Ultrasonics 53 (2013) 837–841

Contents lists available at SciVerse ScienceDirect

Ultrasonics

journal homepage: www.elsevier .com/locate /ul t ras

Study on optimizing ultrasonic irradiation period for thick polycrystallinePZT film by hydrothermal method

Kanako Ohta a,⇑, Gaku Isobe a, Peter Bornmann b, Tobias Hemsel b, Takeshi Morita a

a The University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa, Chiba 277-8563, Japanb Faculty of Mechanical Engineering, University of Paderborn, Pohlweg 47-49, 33098 Paderborn, Germany

a r t i c l e i n f o a b s t r a c t

Article history:Received 16 August 2012Received in revised form 3 December 2012Accepted 4 December 2012Available online 23 December 2012

Keywords:Piezoelectric materialPZT ceramicsHydrothermal methodUltrasonic irradiation

0041-624X/$ - see front matter � 2012 Elsevier B.V.http://dx.doi.org/10.1016/j.ultras.2012.12.003

⇑ Corresponding author. Tel.: +81 4 7136 4615.E-mail addresses: ohta@ems.k.u-tokyo.ac.jp (K. O

(T. Morita).

The hydrothermal method utilizes a solution-based chemical reaction to synthesize piezoelectric thinfilms and powders. This method has a number of advantages, such as low-temperature synthesis, and highpurity and high quality of the product. In order to promote hydrothermal reactions, we developed anultrasonic assisted hydrothermal method and confirmed that it produces dense and thick lead–zircon-ate–titanate (PZT) films. In the hydrothermal method, a crystal growth process follows the nucleationprocess. In this study, we verified that ultrasonic irradiation is effective for the nucleation process, andthere is an optimum irradiation period to obtain thicker PZT films. With this optimization, a 9.2-lm-thickPZT polycrystalline film was obtained in a single deposition process. For this film, ultrasonic irradiationwas carried out from the beginning of the reaction for 18 h, followed by a 6 h deposition without ultrasonicirradiation. These results indicate that the ultrasonic irradiation mainly promotes the nucleation process.

� 2012 Elsevier B.V. All rights reserved.

1. Introduction reaction temperature compared to the conventional deposition

Piezoelectric materials are widely used in devices such as ultra-sonic motors and ultrasonic sensors and so on. For miniaturizingthem, piezoelectric thin films are intensively being studied bya lot of researchers [1–4]. To improve the performance of ultra-sonic medical devices using piezoelectric films, it is important toincrease the piezoelectric constant, decrease the acoustic imped-ance and have the ability to output ultrasonic at an optimumfrequency [5].

PZT has a high piezoelectric constant in general; however, fromthe point of view of acoustic impedance, the values for a human andPZT are respectively about 1.5 � 106 kg/m2s and 30 � 106 kg/m2s[6]. Moreover, in the case of a catheter, 50 lm thickness is requiredfor outputting 30 MHz. However, this thickness range is too thin forfabricating bulk PZT and too thick for depositing as a film.

We have therefore made PZT using the hydrothermal methodbecause it is known to yield thicker and denser PZT without theneed for sintering, compared to PZT formed using conventionalmethods [7]. The hydrothermal method utilizes a solution-basedchemical reaction. Synthesis is carried out around 150� C and pres-sure inside a closed vessel in the case of PZT thin film deposition.Due to the conditions used, it has unique advantages. It is a simpleprocess, yields high-quality material, and offers the possibility ofdeposition on three dimensional substrates. Moreover, the low

All rights reserved.

hta), morita@k.u-tokyo.ac.jp

process allows PZT to be produced without sintering or pollingtreatment [8]. PZT synthesized by this method has therefore beenwidely studied [9,10].

However, for obtaining thick PZT films, the hydrothermal meth-od requires a long deposition time and repeated synthesis. Forobtaining thicker films in a single deposition, we have developedthe Ultrasonic Assisted Hydrothermal Method (UAHTM) for pro-moting the chemical reaction [11,12]. Ultrasonic irradiation ofthe solution is considered to accelerate the chemical reaction,due to cavitation and acoustic streaming [13]. We have alreadyverified the effect of ultrasonic assist for promoting the hydrother-mal reaction [11,14]. However, details of the phenomena stillremain to be clarified for much thicker films. In this study, theobjective was to clarify the optimum conditions to obtain thickerPZT films in a single deposition using UAHTM.

