Objective Diode-pumped alkali-vapor lasers (DPALs) represent a research hotspot in the field of high-power lasers owing to their advantageous properties, such as high efficiency, excellent beam quality, and scalability. However, DPALs are also affected by several issues: (1) alkali-metals are very active and react easily with coating materials of windows, sealant materials, etc., especially at high temperature; therefore, it is very challenging to prepare the sample cell; (2) alkali-metals also react with relaxing gas (methane or ethane) and produce carbon that contaminates windows, thus seriously reducing the windows lifetime of the sample cell. As a consequence, researchers are putting enormous efforts into the exploration of substitutes for alkali-metals. Among these, metastable rare gases are considered to be the most promising candidates. Metastable rare gases have an electronic structure similar to that of alkali-metals. Therefore, optically-pumped metastable rare-gas lasers (OPRGLs) are expected to have similar properties to DPALs, without any of the disadvantages mentioned above. However, metastable rare gases do not readily exist; they are normally prepared trough high-voltage discharge. The density, homogeneity, and stability of metastable rare gases are key factors for OPRGLs.

Methods The discharge system was composed of a simple circuit with a ~100 ns pulsed high-voltage power supply and a pair of parallel plate Cu electrodes (Fig. 2). The flat electrode discharge technique was used to produce a metastable state (5s[3/2]₂) for Kr (here denoted as Kr ^*). The discharge voltage is 15 kV, the distance between electrodes is 1.5 cm, and the electrodes dimensions are 120 mm×10 mm. In order to improve the homogeneity and stability of the discharge, a UV light pre-ionization technique was applied. A stainless-steel structure with three windows was used to contain the ultrahigh purity Kr (99.999%). The front and back windows were made of CaF₂ to ensure that the infrared laser could pass through the cell, whereas the side window was made of polymethyl methacrylate. A tunable continuous infrared laser, based on the Cr∶LiSAF crystal, was used as the probe laser, offering a tunability from 800 nm to 850 nm with a linewidth of 70 MHz. The output power is above 20 mW at 811 nm, with a pump power of 450 mW. Therefore, the absorption profile could be recorded through scanning the laser frequency with a step of 3--5 pm. After the probe laser passed through the cell, the residual light was collected using an integrating sphere, then detected via an avalanche photodiode (APD 120A). An oscilloscope was used to visualize the absorption curve. In order to evaluate the Kr ^* density uniformity, the absorption curves along different paths were recorded separately. Moreover, an Ocean optics HR4000 spectrometer was used to measure the optical emission spectra of the high-voltage DC discharge plasma along the back window.

By means of the Beer theorem, the Kr^* density could be calculated from the residual rate, which is defined as the ratio of the residual probe laser power I_L to its initial value I₀. The value of ln[I₀/I_L(v)] is proportional to the absorption profile g(v). g(v) represents the convolution function of the Doppler broadening profile with the pressure broadening profile. The Doppler broadening profile was obtained by analyzing the optical emission spectra, whereas the pressure broadening profile was derived from the fitting process.

Results and Discussions The absorption curve shape (Fig. 3) could be altered upon tuning the frequency of the probe laser. The shape of the absorption curves indicates that the Kr^* density is firstly increased and then decreased. The maximum Kr^* density occurs 15.8 μs after triggering the discharge. It is noticed that the absorptivity saturates as the frequency of the probe laser approaches the resonant frequency of the transition 5s[3/2]₂→5p[5/2]₃, which could result in a fitting error. The saturation of all curves disappears 23.8 μs after triggering the discharge. Hence, g(υ) could be obtained by fitting the ln[I₀/I_L(v)] curve starting from 23.8 μs, whereas the maximum Kr^* density was calculated using the experimental data at 15.8 μs (Fig. 5). This new fitting technology is here referred to as complex fitting method. Compared with the results obtained from the traditional fitting method (Fig. 4), the proposed complex fitting method has clear advantages.

The results show that under a discharge voltage of 15 kV, the Kr^* density can reach a magnitude of 1.3×10¹² cm^-3 and last for 3 μs (Fig. 6). The results also indicate that the full width at half maximum of the Kr energy profile is 5 GHz, being equally affected by the Doppler and pressure broadening.

Conclusions In this work, a nanosecond high-voltage pulsed power device was fabricated to generate metastable-state rare gases with high density. The time-resolved density of metastable Kr^* was characterized via tunable continuous laser absorption spectroscopy. The results show that under a discharge voltage of 15 kV, Kr^* can reach a magnitude of 1.3×10¹² cm^-3 and last for 3 μs. These results fully meet the requirements for optically-pumped metastable rare gas lasers. In future experiments, the electrodes shape and discharge mode will be explored to further improve the density of metastable Kr.

In addition, it was shown that a complex fitting method can be used to solve the fitting error caused by the saturated absorption. This method can not only improve the calculation accuracy of metastable Kr, but also provides information on the broadening, which can be used in the rate equations for the Kr metastable laser system.