Lithium niobate (LiNbO₃, LN) is currently the preferred electro-optic (EO) Q-switch material for military lasers owing to its ease of growth and processing, low cost, and nonhygroscopic nature. Major national projects such as aerospace projects, lunar exploration, and Beidou satellite navigation demand lasers that can operate over a wide temperature range. However, the performance of LN EO Q switches deteriorates considerably at subfreezing temperatures. Currently, researchers attribute the unsatisfactory performance of cold Q-switching to three reasons: stress birefringence, the pyroelectric effect, and the thermo-optic effect. Most related studies focus on improving cold performance, whereas studies that investigate the mechanism underlying degraded cold Q-switching performance are few. Consequently, the reasons for unsatisfactory cold Q-switching performance are contradictory and confusing, which limits the development of LN EO Q switches with high-temperature stability. In this study, we perform comprehensive investigations to determine the primary factors affecting cold Q-switching performance and fabricate LN Q switches with improved temperature stability.

First, the cold Q-switching performance of several LN Q switches is measured using a flash-lamp pumped Nd∶YAG laser, which is placed in a high?low temperature test chamber. The optical homogeneity of these Q switches at cold temperatures is characterized using high?low-temperature conoscopic interference technology, from which the variation in the birefringence at cold temperatures is clearly observed. Subsequently, we analyze the various factors that may affect the birefringence at cold temperatures and their influence characteristics. The distribution of the electric field created by pyroelectric charges inside the LN crystal is simulated using finite-element-analysis software. Combining the above with the EO effect of the LN crystals, we analyze the spatial-distribution characteristics of birefringence induced by the pyroelectric field and compare them with the experimental results of conoscopic interference. The pyroelectric coefficient of the LN crystals is measured using the dynamic current method, and the conductivity is measured using a high-resistance meter. For further experimental verification, we reduce LN thermochemically in the Li₂CO₃ powder atmosphere at 550 ℃ for 10 h. The Q-switching performance and optical homogeneity of the reduced LN crystals are measured at cold temperatures.

The conoscopic interference patterns at cold temperatures change significantly compared with those at room temperature. Notably, the interference patterns in different areas of the light cross-section of LN differ significantly but show a clear distribution regularity (Figs. 4?6). In the X-axis direction, the interference patterns are symmetrically distributed relative to the crystal center, whereas the interference patterns are the same in the Y-axis direction. The interference ring becomes a distorted ellipse, with the major axis oriented at an angle of ±45° from the X-axis, and the extinction areas transform into two point-like areas separated along the major axis. The greater the distance from the center of the crystal, the greater is the variation in the interference pattern. The interference patterns of multiple LN crystals show the same variation characteristics but with different degrees, which is consistent with the Q-switched output at cold temperatures (Fig. 2 and Table 1). The spatial-distribution characteristics of birefringence at cold temperatures imply that stress and thermo-optic effects are not the primary factors affecting cold Q-switching performance. The results of finite-element simulations indicate that the pyroelectric charges create an uneven electric field inside the LN crystals (Fig. 7). The electric-field components are in the X- and Z-axis directions. The electric-field distribution is independent of the Y-axis and symmetrical within the XZ plane. Combining the above with the EO effect of LN, we conclude that the spatial distribution of birefringence induced by the pyroelectric field is consistent with the experimental results of conoscopic interference. Therefore, the electric field created by the pyroelectric charges is regarded as the dominant factor contributing to the unsatisfactory cold Q-switching performance. The differences in the cold performance of the LN Q switches are attributed to their different pyroelectric coefficients and conductivities (Table 2). The reduced LN crystal (Fig. 9) exhibits a significantly higher conductivity and its cold Q-switching performance and optical homogeneity improve considerably (Table 4, Fig. 10, and Fig. 11), which further confirms the leading role of the pyroelectric field on the cold Q-switching performance.

We conduct a comprehensive investigation to determine the primary factors affecting the cold Q-switching performance of LN EO Q switches. The variation in the birefringence at cold temperatures and its spatial-distribution characteristics are intuitively observed via cold-temperature conoscopic interference experiments. The effects of stress and thermo-optic effects are excluded because their influence on birefringence are inconsistent with the experimental results. The results of finite-element simulations reveal that pyroelectric charges create a spatially variable electric field inside the LN crystal. Furthermore, the spatial distribution of birefringence induced by the electric field via the EO effect is consistent with the experimental results. Therefore, the pyroelectric field is regarded as the primary factor affecting the cold Q-switching performance. LN crystals with significantly higher conductivities are achieved via thermochemical reduction. The cold Q-switching performance and optical homogeneity of the reduced LN improve significantly. However, the reduced LN crystals exhibit significant light absorption, which is undesirable for practical applications. This study is beneficial for guiding the development of LN EO Q switches with high-temperature stability.