Availableonlineatwww.sciencedirect.com °" Science Direct applied ELSEVIER Applied Surface Science 253(2006)1222-1226 Stress driven phase transformation in Zro2 film B. Benali, M. Herbst Ghysel, I. Gallet, A M. Huntz, M. Andrieux Laboratoire dEnude des Materiaux Hors Equilibre(LEMHE) CNRS UMR 8647, Universite de Paris-Sud, Batiment 410, 91405 Orsay, france Received 19 December 2005: received in revised form 30 January 2006: accepted 30 January 2006 Available online 15 March 200 Abstract Zirconia thin layers (250 nm) were deposited on stainless steel substrates using organo-metallic injection chemical vapour deposition (MOCVD) process with zirconium beta-diketonate as precursor at low oxygen pressure and 900C. Low roughness zirconia films were made up of a mixture of tetragonal and monoclinic phases depending on the process conditions. As the zirconia tetragonal phase is known to be stabilized by small grain size and/or internal compressive stresses, tensile and/or compressive external stress fields were applied at room temperature using bending test device. Then, XRD measurements were used to determine tetragonal/monoclinic phase ratio and also residual stresses in the films before and after the tests. The film surface was observed at the various stages of the experiments by field electron gun-scanning electron microscopy (FEG-SEM Under these stress fields, phase transformation occurs in the film, from tetragonal structure to a monoclinic one. Some preferential tetragonal planes give rise to monoclinic ones. The external stress field is also likely to redistribute the internal stresses within the films. C 2006 Elsevier B V. All rights reserved. Keywords: Zirconia thin films; Tetragonal and monoclinic phases; Phase transformation; Bending test; Mechanical properties 1. Introduction effect of stresses on structure transformations In this context recent studies related to zirconia showed that the application of According to its structure, composition and development very high compressive stresses(GPa)may modify the stability conditions, zirconia has properties which enable its use for of the phases [5] various applications. Zirconia powder is used, for instance, Zirconia thin layers deposited by MOCVD present, in most like catalyst or catalyst support in processes such as the cases, a mixture of both monoclinic and tetragonal phases. production of hydrocarbons and/or alcohols from CO and According to the literature, whether it is massive zirconia or H2[1, 2]. Zirconia is also used as element in the fuel cells [3 thin film, the tetragonal phase is metastable atroom like in the matrix of composite ceramics resistant to high temperature. In addition, the stabilization of the metastable tetragonal phase would be managed primarily by the crystallites As thin films, zirconia has important industrial applications size [1, 6, 7] and a critical size for the phase transformation is uclear. ae erospace, electronic,.. )in particular for coating in mentioned, depending on the temperature. the turbine blades due to its good properties as thermal barrier Besides, preliminary determinations by X-ray diffraction and refractory. In service, in addition to the severe thermal (XRD) of the internal stresses(sin y method [8)within environment aggression, these layers of a few microns zirconia layers deposited by MOCVD on a stainless steel, thickness undergo important external mechanical stresses. revealed compressive stresses in the zirconia film with stress The above mentioned applications under severe conditions levels significantly different between the monoclinic and require works and reflexion on the correlations between their tetragonal phases, the tetragonal phase being more constrained. properties and their structure and microstructure, thus on the It is possible that these internal compressive stresses have an effect on the tetragonal phase stabilization and/or transforma- tion in our films Corresponding author. Tel. +33 1 69 15 48 09/69 15 70 20: With the aim of understanding and controlling the mecha fax:+33169154819 nism of phase transformation and stabilization within zirconia films, a tension stress field (stress opposed to that initially 69-4332/S-see front matter C 2006 Elsevier B V. All rights reserved. doi:10.1016 j.apsusc200601060