As described above, the hydrothermal method is a solution-based chemical reaction under high temperature and pressure.The PZT films obtained are comparatively thicker and denser thanPZT formed by conventional methods. However, the long deposi-tion time is a serious problem. For example, Ishikawa showed thatthe deposition of 50-lm-thick PZT films by the hydrothermalmethod required repeating the process 20 times [15]. Such repeti-tion is not realistic for practical applications.

2. Synthesis procedure

Our research group formulated a method using ultrasonic irra-diation and investigated a new autoclave for this purpose. This

838 K. Ohta et al. / Ultrasonics 53 (2013) 837–841

autoclave enables us to irradiate the chemical solution directlywith ultrasonic waves. As shown in Fig. 1, a bolted Langevin PZTultrasonic transducer was attached to a hydrothermal container(Taiatsu Techno Co., Ltd. TAF-SR type 300 ml) in the autoclave. Itshould be noted that this transducer is different from the one usedin our previous study [14] so the irradiation conditions used in thiswork are also different. This means that the effect of the ultrasonicirradiation was not the same in this study as in the previous one.

The driving frequency was controlled to follow the resonantfrequency around 31.0 kHz and the input voltage was 300 Vp-p.Current was measured by a current probe (Tektronix TCPA300)for changing the driving frequency, and the function generator(NF WF1974) was controlled by a PC through a GPIB interface.

Using this autoclave and transducer, PZT was deposited on oxi-dized titanium substrate oxidized at 600� C for 5 min; the startingmaterials are shown in Table 1 [14]. The gap between the tip of thetransducer and the substrate was 5 mm. Our objective thereforewas to clarify the optimum irradiation period for a thick film.

Fig. 2. Surface images of PZT (a) without ultrasonic assist and (b) with ultrasonicassist.

3. Effect of UAHTM

3.1. Irradiating ultrasonic beginning or later

As mentioned in introduction, the ultrasonic irradiation pro-motes the chemical reaction. The PZT film deposited by UAHTMwas denser as shown in Fig. 2. In this study, we focused on the dif-ference in grain size, and it can be seen that the grains in Fig. 2b aresmaller than those in Fig. 2a. The film becomes thicker by contin-uous piling up of grains. Therefore, when the deposition time isconstant, the number of grains and their size are important for athicker film. In this case, the ultrasonic assist makes the grains

Fig. 1. Ultrasonic assisted hydrothermal method.

Table 1Conditions for depositing PZT.

Hydrothermal method

ZrCl2O�8H2O 0.604 gPb(NO3)2 2.070 gTiO2 (rutile type) 0.100 gH2O 37.50 mlK0H (8 N) 12.50 mlSolution volume 50 mlTitanium substrate 17 mm � 25 mm � 50 lmTemperature 140� C

Fig. 3. Surface images of PZT (a) irradiated at the first of the deposition and (b)irradiated at the end of the deposition.

smaller while the number of grains increases. Moreover, it isexpected that changing the irradiation time affects the numberand size of the grains.

Table 2Ultrasonic assist conditions.

K. Ohta et al. / Ultrasonics 53 (2013) 837–841 839

Under our standard conditions, ultrasonic irradiation was car-ried out during the entire deposition period. The hydrothermalmethod is divided into two processes, a nucleation process and acrystal growth process [16]. From the experiments, PZT grains irra-diated ultrasonic irradiation was small, and it is expected that theultrasonic-assist promotes the nucleation process. To clarify thiseffect, two PZT films were produced with the same condition asshown in Table 1. The first was irradiated with ultrasonic wavesat the beginning of the deposition while the other was irradiatedin the end period of deposition. As shown in Fig. 3, there werefew PZT crystals on the surface in Fig. 3b when irradiated at theend of the deposition. From this result, it was confirmed that theultrasonic assist effects the beginning period.

