Open Access
Add to BibSonomy
Abstract
The complex excited energy levels in the diode-pumped metastable Ar laser may induce harmful effects in laser cycling. Significantly, the influence of the population distribution in 2p energy levels on the laser performance is unclear yet. In this work, the absolute populations in all the 2p states were measured online by the simultaneous applications of tunable diode laser absorption spectroscopy and optical emission spectroscopy. The results showed that most atoms were populated to the 2p8, 2p9, and 2p10 levels while lasing, and the majority of the 2p9 population was efficiently transferred to the 2p10 level with the aid of helium, which was beneficial for the laser performance.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Diode-pumped metastable rare gas lasers (DPRGLs) use the metastable state of rare gas atoms (Rg*: Ne*, Ar*, Kr*, and Xe*) generated by gas discharge as the laser gain medium. The DPRGLs are anticipated to become candidates for high-energy lasers analogous to the diode-pumped alkali vapor lasers (DPALs). Due to the similarity of atomic configuration and optical properties, the DPRGLs can potentially inherit the power scaling ability of DPALs. At the same time, this laser is chemically stable and easy to operate because its gain medium is an inert gas.
In 2012, Han and Heaven first proposed the concept of DPRGL and demonstrated lasing by optically pumping Ar*, Kr*, and Xe* [1]. In 2015, Rawlins et al. realized efficient CW lasing using a microwave-driven micro-discharge scheme [2]. In 2017, a diode-pumped Ar* laser was demonstrated by Han and Heaven, which realized 4 W output in a highly repetitive nanosecond pulsed DC discharge [3]. For increasing the gain volume, the dielectric barrier discharge method was explored in [4,5]. In 2018, a power scaling analysis was made by Eshel et al. based on a five-level model, which theoretically demonstrated the possibility of a 100 kW Ar* laser with high efficiency [6]. Recently, a new type of DPRGL using the plasma jet as the gain medium was demonstrated by Wang et al., which was likely to overcome the limitations in traditional space-constraint discharge and enhance the power scaling ability of DPRGLs [7,8].
In this paper, Paschen notation is used to label all energy levels, except for special instructions. Figure 1 depicts the operation mechanism and primary energy levels of Ar in DPRGL. There are 4 energy levels in 1s multiple states and 10 energy levels in 2p multiple states. Typically, the outermost electron of Ar is excited to the 1s5 level through electron collisions in a gas discharge. Then the diode laser centered at ∼811.5 nm pumps the atom of 1s5 to 2p9 level, followed by a rapid relaxation between the 2p9 and 2p10 with the aid of helium (He) as the buffer gas. Finally, the ∼912.3 nm laser is obtained from the stimulated emission of 2p10→1s5.
Ideally, the DPRGL is expected to run as a three-level laser system (1s5, 2p9 and 2p10). However, due to adjacent energy levels and complex transition channels, atoms involved in the three-level laser cycle will relax to other energy levels, somewhat deteriorating the laser performance.
Among the 1s multiple energy levels, the 1s4 state is considered as an interference level with noticeable adverse effects on the laser performance. A considerable population (∼20%) in the upper laser level of 2p10 will spontaneously radiate to 1s4 level. In contrast, the 1s2 and 1s3 levels have a minimal effect. Han et al.'s studies suggested that the negative impact caused by 1s4 level could be reduced under an elevated gas temperature [2,9]. Emmons and Weeks also concluded that the laser performance could be better if the number of atoms transferred to the 1s4 were decreased [10]. Yang et al.'s theoretical calculation indicated that the 1s4's negative influence could be resolved by elevating the metastable densities [11]. And in a subsequent kinetics study experiment, the Ar* laser experienced a transition from pulsed to CW lasing by increasing the 1s5 density and indirectly reducing the adverse effects of 1s4 [12]. In order to minimize the 1s4 accumulation, a dual-wavelength pumping scheme was proposed by Sun et al., which could dramatically increase laser efficiency [13]. In summary, the influence of 1s4 could be controlled by increasing the density and generation rate of Ar* or utilizing a new secondary pumping scheme.
