phenol sealed condition (Fig. 6a), where fine cracks are observed below the plastically deformed zones and adjacent to rougher areas, which represents areas where part of the coating surface had been deformed, fractured and removed from samples. In the phosphoric sealed condition (Fig. 6c), small tearings can be observed running parallel to the sliding directions which indicates that gross plastic deformation of the coating has occurred. This behavior of plastic deformation has been reported for bulk ceramic-on-ceramic wear test and associated to compressive residual stress conditions [12,13]. The second mechanism, present in epoxy sealed coating, is associated with tensile residual stress condition and it is revealed by a network of coarse cracks within the wear track, which is probably due to greater resistance to detachment of areas of coating. All this can be seen in Fig. 6b.

4. Conclusions From the above results, it can be concluded that sealing treatment has improved the abrasion wear resistance of the as-deposited coating– substrate system by decreasing the open porosity and developing a better joint between lamellae. It was shown that the sealing treatment with phosphoric and epoxy based sealants was more effective in improving the abrasion wear resistance. In the sliding

0,8

As-sprayed

Coefficient of friction, µ

0,7

Phenol sealed

0,6 Phosphoric sealed

0,5

0,4 Epoxy sealed

0,3

0,2

0

100

200

300

400

500

600

700

800

900

1000

Sliding distance, m

Fig. 5. Variation of coefficient of friction on sliding surface during ball-on-disc test. Roughness values after ball-on-disc test are as follows: (a) as-prayed, 10.4 Am; (b) epoxy sealed, 0.6 Am; (c) phenol sealed, 4.1 Am; (d) phosphoric sealed, 4,9 Am.

1314


surface, obtained after ball-on-disc test, damages dominated by plastic deformation and surface cracking were observed, and these were associated with the level of residual stress. By using scan microhardness, the penetration depth of sealants throughout alumina – 13% titanium coating was determined; it turned out to be 400 Am in phosphoric and epoxy sealed condition, and 150 Am in phenol sealed condition.

Acknowledgment The authors wish to acknowledge the great help and technical advice from O. Leon of the UNEXPO Wear Laboratory.

References

Fig. 6. Scanning electron microscope (SEM) micrographs of the wear traces of ball-on-disc tested: (a) unsealed Al2O3 – 13%TiO2 plasma-sprayed coating; (b) epoxy sealed coating; (c) phosphoric sealed coating.

[1] V. Wilms, H. Herman, Thin Solid Films 39 (1976) 251. [2] Th. F. Weber, ASM Heat Treat. Surf. Eng. 2 (1994) 573. [3] H. Herman, S. Sampath, in: K.H. Stern (Ed.), Metallurgical and Ceramic Protective Coatings, Chapman & Hall, London, 1996, p. 263. [4] K. Niemi, P. Sorsa, P.P. Vuoristo, T. Ma¨ntyla¨, Proc. 7th Int. Thermal Spraying Conf., Massachusetts, 1994, p. 533. [5] A.A. Abdel-Samad, A.M.M. El-Bahloul, E. Lugscheider, S.A. Rassoul, J. Mater. Sci. 35 (2000) 3127. [6] J. Knuutila, S. Ahmaniemi, E. Leivo, P. Sorsa, P.P. Vuoristo, T. Ma¨ntyla¨, Proc. 15th Int. Thermal Spraying Conf., France, 1998, p. 145. [7] S. Liscano, L. Gil, M. Staia, Surf. Coat. Technol. 188 – 189 (2004) 135. [8] Standard Test Method for Measuring Abrasion using the Dry Sand/Rubber Wheel Apparatus, G 65-00-1, ASTM, United States, 2001. [9] H. Kim, C. Lee, Y. Kweon, Surf. Coat. Technol. 139 (2001) 75. [10] J. Knuutilla, P. Sorsa, T. Ma¨ntyla¨, J. Therm. Spray Technol. 8 (1999) 249. [11] E.M. Leivo, M.S. Vippola, P.P.A. Sorsa, P.M.J. Vuoristo, T.A. Ma¨ntyla¨, J. Therm. Spray Technol. 6 (1997) 205. [12] S.J. Bull, R. Kingswell, K.T. Scott, Surf. Coat. Technol. 82 (1996) 218. [13] D. Chuanxian, H. Bingtang, L. Huiling, Thin Solid Films 118 (1984) 485.