3.2. Changing irradiation period

Based on the above results, we decided to start the ultrasonicirradiation from the beginning of the hydrothermal method, andwe modified the ultrasonic assist irradiation period. Even afterstopping the ultrasonic assist treatment, the hydrothermal processwas continued for 24 h total reaction time. As shown in Table 2,Film A was deposited without ultrasonic assist. For Films B–E,ultrasonic irradiation was started from the beginning of the depo-sition. Film E was synthesized with ultrasonic assist for 24 h, whichmeans it was irradiated throughout the deposition time.

Scanning electron microscopy (SEM) images of each of the PZTfilms are shown in Fig. 4. It was obvious that the grain size of eachfilm was different. Without the ultrasonic irradiation, the titaniumsubstrate was not totally covered with the PZT film, as shown inFig. 4a. On the other hand, the PZT films became dense with ultra-sonic irradiation. The grain size after the first 12 h of ultrasonicirradiation was found to be the largest. If the number of PZT crys-tals was the same, it is expected that this film is the thickest.

Close-up cross sectional images of the films are also shown inFig. 4. The relationship between the ultrasonic irradiation periodsand the thickness is shown in Fig. 5. In terms of thickness, Film Awithout the ultrasonic assist was very thin, and with longer ultra-sonic assist times the thickness increased. The maximum thicknesswas obtained for Film D which was deposited with 18 h of ultra-sonic assist, yielding a thickness of 9.2 lm. From this result, itwas found that an optimum ultrasonic irradiation period existsfor large thickness. This result was somewhat different from ourexpectation that Film C would be the thickest.

The results indicated that both of the nucleation process and thecrystal growth process are important for thick film. It is supposed

that the ultrasonic irradiation promotes only nucleation processand during the ultrasonic irradiation, the number of grainsincreases.

When we compare Film C and Film D, Film D was thicker be-cause it had sufficient nuclei owing to large ultrasonic irradiationperiod. On the other hand, Film E had no period to grow up thenuclei, meaning the film was composed of many small grains.Therefore, the Film E was thinner than Film D that had period ofthe crystal growth process.

In our previous study [14], a PZT film was deposited withultrasonic assist during the whole reaction time, yielding athickness of 9.9 lm, which is larger than obtained in the presentstudy. This inconsistency comes from the different ultrasonicirradiation conditions used. This means that the PZT film thick-ness is a function of the ultrasonic parameters. The relationshipbetween the ultrasonic parameters and the thickness is on-goingwork.

4. Mechanism of UAHTM

In the previous chapter, we clarified that there is an optimumirradiation period for the thicker PZT film. For these depositions,the total reaction time was fixed to 24 h and the ultrasonic irradi-ation period was changed. As a result, it was suggested that theultrasonic irradiation promotes the nucleation process and doesnot the crystal growth process.

If this hypothesis is correct, the grain size just after stoppingultrasonic irradiation must be small regardless of the reaction per-iod. Therefore, we prepared these kinds of films deposited by dif-ferent total reaction time, and during all the reaction time, theultrasonic was irradiated.

The reaction time for these Films (F, G and H) are shown in Ta-ble 3. These new Films were all irradiated with ultrasound duringthe synthesis and only the deposition time was varied. The deposi-tion time of Film F was 6 h, that for Film G was 12 h and that forFilm H was 18 h.

SEM images showed that the grain sizes were indeed almost thesame, as can be seen in Fig. 6. Moreover the number of grains gotincrease with larger reaction periods as we expected. This resultsuggests that ultrasonic irradiation promotes the nucleation pro-cess. We thus concluded that in order to obtain a thick film, irradi-ating with ultrasonic during only the nucleation process is theoptimum condition. First, the number of crystal nuclei was in-creased by ultrasonic irradiation then the cores were grown at

Fig. 5. Thickness dependence on ultrasonic irradiation period.

Fig. 4. Surface and cross-sectional images of PZT. (a) Film A (no ultrasonic assist), (b) Film B (ultrasonic assist for first 6 h), (c) Film C (ultrasonic assist for first 12 h), (d) Film D(ultrasonic assist for first 18 h) and (e) Film E (ultrasonic assist for first 24 h).

