Realization of high performance PZN-PT single crystal based piezoelectric flexural mode hydrophone for underwater sensor applications

In this article, attempts are made to grow large size PZN-PT single crystals using high temperature solution growth method by implementing novel bottom cooling technique. The grown crystals are oriented and poled along 〈001〉 direction and obtained larger piezoelectric strain coefficient (d33 > 2000 pm V−1) suitable for development of underwater acoustic sensor requirements. Flexural mode hydrophone is realized using the oriented PZN-PT single crystal discs. Finite element modeling is employed to examine the design of the flexural mode hydrophone and an equivalent circuit model is also applied to study its acoustic characteristic at two extreme boundary conditions like simply supported and clamped edge condition. The underwater acoustic response of the PZN-PT single crystal based flexural mode hydrophone is evaluated over the frequency range (100 Hz to 12 kHz) and its responses are compared with the FEM and equivalent circuit model results. The predicted results from FEM and equivalent circuit model are found to be in good agreement with the experimental results. The receiving sensitivity of the PZN-PT single crystal-based hydrophone is 12 dB higher than the PZT 5A based hydrophone in the frequency range of 2 kHz to 6 kHz. The fabricated PZN-PT single crystal-based hydrophone offers better performance than the conventional piezo ceramic based flexural hydrophone.


Introduction
Hydrophone that senses pressure variations of acoustic waves in underwater is indispensable element in underwater acoustic systems [1,2]. It is significantly used in wide range of applications for both military and civilian purposes like acoustic communication, detection, imaging, monitoring, surveillance and exploration etc, [3][4][5][6][7][8][9][10][11][12]. Depending on how the acoustic pressure is converted into electrical signal, hydrophones are categorized into piezoresistive, piezoelectric and optical systems [13][14][15]. Among them piezoelectric hydrophones are widely used for its versatility and optimal performance. The comprehensive performance of the hydrophone is strongly associated with the functionality of an active material [1]. Till now most of the piezoelectric hydrophones are utilizing PZT based ceramic as an active element and it has reached their performance plateau in terms of its piezoelectric functionality. Also, much of the research to improve the comprehensive performance of the hydrophone is focused on the signal processing, measurement techniques and physics of the hydrophone [16][17][18]. Therefore, a material with superior piezoelectric functionality than PZT ceramics is highly required for enhancing the performance of the hydrophone and in turn, the underwater acoustic systems [19,20].
The advent of relaxor-PbTiO 3 single crystals has enticed extensive interests among researchers in the field of underwater acoustics due to its incredible piezoelectric properties. Relaxor-PbTiO 3 single crystals such as PMN-PT and PZN-PT has longitudinal piezoelectric coefficient d 33 > 2000 pC/N, electromechanical coupling coefficient k 33 > 0.9 and low dielectric loss, tan δ < 0.1 which are very much essential to enhance the performance of the piezoelectric hydrophones [21,22]. Both PZN-PT and PMN-PT single crystals have almost similar material properties but PMN-PT crystals show larger variations and inconsistencies in their properties, especially for the crystals of compositions around the Morphotropic Phase Boundary (MPB). This undesirable effect is mainly due to the compositional segregation during the growth process. However, flux grown PZN-PT single crystal show better characteristics than melt grown PMN-PT single crystals, like better compositional homogeneity, larger and consistent dielectric and electromechanical properties and higher depolarization temperature [23,24].
Both PMN-PT and PZN-PT single crystals are already used as active materials in the hydrophones [25,26]. However, the design of the hydrophone is complex and uses the multiple single crystal elements positioned in a specific way which makes them difficult in the hydrophone construction. In this article, the piezoelectric hydrophone is constructed in a simpler twin bilaminar configuration where the PZN-PT single crystal is used as sensing element. PZN-PT single crystal is grown using inhouse developed modified flux growth method in a larger dimension (edge length > 20 mm). The optimal piezoelectric properties are obtained by orienting and poling the PZN-PT single crystal along the 〈001〉 direction. Finite element analysis package ATILA is used to design and optimize the various parameters in constructing flexural mode hydrophone. The performance of fluid backed flexural mode configuration of PZN-PT based hydrophone is analyzed using equivalent circuit model. The fabricated PZN-PT single crystal-based hydrophone shows better performance than the conventional PZT ceramics-based hydrophone.

