These process conditions had influence on the color (thickness) of the film ... on the heat treatment conditions, however the transformation always follows the ...
Supplemental Proceedings: Volume 2: Materials Characterization, Computation, Modeling and Energy TMS (The Minerals, Metals & Materials Society), 2010
WEAR RESISTANCE AND ADHERENCE OF TiO2 SOL-GEL THIN FILMS M. A. Alterach1, 2, P. C. Favilla1, 3, M. R. Rosenberger1, 2, A. E. Ares1, 2, C. E. Schvezov1, 2, 3 1
2
Consejo Nacional de Investigaciones Científicas y Técnicas – CONICET Rivadavia 1917 (1033). Buenos Aires, Argentina. Facultad de Ciencias Exactas Químicas y Naturales, Universidad Nacional de Misiones Félix de Azara 1552 (3300). Posadas, Misiones, Argentina. 3 CEDITec - CEDIT, Félix de Azara 1890, (3300) Posadas, Misiones, Argentina. Keywords: wear, adherence, films, titanium oxide Abstract
The wear resistance and adherence of TiO2 films was studied; the films were synthesized by the sol-gel dip-coating technique on a grade 5 titanium alloy. Monolayer and multilayer films were deposited by varying: dip-coating velocity, aging time of sol and the heat treatment parameters. These process conditions had influence on the color (thickness) of the film and the cracks formation. The wear resistance was measured on a ball-on-flat machine using a rotating glass ball as counterface. The worn films were evaluated by the size of the superficial scare, which can be noted by the change of color on the surface. The adherence was measured on a scratch test machine and examined by optical microscope to determine the adherent critical load. The best wear resistance and adherence of the films was measured on a trilayer film fabricated with a heat treatment at 500ºC for 1h for each layer. Introduction At present the main material used in prosthetic cardiac valves is pyrolitic carbon which has good haemocompatibility. The alternative materials require surface modifications to improve haemocompatibility usually procured by deposition of different films on the substrate material, such as diamond like carbon (DLC) [1], silicon carbide (SiC) [2], titanium nitride (TiN) [3], titanium oxide (TiO2) [4, 5], aluminium oxide (Al2O3) [6] and low temperature pyrolitic carbon (LTIC) [7, 8]. Zhang et al. [7] obtained films of titanium oxide deposited by ion implantation and showed that the performance is better than that of LTIC, with longer clotting time and less plaquete adhesion. Velten et al. [9] synthesized films of titanium oxides on titanium CP and Ti-6Al–4V alloy using thermal oxidation, anodic oxidation and by the sol-gel technique and compared the corrosion resistance of the films deposited by the differents techniques. The coatings obtained by anodic and thermal oxidation showed a relation between color and thickness of the film. In addition films of 100nm and thicker were a protective barrier against corrosion independently of the synthesis technique. The sol-gel process produces an amorphous TiO2 coating, so a heat treatment is performed to obtain crystalline TiO2 with anatase, rutile or both structures. The amount of each phase depend on the heat treatment conditions, however the transformation always follows the route amorphous – anatase – rutile, and at higher temperatures accelerate the transitions between structures [9].
609
Zhang et al. [10] synthezed films of TiO2, SiO2 and hydroxiapathite by the sol-gel technique on grade 5 titanium alloy in order to study the tribological behavior of the different films. They found that the titanium oxide films experienced fragile fracture for loads between 1 and 3 N. For loads below 1 N there is no wear of the films whereas at 3N there is extensive wear with fragile fracture of the film. Nevertheless the resistance was very low which was attributed to high rugosity of the surface of the films. The objective of the present research was to characterize the adherence and wear resistance of films of titanium oxide deposited on titanium alloy by the sol-gel technique. Experimental Procedure Samples of grade 5 titanium were prepared with a mirror surface obtained with a silice suspension 0.03µm with 10% in volume of H2O2. The deposition of the films were produced employing a sol-gel dip-coating technique. The sol is prepared from titanium tetrabutoxide, isopropylic alcohol, water, chloridric acid and ethyl acetoacetate [11]. The films were obtained at room temperature of 29 ± 5ºC and relative humidity of 60 ± 10%. The samples were withdrawn from the sol at 1, 2 and 3cm/min. After this the samples were dried in air at room temperature during 1h and then heat treated at 500ºC during 1h. The temperature increased from room temperature at 10ºC/min and cooled at a rate lower than 10ºC/min. The heat treatment for the monolayer coated samples consisted in a heating period at 10ºC/min or 3ºC/min up to 500ºC or 600ºC, maintained for 1h and then cooled to room temperature at a rate lower than 10ºC/min. The multilayered films were obtained in three ways; first the three layers were deposited, dried and then heat treated for 60min at 500ºC (heating rate of 10ºC/min); second, after each layer deposition a heat treatment was applied during 30min at 200ºC (heating rate of 5ºC/min) and at the end the sample was heat treated for 60min at 500ºC (heating rate of 10ºC/min); and third, after each layer deposition, the sample was heat treated at 500ºC during 60min (heating rate of 10ºC/min). The experimental sequence employed is shown schematically in Figure 1.
