Vibration Velocity Limitation of Transducer Using Titanium-Based Hydrothermal Lead Zirconate Titanate Thick Film

Transcription

1 Jpn. J. Appl. Phys. Vol. 42 (2003) pp Part 1, No. 5B, May 2003 #2003 The Japan Society of AppliedPhysics Vibration Velocity Limitation of Transducer Using Titanium-Based Hydrothermal Lead Zirconate Titanate Thick Film Takefumi KANDA,YutakaKOBAYASHI 1, Minoru Kuribayashi KUROSAWA 1 andtoshirohiguchi 2 Graduate School of National Science and Technology, Okayama University, Tsushima-naka, Okayama , Japan 1 Department of Advanced Applied Electronics, Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama , Japan 2 Department of Precision Machinery Engineering, Graduate School of Engineering, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo , Japan (ReceivedNovember 18, 2002; acceptedfor publication February 10, 2003) High-intensity vibration operation of a titanium basedleadzirconate titanate (PZT) thick film longitudinal vibrator fabricated using a hydrothermal method is described in this paper. Higher limitation of vibration velocity is very important for high output power actuators. For sensors, linearity of the frequency response is an important factor. To investigate the performance of the PZT film material, we changed the deposition process of the hydrothermal method and increased the thickness of the PZT film. As a result, the maximum vibration velocity of the longitudinal vibrator was 2.5 m/s (0 to peak value). This value was almost 3-foldthat of bulk PZT material. The vibration velocity was limitedby the value of maximum stress. The maximum stress of the hydrothermal PZT transducer was larger than that of the bulk PZT transducer. At such high-intensity vibration of over 2 m/s or driving voltage of 150 V p-p, the frequency response curves were not linear andthe jumping phenomenon was observed. [DOI: /JJAP ] KEYWORDS: PZT, thick film, hydrothermal method, vibration velocity, frequency response, polycrystal structure 1. Introduction The hydrothermal method for deposition of lead zirconate titanate (PZT) film 1,2) has been receiving increasing attention for fabricating micro size devices such as micro electro mechanical systems (MEMS). This is because, with this method, it is possible to realize a piezoelectric film on a curvedsurface, a thicker film of over 10 mm, automatic polarization, andthe elimination of the annealing process. This deposition process was utilized to fabricate micro ultrasonic motors, 3,4) vibro touch probe sensors, 5,6) fluidic devices, 7) vibrating gyroscopes, 8) andtuning type vibrators. 9) We have already reported that, by using the hydrothermal deposition process for PZT, we can fabricate vibrators that have higher limitation of vibration velocity andhigher linearity of the frequency response under high-intensity operation. 10,11) It is important to know how the limitation of the vibration velocity is. The static factors such as piezoelectric constants, d 31, d 33, and e 31, were measuredto evaluate the performance of the material in most of the cases for the evaluation of piezoelectric vibrators. However they indicated only the proportional ratio between an electric port anda mechanical port. They do not give sufficient information on the performance of working conditions. Therefore, it is important to know the limit of the vibration velocity and the frequency response curves at high-intensity operation. 10) In this paper, high-intensity vibration operation of PZT film longitudinal vibrators fabricated using a hydrothermal method is described. To investigate the performance of the PZT film material, we changedthe deposition process of the hydrothermal method and increased the thickness of the PZT film. The characteristics of the film as a piezoelectric material, the limitation of vibration velocity andthe linearity of frequency response of longitudinal vibrator were evaluatedby experiments. address: kanda@sys.okayama-u.ac.jp 2. Hydrothermal Deposition Process and Sample Transducers The Hydrothermal method utilizes the chemical process repetition between titanium base substrate andthe melted ions in an autoclave. 1) The process has two steps, both of which are carriedout in the autoclave at temperatures over 100 deg Celsius, and pressures above 0.1 MPa. These values of temperature andpressure were lower than those of most PZT film deposition processes. The hydrothermal deposition of PZT film consists of two different processes. 