Stress driven phase transformation in ZrO2 film B. Benali, M. Herbst Ghysel, I. Gallet, A.M. Huntz, M. Andrieux * Laboratoire d’Etude des Mate´riaux Hors Equilibre (LEMHE), CNRS UMR 8647, Universite´ de Paris-Sud, Baˆtiment 410, 91405 Orsay, France Received 19 December 2005; received in revised form 30 January 2006; accepted 30 January 2006 Available online 15 March 2006 Abstract Zirconia thin layers (250 nm) were deposited on stainless steel substrates using organo-metallic injection chemical vapour deposition (MOCVD) process with zirconium beta-diketonate as precursor at low oxygen pressure and 900 8C. Low roughness zirconia films were made up of a mixture of tetragonal and monoclinic phases depending on the process conditions. As the zirconia tetragonal phase is known to be stabilized by small grain size and/or internal compressive stresses, tensile and/or compressive external stress fields were applied at room temperature using a bending test device. Then, XRD measurements were used to determine tetragonal/monoclinic phase ratio and also residual stresses in the films before and after the tests. The film surface was observed at the various stages of the experiments by field electron gun–scanning electron microscopy (FEG-SEM). Under these stress fields, phase transformation occurs in the film, from tetragonal structure to a monoclinic one. Some preferential tetragonal planes give rise to monoclinic ones. The external stress field is also likely to redistribute the internal stresses within the films. # 2006 Elsevier B.V. All rights reserved. Keywords: Zirconia thin films; Tetragonal and monoclinic phases; Phase transformation; Bending test; Mechanical properties 1. Introduction According to its structure, composition and development conditions, zirconia has properties which enable its use for various applications. Zirconia powder is used, for instance, like catalyst or catalyst support in processes such as the production of hydrocarbons and/or alcohols from CO and H2 [1,2]. Zirconia is also used as element in the fuel cells [3] like in the matrix of composite ceramics resistant to high temperatures [4]. As thin films, zirconia has important industrial applications (nuclear, aerospace, electronic, ...) in particular for coating in the turbine blades due to its good properties as thermal barrier and refractory. In service, in addition to the severe thermal environment aggression, these layers of a few microns thickness undergo important external mechanical stresses. The above mentioned applications under severe conditions require works and reflexion on the correlations between their properties and their structure and microstructure, thus on the effect of stresses on structure transformations. In this context, recent studies related to zirconia showed that the application of very high compressive stresses (GPa) may modify the stability of the phases [5]. Zirconia thin layers deposited by MOCVD present, in most cases, a mixture of both monoclinic and tetragonal phases. According to the literature, whether it is massive zirconia or thin film, the tetragonal phase is metastable at room temperature. In addition, the stabilization of the metastable tetragonal phase would be managed primarily by the crystallites size [1,6,7] and a critical size for the phase transformation is mentioned, depending on the temperature. Besides, preliminary determinations by X-ray diffraction (XRD) of the internal stresses (sin2 c method [8]) within zirconia layers deposited by MOCVD on a stainless steel, revealed compressive stresses in the zirconia film with stress levels significantly different between the monoclinic and tetragonal phases, the tetragonal phase being more constrained. It is possible that these internal compressive stresses have an effect on the tetragonal phase stabilization and/or transformation in our films. With the aim of understanding and controlling the mechanism of phase transformation and stabilization within zirconia films, a tension stress field (stress opposed to that initially www.elsevier.com/locate/apsusc Applied Surface Science 253 (2006) 1222–1226 * Corresponding author. Tel.: +33 1 69 15 48 09/69 15 70 20; fax: +33 1 69 15 48 19. E-mail address: michel.andrieux@lemhe.u-psud.fr (M. Andrieux). 0169-4332/$ – see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.apsusc.2006.01.060
B. Benali et aL./Applied Surface Science 253(2006)1222-1226 223 present in material) was applied at room temperature to our stress decreased steadily from 0 to -250 MPa during 60 S. zirconia films deposited by MOCVd on stainless steel in order Then the sample is rapidly released(during 20 s) to modify the intemal stresses level and to possibly modify the phase equilibrium It is important to remark that when a film is subjected to tensile stress along the large (longitudinal) direction, it 2. Experimental procedures undergoes compressive stresses in the transverse direction at The samples used in this study were obtained by MOCVD The X-ray diffraction patterns were obtained with Cu Ko deposition of a zirconia film at 900C on a stainless steel. radiation(a=0. 