Possible Phase Transition and Band Gap Closing in Photoexcited Semiconductor CeZn3P3

Abstract

In f -electron systems, external fields, such as a magnetic field, and pressure play crucial roles in changing the electronic state, for example, in magnetic-field-induced valence transition and pressure-induced superconductivity. Optical light is another attractive external field source. For d-electron systems, many interesting phenomena, i.e., photoinduced metallic states in Mott insulators, photoinduced magnetization, and a photoinduced change in spin configuration, have been reported. The effects of light illumination on f -electron systems have been extensively studied for Eu chalcogenides. A photoinduced metal– insulator transition and a photoinduced magnetization have already been reported. However, the effects of light illumination on other f -electron systems have not been thoroughly studied. In this paper, the investigation of photoinduced phenomena in a Ce-based semiconductor, CeZn3P3, 12) is reported. CeZn3P3 crystallizes into a hexagonal ScAl3C3-type structure, in which Ce atoms form a two-dimensional triangular lattice. CeZn3P3 and other ScAl3C3-type compounds, such as YbAl3C3, are attracting much attention as low-dimensional quantum spin systems. A sintered polycrystalline sample was prepared by a standard solid-state-reaction technique. The product was evaluated using the powder X-ray diffraction pattern and found to show an almost single phase with a small amount of unknown impurities. The fundamental band gap of CeZn3P3 was estimated using the diffuse reflectance Rd spectrum (0.5–4 eV), which was obtained using a UV–visible spectrometer (Shimadzu UV-3600). After calculating the Kubelka–Munk function f ðRdÞ corresponding to the absorption coefficient, 1⁄2h f ðRdÞ 1=n (h : photon energy) was obtained, as shown in the inset of Fig. 1. Following the report on the indirectgap semiconductor SmZn3P3, 18) we employed n 1⁄4 2. The band gap of 0.365 eV was determined by the tangential line (see the solid line in the inset of Fig. 1) at the initial increase in 1⁄2h f ðRdÞ . The temperature dependence of electrical resistivity ðT Þ between 30 and 300K under light illumination was measured by a conventional DC four-probe method using a closed-cycle He gas cryostat. The optical source was a laser diode with a photon energy of 1.85 eV, which is sufficient for exciting electrons across the band gap. A parallelepiped sample with dimensions of 1:25 1:6 10 mm was cut from a pellet of CeZn3P3. The distance between the voltage electrodes was 3.5mm. All electrodes were covered by metal plates to reduce extrinsic photovoltaic effects. The optical light from the laser diode was focused on the area between the voltage electrodes. Figure 1 shows the measurement history of ðT Þ. Three alternating measurements of ðT Þ with light turned off and on were carried out. All ðT Þ curves show no thermal hysteresis. We kept the light on for data acquisition (about 7 h for one cycle between 300 and 30K). The illumination with a fluence rate of 4.5W/cm transformed the ðT Þ of the virgin sample denoted as 1st (0W/cm) into a nonsemiconducting ðT Þ with a sharp anomaly at Ta 1⁄4 240K. When we turned off the light, although the semiconducting behavior was recovered, the ðT Þ value was not perfectly reproduced [see ðT Þ denoted as 2nd (0W/cm)]. The nonsemiconducting behavior with a sharp anomaly appeared again when we illuminated the sample; however, the optical power necessary to reproduce the anomaly increased and Ta decreased. The third measurement without light also showed a decrease in ðT Þ compared with that of the second one. In the third measurement with light [3rd (15W/cm)], we could not observe the sharp anomaly, even under the maximum fluence rate. Although intense light gradually damaged the sample, ðT Þ under light illumination exhibited a marked increase in carrier number. The ðT Þ with the sharp anomaly at Ta suggests that a photoinduced phase transition is expected to occur. Further studies are required to clarify the origin of the anomaly. The increase in the temperature of the sample caused by the optical illumination is expected to be small, at least above 50K, since ðT Þ for 2nd (15W/cm) is nearly identical to that for 1st (4.5W/cm) with a different fluence rate. Between the measurements of ðT Þ for 2nd (0W/cm) and ðT Þ for 2nd (15W/cm), the fluence rate dependence of ðT Þ was measured (see Fig. 2). Below 11W/cm, the recovery of ðT Þ without light was carefully examined. ðT Þ with 3W/cm showed a weak anomaly at approximately 240K, which may be a precursor of a phase transition. The temperature of the anomaly decreased to 210K with increasing fluence rate (see the arrows in Fig. 2). When the fluence rate was increased, the gradual inhibition of the semiconducting behavior, resulting in decreased ðT Þ Fig. 1. (Color online) Measurement history of ðT Þ of CeZn3P3. The inset shows the plot of 1⁄2h f ðRdÞ 1=2 vs h . Journal of the Physical Society of Japan 82 (2013) 125001