Growth of PZN-PT single crystal
Single crystals of (1-x) Pb(Zn 1/3 Nb 2/3 )O 3 -xPbTiO 3 at MPB compositions have shown exceptional piezoelectric coefficients that are expected to offer significant improvement in the performance of polycrystalline PZT ceramic based hydrophone [27,28]. However, developing large high-quality single crystal is difficult because of the intricacies and complex thermodynamic behavior involved during the growth process. The important aspects to be taken care in growing PZN-PT single crystal are (i) high vapour pressure of PbO, (ii) incongruent melting & (iii) compositional inhomogeneity. In addition, the perovskite phase of PZN-PT single crystal is not stable at high temperature which eventually decomposes to non-piezoelectric pyrochlore phase. These factors suggest that PZN-PT single crystal can be grown only from high temperature solution growth method [29]. The use of flux guarantees the reduction in crystal growth temperature thereby reducing the PbO volatility which increases the chance of phase pure single crystal formation. Further, the flux grown PZN-PT single crystals possess good chemical homogeneity in comparison to the melt grown PMN-PT crystals [24]. In this work, PZN-PT single crystals are grown by modified flux method using PbO as a solvent by spontaneous nucleation process. The experimental procedure and details of the modified flux growth used for growing PZN-PT single crystals at morphotropic phase boundary can be found in [28]. The large size PZN-PT single crystal grown using modified flux growth method is shown in figure 1.

Fabrication of PZN-PT single crystal discs
PZN-PT single crystal has to be realized in the shape of a circular disc to fabricate the hydrophone in a flexural mode configuration. As grown single crystal are fabricated (made into) into disc shape by subjecting it to consecutive processes such as orienting, dicing, lapping, polishing, electrical contact making and poling and the optimized process flow for the fabrication of PZN-PT discs are shown in the figure 2. Since, the material properties of the single crystals are direction dependent and it is necessary to orient the single crystal in the particular direction to obtain the desired properties. Initially, the as grown single crystals are oriented in the 〈001〉 direction using Laue diffraction method. Then, the oriented single crystals are diced and circular disc shaped using 0.125 mm thick diamond wire saw cutting system (STX-203, MTI Corporation, California, USA) at the rate of 0.15 mm min −1 . Circular discs are then lapped and polished using series of silicon carbide (SiC) sheets with the roughness ranging from 8 μm to 1 μm and alumina colloidal solution of particle size ranging from 0.5 μm to 0.05 μm. Lapping and polishing process ensures that the surface of single crystal is without the layer of deformed crystal structure which is formed due to high strain produced during cutting, dicing and lapping processes. Then, using DC magnetron sputtering, Au/Cr electrodes of thickness 300/30 nm are deposited on both sides of the polished single crystal surfaces.
PZN-PT single crystals have strong anisotropic characteristics where the application of external electric field along different directions leads to different domain engineered configurations. PZN-PT single crystals poled along non-polar direction leads to multi-domain state which displays higher longitudinal piezoelectric coefficient. The electric field strength and temperature are the two important parameters that determine the outcome of poling process which in turn affect its piezoelectric properties. Single crystals are poled in silicone oil at room temperature by DC contact poling method. The process of poling is sensitive to the applied electric field. Low electric field amplitude results in non-poling of the entire sample, while larger magnitude of electric field can lead to degradation of its properties termed as 'overpoled' [24].
PZN-PT single crystals are subjected to the conventional Direct Current (DC) poling process using aixACCT poling instrument. Since the T c /T R-T for the PZN-PT single crystal is relatively low, poling is carried out at room temperature. In general, for the PZN-PT single crystal, maximum piezoelectric properties are obtained for the poling field of 4. 0-6.0 kV cm −1 . Constant electric field i.e., direct current (DC) is applied between the electrodes which are aligned along the [001] direction i.e., plane parallel to the thickness direction. In the present investigation, PZN-PT single crystals are optimally poled at the electric field of 5-5.5 kV cm −1 .
The piezoelectric strain coefficient (d 33 ) is calculated from electric field induced unipolar strain loop which is measured using laser interferometer (SIOS SP-S 120 E) with ferroelectric test system (TF analyser 2000, aixACCT) Unipolar strain of the poled PZN-PT single crystal disc is measured after 24 h of poling process which is shown in the figure 3. The unipolar strain with respect to electric field shows a non-hysteric behavior.
The piezoelectric strain coefficient (d 33 ) of PZN-PT single crystal discs displays high value ∼ 2200 pm V −1 which is very essential to enhance the receiving sensitivity (M) of the hydrophone. The hydrophone with high receiving sensitivity is greatly required, which reveals how much electrical signal can be generated per unit acoustic pressure incident normally on to the active element [2], where g 33 is the piezoelectric voltage coefficient and t is the thickness of the disc. PZN-PT single crystal disc shows high dielectric constant ∼ 5140 and a low dielectric loss 0.059 at the frequency of 1 kHz, as shown in the figure 4. The high dielectric constant facilitates better electrical matching with low electrical impedance. Low dielectric loss is essential for the hydrophones particularly operating at far below the resonance frequency where the energy dissipation is predominantly due to the dielectric loss of the active material. PZN-PT single crystal disc exhibits higher piezoelectric voltage coefficient (g) which has direct impact on the figure of merit (FoM) of the hydrophone.
The figure of merit of the hydrophone is directly proportional to the voltage coefficient (g) and charge coefficient (d) that also includes the noise generating factor, tan δ, to highlight its importance [2],

