Planar microphone based on piezoelectric electrospun poly(γ-benzyl-α,L-glutamate) nanofibers

Since the first piezoelectric crystal microphone developed by Nicolson using Rochelle salt in 1919,¹ piezoelectric materials have been widely used in various transducers.^2,3 Low sensitivity and inconsistent frequency characteristics have limited the commercial use of such transducers for airborne sound; however, their use for underwater sound became popular because they can operate over a wide range of static pressure.⁴ Although the piezoelectric property of poly(vinylidene fluoride) (PVDF) first reported in 1969 stimulated interests in its use for transducers having a broad frequency range,⁵ low depolarization temperature (80 °C) limited its widespread use, while attempts to increase the depolarization temperature were met with only moderate improvements.⁶ In 1996, space charged low density polypropylene (LDPP) electrets were developed that exhibited large piezoelectric coefficient (>150 pC/N) equaling those in crystalline and ceramic materials.⁷ The mechanical properties and polarization of LDPP electrets have been extensively studied for applications in flexible field effect transistors and ferroelectret accelerometers.⁸ Because fabrication of piezoelectric LDPP is relatively simple, it was once thought that this material could replace the crystal and ceramic materials used in microphones;⁷ however, the depolarization temperature (60 °C) of this material turned out to be even lower than that of PVDF and therefore could only be used under limited environmental conditions. Due to these reasons, there has been a long-standing need for improved piezoelectric materials that can be made into simple transducers and are also capable of working at high temperatures. Recently, Sessler and co-workers reported micro-patterned fluorethylene-propylene ferroelectret transducers with high piezoelectricity that can operate over a wide range of frequencies and temperatures.⁹ 

We recently reported fabrication of thermally stable piezoelectric nanofibers by electrospinning, poly(γ-benzyl-α,L-glutamate) (PBLG), a synthetic poly(amino acid) with stable alpha helical conformation.¹⁰ This helical polymer contains a large electrical dipole due to a collection of hydrogen bonds pre-aligned in the direction of the helical axes. When the polymer solution is subjected to electrospinning condition, the dipole of the PBLG interacts with the external electric field, and nanofibers are formed with individual PBLG helices poled in the direction of the fiber. This process created fibers with permanent polarity as evidenced by second harmonic generation microscopy and electric field-induced fiber bending experiment. After the electrospinning process, the macroscopic dipoles of helical PBLG molecules are tightly held together by intermolecular forces that are reported to be stable up to 130 °C in the solid state.¹¹ The fibers showed relatively large piezoelectricity: Compressive piezoelectric coefficient (d₃₃) of 20 pC/N and shear piezoelectric coefficient (d₁₄) of −1 pC/N.¹² The piezoelectricity was stable at 100 °C for more than 24 h; this, to the best of our knowledge, is one of the highest thermally stable piezoelectric coefficient reported for poled polymers.¹² The fibers can be hot pressed at 100 °C into continuous film without loss of piezoelectricity.

The thermal stability, strong permanent dipole, and ease of fabrication make the PBLG-based piezoelectric fibers and film an excellent candidate for simple planar piezoelectric transducers, especially for potential use as underwater microphone and vector sensors. Both the flat frequency response of pressure microphone and high sensitivity of velocity microphone are critical for making a vector sensor.

In this paper, we demonstrate velocity and pressure microphones fabricated by compressed electrospun PBLG nanofibers. A film fabricated by compression of directionally aligned electrospun PBLG fibers is attached to a Mylar film in such a way that the deformation of Mylar film induces the shear piezoelectricity (d₁₄) in PBLG film (see Sec. II, and Fig. 5b in Ref. 12). The frequency response of pressure microphone and beam pattern of velocity microphone are presented and compared with a commercial B&K microphone.

As previously described, a custom-made electrospinning setup consisted of a syringe pump (KD Scientific, Holliston, MA), a power supply (Gamma High Voltage Research, Ormond Beach, FL) and a rotating mandrel.¹⁰ A 1 ml TB syringe (Beckton Dickinson, Franklin Lakes, NJ) with a PrecisionGlide needle (0.5 in., 27 gauge) was loaded with PBLG solution (MW = 162 900, 100 mg/ml in dichloromethane) in the syringe pump. Electrospinning was conducted by applying −12 kV between the needle tip and the grounded rotating mandrel (2500 rpm) wrapped with aluminum foil target. The distance between the needle and mandrel target was 5 cm and flow rate of syringe pump was 2 ml/h.

