4. Experimental Result and Analysis

4.2 Temperature robustness comparison between CPPLN and LBO

When the input 1064nm light is 22.53W, the curves of the frequency-doubled optical power generated by CPPLN (www.wisoptic.com) and LBO (www.wisoptic.com) with temperature are shown in Figure 5(a) and Figure 5(b). The half-maximum full width of the frequency-doubled optical power of CPPLN with respect to temperature is 8.40℃, ranging from 24.19℃ to 32.59℃. The half-maximum full width of the frequency-doubled optical power of LBO with respect to temperature is 6.12℃, ranging from 15.54℃ to 21.66℃, which is lower than the half-maximum full width of CPPLN. Within the half-maximum width range, LBO shows a monotonically increasing and monotonically decreasing trend, while CPPLN shows an oscillating trend, not a simple monotonically increasing or decreasing trend. There are three peaks in this range, which are 148mW, 120mW and 105mW respectively.

From equations (4) and (5), we know that the temperature range of the crystal adaptation is related to factors such as the refractive index of the crystal, the structure of the polarized crystal (including duty cycle, polarization period, chirp, etc.). Since there will be certain processing errors between the designed crystal structure and the actual processed crystal structure, and there will inevitably be certain errors between the refractive index calculated by the Sellmeier Equation and the actual refractive index, these factors lead to the difference between the temperature range of the crystal reciprocal lattice adaptation and the simulation results. In this case, the absolute values of theory and practice are not necessarily exactly the same, but the relative trends should be consistent. Therefore, in the experiment, we are more concerned about the shape of its characteristic curve than the absolute values.

We believe that the reason why the CPPLN crystal curve shows such a trend is related to the polarization period accuracy that can be achieved by the processing manufacturer of the polarized crystal. The minimum processing accuracy of the polarization period of the CPPLN crystal used in this experiment is 10nm, that is, the error of the polarization period is about 0.14%. Therefore, the actual structure of CPPLN is not consistent with the ideal structure, but there is a deviation. The actual distribution of the crystal reciprocal lattice vector is not exactly the same as the simulation, and it is not uniform. At 27℃ and 30℃, there is a significant drop in power due to the lack of adapted reciprocal lattice vectors caused by processing errors. At 25℃, 29℃ and 32℃, the reciprocal lattice vector of the crystal can just compensate for the phase mismatch of the frequency doubling process, resulting in a significant increase in power, leading to oscillation of the temperature characteristic curve in the temperature range of 24.19℃ to 32.59℃. To further illustrate the problem, we simulated the temperature and frequency doubling efficiency of CPPLN crystals with different duty cycles, and the results are shown in Figure 6. It can be seen that when the duty cycle of the polarized crystal has a deviation of 0.01%, the relationship curve between temperature and frequency doubling efficiency will show a very significant change: when the duty cycle is 49.61%, the curve is relatively flat, and the frequency doubling efficiency of the middle and two wings is close; when the duty cycle becomes smaller, the frequency doubling efficiency of the middle part is enhanced, and the frequency doubling efficiency of the two wings is weakened; when the duty cycle becomes larger, the frequency doubling efficiency of the middle part is weakened, and the frequency doubling efficiency of the two wings is enhanced, and three peaks appear at different positions. When the duty cycle is 49.62%, the trend of the curve is basically consistent with the results obtained in the experiment, indicating that the duty cycle of the CPPLN crystal actually processed should be slightly larger than the designed duty cycle.