1) Table I lists the process repetition conditions of the hydrothermal method. The first step is the nucleation process. During this process, nuclear PZT crystals are deposited on the titanium base. The mechanism of crystal generation using the base material, titanium, has an advantage with regard to the bonding of the film to the base substrate. The secondstep is the crystal growth process. During this secondprocess, PZT crystals are expectedto grow. By repeating this secondprocess, a thicker film can be Table I. Process repetition conditions of the hydrothermal method for PZT film. 1st Process (Nucleation process) ZrOCl 2 8H 2 O [g] meltedinto 2.00 [ml] H 2 O TiCl [ml] 1.95 [mol/l] TiCl 4 Pb(NO 3 ) [g] meltedinto 6.20 [ml] H 2 O KOH 2.69 [g] meltedinto 11.0 [ml] H 2 O Temperature: 120 deg Celsius Process repetition: 24 h 2ndProcess (Crystal growth process) ZrOCl 2 8H 2 O [g] meltedinto 2.00 [ml] H 2 O TiCl [ml] 1.95 [mol/l] TiCl 4 Pb(NO 3 ) [g] meltedinto 6.60 [ml] H 2 O KOH 2.69 [g] meltedinto 11.0 [ml] H 2 O Temperature: 120 deg Celsius Process repetition: 24 h n times 3014

2 Jpn. J. Appl. Phys. Vol. 42 (2003) Pt. 1, No. 5B T. KANDA et al mm Electrode (Au) Ti: 100 µm 7.9 mm PZT Film Fig. 1. Schematic view of the rectangular PZT transducer for longitudinal vibration; half wavelength. obtained. 10,11) To investigate the influence of film thickness, rectangular shapedlongitudinal vibrators were usedfor the estimation of the vibration conditions. These rectangular shaped longitudinal vibrators, which were 7.9 mm long and 1 mm wide, were fabricated. The schematic of the longitudinal vibrator is shown in Fig. 1. The vibration mode was half wavelength longitudinal vibration. The element was supported at the central nodal point of the axial vibration component. The base material was titanium on which the PZT film was deposited which was followed by deposition of the Au electrode. 3. Influence of Increased Film Thickness To investigate the performance of the hydrothermal PZT film material at high-intensity operation, we increasedthe thickness of the PZT film by repeating the crystal growth process. We repeatedthe process repetition process for a maximum of 30 times. That is, we repeatedthe secondstep process 29 times after the 1st step process. Figure 2 shows the surface images of hydrothermally deposited PZT film by the scanning electron microscope (SEM). In Fig. 2, process was repeated2, 10 and30 times. As shown in these images, hydrothermally deposited PZT film has a polycrystalline structure. When the process repetition times increased, smaller crystals were fabricatedon the surface of the larger crystals. We measuredthe thickness of the PZT film by observing the SEM image of the cross section as shown in Fig. 3. Figure 4 shows the enlargedpictures of the cross-sectional images. As shown in these SEM images in Fig. 3, the thickness of the polycrystalline film increasedas the process repetition times increased. Figure 5 shows the relationship between the film thickness andthe process repetition times. As shown in Fig. 5, the thickness of the PZT film was proportional to the process repetition times. However the increasing ratio differed depending on autoclaves, which were usedfor the process repetition. The maximum thickness of the film was 55 mm when the process repetition time was 30. Figure 6 shows the X-ray diffractometer (XRD) patterns of the PZT film when the process repetition process times are 2 and30. Each XRD pattern indicates that the crystal structure is that of PZT. In addition, we estimated the atomic weight ratio of Zr/(Zr+Ti) of PZT crystal in the film. The atomic weight ratio can be estimatedby using the wavelength dispersible X-ray spectrometer (WDS). Figure 7 shows the relationship between the atomic weight ratio of Zr/(Zr+Ti) andthe process repetition times. As is well known, the optimal piezoelectricity of PZT material is Fig. 2. SEM image of the hydrothermal PZT surface. obtainedwhen the crystals are in the morphotorophic phase boundary (MPB). When the process repetition time is 2, the Zr/(Zr+Ti) ratio was about 0.5 although the ratio is 0.52 when the crystals are in MPB. However, when the process repetition times increased, the ratio was about 0.7 and this ratio did not change by the process repetition times when the process repetition time is more than 10. These results show that the chemical composition of the PZT thick film was constant except for the part near the titanium substrate. The thickness of the layer, in which the chemical composition is not constant, is estimatedto be 5 to 10 mm. Figures 8 and 9 show the admittance and phase change near the resonance frequency of the longitudinal vibrator when the process repetition times are 2 and30. The resonance frequencies were khz and273.9 khz. At these resonance frequencies, the admittance values were 6.12 ms and 1.07 ms. The value of admittance is decreased when the process repetition times are increasedandthe thickness of the PZT film is increased. The characteristics of the PZT film as a piezoelectric material were also estimated. Table II lists the experimental results andthe comparison with those of bulk PZT materials. 