Sol
Mirror Finishing of Titanium
bilayer , trilayer
Dip-coating
Air Drying
Heat Treatment
Mechanical Tests
Figure 1. Block diagram of the experimental procedure. The adherence of the films was determined by scratch test method, using a CSEM Revetest equipment with a Rockwell C diamond indenter 0.2mm in radius. The tests were carried out at a
610
maximum load of 20N which was increased from 0N at a speed of 4N/mm. Two tests were performed on the center region of each sample. A total of 11 samples were tested. The wear tests were performed in a ball-on-flat equipment build for this purpose. The ball rotates on the sample surface with the rotating axis forming an specified angle with the sample surface. The test conditions were: a ball velocity of 7rpm, the applied load of 1N, the test time of 5min, lubricated with ethylenglicol, and a ball made of borosilicate glass of 6mm in diameter. The resulting scars were observed under the optical microscope and categorized in four categories; very worn films (VW), worn films (W), hardly worn films (HW) and non-worn films (NW), see Figure 2. Contact Zone
Contact Zone Wear Scar
NW
Contact Zone
Wear Scar
Wear Scar
W
HW
VW
Figure 2. Categories of the wear scars. The wear scale is based on observations and the need to define an appropriate scale associated to the degree of wear of the film; very worn (VW) defines the case of a film completely lost in the area of contact that is, the scar has the same area as the contact area; on the other hand non-worn (NW) means that non of the film has been lost despite the fact that the film itself could have experienced some wear. Results and Discussion A number of eleven samples were prepared, nine with a withdrawal velocity of 2 cm/min, one at 1 cm/min and one at 3 cm/min, 5 samples were coated with a monolayer (see Table I), 3 bilayer and 3 trilayer of titanium oxide (see Table II). Scratch Test One sample (number 3) with a trilayer film and a withdrawal velocity of 2 cm/min was tested at a load rate of 2N/mm in order to adjust the optimal test conditions, particularly the load to failure of the film. Two scratches were performed increasing the load from 0 to 10 N. The results for both scratches are shown in Figure 3. The load increases from left to right and the first observation is an increase of the plastic deformation also some cracks appear in the coating along the scratch (cohesive failure), and the number of cracks increases as the load increases but there is no spalling of the film.
611
1 mm
1 mm
Figure 3. Twins scratches on sample 3 without spalling of the film. Load conditions: 0 to 10 N at rate of 2N/mm. As a result two other tests under more severe conditions were performed; in both, the final load was increased to 20N and in one of them the load rate was also increased from 2 to 4N/mm. The results of the tests showed that the load rate was important, since at 2N/mm there was no failure of the coatings as occurred in the test at 4N/mm. Therefore a load of 20N and a load rate of 4N/ mm were selected for the scratch tests. The results of each test were generally classified in three types of sections depending on the effect which is observed (see Figure 4-a)). The first correspond to a section in which both the coating and the substrate are deformed with the presence of some cracks (No Failure), the second where the coating presents spalling in part of the width of the scratch (Partial Failure) and the third where there is complete spalling (Total Failure) across the scratch. In order to compare the different performance and quality of the coating, the load at the beginning of the Partial Failure section is defined as the critical load (Section D in Figure 4). No Failure
A a)
B
Total Failure
Parcial Failure
C
D
E
F
Spalling A
B
D
C 3
E 3
F
b) Cracks
Coating
Substrate
Figure 4. Optical Micrographs. a) Full scratch length on sample 2 and b) the three categories in the scratch test results.
612
Sample 7 8 9 10 11
Table I. Critical loads for the monolayer films. Critical load (N) Withdrawal rate Color Scratch Scratch (cm/min) 1 2 2 Dark Blue 6 2 Dark Blue 0 0 3 Light Blue 0 1 Purple >20 >20 2 Dark Blue 9 0
Heat Treatament D*-HT600** D-HT500** D-HT500 D-HT500 D-HT500a***
*D: deposition and drying, **HT500 and HT600: heat treatment at 500ºC or 600ºC during 1h and a heating rate of 10ºC/min , *** HT500a: heat treatment at 500ºC during 1h and a heating rate of 3ºC/min.