Construction of PZN-PT flexural disc hydrophones
PZN-PT single crystal hydrophone is fabricated based on the flexural mode configuration which produces high voltage output for small mechanical excitations and conversely produces high strain for an applied electrical excitation. The Flexural disc hydrophone design offers a convenient choice for utilizing the disc-shaped piezoceramic samples [30]. In order to demonstrate the capabilities of PZN-PT piezoelectric material, a flexural disc hydrophone is constructed and evaluated. In this configuration, two discs of PZN-PT samples are bonded on either side of a Be-Cu substrate using epoxy resin. The diameter and thickness of the PZN-PT discs are 10 mm and 0.5 mm, respectively, and those of the metal substrate are 25 mm and 1.2 mm, respectively. A coaxial  cable is attached to the discs to provide electrical connection. The composite flexural disc is attached to one end of a brass cup with 25 mm outer diameter and 2 mm wall thickness. This assembly is mounted inside a prefabricated rubber cup. The front face is moulded with 1 mm thick polyurethane to make it water worthy. Figure 5. shows the schematic diagram (a) of the hydrophone and the photograph (b) of the assembled and moulded hydrophone.

Finite element analysis
A Finite element analysis model is developed using ATILA code to verify the functional design of the PZN-PT hydrophone which has direct impact on its receiving sensitivity. ATILA is a finite element software package specifically developed by the analysis of two-or three-dimensional structures that contain piezoelectric and magnetostrictive materials. ATILA software package is developed by ISEN France and marketed by    [31]. The geometry of the hydrophone used for the modelling studies are shown in figure 6. A 2-dimensional axisymmetric model is implemented because of the simplicity in the geometry and it enables the use of only the one half of the cross-section shown in the figure which is sufficient for the FEM analysis. Rotation about the symmetry axis generates the complete model in 3-dimension. The electrodes on the top and bottom surfaces of the PZN-PT discs are assigned with 1 V and 0 V of electrical excitation, respectively. Polarization is defined along the thickness direction. Acoustically transparent rubber is used for encapsulating the hydrophone which is surrounded by water medium on all sides. The outer surface of water region is assigned with non-reflecting boundary condition simulating infinite extent of the water medium. All the external surfaces are assigned with stress-free boundary condition. All the faces are assigned with eight-nodded QUAD elements with structured mesh configuration. The size of the mesh is smaller than λ/5 at the highest frequency of interest and the aspect ratio is maintained smaller than 1:5 in order to ensure good quality mesh in all material regions. Harmonic analysis is performed with 100 frequency points in the band 100 Hz to 12 kHz. The complete finite element mesh of the hydrophone surrounded by the water medium is shown in figure 7. The material parameters of various material faces shown in figure 6 and used in the FEM studies are presented in table 2.

Equivalent circuit modelling
Electrical equivalent circuit modelling is presented to substantiate the results obtained from the Finite Element Modelling (FEM) results and also the experimental results of the fabricated PZN-PT flexural disc hydrophone. Using this model, the upper-limit and lower-limit of the receiving sensitivity is predicted for two extreme boundary conditions, namely, simply supported and clamped edge conditions. In practice, the acoustic response of a hydrophone will always lie in between the responses corresponding to these two extreme boundary conditions.
The lumped-parameter equivalent circuit of the flexural disc hydrophone is shown in figure 8 [30]. The parameters C 0 and R 0 are the capacitance and resistance, respectively, the parameters R 1 , L 1 and C 1 are the  electrical equivalent of mechanical parameters, namely, loss factor, mass and compliance, respectively and C b is the compliance of backing volume. These circuit parameters are required as input to the model, in addition to the dimensional details of the hydrophone. They are determined from the measured electrical impedance spectrum shown in figure 8, using the method described elsewhere [33]. Dimensions, material parameters and equivalent circuit parameters used in the equivalent circuit model are given in table 3. As shown in figure 10, P is the pressure amplitude of the incident acoustic wave and V oc is the voltage developed by the hydrophone in response to the incident acoustic wave. The ratio between these two quantities gives the Receiving Sensitivity of the hydrophone in the frequency band that includes the first resonance.