12) The piezoelectric constants are evaluatedby the measurement of vibration displacements. 10,13) As shown

3 3016 Jpn. J. Appl. Phys. Vol. 42 (2003) Pt. 1, No. 5B T. KANDA et al. Fig. 3. SEM image of the hydrothermal PZT film cross section. Fig. 4. SEM image of the hydrothermal PZT film cross section; enlarged images of Fig. 3. Table II. Characteristics of the PZT film as piezoelectric material. Hydrothermal PZT film Bulk PZT: Atomic ratio of Zr/Ti = 52/48 15) " T 33 =" k (k 31 ) e 31 (N/V) 0:12 3:06 d 31 (pc/n) 17 93:6 in Table II, the characteristics of the film do not exceeded those of the bulk material. However, the listedparameters indicate the characteristics under static conditions. Then these experimental results do not indicate whether the hydrothermal PZT film is inferior to the bulk PZT as a component of vibrating devices. 4. Vibration Velocity We measuredthe vibration velocity of the transducers. Using a laser Doppler vibrometer, we measuredthe vibration velocity of the tip of the longitudinal vibrator at its resonance frequency. Figure 10 presents the relationship between the vibration velocity of the longitudinal vibrator Fig. 5. Thickness of the PZT film createdby each repetition process; each marker indicate a different container (autoclave). andthe driving voltage. As shown in Fig. 10, the vibration velocity was linearly increasedagainst the driving voltage andnot saturatedbeneath the breakdown voltage. The vibration velocity at the breakdown voltage increased against the thickness of the PZT film. The maximum vibration velocity was obtainedfrom the transducer whose PZT film was deposited by 15 repetition processes. The maximum vibration velocity was 2.5 m/s at a driving voltage of 114 V p-p. This vibration velocity value andthose in the

4 Jpn. J. Appl. Phys. Vol. 42 (2003) Pt. 1, No. 5B T. KANDA et al Fig. 8. Admittance of the longitudinal vibrator; process repetition time was 2. Fig. 6. XRD patterns of PZT film; process repetition times are 2 and30. Fig. 9. Admittance of the longitudinal vibrator; process repetition time was 30. Fig. 7. Chemical composition of hydrothermally deposited PZT film; relationship between the mol ratio of Zr/(Zr+Ti) andthe process repetition times. following pages are 0 to peak values. However even if the thickness of the PZT film was more than 30 mm, the vibration velocity just below the breakdown voltage did not increase. The maximum vibration velocity was 2.5 m/s, which was 3-foldthat of the vibrators which consist of bulk PZT material. 14) It has been reportedthat the vibration velocity is dependent on the maximum stress when the piezoelectric transducer is vibrating under a high-intensity operation. 15) In the case of compoundvibrators which consist of a titanium base andpiezoelectric film, the stress is dependent on the strain andthe Young s modules of the PZT material. In the case of the vibrator fabricatedof bulk PZT material, N81 (Tokin Corp.), vibration velocity was maximum when the stress was 41 MPa. 15) On the other hand, in the case of the hydrothermal PZT, the stress was 47 MPa when the vibration velocity was maximum at 2.5 m/s, Young s modulus was 4: N/m 2, the density was 4: kg/m 3, 13) the thickness of titanium was 100 mm, andthe thickness of PZT was 10 mm on each side. In this case, the compoundyoung s modulus was 8: N/m 2 andthe maximum strain was 5: This result indicates that the stress at maximum vibration velocity of hydrothermally deposited PZT was larger than that of bulk PZT. To compare with the experimental results of maximum stress on the vibrator using the titanium basedhydrothermally deposited PZT film, we estimated the maximum stress of the vibrator using titanium base andbulk PZT. The Young s modulus values of PZT in the morphotropic phase boundary and PZT type piezoelectric material N81 are 7: N/m 212) and 1: N/m 2, 15) respectively.