15406 nm) in an Xpert PHILIPS PW 1840 These specimens are made up of diffractometer. The patterns were obtained by step scanning from27°to37°in26, with an increment of0.025°anda a stainless steel substrate AlSI 304 mirror finished (austenitic counting time of 100 s per step. They were performed ex situ, FeNig Crig) on the samples before and after bending a chromium oxide(Cr2O3) layer of a100 nm naturally The diffraction peaks were identified according to JCPDS formed during the heating of the substrate in the reactor, files no. 37-1484 for the monoclinic ZrO, and 17-923 for the oating whose thickness is approximately 250 al zro deposited by MOCVD at 900C, Ptot=100 Pa and a low A low incidence angle (2) was used so that the X-ray beam oxygen pressure (5% O2, gsp N2) did not reach the substrate nor the chromium oxide(Cr2O3) These last conditions are suggested by [6], where it was intermediate layer. In our experiences we used 15 y between shown that a low oxygen pressure at high temperature promotes 60°and+60° conditions were chosen here in order to have sufficient 3. Results and discussion tetragonal phase for the study The samples dimensions are (length x width x thickness) The X-ray diffraction patterns and the ratio of tetragonal =(25mm×4mm×0.5mm) phase of these samples are gathered in Fig. 2. The samples were tested with a four-points bending device The formula used to calculate the phase amount is: %(0)=1 which is described in an other work [9]. A load cell in the (11 1)/((1 11)+1(11 1)m+/(1 1 I)m)[10]. In fact, this center of the sample provides the force applied at the central formula is available for powder and essentially non-textured loading points whereas displacement is provided with the material which is often assumed in the literature. To be sure upper loading points(Fig. 1). The applied stress level is of this point, the evolution of the intensities of these three calculated assuming the strain continuity at the film-substrate interface and considering elastic the film behaviour 91 %@(1=81% Bending experiments were conducted at room temperature according to three different conditions Sample 1: subjected to tension stress(film above the sub- strate, Fig. 1). The stress climbed up steadily from 0 MPa to a value just above the substrate yield strength(205 MPa) I tensile evele %q(t)=59% during 130 s. Then the sample is rapidly released(during 20 s) Sample 2: subjected to 16 tensile cycles without exceeding the substrate yield stress. For the first cycle, the stress in the film climbed up from 0 to 230 MPa then it decreased to 30 MPa. This cycle is repeated 16 times between 30 and Sample 2 230 MPa. Each cycle has a duration of 110 s. %p(t)=47% Sample 3: tested under compression(film under the substrate, Fig. 1)without exceeding the substrate yield strength. The 1000116 tensile cycles %q()=67% 17% film under tension 八 Fig. 1. Scheme of the four-point bending tests Fig. 2. X-ray diffraction patterns of the four bending samples
present in material) was applied at room temperature to our zirconia films deposited by MOCVD on stainless steel in order to modify the internal stresses level and to possibly modify the phase equilibrium. 2. Experimental procedures The samples used in this study were obtained by MOCVD deposition of a zirconia film at 900 8C on a stainless steel. These specimens are made up of: a stainless steel substrate AISI 304 mirror finished (austenitic FeNi9Cr19), a chromium oxide (Cr2O3) layer of 100 nm naturally formed during the heating of the substrate in the reactor, a ZrO2 coating whose thickness is approximately 250 nm, deposited by MOCVD at 900 8C, Ptot = 100 Pa and a low oxygen pressure (5% O2, qsp. N2). These last conditions are suggested by [6], where it was shown that a low oxygen pressure at high temperature promotes the formation of the tetragonal phase in zirconia films. These conditions were chosen here in order to have sufficient tetragonal phase for the study. The samples dimensions are (length width thickness) = (25 mm 4 mm 0.5 mm). The samples were tested with a four-points bending device which is described in an other work [9]. A load cell in the center of the sample provides the force applied at the central loading points whereas displacement is provided with the upper loading points (Fig. 1). The applied stress level is calculated assuming the strain continuity at the film–substrate interface and considering elastic the film behaviour [9]. Bending experiments were conducted at room temperature according to three different conditions: Sample 1: subjected to tension stress (film above the substrate, Fig. 1). The stress climbed up steadily from 0 MPa to a value just above the substrate yield strength (205 MPa) during 130 s. Then the sample is rapidly released (during 20 s). Sample 2: subjected to 16 tensile cycles without exceeding the substrate yield stress. For the first cycle, the stress in the film climbed up from 0 to 230 MPa then it decreased to 30 MPa. This cycle is repeated 16 times between 30 and 230 MPa. Each cycle has a duration of 110 s. Sample 3: tested under compression (film under the substrate, Fig. 1) without exceeding the substrate yield strength. The stress decreased steadily from 0 to 250 MPa during 60 s. Then the sample is rapidly released (during 20 s). It is important to remark that when a film is subjected to tensile stress along the large (longitudinal) direction, it undergoes compressive stresses in the transverse direction at a level depending on the Poisson’s coefficient. The X-ray diffraction patterns were obtained with Cu Ka radiation (l = 0.15406 nm) in an Xpert PHILIPS PW 1840 diffractometer. The patterns were obtained by step scanning from 278 to 378 in 2u, with an increment of 0.0258 and a counting time of 100 s per step. They were performed ex situ, on the samples before and after bending. The diffraction peaks were identified according to JCPDS files no. 37-1484 for the monoclinic ZrO2 and 17-923 for the tetragonal ZrO2. A low incidence angle (28) was used so that the X-ray beam did not reach the substrate nor the chromium oxide (Cr2O3) intermediate layer. In our experiences we used 15 c between 608 and +608. 3. Results and discussion The X-ray diffraction patterns and the ratio of tetragonal phase of these samples are gathered in Fig. 2. The formula used to calculate the phase amount is: % (t) = I (1 1 1)t/(I (1 1 1)t + I (1 1 1)m + I (1 1 1)m) [10]. In fact, this formula is available for powder and essentially non-textured material which is often assumed in the literature. To be sure of this point, the evolution of the intensities of these three B. Benali et al. / Applied Surface Science 253 (2006) 1222–1226 1223 Fig. 1. Scheme of the four-point bending tests. Fig. 2. X-ray diffraction patterns of the four bending samples.
B. Benali et al. /Applied Surface Science 253(2006)1222-1226 (-111m (111t (111m ◆(111m/(11t▲lm/l(111 l(11lm/-111 0.8 6000 4000 3000 8%4 1000 -80-60-4 Fig. 3. Evolution of the three peaks intensities vs. y angles used for the Of tension calculation of the phase ratio and the internal stresses. Sample state Fig 4. Evolution of the intensities ratio of the significant peaks according to the peaks versus y, the tilt angle of the sample surface as defined sample state by the sin y method [8, is given in Fig 3. It can be seen that the intensities of the peaks do not evolve on a wide range However, by looking at the relative evolution of the plane so that it can be concluded that the film is not significantly intensities corresponding to the six peaks located between 33 textured along the diffraction directions and 36(more sensitive to texture)in the X-ray diffraction From the results in Fig. 2, it appears that the film structure patterns(compared to that of the (111))(Fig. 5), it can be is modified according to the applied external stresses and to seen that all the slopes are close to the one obtained for the the number of cycles, the tetragonal phase disappearing (111)m orientation with an evolution al ways less significant progressively, even for the sample subjected to a compressive than the one obtained for the (-l 1 1)m, except for the(20 O)m field along its length. This last result is not so surprising since, plane which presents an intermediate slope. Would this when applying a bending to the sample with compressive tendency indicate that the t-m transformation is more stresses on the film along its length, transverse tensile stresses complex than what it was envisaged above and would it be of connected to Poisson coefficient appear. Thus, probably the the type; (111)-(-1 1 I)m +(200)m? Further experiments tensile stresses could be responsible for the transformation will be conducted in this way of phases occurring during this experiment. The reduction in Concerning internal stresses, the level of stresses remains the tetragonal phase ratio (nearly 17%)after compression, almost unchanged after 16 tensile cycles( see Table 1). Indeed lower than the reduction obtained after tension (27%0), con- after the test, the untransformed tetragonal phase keeps the firms these interpretations same stress level as before. The transformed tetragonal phase Concerning the diffraction directions, let us note that the into monoclinic phase is associated to a positive volume (11 1), peak is systematically concemed by an intensity variation. As a co the new monoclinic phase reduction. It seems to be transformed into monoclinic phase by submitted to compressive stresses. The thermal stress has been ng birth preferentially to the(-l ll)m peak. estimated as N-l GPa taking into account the differences To check this assumption, on the graph of Fig 4, the relative between the thermal expansion coefficients of the film and its tensity of the planes was plotted for the various conditions substrate and considering the monoclinic to tetragonal phase pplied to the samples. It can be seen that the relative ratio in the film. Thus, the residual stress is much lower than the intensities of the two monoclinic planes(-1 11)m and(11 1)m ncrease with the applied stress and the number of cycles x(200)m/(1 However, as shown by the slope of these curves, the intensity 002)mo 11200/(11 0(20m/U11M of the(-l 11)m plane increases in a more significant way than that of the(111)m plane. All these results