Experimental results and discussion
The Impedance analyser (Agilent 4294A) is used for measuring the resonance characteristics of the fabricated PZN-PT hydrophone. The electrical admittance spectrum is measured for the frequency range from 100 Hz to 12 kHz and recorded with 1600 points in the measured frequency range. The fabricated PZN-PT hydrophone is found to resonate at about 8 kHz. The electrical admittance spectrum is shown in figure 9.
The underwater acoustic characteristics of the PZN-PT hydrophone are determined experimentally in an open acoustic tank. The schematic of the acoustic measurement setup used for measuring the receiving voltage Table 2. Material parameters of various components of the hydrophone shown in figure 3 and used in FEM studies.
Youngs modulus (10 9 Pa) Y 125 Poisson's ratio σ 0.33 Density (kg m −3 ) ρ 8200 Rubber [2] Youngs modulus (10 9 Pa) sensitivity is shown in figure 10. The standard, comparison-calibration method is used to measure the frequency dependent receiving sensitivity of the hydrophone. The test hydrophone (fabricated PZN-PT hydrophone), the standard hydrophone (B&K, 8104) are positioned collinearly at a depth of 1.5 m in the acoustic tank. The  standard projector (Massa Corp., TR25) is used as an acoustic source which is placed in line with the test and standard hydrophone. A Linear Frequency Modulation (LFM) signal is generated and amplified using power amplifier (Instruments Inc., L6) and fed as an input to the projector. The projector transmits the acoustic signals which are received by the test and the standard hydrophones as shown in figure 10. The transmitted signal received by the test and standard hydrophone are then amplified by pre-amplifier and recorded in the automatic data acquisition system (National instruments). The LabVIEW software is used for performing the operations such as signal generation, processing and data acquisition. The Receiving Sensitivity is determined from the measured voltage across the test hydrophone for unit input acoustic pressure. The measured Receiving Sensitivity of the PZN-PT hydrophone is verified and compared with those obtained by finite element modelling and is shown in figure 11. The low frequency sensitivity is about −180 dB (1V/μPa) and the peak value at resonance is about −170 dB. These values are much higher than those normally observed in PZT-based hydrophones of comparable size. It is seen from the figure that the agreement between the experimental and finite element modelling results of sensitivity is fairly good at frequencies below resonance, which is usually the receiver operating band in underwater applications [2]. The resonance frequencies in both cases are comparable, however, the predicted value of sensitivity is slightly higher than the measured values in the vicinity of resonance. This is because, the Finite Element Model assumes that the bonding between piezoceramic disc and the substrate is ideal and lossless, whereas in practice, the bonding glue is a lossy material which is found to reduce the sensitivity of the hydrophone at resonance.