840 K. Ohta et al. / Ultrasonics 53 (2013) 837–841

the condition without ultrasonic irradiation. Irradiating ultrasonicirradiation at nucleation process enables the formation of thick PZTfilms.

5. Summary

In this study, it was verified that ultrasonic irradiation promotesthe nucleation process in the hydrothermal synthesis of PZT thinfilms. This means that the ultrasonic irradiation increases thenumber of crystal nuclei and does not have effect to increasingthe grain size. To obtain a thick film, increasing the number ofPZT crystals and obtaining large size crystals are important. Weverified that there is an optimum ultrasonic irradiation period,and in the case of deposition for 24 h, the total thickness of a PZTpolycrystalline film with ultrasonic irradiation during the first18 h was the largest at 9.2 lm.

Table 3Condition of deposition.

Fig. 6. Surface images of PZT (a) Film F (deposition for 6 h), (b) Film G (deposition for 12 h) and (c) Film H (deposition for 18 h).

K. Ohta et al. / Ultrasonics 53 (2013) 837–841 841

Acknowledgements

This research was supported by the New Energy and IndustrialTechnology Development Organization (NEDO). We appreciateDr. Daeyong Jeong for fruitful discussions. The kind support ofTaiatsu Techno Co., Ltd. is highly appreciated.

References

[1] T. Morita, M.K. Kurosawa, T. Higuchi, IEEE transactions on ultrasonics,Ferroelectr. Freq. Cont. 45 (1998) 1178–1187.

[2] T. Morita, Sens. Actuators, A 103 (2003) 291–300.[3] T. Kanda, T. Morita, M.K. Kurosawa, T. Higuchi, Sens. Actuators, A 83 (2000) 67–

75.[4] T. Kanda, Y. Kobayashi, M.K. Kurosawa, T. Higuchi, Jpn. J. Appl. Phys. 42 (2003)

3014–3019.[5] Yoshifumi Saijyo, Med. Imaging Technol. 21 (2003) 101–105 (in Japanese).[6] C.T. Lancee, J. Souquet, H. Ohigashi, N. Bom, Ultrasonics 23 (1985) 138–142.

[7] M. Ishikawa, M.K. Kurosawa, A. Endoh, T. Ushida, S. Takeuchi, N. Kawashima,in: Proc. of the IEEE Ultrasonics Symposium, 2006, pp. 2369–2372.

[8] T. Kikuchi, T. Tsurumi, Y. Ohba, M. Daimon, Jpn. J. Appl. Phys. 31 (1992) 3090–3093.

[9] Y. Ohba, M. Miyauchi, T. Tsurumi, M. Daimon, Jpn. J. Appl. Phys. 32 (1993)4095–4098.

[10] K. Shimomura, T. Tsurumi, Y. Ohba, M. Daimon, Jpn. J. Appl. Phys. 30 (1991)2174–2177.

[11] Y. Kadota, M. Ishikawa, H. Hosaka, T. Morita, IEEE transactions on ultrasonics,Ferroelectr. Freq. Cont. 56 (2009) 9–13.

[12] P. Bornmann, T. Hemsel, R. Ageba, Y. Kadota, T. Morita, J. Korean Phys. Soc. 57(2010) 1122–1126.

[13] A. Rujiwatra, C. Wongtaewan, W. Pinyo, S. Ananta, Mater. Lett. 61 (2007)4522–4524.

[14] R. Ageba, Y. Kadota, T. Maeda, N. Takiguchi, M. Ishikawa, P. Bornmann, T.Hemsel, T. Morita, J. Korean Phys. Soc. 57 (2010) 918.

[15] M. Ishikawa, M.K. Kurosawa, A. Endoh, S. Takeushi, Jpn. J. Appl. Phys. 44 (2005)4342–4346.

[16] Y. Ohba, K. Arita, T. Tsurumi, M. Daimon, Jpn. J. Appl. Phys. 33 (1994) 5305–5308.