are representative of a phase transformation()-(m) where the type(111), planes would give birth mainly to planes of the type(-1 11)m 111)1→(-111) Chraska [ll], already affirms that the (111) plane gives 二 preferentially rise to the two planes(-l l I)m and (11 l)m This is due to the martensitic character of the t-m ransformation in relation to the orientation relationship which exists between crystalline directions: for temperature lower than 1000C, it was shown that (100)mll(100),and pression [001 mlI[o 0 1]r. They add that the (11 1)m plane is less Sample state stable than the(-1 II)m plane because of its higher surface Fig. 5. Evolution of the intensities ratio of all peaks according to the sample energy. All their results agree ours
peaks versus c, the tilt angle of the sample surface as defined by the sin2 c method [8], is given in Fig. 3. It can be seen that the intensities of the peaks do not evolve on a wide range so that it can be concluded that the film is not significantly textured along the diffraction directions. From the results in Fig. 2, it appears that the film structure is modified according to the applied external stresses and to the number of cycles, the tetragonal phase disappearing progressively, even for the sample subjected to a compressive field along its length. This last result is not so surprising since, when applying a bending to the sample with compressive stresses on the film along its length, transverse tensile stresses connected to Poisson coefficient appear. Thus, probably the tensile stresses could be responsible for the transformation of phases occurring during this experiment. The reduction in the tetragonal phase ratio (nearly 17%) after compression, lower than the reduction obtained after tension (27%), con- firms these interpretations. Concerning the diffraction directions, let us note that the (1 1 1)t peak is systematically concerned by an intensity reduction. It seems to be transformed into monoclinic phase by giving birth preferentially to the (1 1 1)m peak. To check this assumption, on the graph of Fig. 4, the relative intensity of the planes was plotted for the various conditions applied to the samples. It can be seen that the relative intensities of the two monoclinic planes (1 1 1)m and (1 1 1)m increase with the applied stress and the number of cycles. However, as shown by the slope of these curves, the intensity of the (1 1 1)m plane increases in a more significant way than that of the (1 1 1)m plane. All these results are representative of a phase transformation (t) ! (m) where the type (1 1 1)t planes would give birth mainly to planes of the type (1 1 1)m: (1 1 1)t ! (1 1 1)m. Chraska [11], already affirms that the (1 1 1)t plane gives preferentially rise to the two planes (1 1 1)m and (1 1 1)m. This is due to the martensitic character of the t ! m transformation in relation to the orientation relationship which exists between crystalline directions: for temperature lower than 1000 8C, it was shown that (1 0 0)mjj(1 0 0)t and [0 0 1]mjj[0 0 1]t. They add that the (1 1 1)m plane is less stable than the (1 1 1)m plane because of its higher surface energy. All their results agree ours. However, by looking at the relative evolution of the plane intensities corresponding to the six peaks located between 338 and 368 (more sensitive to texture) in the X-ray diffraction patterns (compared to that of the (1 1 1)t) (Fig. 5), it can be seen that all the slopes are close to the one obtained for the (1 1 1)m orientation with an evolution always less significant than the one obtained for the (1 1 1)m, except for the (2 0 0)m plane which presents an intermediate slope. Would this tendency indicate that the t ! m transformation is more complex than what it was envisaged above and would it be of the type; (1 1 1)t ! (1 1 1)m + (200)m? Further experiments will be conducted in this way. Concerning internal stresses, the level of stresses remains almost unchanged after 16 tensile cycles (see Table 1). Indeed, after the test, the untransformed tetragonal phase keeps the same stress level as before. The transformed tetragonal phase into monoclinic phase is associated to a positive volume variation. As a consequence, the new monoclinic phase is submitted to compressive stresses. The thermal stress has been estimated as 1 GPa taking into account the differences between the thermal expansion coefficients of the film and its substrate and considering the monoclinic to tetragonal phase ratio in the film. Thus, the residual stress is much lower than the 1224 B. Benali et al. / Applied Surface Science 253 (2006) 1222–1226 Fig. 3. Evolution of the three peaks intensities vs. c angles used for the calculation of the phase ratio and the internal stresses. Fig. 4. Evolution of the intensities ratio of the significant peaks according to the sample state. Fig. 5. Evolution of the intensities ratio of all peaks according to the sample state.