The experimental and FEM results are further validated by using electrical equivalent circuit model described in section 6. The receiving sensitivity of the PZN-PT hydrophone is determined under two boundary conditions, namely, simply supported edge and clamped edge. These two conditions set the theoretical upper limit and lower limit of receiving sensitivities, respectively [30]. A comparison of experimental receiving sensitivity plot with the theoretical limits predicted by equivalent circuit modelling is given in figure 12. It is seen from the figure that the measured plots lie in-between these two limits, but closer to the lower limit that corresponds to clamped boundary condition. This indicates that the flexural disc is partially clamped by the process of assembly and moulding. The results obtained by the three methods described in the present work are consistent, mutually supportive and validate each other.
The relative performance of the PZN-PT hydrophone is also studied in comparison with that of the conventional PZT-based hydrophone. Figure 13 shows the measured Receiving Sensitivity plots of two hydrophones; PZN-PT single crystal discs and PZT 5A ceramic discs. Same set of passive components are used in both the hydrophones and the only difference being the type of piezo discs. It can be seen from the figure that the Receiving Sensitivity of the PZN-PT single crystal hydrophone is ∼ 12 dB higher than that of PZT 5A hydrophone at low frequencies. It is demonstrated that PZN-PT SC hydrophones have superior acoustic performance over conventional piezoceramics and are found to be very useful for underwater applications.   The performance of the proposed hydrophone can be improved by using different crystal orientation which allows to engineer the piezoelectric domains in a specified direction to maximize the material properties. Improvement in the material design can improve the performance of the proposed hydrophone. The limitation of the proposed hydrophone is that the operating temperature is below 160°C that is the curie temperature of PZN-PT single crystals.

Conclusions
Device-quality PZN-PT single crystals are grown by modified flux growth technique. Subsequently, PZN-PT single crystals are subjected to a series of successive processes such as dicing, lapping and polishing to fabricate into circular discs and then poled along 〈001〉 crystallographic direction to achieve high piezoelectric coefficients. The dielectric constant (ε 33 T /ε 0 ) and the piezoelectric charge coefficient (d 33 ) of these crystals are about 5140 and 2200 pm V −1 , respectively. A flexural mode hydrophone is designed and fabricated using these PZN-PT single crystal discs as active material. The hydrophone design is analyzed by Finite Element Modelling studies using ATILA code. In order to verify the design and the FEM results, an electrical equivalent circuit model is also performed. The underwater acoustic characteristics of the fabricated PZN-PT single crystal hydrophone is experimentally measured and compared with the results obtained from Finite Element Modelling and equivalent circuit modelling studies. The model results agree well with that of the experiments. The fabricated PZN-PT single crystal hydrophone shows Receiving Sensitivity values above −180 dB (1V/μPa) over a wide band, which is 12 dB higher than that of the conventional PZT-based hydrophone, and shows a greater potential for use in next generation underwater acoustic devices.