With the advancement of the era, the demand for ultraviolet photodetectors in environmental monitoring and communication security continues to grow, leading to increasingly stringent performance requirements. In this case, self-driven ultraviolet photodetectors emerge to meet the needs of energy conservation and device miniaturization. However, conventional self-driven ultraviolet photodetectors still face some challenges such as low efficiency in photo-generated carrier separation, poor photoresponse performance, and limited response speed. Ferroelectric thin films with a high remnant polarization can form a depolarization field that penetrates the entire bulk material, enabling an effective separation of internally generated photo-generated electrons and holes. This provides a promising solution to the aforementioned issues. Pb(Zr_xTi1-x)O₃(PZT) is a typical ABO₃-type perovskite ferroelectric material. This material is widely used in ferroelectric memories, micro-electromechanical systems, and photodetectors due to its excellent ferroelectric, piezoelectric, and photoelectric properties. Extensive studies show that the composition significantly affects the crystal phase structure and ferroelectric properties of PZT thin films. However, its impact on the photoelectric performance requires a further systematic investigation. This work was to analyze the photoelectric characteristics of PZT thin films with different Zr/Ti ratios, in order to elucidate the influence of compositional modulation on the photoelectric effect.

Pb(Zr_xTi₁-x)O₃ ferroelectric thin films with different Zr/Ti ratios were prepared by a sol-gel method. The selected precursors and solvents were high-purity lead acetate [Pb(CH₃COO)₂·3H₂O], zirconium n-propoxide (C₁₂H₂₈O₄Zr), and titanium isopropoxide (C₁₂H₂₈O₄Ti) as sources for lead, zirconium, and titanium, respectively, glacial acetic acid as a chelating agent, and n-propanol as a stabilizer. Semitransparent gold electrodes were deposited on the surface of the thin films by a model VZZ-300 high-vacuum thermal evaporation system (VANNO Co., China) to fabricate self-driven ultraviolet photodetectors with an Au/PZT/FTO vertical structure. The crystal structure of the films was characterized by a model D8 Advance X-ray diffractometer (XRD, Bruker Co., USA). The surface roughness of the films was determined by an atomic force microscope (AFM, Bruker Dimension Edge Co., USA). The optical properties of the films were analyzed by a model UV-3600 Plus ultraviolet-visible-near-infrared spectrophotometer (UV-Vis-NIR, Shimadzu, Japan). The ferroelectric properties were measured by a model Precision LC II ferroelectric test system (Radiant Co., USA). The current-time (I-t) curves were obtained by a model Keithley 2400 source meter, with a 150 W ultraviolet-enhanced xenon lamp as a light source.

The XRD patterns indicate that the three prepared PZT thin films with different Zr/Ti ratios all exhibit a typical perovskite structure, and no diffraction peaks from impurity phases appear aside from those originating from the FTO substrate. As the Zr content increases, the surface morphology of the films transitions from elongated needle-like structures to island-like structures, and finally to wavy undulations. The lowest root mean square (RMS) roughness of 2.62 nm is obtained at a Zr/Ti ratio of 0.52:0.48. The remnant polarization first increases from 27.9 μC/cm² (Zr/Ti=0.49/0.51) to a maximum of 33.2 μC/cm² (Zr/Ti=0.52/0.48), and then decreases to 31.1 μC/cm² (Zr/Ti=0.55/0.45) as the Zr content increases. A higher remnant polarization is beneficial to forming a stronger built-in electric field, thereby improving the separation efficiency of photogenerated carriers. The PZT films with different Zr/Ti ratios are all wide-bandgap semiconductors (>3.6 eV). As the Zr content increases, the bandgap widens from 3.60 eV to 3.68 eV, showing a blue-shift trend. When a negative poling voltage is applied, the depolarization field inside the PZT aligns with the built-in field induced by the interfacial Schottky barrier, synergistically enhancing the driving force for carrier separation and leading to a significant increase in photocurrent. For the sample with a Zr/Ti ratio of 0.52:0.48 at a poling voltage of -2 V, the responsivity and detectivity reach 3.2 mA/W and 0.33×10¹¹ Jones, respectively. Even under a weak illumination of as low as 0.17929 mW/cm², the device still generates a photocurrent of 1.02 nA, demonstrating the excellent detection sensitivity.

Pb(Zr_xTi₁-x)O₃ ferroelectric thin films with different zirconium-to-titanium ratios (i.e., Zr/Ti=0.49:0.51, 0.52:0.48, 0.55:0.45) were fabricated by a sol-gel method, and self-driven ultraviolet photodetectors with an Au/PZT/FTO structure were constructed. The structural characterization revealed that all PZT thin films exhibited a pure perovskite phase with a good crystalline quality. The AFM analysis indicated that the film surfaces were smooth, dense, and displayed a uniform grain distribution. The ferroelectric property measurements further confirmed that all films with different Zr/Ti ratios showed characteristic ferroelectric hysteresis loops and possessed a high remnant polarization, having a maximum value of 33.2 μC/cm² at Zr/Ti=0.52:0.48. The results of photoelectric tests demonstrated that the device based on this optimal composition exhibited stable and reproducible photocurrent responses. At a poling voltage of -2 V, its responsivity and detectivity were significantly enhanced. In summary, the rational adjustment of the Zr/Ti ratio could effectively optimize both the ferroelectric properties of PZT thin films and the photoelectric characteristics of the corresponding devices. This work could innovatively utilize the inherent bulk depolarization field of PZT ferroelectrics as a driving force to achieve an efficient separation of photogenerated electron-hole pairs. Moreover, it could provide a systematic optimization strategy from the perspectives of compositional design and polarization modulation, offering an effective material- and physics-based solution to overcome the key bottlenecks of low responsivity and detectivity in self-driven ultraviolet photodetectors.