Among the 2p multiple energy levels, the energy difference between 2p8 and 2p9 is as small as 154.5 cm−1 (∼0.02 eV), which means that a considerable number of atoms may be transferred from 2p9 to 2p8 by collisional relaxation process. Emmons and Weeks predicted that the argon atoms in 2p9 level would be transferred to 2p8 and 2p10 levels in nearly the same proportion by He collision at 300 K [14]. Besides, atoms in the upper levels of the pump and lasing transitions can also be excited to higher 2p states through electron collisions. But the relaxation kinetics and influence of other 2p levels in the DPRGL were unclear.
This work aims to establish the quantitative population distribution in DPRGL of Ar* for related levels, including the 1s5 state and all the 2p energy levels. First, tunable diode laser absorption spectroscopy (TDLAS) method was used to obtain the absorption profile of the transition 2p10→2s5 at 740 Torr. The time-dependent number densities of 2p10 and 1s5 were simultaneously measured to correlate the densities of these two levels, and the kinetics was studied. Second, optical emission spectroscopy (OES) was utilized to obtain the population distribution in all the 2p energy levels. The number density of 2p10 in the first step was utilized to calculate all the 2p energy levels’ absolute density values based on their relative emission intensity. Then they were finally shown as the ratio values to the 1s5 level to account for their proportion to the total Ar* density. The population distributions were compared in three conditions: ‘discharge only’, ‘pumping without a resonator’ and ‘pumping with lasing’. The analysis further revealed the kinetics of the DPRGL.
2. Experimental setup
The gas discharge setup includes a pulsed DC discharge system and a gas chamber. The pulsed DC discharge power source used in the experiment could run at a voltage of up to 5.4 kV with ∼80 ns pulse duration and a repetition rate of 20 kHz. The pulsed DC discharge method could efficiently generate Ar metastables with high density. Detailed information on the pulsed DC discharge system, the construction of sealed cubic gas chamber and gas control devices were described in [12,15]. Ar/He (molar ratio 10:90, and the gas purity is 99.999%) mixed gas was utilized in this experiment. The mixed Ar/He gas flowed at a flow rate of 2 L/min and the chamber pressure was maintained at ∼740 Torr. In Fig. 2, two round electrodes were made of tungsten copper alloy. The cross-section glow discharge area was 5 mm in diameter and the discharge gap size between two electrodes was 4 mm. The glow came from multiple transitions that radiated from the 2p and upper energy levels to the 1s multiple states.
The optical setup is shown in Fig. 3(a). Along the Y direction, a volume-Bragg grating narrowed diode pump source was used to pump the Ar metastables [16]. It was locked to the 1s5→2p9 transition with a center wavelength at 811.53 nm and a linewidth of 0.15 nm (FWHM). It could produce up to 50 W output power, and the focused spot size was measured to be ∼2 mm2 (90% power included, and the pump intensity was ∼2.5 kW/cm2). The laser resonator consisted of two dichroic plane mirrors with high reflectance (>99%) at the laser wavelength and high transmittance at the pump wavelength (>97%). The reason for choosing two highly reflective mirrors for the resonator was to enhance the intracavity circulating laser intensity. When the 912 nm laser was generated, it first passed through a 45°beam splitter (BS, > 99% reflection at 811 nm and >65% transmission at 912 nm) to eliminate the pump light and then passed through a bandpass filter (BF, Thorlabs, center wavelength at 910 nm with a FWHM of 10 nm) to further eliminate the pump light and discharge fluorescence, and was finally detected by a photodetector (PD1, Thorlabs, PDA100A2).
Fig. 3. (a) Schematics of the optical setup. The pathway of the pump source is in the Y direction, and the pathways of the two probe lights are in the X direction. DL1: 1067.356 nm diode laser, DL2: 801.478 nm diode laser, L - lens, HR - high reflector, BS - beam splitter, BF - bandpass filter, PD - photodetector, HP - half-wave plate, PBS - polarization beam splitter (b) Schematics of the fluorescence measurement devices.