A mat (15 μm) of directionally aligned PBLG fibers was peeled from the aluminum foil target after the electrospinning. It was loaded on a hot press machine, Wabash G30H-15-BP (Wabash, IN), and 1000 lb stress was applied for 30 min at 100 °C.¹² The microphone test sample was prepared by cutting the resulting film (15 μm) along the shear direction (45° to fiber direction) in a rectangular shape (see Fig. 5b in Ref. 12) and coating both the top and bottom surfaces with evaporated Al. This film (4 mm × 19 mm) was attached to a Mylar film (25 μm thicknesses and 25.4 mm diameter) only by its two opposing sides of long axis to create a diaphragm structure which produces shear stress (d₁₄) to the PBLG molecules when deformed. For double layer film, the two electrode PBLG films were glued using Spurrs^® low viscosity epoxy (Polysciences, Warrington, PA).

The chamber of pressure and velocity microphones was 2.54 cm in diameter and 1.7 cm in height. A diaphragm sample with PBLG attached on a layer of Mylar film was held inside the chamber by the top open ring of the microphone holder (Fig. 1). The velocity chamber was designed with an open backside exposing the PBLG film and diaphragm to sound, while the pressure microphone had a completely sealed back chamber. In both sensors, the PBLG thin film is deformed by the deflection of the Mylar diaphragm thus producing a voltage proportional to the diaphragm deflection. A custom-made microphone test setup consisted of a small (10 cm diameter) loud speaker, a computer with audacity software, a SRS 560 preamplifier (Stanford Research Systems) and B&K dual microphone supply 5935 (B&K, Naerum, Denmark). The frequency responses of the PBLG microphones were tested using this setup in a semi-anechoic chamber where the microphones were set 1 m away from the loud speaker. To calibrate the system, a standard B&K microphone (type 4165, $12$-in., −26 dBV/Pa) was used at the same location as the PBLG microphones. During the measurement, the software generated sound signals in discrete frequencies from 200 Hz to 10 kHz. The electrical signals detected by the PBLG microphones and B&K microphone were fed to the SRS 560 and B&K 5935 preamplifiers, respectively, and recorded by the computer. Background noise from the semi-anechoic chamber was removed by averaging 200 adjacent data points.

To achieve high signal to noise ratio, piezoelectric microphone needs to be built with materials of relatively large capacitance and high piezoelectric coefficient. We reported previously that electrospun PBLG fibers show the highest piezoelectric coefficient when deformed in the direction of the fiber axes (d₃₃ mode), which is also the direction of the poled dipoles. Unfortunately, this mode could not be used for microphone application because capacitance in the direction of the fiber is extremely small. When the fibers were compressed into thin film, its capacitance across the top and bottom surface of the film was relatively large (150 pF), and we accordingly thought that the film's d₁₄ mode or d₃₁ could be used for microphone construction. Because the shear piezoelectric coefficient d₁₄ is approximately −1 pC/N, which is an order of magnitude higher than that of the d₃₁ mode, we chose to explore piezoelectricity in d₁₄ mode for the microphone application. The poled PBLG film was cut at an angle 45° to the fiber direction as previously described¹² and glued to a Mylar film to create diaphragm structure. Compared with the Mylar film, the PBLG film features a smaller dimension, and it is held by the microphone holder only at two opposing ends of the film and not on its entire periphery. Thus sound waves only stretch the PBLG film along the long axis of the film. Because electrodes are deposited on the top and bottom surfaces of the PBLG film, the stretching motion creates shear piezoelectricity (d₁₄ mode) in the PBLG film. For microphone design (see Sec. II for details), we followed Watakabe, who reported that a cylindrical chamber of at least 10 mm diameter and 15 mm height is needed to obtain flat frequency response and high gain.¹³ 

In a piezoelectric diaphragm microphone, the open-circuit voltage from the piezoelectric film is given by

where V_g is the open circuit voltage in volts, Q_p is the charge generated by piezoelectric plate, and C_p is the capacitance of the piezoelectric plate.

Ichinose optimized the design of a piezoelectric diaphragm consisting of a piezoelectric plate and a base plate for transducer application.¹⁴ The charge generated in the piezoelectric plate can be given by

where k is a parameter related to the Young's modulus, thickness of piezoelectric plate and base plate; P_m is the sound pressure; d₃₁ is the piezoelectric coefficient; and D is the combined flexural rigidity of the piezoelectric plate and the base plate. This expression can be also written in d₁₄ mode as follows:

From Eq. (3), we can see that the output voltage from the microphone is independent of frequency and is proportional to the piezoelectric coefficient of piezoelectric material and diaphragm structure. We found that even a slight change in the film condition (due to wrinkles and thickness variation) can affect the resonance frequency of microphone devices. Therefore in this application, a layer of Mylar film was glued to the back of PBLG film to control the stiffness of PBLG samples.