B. Benali et aL./Applied Surface Science 253(2006)1222-1226 225 It is not impossible that such a correlation takes place. Internal stresses in films of the as-deposited sample and sample 2(after 16 Indeed. the level of internal stresses, which is different cycles of tension) between the two monoclinic and tetragonal (more con- Phases a(MPa) strained)phases, could be responsible for this possible As-deposited After 16 cycles correlation. In addition, the experiment of Bouvier et al. [51 (-111)m seems to agree this point since the strong compression -151±53 -236±60 applied to the samples could be responsible for the shape (111) 326±50 -336+52 and/or the size of the grains. By this way, the crystallite size could be hold below the critical size for the phase transformation [1. This critical size varies according to thermal stress induced during cooling so that most of the the physical nature of zirconia: 50-55 nm for ZrO2 deposited residual stresses come from the growth process by MOCVD [6], 18 nm in powder Zro2 [11] and 30 nm in It is likely, according to the martensitic character of the bulk Zro2[7] transformation [12, 13] that the t- m transformation is due to Therefore. in order to corroborate these research sugges- the fact that tension applied to our films induces a finite tions, it is necessary to look at the microstructure of displacement, without diffusion, of a great number of atoms our samples. This was done by FEG-SEM analyzes of our associated to a lattice shear It was shown that the application of a tensile stresses The micrographs of Fig. 6 Suggest that the grains take a generates a t- m phase transformation within zirconia films. structure increasingly facetted with the stress and the cycles The question which arises at this stage concerns a possible applied. Our feeling is that the microstructure of the as correlation between the stress application and the crystallite deposited sample corresponds to an agglomerate of grains size, which is given, according to the literature, as responsible which de-aggregates afterwards when their size increases due for the tetragonal stabilization. to the applied str 卧10wB·5A;城门@d50kX1ur As-deposited After compression 1的wD6myA城@:50oxe After tension After 16 cycles of tension ig. 6. FEG-SEM micrograph (50,000x)of the four samp
thermal stress induced during cooling so that most of the residual stresses come from the growth process. It is likely, according to the martensitic character of the transformation [12,13] that the t ! m transformation is due to the fact that tension applied to our films induces a finite displacement, without diffusion, of a great number of atoms associated to a lattice shear. It was shown that the application of a tensile stresses generates a t ! m phase transformation within zirconia films. The question which arises at this stage concerns a possible correlation between the stress application and the crystallite size, which is given, according to the literature, as responsible for the tetragonal stabilization. It is not impossible that such a correlation takes place. Indeed, the level of internal stresses, which is different between the two monoclinic and tetragonal (more constrained) phases, could be responsible for this possible correlation. In addition, the experiment of Bouvier et al. [5] seems to agree this point since the strong compression applied to the samples could be responsible for the shape and/or the size of the grains. By this way, the crystallite size could be hold below the critical size for the phase transformation [1]. This critical size varies according to the physical nature of zirconia: 50–55 nm for ZrO2 deposited by MOCVD [6], 18 nm in powder ZrO2 [11] and 30 nm in bulk ZrO2 [7]. Therefore, in order to corroborate these research suggestions, it is necessary to look at the microstructure of our samples. This was done by FEG-SEM analyzes of our films. The micrographs of Fig. 6 Suggest that the grains take a structure increasingly facetted with the stress and the cycles applied. Our feeling is that the microstructure of the asdeposited sample corresponds to an agglomerate of grains which de-aggregates afterwards when their size increases due to the applied stress. B. Benali et al. / Applied Surface Science 253 (2006) 1222–1226 1225 Table 1 Internal stresses in films of the as-deposited sample and sample 2 (after 16 cycles of tension) Phases s (MPa) As-deposited After 16 cycles (S1 1 1)m 0 100 0 100 (1 1 1)m 151 53 236 60 (111)t 326 50 336 52 Fig. 6. FEG-SEM micrograph (50,000) of the four samples