Along the X direction, two tunable single-frequency diode lasers with the central wavelength of 801 nm (DL2) and 1067 nm (DL1) were used as probe lights to simultaneously measure the number densities of 1s5 (1s5→2p8) and 2p10 (2p10→2s5). Their wavelengths were monitored by a wavelength meter (Highfinesse WS7, not shown in our figure). The beam of DL1 was adjusted as horizontally polarized by a half-wave plate (HP1) and then passed through a polarization beam splitter (PBS1), the discharge region, and PBS2. It was detected by a photodetector (PD2, Thorlabs, DET10A2) in sequence. The beam of DL2 was adjusted as vertically polarized and was detected by PD3 (Thorlabs, PDA100A2). All photodetectors (PD1, PD2 and PD3) were connected to a multichannel oscilloscope for data record and analysis.
The fluorescence measurement devices are shown in Fig. 3(b). Along the probe light direction (X direction), two circular apertures with a diameter of 1 mm and a separation distance of 8 mm were installed and placed in front of the discharge chamber’s window. A 600-µm fiber was installed after the two apertures to collect the plasma fluorescence, which was then delivered to a spectrometer. The two apertures were used to isolate the stray light that came from the discharge chamber and external environment. And it also ensured that the collected fluorescence mainly came from the area where the probe light passed through. The spectrometer has two channels, which could cover the concerned emission lines of 2p energy levels. One channel has a spectral range of 745∼815 nm with a resolution of 0.039 nm, and the other has a spectral range of 805∼926 nm with a resolution of 0.071 nm.
3. Results and discussion
3.1 Simultaneous measurement of the population in 1s5 and 2p10 levels
First, the TDLAS method was used to measure the absorption profile of the 2p10→2s5 (4p2[1/2]1→5s2[3/2]°2, in Racah notation) transition. The single-frequency (FWHM ∼1 MHz) probe laser is a Littrow configuration centered at the transition line (∼1067.35 nm). The wavelength tuning was realized by rotating a grating. Data points in the wavelength range from 1067.00 nm to 1067.65 nm were recorded, see Fig. 4. At atmospheric pressure, the linewidth broadening was mainly determined by the collisional process, and the line shape could be reasonably fitted by the Lorentzian profile. From the fitted curve, the linewidth was measured as ∼29.9 GHz (FWHM), and the peak atomic absorption cross-section was ∼7.9 × 10−14 cm2 at the central wavelength of 1067.3565 nm. Then, using this data, the 2p10 density could be calculated from the Beer-Lambert law by measuring the incident and transmitted powers of the probe laser.
Fig. 4. Absorption coefficient of transition 2p10→2s5 at a pressure of 740 Torr. Ii represents the incident intensity, and It represents the transmitted intensity.
To correlate the densities of 2p10 and 1s5, another probe laser centered at 1s5→2p8 (∼801.4786 nm) was used to simultaneously measure the 1s5 number density. The probe laser was detuned ∼0.013 nm from the central wavelength to avoid absorption saturation. Because the probe and pump directions were orthogonal, the absorption pathway of the probe light was divided into a pump region (∼2 mm) and a discharge region (∼3 mm), and the method to achieve the number density in the pump region was described in our previous work [12].
Figure 5 depicts the number densities of 1s5 and 2p10 evolving with time. First, when only gas discharge existed (‘discharge only’), the 1s5 density had a peak value of ∼1.2 × 1014 cm−3, while the peak 2p10 density was as small as 1.4 × 1012 cm−3. Then the discharge region was pumped by the 811.5 nm high power diode laser without a resonator (‘pumping without a resonator’), the peak 1s5 density decreased to ∼9 × 1013 cm−3, and the 2p10 density increased to ∼1.4 × 1013 cm−3. When a resonator was added, the 912 nm laser was generated with an averaged intracavity power of ∼80 mW (pumping with lasing). In this condition, a decrease in 2p10 peak density was observed due to the population consumption by stimulated emission.
Fig. 5. Time evolution of 2p10 and 1s5 number densities in different conditions at the pressure of 740 Torr. (a1) ∼(a3) represent the 2p10 densities and (b1) ∼(b3) represent the 1s5 densities.