5 3018 Jpn. J. Appl. Phys. Vol. 42 (2003) Pt. 1, No. 5B T. KANDA et al. Fig. 10. Vibration velocity of the vibrators using films deposited by each repetition process; Relationship between the vibration velocity andthe driving voltage. Fig. 11. Vibration velocity of the vibrators using films deposited by each repetition process; Relationship between the vibration velocity andthe electric field. When the thickness of titanium is 100 mm, andthickness of PZT is 10 mm on each side (this condition is similar to that of the titanium basedhydrothermal PZT film vibrator), the compoundyoung s modulus values are 10 N/m 2 and12 N/ m 2. When the maximum vibration velocity is 2.5 m/s, the maximum stress values of vibrators, using morphotropic phase boundary PZT andn81 are 53 MPa and62 MPa. These maximum stress values are larger than those of hydrothermal PZT film vibrator and the N81 bulk PZT vibrator stress limitation. These calculation results show that the vibration limitation of the vibrator using hydrothermally deposited PZT film is higher than that of the bulk PZT vibrator andthe higher limitation does not depend on the compound vibrator structure. Hence we can conclude that the hydrothermal PZT film transducer can realize much higher vibration velocity under high-intensity operation than the bulk PZT vibrator. We have reportedthat the increasing ratio of the vibration velocity against the driving voltage was constant. 10,11) At that time, the process repetition time was 15 andthe film thickness was 30 mm at the most. However Fig. 10 shows different results. The increasing ratio was not constant. Figure 11 shows the relationship between the vibration velocity andthe electric field. In Fig. 11, the slope of the graph was constant when the electric fieldwas lower except when the process repetition times were 15. We can report that the vibration velocity is in proportion to the driving voltage when the thickness of the PZT film is not so large andthe vibration velocity is in proportion to the electric field when the film thickness is larger. These results indicate the influence of the vibrator base substrate on the vibration velocity. As shown in Fig. 12, the breakdown voltage of each transducer was saturated at about 150 V p-p. The breakdown electric field of the PZT films deposited on each transducer was saturatedat 4 MV/m. Fig. 12. Breakdown voltage of the vibrators using films deposited by each repetition process; each marker indicates a different container (autoclave). 5. Frequency Response Under the conditions of a constant driving voltage, the vibration velocity was measuredby changing the driving frequency. In the case of bulk materials, it is well known that the response curves are distorted and jumping phenomenon is observed. 16,17) In the case of hydrothermal PZT film devices, we have reported that the response curves were smooth andcontinuous at a comparatively high-intensity operation. In order to estimate the influence of the film thickness on the linearity, we measuredthe frequency response of the thicker film vibrators. Figures 13 and 14 show the frequency response near the resonance frequency. The frequency scanning was carried out in up and down directions in Figs. 13 and 14. As shown in these figures, the response curves were smooth andcontinuous at a comparatively highintensity operation of 1.5 m/s. However, at the higher