1226 B. Benali et al. /Applied Surface Science 253(2006)1222-1226 4. Conclusion [3]RJ. Gorte, J.M. Vohs, J. Catal. 216(2003)477-486 4]IAH. Al-Dawery, E.G. Butler, Composites Part A: Appl Sci. Manuf. Part The zirconia thin layers (250 nm) deposited on stainless A32(8)(2001)100 5 P. Bouvier, J Godlewski, G. Lucazeau, J Nucl. Mater. 300(2002)118- steel substrates by MOCVd process at low oxygen pressure and 900C, are made up of tetragonal and monoclinic phase [6O. Bernard, Doctor Thesis, LEMHE, Universite Paris Xl, Orsay, 2004 whose ratio depends on the process parameters. When Thin Solid Films, submitted for publication. applying an external stress and more precisely an externa [7] G. Pengtao, LJ. Meng, M.P. dos Santos, V. Teixeira, M. Andritschky tensile stress using a bending apparatus, the structural and Vacuun6402002)267-273. [8 G. Maeder, J. L. Lebrun, in R. Caplain(Eds ) Methodes mechanical state of our zirconia films, initially subjected to usuelles de caracterisation des Eyrolles, 1988 compressive residual stresses, is modified. It induces a [9]O.Bernard,GAmiri,CHaut, A. M. Huntz, M. Andrieux. Mater. partially tetragonal to monoclinic phase transformation, Sci.Eng.A335(2002)32-42. mainly(11 1),(-1 11)m, function of the stress level and [10] R.C. Garvie. P.S. Nicholson, J. Am. Ceram Soc. 55(6)(1972)303- the number of applied cycles [11] T. Chraska, A H. King, CC Berndt, Mater. Sc Eng. A 286(2000)169- References [12] G.K. Bansal, A H. Heuer, Acta Metall. 20(1972)1281-1289 [13]S. Deville, G. Guenin, J. Chevalier, Acta Mater. 52(2004)5697- [1 A. Methivier, Doctor Thesis, ENSM and INPG, Saint Etienne, 1992. 5707 [2] A. Methivier, M. Pijolat, J Catal. 139(1993)329-337
4. Conclusion The zirconia thin layers (250 nm) deposited on stainless steel substrates by MOCVD process at low oxygen pressure and 900 8C, are made up of tetragonal and monoclinic phases whose ratio depends on the process parameters. When applying an external stress and more precisely an external tensile stress using a bending apparatus, the structural and mechanical state of our zirconia films, initially subjected to compressive residual stresses, is modified. It induces a partially tetragonal to monoclinic phase transformation, mainly (1 1 1)t ! (1 1 1)m, function of the stress level and the number of applied cycles. References [1] A. Methivier, Doctor Thesis, ENSM and INPG, Saint Etienne, 1992. [2] A. Methivier, M. Pijolat, J. Catal. 139 (1993) 329–337. [3] R.J. Gorte, J.M. Vohs, J. Catal. 216 (2003) 477–486. [4] I.A.H. Al-Dawery, E.G. Butler, Composites Part A: Appl. Sci. Manuf. Part A 32 (8) (2001) 1007–1012. [5] P. Bouvier, J. Godlewski, G. Lucazeau, J. Nucl. Mater. 300 (2002) 118– 126. [6] O. Bernard, Doctor Thesis, LEMHE, Universite´ Paris XI, Orsay, 2004. Thin Solid Films, submitted for publication. [7] G. Pengtao, L.J. Meng, M.P. dos Santos, V. Teixeira, M. Andritschky, Vacuum 64 (2002) 267–273. [8] G. Maeder, J.L. Lebrun, in: D. David, R. Caplain (Eds.), Me´thodes usuelles de caracte´risation des surfaces, Eyrolles, 1988. [9] O. Bernard, G. Amiri, C. Haut, B. Feltz, A.M. Huntz, M. Andrieux, Mater. Sci. Eng. A 335 (2002) 32–42. [10] R.C. Garvie, P.S. Nicholson, J. Am. Ceram. Soc. 55 (6) (1972) 303– 305. [11] T. Chraska, A.H. King, C.C. Berndt, Mater. Sc. Eng. A 286 (2000) 169– 178. [12] G.K. Bansal, A.H. Heuer, Acta Metall. 20 (1972) 1281–1289. [13] S. Deville, G. Gue´nin, J. Chevalier, Acta Mater. 52 (2004) 5697– 5707. 1226 B. Benali et al. / Applied Surface Science 253 (2006) 1222–1226