In order to analyze the total generated populations in one discharge period, the averaged number densities of 2p10 and 1s5 were calculated by integration as:
The averaged number densities of 2p10 and 1s5 are depicted in Fig. 6. The 1s5 density was ∼6.5 × 1012 cm−3 under discharge only, sufficient to support lasing. When the pump laser was turn on without a resonator, the 1s5 density decreased by 4.26 × 1012 cm−3 (6.5 × 1012→2.24 × 1012 cm−3), and the 2p10 density increased by 0.996 × 1012 cm−3 (0.10 ×1012→1.10 × 1012 cm−3). When the resonat cavity was built, the 2p10 population decreased by 8.2% (1.10 × 1012→1.01 × 1012 cm−3) due to the relatively weak intracavity 912 nm laser circulating. The reason that only a small amount of 1s5 population was transferred to 2p10 was complex, including the relaxation process between 2p9 and 2p10, the loss of 2p9 population due to spontaneous emission and collisional relaxation processes. Especially, atoms at 2p9 partly could relax to other higher 2p levels, and the 2p multiple population distribution will be measured in the next section.
3.2 Population distribution in the 2p states
The time-integrated level populations are proportional to their line intensities of fluorescence [17–19]. The method of OES was utilized to determine the 2p population distribution. In this measurement process, three factors were considered in advance. First, proper 2p→1s transition lines were chosen, see Table 1. The fluorescence of each 2pj→1s transition (j = 1, 2, …, 10) may have several radiative channels (decided by the transition selection rule). The emission spectral lines were chosen from relatively strong spectral lines that are not mixed with other adjacent spectral lines. Second, the 2p9 energy level has only one spontaneous transition (2p9→1s5), so the fluorescence cannot be separated from the pump light. However, the 2p9→1s5 fluorescence signal could be extracted by subtracting the pump stray light signal (determined when the pump light was turned on and discharge was turned off) from the fluorescence signal (determined when the pump light and discharge were turned on). Third, two circular apertures were utilized to isolate most of the stray light from pump reflection and surrounding environment.
Table 1. Spectroscopic data of selected line for 2p→1s transition in Ar [20].
The fluorescence emission lines are shown in Fig. 7. The strongest line intensity was the 2p10→1s5 (∼912 nm, presented by multiplying a factor of 0.1), representing a large population in the 2p10 level. Two moderate-intensity lines were the 2p9→1s5 (811.53 nm) and 2p8→1s4 (842.46 nm). Other lines were relatively weak compared to the 2p8, 2p9 and 2p10 levels.
Fig. 7. Emission spectroscopy at the pressure of 740 Torr. The strong spectral lines of 811.53 nm and 842.46 nm are marked with arrows. Data between 909.63 nm and 914.28 nm has been multiplied with a factor of 0.1.
The population distribution ratios of all the 2p states were obtained by using the OES method. The absolute averaged number density of each 2p level could be calculated from the averaged density of 2p10. It should be noted that the 2p10 population measured by TDLAS and the 2p distribution ratios measured by OES were conducted simultaneously. In order to reveal the fraction of the 2p population relative to the 1s5 level, the population of each 2p level was expressed as a ratio to the 1s5 averaged density (discharge only). The complete population distribution in 2p states was shown in Fig. 8. It could be seen that most excited atoms located at the 2p8, 2p9 and 2p10 levels, while other higher levels’ (2p1∼2p7) effect on the laser performance could be ignored. When only gas discharge existed, the population was mainly located at the 2p10 level, with a small fraction of 1.82% relative to the 1s5. When pumping was on, the 2p population was dramatically changed, showing a significant increase in the 2p8 to 2p10 levels. For example, 2p8, 2p9 and 2p10 were 96.92% relative to the total 2p states under pumping without a resonator. An ideal distribution was that most of the 2p9 population was transferred to the 2p10 with the aid of He, which was beneficial for the laser performance. Due to the small energy difference between 2p8 and 2p9 (154.5 cm−1), a certain amount of population was observed at the 2p8 level, which could not be ignored, but still not so significant as compared with the 2p10. When the laser circulation was built with a resonator, the atoms in 2p10 were partially consumed (1.03% relative to the 1s5), and an increase of atoms populated in the 2p6 to 2p9 level was observed. Because lasing could enhance the pump absorption and stimulated emission, more atoms in 1s5 were pumped to 2p9 level and then relaxed to those adjacent levels. In the measurement of 2p population distribution, the error was estimated to be around 10% which mainly arise from the non-uniform spatial distribution of the plasma, the fluctuations of pump source power, the noise on the optical spectrum, the tiny drift of experimental conditions and the measuring deviation.