The results of critical loads for 5 samples with a monolayer film are shown in Table I. It is observed that the highest load, and therefore best adhesion, corresponds to sample number 10 which was coated at a withdrawal rate of 1 cm/min. At this low speed and considering that the drainage time is the longest, the resultant film may be the thinniest. A thin film could be more able to withstand the load by following the deformation of the substrate [12]. The second possibility is that the better adherence of the film could be due to a graded interface between film and substrate [13]. The critical loads obtained for bi- and trilayers films are shown in Table II. It is observed that the highest load corresponds to sample 5 which shows the best adherence with a trilayer coating which is as good as the performance of sample 10 with a monolayer coating. Table II. Critical loads of the bi and trilayer samples. Withdrawal Critical Load (N) Nº of Sample rate Color Heat Treatment Scratch Scratch layers (cm/min) 1 2 1 2 2 Light Blue 0 0 D*-HT200**, D-TT500 D-HT200, D-HT200, D2 3 2 Yellow 11 8 HT500*** 3 3 2 Yellow 15 D, D, D-HT500 4 2 2 Light Blue 0 0 D-HT500, D-HT500 D-HT500, D-HT500, D5 3 2 Yellow >20 >20 HT500 6 2 2 Light Blue 0 D, D-HT500 *D: deposition and drying, **HT200: heat treatment at 200ºC during 30min and a rate of 5ºC/min, *** and HT500: heat treatment at 500ºC during 1h and a heating rate of 10ºC/min.
In general the samples with a trilayer coating present better adherence than the bilayer coatings. Among the bilayer coatings there was no appreciable differences in adherence. Samples 1, 4 and 6 have the poorest adherence with scratches which show complete failure right from the beginning of the test. On the other hand, the best behavior corresponds in decreasing degree of adherence to samples 5, 2 and 3. The good adherence of the trilayer coatings may not be attributed to the thickness of the coatings since all of them have similar thickness which is concluded by the same yellow colour of the films. The better behaviour could be attributed to the heat treatment. The best adherence obtained for sample 5 in Table II corresponds to a heat treatment of 500ºC for 60min after each layer deposition which may have produced better graded interfaces between layers [13].
613
Wear Tests The results of the wear tests show completely different behaviour between the monolayer and the multilayer coatings. The monolayer films at the end of the test were completely worn (Table III), given by a scar of the size of the contact area. The conclusion for these films is that they were not able to withstand the wear which under the test condition could be defined as severe. Table III. Wear test results of the monolayer samples. Test time: 5min. Load: 1N Withdrawal Heat Sample rate Treatment Color Scar*** (cm/min) 7 2 D*-HT600** Dark Blue VW 8 2 D-HT500** Dark Blue VW 9 3 D-HT500 Light Blue VW 10 1 D-HT500 Purple VW *D: deposition and drying, **HT500 and HT600: heat treatment at 500ºC or 600ºC during 1h and a heating rate of 10ºC/min , *** according to Figure 2.
On the other hand, the wear resistances of the multilayer coatings were much better than for the monolayer films as shown in Table IV. The highest wear resistances correspond to the coatings which were heat treated at 500ºC at the end of the two or three depositions (samples 3 and 6) and those with intermediate heat treatment at 500ºC after each deposition (samples 4 and 5). The lowest wear resistances correspond to samples 1 and 2 which have bi- and trilayer coatings respectively and intermediate heat treatments after each deposition at 200ºC. The better wear resistance may not be attributed to the coating thickness since the bilayer coatings on one hand and the trilayer coatings on the other hand have similar thickness given by the same color of the films. Table IV. Wear test results of the bi and trilayer samples. Test time: 5min Load: 1N Withdrawal Nº of Sample rate Heat Treatment Color Layers (cm/min) 1 2 2 D*-HT200**-D-HT500*** Light Blue 2 3 2 D-HT200- D-HT200- D-HT500 Yellow 3 3 2 D-D-D-HT500 Yellow 4 2 2 D-HT500 - D-HT500 Light Blue 5 3 2 D-HT500-D-HT500-D-HT500 Yellow 6 2 2 D-D-HT500 Light Blue
Scar**** VW HW NW NW NW NW
*D: deposition and drying, **HT200 heat treatment at 200ºC during 30min and a heating rate of 5ºC/min. *** HT500 heat treatament at 500ºC during 1h and a heating rate 10ºC/min,****according to Figure 2.