Frequency response of the PBLG pressure microphone is shown in Fig. 2(a). It shows a flat frequency response (±3 dB) from 200 Hz to 4 kHz; this is consistent with the theoretical expectation from Eq. (3). When compared to B&K calibrated microphone, the PBLG microphone has 30 dB lower sensitivity. This is mainly because of the relatively low d₁₄ piezoelectric coefficient (−1 pC/N) of the PBLG film as compared to the high piezoelectric coefficient of LDPP electret polymers (>150 pC/N). However, due to PBLG's remarkably stable dipoles, the PBLG-based microphone is thermally stable and can withstand heat up to 100 °C while LDPP microphone loses its function around 65 °C. In addition, the simple planar structure of the PBLG-based microphone design makes the microphone fabrication process extremely simple, while the microphone's insensitivity against static pressure makes it ideal for under water use. All these factors make the PBLG-based microphone a very unique transducer that can potentially be utilized in harsh environmental conditions.

The velocity microphone detects the phase difference of impinging sound waves between the two sides of the Mylar diaphragm. The lightweight PBLG film produces an output voltage that is directly proportional to the motion of the Mylar diaphragm. The equation for velocity microphone is shown as follows:¹⁵ 

where Δp and l are the sound pressure difference and the sound path length from the front to the back of microphone chamber, respectively; P_m is the amplitude of impinging sound pressure; ω is the frequency; and θ is the angle between the incident sound direction and the direction normal to the plane of microphone sample. The main requirement for a velocity microphone is the directivity, which is the frequency response of microphone that varies in response to change in the impinging sound direction. From Eq. (4), it can be seen that the frequency response of velocity microphone should be a cosine function of angle of incidence of sound wave. So the minimum and maximum frequency response must be achieved at 90° and 0/180°, respectively.

The directional characteristics of the PBLG velocity microphone at three different frequencies (500 Hz, 1 kHz, and 2.5 kHz) were measured using the acoustic measurement setup where the facing angles between the microphone and the speaker were varied from 0° to 360°. The output signal at 0° was set to 0 dBV/Pa (baseline), and the values at other angles were determined relative to the 0° value. The directional characteristics of the velocity microphone is shown in Fig. 2(b), where the solid line drawing is a sine wave fitting curve and the experimental data are plotted at individual facing angles of 0°, 25°, 45°, 70°, 90°, 115° and then the other side of the lobe at identical facing angles. The resulting beam pattern is similar to the one from commercial velocity microphone. The maximum and minimum responses are detected at 0° and 90° facing angle, respectively; this is consistent with the theoretical expectation from Eq. (4).

It is known that the sensitivity of a velocity microphone is one of the critical parameters for the design of vector sensors. The PBLG velocity microphone shows an approximately 30 dBV/Pa difference between the sound waves impinging from two perpendicular directions; this meets the requirement for a velocity microphone. The results clearly show that the PBLG microphone is a true velocity microphone, and that it can be used for measuring the particle velocity of sound wave and for detecting the direction of sound wave by functioning as a vector sensor. Additionally, the similar chamber design for both the velocity microphone and the pressure microphone makes this microphone easy to be integrated into a vector sensor.

Microphones composed of multilayer films typically show higher sensitivity compared to monolayer system because of improvement in capacitance and mechanical strength. We fabricated a double layer PBLG film by laminating two electroded PBLG films in parallel using low viscosity epoxy. The frequency response of the double layer PBLG microphone was measured using the same measurement setup, and the result is shown in Fig. 3. The measurements show that the sensitivity of the microphone was improved by 1 dB, between 300 and 500 Hz. Theoretically, every doubled capacitance should result in 3 dB improvement in the output power of the microphone. The modest improvement could be due to the epoxy glue, which is significantly stiffer than the PBLG film, reducing the strain level of the PBLG film. Thus the experimental result is consistent with the theoretical expectation. This result also suggests that the sensitivity of PBLG microphone can be greatly improved by using laminated multilayer samples. However, damping of the sensitivity is seen starting from 4 kHz; this could be due to the weak mechanical strength of the PBLG diaphragm structure. We believe that this issue can be resolved by using more layers of laminated PBLG film and a thicker Mylar film, which will enhance both the mechanical property and the sound sensitivity of the PBLG microphone.

We fabricated a simple planar microphone from piezoelectric PBLG film that was produced by hot pressing, electrospun PBLG fibers as previously reported.^10,12 The PBLG film was cut at a 45° angle to the fiber direction and coated with Al on both the top and bottom surfaces before it was attached to a Mylar film (25 μm) to create a diaphragm structure. The microphone's responses as velocity and pressure sensors in the audio frequency range have been tested, and in both cases, the sensitivity was around −60 dBV/Pa, which is about 30 dB below that of a typical electret microphone. The directional response of the PBLG microphone showed that the PBLG microphone is a true velocity microphone that can be potentially utilized to fabricate small scale vector sensors. The thermal stability of the poled dipoles in the PBLG film and ease of fabrication make PBLG-based material a good candidate for a planar piezoelectric microphone for use in under water, which requires constant sensitivity independent of depth.

This work was supported by a grant from the Office of Naval Research (N00014-13-1-0167). The authors thank Steven E. Fick for his help in design and fabrication of impedance match circuit for the microphone measurement.