6 Jpn. J. Appl. Phys. Vol. 42 (2003) Pt. 1, No. 5B T. KANDA et al such high-intensity vibration levels of over 2.0 m/s or driving voltages of over 150 V p-p, the frequency response curves were not linear andthe jumping phenomenon was observed. The experimental results indicated that the vibration velocity of the vibrator is in proportion to the electric fieldwhen the PZT film is thicker. Fig. 13. Frequency response of the vibrator; the frequency scanning direction was upward. Acknowledgements This work was supportedby the Grant-in-aidfor general scientific research of the Ministry of Education, Culture, Sports, Science andtechnology, andby the Proposal-Based New Industry Creative Type Technology R&D Promotion Program from the New Energy andindustrial Technology Development Organization (NEDO) of Japan, andby the Grant-in-aidfor Research Fellowship for Young Scientists of the Japan Society for the Promotion of Science. The authors wouldlike to thank Mr. Yasui of The University of Tokyo, Mr. Igi andms. Sasaki of the Tokyo Institute of Technology for valuable advice on and their assistance with the hydrothermal deposition of PZT thin film. Fig. 14. Frequency response of the vibrator; the frequency scanning direction was downward. intensity operation of more than 2.0 m/s, the response curves were not linear. Even if the vibration velocity was less than 1.5 m/s, at the high driving voltage of 150 V p-p, the response curves of the vibrator of which the film thickness was 40 mm were not smooth andcontinuous. 6. Conclusion In order to fabricate the high power output actuator, a PZT thick film was deposited by the hydrothermal method and the dynamic performance of hydrothermal PZT film vibrators was determined. The thickness of the PZT film ranged up to 55 mm. The maximum vibration velocity was 2.5 m/s, which value was 3-foldthat of vibrators using bulk PZT. In 1) K. Shimomura, T. Tsurumi, Y. Ohba andm. Daimon: Jpn. J. Appl. Phys. 30 (1991) ) T. Morita, T. Kanda, M. Kurosawa and T. Higuchi: Jpn. J. Appl. Phys. 36 (1997) ) T. Morita, M. K. Kurosawa andt. Higuchi: IEEE Trans. Ultrason. Ferroelectr. & Freq. Control 45 (1998) ) T. Morita, T. Kanda, M. Kurosawa and T. Higuchi: Jpn. J. Appl. Phys. 36 (1997) ) T. Kanda, T. Morita, M. K. Kurosawa and T. Higuchi: Sens. & Actuat. A 83 (2000) 67. 6) T. Kanda, T. Morita, M. K. Kurosawa and T. Higuchi: Jpn. J. Appl. Phys. 40 (2001) ) H. Yasui, M. K. Kurosawa andt. Higuchi: Sens. & Actuat. A 96 (2002) 28. 8) H. Sato, F. Arai, T. Fukuda, K. Itoigawa and Y. Tsukahara: Trans. IEE Jpn. 121E (2001) 281 [in Japanese]. 9) J. Terada: Trans. IIEJ J83 (2000) 405 [in Japanese]. 10) T. Kanda, M. K. Kurosawa, H. Yasui and T. Higuchi: Sens. & Actuat. A 89 (2001) ) T. Kanda, Y. Kobayashi, M. K. Kurosawa, H. Yasui and T. Higuchi: Jpn. J. Appl. Phys. 40 (2001) ) B. Jaffe, W. R. Cook andh. Jaffe: Piezoelectric Ceramics (Academic Press, London, 1971) p ) Y. Ohba, M. Miyauchi, T. Tsurumi andm. Daimon: Jpn. J. Appl. Phys. 32 (1993) ) S. Hirose, K. Nakamura, Y. Adachi and H. Shimizu: Proc. Ann. Meet. Acoust. Soc. Jpn. (1991) 845 [in Japanese]. 15) M. Umeda, S. Takahashi, Y. Sasaki, K. Nakamura and S. Ueha: Trans. IEICE. J82-C-1 (1999) 762 [in Japanese]. 16) T. Wada: Trans. IIEJ 119 (1999) 246 [in Japanese]. 17) J. Nosek: IEEE Trans. Ultrason. Ferroelectr. Freq. Control 46 (1999) 823.