The improvements are attributed to the effects of the heat treatments in particular those which include high temperatures (500ºC) during final and intermediate heat treatments. The heat treatment at 200ºC did not produce resistant coating as the other ones which could be the result of an inadequate evaporation and sintering process due to the low temperature which may not be recovered by the final heat treatment at 500ºC.
614
Conclusions Mono-, bi- and trilayers coatings of TiO2 were produced by the sol-gel dip-coating technique on grade 5 titanium alloy used as substrate. The coated samples were heat treated following different procedures and temperatures. The coatings were tested for adherence using the scratch test and for wear resistance using a ball-on-flat procedure. The results show that the best adherence and wear resistance is obtained for the trilayer coating heat treated at 500ºC with a heating rate of 10ºC/min, as a final or intermediate procedure. This behaviour may be attributed to a better sintering process. In the case of monolayer coatings the same heat treatment procedure the best adherence and wear resistance. Acknowledgments The authors wish to thank Elena Forlerer and Fernando Rodriguez from Laboratorio de Plasma del Centro Atómico Constituyentes, Comisión Nacional de Energía Atómica, Argentine, for the scratch test measurements. The authors acknowledge the financial support of the ANPCyT, CEDIT and CONICET of Argentine. References 1. J. H. Sui, W. Cai, “Effect of diamond-like carbon (DLC) on the properties of the NiTi alloys”, Diamond & Related Materials, 15 (2006), 1720–1726. 2. P. Baurschmidt, M. Schaldach, “Alloplastic materials for heart-valve prostheses”, Med. Biol. Eng. Compu, 18 (1980), 496-502. 3. I. Dion, F. Rouais, L. Trut, C. Baquey, J. Monties, P. Havlik, “TiN coating: surface characterization and hemocompatibility”. Biomaterials, 14 (1993), 169–76. 4. J. Liu, D. Yang, F. Shi and Y. Cai, “Sol-Gel deposited TiO2 film on NiTi surgical alloy for biocompatibility improvement”; Thin Solid Films, 429 (2003), 225-230. 5. E. Eisenbarth, D. Velten, K. Schenk-Meuser, P. Linez, V. Biehl, H. Duschner, J. Breme and H. Hildebrand. “Interactions between cells and titanium surfaces”; Biomolecular Engineering, 19 (2002), 243-249. 6. T. Yuhta, Y. Kikuta, Y. Mitamura, K. Nakagane, S. Murabayashi, I. Nishimura, “Blood compatibility of sputter-deposited alumina films”. Journal Biomedical Mater. Research, 28 (1994), 217–24. 7. F. Zhang, X. Liu, Y. Mao, N. Huang a, Y. Chen, Z. Zheng, Z. Zhou a, A. Chen, Z. Jiang; “Artificial heart valves: improved hemocompatibility by titanium oxide coatings prepared by ion beam assisted deposition”, Surface and Coatings Technology, 103–104 (1998), 146–150. 8. N. Huang, P. Yang, “hemocompatibility of titanium oxide films”, Biomaterials, 23 (2003), 21772187. 9. D. Velten, V. Biehl, F. Aubertin, B. Valeske, W. Possart, J. Breme; “Preparation of TiO2 layers on cp-Ti and Ti-6Al-4V by thermal and anodic oxidation and by sol-gel coating techniques and their characterization”, Journal of Biomedical Materials, 59 (2001), 18-28.
615
10. W. Zhang, C. Wang, W. Liu, “Characterization and tribological investigation of sol–gel ceramic films on Ti–6Al–4V”, Wear 260 (2006), 379–386. 11. P. Favilla, M. Alterach, M Rosenberger, A. Ares, C. Schvezov, O. Amerio, “Obtención y caracterización de recubrimientos de oxido de titanio via deposición sol-gel”, (Paper presented at the SAM-CONAMET 2007 Congress, San Nicolás, Buenos Aires, Argentine, 4 September 2007), 6. 12. Kenneth Holmberg, Allan Matthews, Coatings Tribology. Properties, Techniques and Applications in Surface Engineering. (Amsterdam, The Netherlands, Elsevier Science, 1998). 13. S. Suresh, “Graded Materials for Resistance to Contact Deformation and Damage”, Science`s Compass, 292 (2001), 2447 – 2451.
616
