**3. Results and discussion**

comparative study of materials is relatively difficult. From the perspective of minimizing the environmental load by avoiding the use of lead-containing materials, consideration of leadfree piezoelectric materials is valuable for energy-harvesting devices [13]. Therefore, in addition to PZT-based ceramics, barium titanate (BT)-based ceramics was evaluated for piezoelectric materials in this chapter. The performance of piezoelectric energy-harvesting devices that captured frequencies of 60 Hz for PZT-based and BT-based ceramics was evalu‐ ated and the figures of merit of the materials have been discussed in order to provide the guidelines of the piezoelectric material selections. The results using power management circuit

The commercial piezoelectric bimorph is used for basic operations in the power management circuit. And the piezoelectric unimorph were fabricated by remodeling the commercial bimorphs. The experimental setup for measuring the unimorph generator is shown in **Figure 6**. Details were described in our previous study [16]. Commercial bimorph was produced by using a patterned fiber-reinforced plastic (FRP) plate (105 mm × 10 mm × 1.6 mm) that was used as the base beam. Piezoelectric ceramics was adhered to both the side of the beam. The unimorph structure was fabricated by removing the piezoelectric ceramics from both the side and adhering PZT- or BT-based ceramics (10 mm × 18 × 0.5 mm) to one side of the FRP beam. The unimorph was attached to the vibration generator and oscillated with various frequencies and accelerations. And the characteristic frequency of the beam was 57

**Figure 6.** The experimental setup for measuring the unimorph generator. The unimorph constructed by the piezoelec‐ tric ceramics and FRP beam was attached to the vibration generator and oscillated with various frequencies and accel‐ erations. The displacements and accelerations of the unimorph were monitored by the acceleration sensor attached to

are included for evaluating the performance as the power source.

**2. Experimental**

134 Piezoelectric Materials

the other end during measurement [16].

The frequency dependence of the Vcharge from the piezoelectric unimorph is shown in **Fig‐ ure 7**. The Vcharge shows the maximum at 57 Hz, indicating that the maximum displacement of the bimorph provides the largest voltage output. The output voltages across the various load resistors were measured. **Figure 8** shows the load voltage delivered to load resistance by oscillating the unimorph with a relative displacement of 0.2, 0.4, and 0.8 mm and a frequency of 60 Hz. It is found that the voltage increases proportionally to the relative displacement. From the voltage, the output electric powers, P, at the load resistance were calculated using P=1/2(V2 /R) equation. The results are shown in **Figure 9**. It is noted that the electric powers varied with the load resistance. In all samples, the power increases with increasing resistance and showed that their peak power decreases with increasing resistance. The resistance that yielded maximum power varied with the samples, this resistance roughly correspond to the resistance of the samples. These results were very close to our previous results; however, the hard PZT ceramics yielded higher output voltage but smaller power in this study. The reason for this phenomenon is not yet clear. In the case of the hard PZT, the electrical damping may be larger [13]. Unimorph that used PZT with higher Tc generates larger voltage and power. Concerning generators that used BT ceramics, Mn-doping was effective. Mn-addition raises *Q*m in BT ceramics, and PZT with higher Tc exhibits higher *Q*m. The results of this study suggest that hard piezoelectric properties would be favorable for a piezoelectric generator. The relation between maximum voltage or power and the materials' parameter from the unimorph using various ceramics is shown in **Figures 10** and **11**, respectively. For maximum voltage, *g*<sup>31</sup> is considered to be good parameter. For maximum power, *d*31*g*31/tanδ, *k*312 *Q*m, and *d*31*g*31 are close to the behavior of the maximum power; however, hard PZT ceramics yielded smaller power, compared with the expected values estimated from the material parameters.

**Figure 7.** Frequency dependence of Vcharge voltages from the unimorph generator.

P=1/2(V2

136 Piezoelectric Materials

/R) equation. The results are shown in **Figure 9**. It is noted that the electric powers

*Q*m, and *d*31*g*31 are close

varied with the load resistance. In all samples, the power increases with increasing resistance and showed that their peak power decreases with increasing resistance. The resistance that yielded maximum power varied with the samples, this resistance roughly correspond to the resistance of the samples. These results were very close to our previous results; however, the hard PZT ceramics yielded higher output voltage but smaller power in this study. The reason for this phenomenon is not yet clear. In the case of the hard PZT, the electrical damping may be larger [13]. Unimorph that used PZT with higher Tc generates larger voltage and power. Concerning generators that used BT ceramics, Mn-doping was effective. Mn-addition raises *Q*m in BT ceramics, and PZT with higher Tc exhibits higher *Q*m. The results of this study suggest that hard piezoelectric properties would be favorable for a piezoelectric generator. The relation between maximum voltage or power and the materials' parameter from the unimorph using various ceramics is shown in **Figures 10** and **11**, respectively. For maximum voltage, *g*<sup>31</sup> is

to the behavior of the maximum power; however, hard PZT ceramics yielded smaller power,

considered to be good parameter. For maximum power, *d*31*g*31/tanδ, *k*31<sup>2</sup>

**Figure 7.** Frequency dependence of Vcharge voltages from the unimorph generator.

compared with the expected values estimated from the material parameters.

**Figure 8.** Load resistance dependence of the output voltage from the unimorph using (a) PZTh325 and (b) PZTs140 ceramics.

**Figure 9.** Load resistance dependence of the output power from the unimorph using various ceramics. The output electric powers, P, at the load resistance, R, were calculated using P=1/2(V2 /R) equation from the output voltage, V.

**Figure 10.** The relation between maximum voltage and various material parameters.

**Figure 11.** The relation between maximum power and various material parameters.

The charging voltage (*V*charge), oscillation signal (vibration), and Pgood were measured by oscillating the unimorphs consisted with PZTs330 and BT-Mn at 60 Hz for 45 sec. The results are shown in **Figure 12(a and b)**. The Vcharge keeps constant voltage during oscillation, and the voltage damped shortly after the quitting the oscillation. Time constant of the damped oscillation was 3.4 s for both measurements. Considering that the capacitance (*C*) of the charging capacitor is 22 μF, the damping behavior is reasonable. Pgood keeps constant during oscillation and a few seconds after quitting the oscillation for both measurements, showing the unimorph EH generators used in this study have the capability of the power sources. The stored energy is calculated by using the (1/2)*CV*in <sup>2</sup> equation. The stored power is calculated by dividing the stored energy by an oscillation time of 45 sec. The maximum output voltage and power across the load resistance, stored power, and energy in the capacitor and Pgood time are summarized in **Table 3**. The relations among voltage, power, and energy are roughly related; however, the relations are not so simple with careful evaluation. For example, maximum powers are dependent on the load resistance, while the stored power and energy are more closely related to the output voltage. The operation of the power supply integrated circuit might require voltage to the high impedance input from the piezoelectric generators.

**Figure 12.** VCharge, oscillation signal (vibration), and Pgood measured by oscillating the unimorphs consisted with (a) PZTs330 and (b) BT-Mn at 60 Hz for 45 sec.


**Table 3.** Performances of the piezoelectric generator with piezoelectric ceramics [16].

**Figure 10.** The relation between maximum voltage and various material parameters.

138 Piezoelectric Materials

**Figure 11.** The relation between maximum power and various material parameters.

The charging voltage (*V*charge), oscillation signal (vibration), and Pgood were measured by oscillating the unimorphs consisted with PZTs330 and BT-Mn at 60 Hz for 45 sec. The results
