The sintering temperature can be determined by DSC in Fig. 1. The ceramic powder was evenly spread into a uniform thin layer at the bottom of the crucible. In the dynamic atmosphere of nitrogen, the gas flow was 20 mL/min, and the temperature rose from room temperature to 1300 ˚C at a heating rate of 10 ˚C /min. The endothermic peak around 880 ˚C is due to the dissolution of PbO (the melting point of PbO is 886 ˚C). In the range of 1000 ~1150 ˚C, there are many exothermic peaks and obvious weight loss, indicating that the ceramics begin to react and crystallize within this temperature range. There is an endothermic peak near 1300 ˚C, and a lot of weight loss phenomenon. Combined with the analysis of the actual sintering process, the temperature is too high that the ceramics appear over melting phenomenon. Therefore, the sintering temperature range of ceramics was determined to be 1150 ~1200 ˚C.
In Fig. 2(a), the main crystalline phase of all samples is perovskite structure((ICSD-89-8012). And no obvious impurity phase was found, which means the addition of appropriate amount of PbO can offset the dissolution of PbO in sintering process and maintain the pure perovskite structure. In Fig. 2(b), the diffraction peak near 44°was split into two diffraction peaks of (200) and (002), indicating that all samples are orthonormal phase. Compared with the undoped material, the intensity of the diffraction peak of the composite sample is higher, indicating that the crystallinity is higher. It can be predicted that the sample structure becomes denser. There is no obvious offset in the diffraction peak, so it can be inferred that Al3+ does not enter the main lattice of perovskite structure in large quantity, and Al2O3 is mainly distributed in the grain gap.
Figure 3 shows the surface microscopic morphology of the samples sintered in lead-free atmosphere and lead atmosphere respectively. Obviously, sintered in lead atmosphere, samples have lower porosity and higher density characteristics. However, there are still a few holes on the surface of all samples, which are caused by the inevitable volatility of lead. The average grain size of all samples is less than 3µm, and the density is increased as the composition of Al2O3 increases. This is particularly evident in the sample with x= 0.08(Figure 3(d, h)). With the increase of composite concentration Al2O3, there are more and more flake structure at grain spacing.
Figure 4 shows Al2O3 doping in PLZST materials and the EDS element distribution of the composite ceramic sample x=0.1. The collection point is located near the grain gap (Figure 4(b)). Thus, the area with pores has no element distribution. The distribution of Pb (Figure 4(c)), Zr (Fig. 4d), and O (Figure 4(e)) on the grain is relatively uniform, mainly came from pre-sintered perovskite structure PLZST. However, the distribution of Al element (Figure 4(f)) is uneven. A small part of Al element distributed in PLZST grain. Most Al elements are enriched in a horizontal line, corresponding to the flake in the SEM image in Figure 4(b). According to the XRD results above, although aluminum atoms are not involved in the phase formation of perovskite structure, but Al2O3 exists in the grain gap in the form of flake structure, effectively restraining abnormal grain growth. As can be seen from the Figure 4(a), Al2O3 realizes denser and smaller grain size PLZST materials.
Figure 5(a-f) show that the permittivity of all samples increases first and then decreases slowly with increasing temperature and reach the maximum value near Curie temperature (Tc). The two temperature platforms in the intermediate temperature curve are typical orthogonal opposite ferroelectric temperature transformation curves. The ortho phase changes from normal temperature to tetragonal antiferroelectric phase and then to cubic paramagnetic phase[23, 24]. The Curie temperature peak region of all samples is relatively broad. The contrast between samples x=0.01~0.03 and samples x=0, 0.04, 0.08 indicates that the proper combination of Al2O3 can delay the temperature relaxation phenomenon of PLZST. The dielectric constant of composite Al2O3 samples decreases since the Al2O3 has a low dielectric constant (~10). And the change value of ε is less than 500, which is beneficial to improve the temperature stability of ceramic materials. The dielectric loss increases during the heating process, which is mainly due to the relaxation loss. With the increase of Al2O3 composite amount, the loss of ceramic sample becomes stable at high temperature, indicating Al2O3 composite is beneficial to improve the thermal stability at high frequency. But with the increase of composite amount, the loss also increases. Figure 5(g) shows the dielectric spectrum at room temperature. With the increase of alumina compound amount, the peak value of dielectric constant decreases from 200 at x=0 to 125 at x=0.1. At the same time, the permittivity is stable within the frequency range of 100 Hz~1MHz. The change of Al2O3 compound amount has little effect on dielectric loss. Dielectric loss of all samples is less than 0.025 below 1 MHz.
Figure 6(b) and (c) show the energy storage parameters of each sample based on data collected from Figure 6(a) P-E loop results. In Figure 6(b), all composite samples show an increase in BDS compared to the base material. BDS increased from 16.04 kV/mm at x=0 to 25.57 kV/mm at x =0.02, and BDS increased by 60%. However, BDS decreased as x continue to increase, which is corresponding to other studies[19, 20, 22]. An excessive addition of Al2O3 will lead to more Al2O3 crystallites (confirmed in Fig. 3(g), (h) and Fig. 4(b)), causing the failure of the microstructure to be completely grown into ceramics. The maximum polarization intensity Pmax has the same trend as BDS. Pmax is 4.9 µC/cm2 at x=0 and 26.2 µC/cm2 at x=0.02. When the compound content is more than 0.03, Pmax decreases obviously. This means the increased BDS can’t make the polarization intensity reach the saturation intensity. In addition, when x=0.1, BDS and Pmax curves no longer fit. It can be inferred that the dielectric constant drops from 200 to 125 due to excessive Al2O3. At the end of the curve, the maximum polarization intensity finally reaches the saturation polarization intensity of the material.
In Fig. 6(c), for x=0, 0.01, 0.02, the maximum energy storage density Wst is 0.51, 1.92 and 4.92 J/cm3, and the effective energy storage density Wre is 0.32, 1.15 and 2.95J/cm3. The energy storage efficiency of all samples fluctuates around 60%. The performance of doped PLZST is generally better than pure PLZST. In the range of x =0.01-0.03, the performance is improved. And at x=0.02, the performance is improved the most. The energy storage efficiency is generally not high, less than 65%. The composite ceramics has the potential to be applied to high energy storage density dielectric ceramics, but it needs to improve the low energy storage efficiency.