Fig. 8. Ar(2p) energy levels’ population density (relative to the 1s5 level) at the pressure of 740 Torr.
The diode laser pumped the atoms from the 1s5 to 2p9 level, and then the collisional process redistributed the 2p9 population. The collisional process was mainly affected by the gas pressure. To study the effect on 2p population distribution by the gas pressure, we increased the Ar/He mixture gas pressure while keeping the mole fraction constant. The result is shown in Fig. 9. The 2p10 population experienced a near-linear growth as pressure increased from 721 to 825 Torr. At the same time, the 2p9 and 2p8 populations experienced a relatively slow increase. The phenomenon confirmed that the effective population was transferred from 2p9 to 2p10 at a buffer pressure close to the atmosphere.
Fig. 9. The population of 2p8, 2p9 and 2p10 at different pressure (The discharge parameters were 2 kV and 10 kHz).
3.3 Discussion
In this experiment, the averaged population distribution of the 2p states was measured and analyzed. At the gas pressure of ∼740 Torr, the number density of 1s5 decreased from ∼6 × 1012 cm−3 to ∼2.2 × 1012 cm−3 under the action of pumping. And the 2p population distribution showed an optimistic result that most of the population in 2p9 level was transferred to 2p10 level. At the same time, the 2p8 had a moderate value that could be accepted. And the populations at higher 2p levels (2p1∼2p7) could be ignored.
The laser power in this experiment was low, with an intracavity power estimated to be ∼80 mW compared to the ∼50 W pump power. The low efficiency was due to several reasons, including the low 1s5 density at the current devices (∼6.5${\times}$ 1012 cm−3), the unmatched pump linewidth and the low pump intensity, which was not the physical limitation of the laser itself and could be improved in the future. In Eshel et al.'s prediction, an efficient 100 kW Ar* laser requires a pump intensity of 20 kW/cm2 and 1s5 density of 1.0${\times}$ 1014 cm−3 [6]. The population distribution in this preliminary experiment could not be extrapolated for such a power-scaled condition. As atomic density and energy deposition increase, more atoms will be populated in higher levels [21–23], which may deteriorate the laser performance. Such work needs to be conducted in the next step.
4. Conclusion
In this paper, the 2p population distribution of the diode-pumped metastable argon laser was experimentally measured. By the combination of TDLAS and OES methods, the absolute densities of the population distribution in 2p states (2p1-2p10) were obtained. The system has been studied in three conditions: ‘gas discharge’, ‘pumping without a resonator’ and ‘pumping with lasing’. The population distributions in the above three states were measured and analyzed. The conclusions were optimistic within the range of our experimental conditions (1s5 was ∼1012 cm−3, and pump intensity was ∼2.5 kW/cm2). Most excited atoms were at the 2p8, 2p9 and 2p10 levels. The majority of 2p9 population was efficiently transferred to the 2p10 with the aid of He, which was beneficial for the laser performance. The 2p8 occupied a certain number of atoms, mainly by the collisional transfer from the 2p9, but not significant and was tolerable. Even with low laser power, the stimulated emission induced 2p10 population consumption was successfully observed. This work revealed the population distribution of 2p energy level in experiment, which was essential for diagnosing and developing DPRGL. The population distribution and the correlation with laser performance need to be further investigated for power scaling with much higher atomic density and pump intensity.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. J. Han and M. C. Heaven, “Gain and lasing of optically pumped metastable rare gas atoms,” Opt. Lett. 37(11), 2157–2159 (2012). [CrossRef]
2. W. T. Rawlins, K. L. Galbally-Kinney, S. J. Davis, A. R. Hoskinson, J. A. Hopwood, and M. C. Heaven, “Optically pumped microplasma rare gas laser,” Opt. Express 23(4), 4804–4813 (2015). [CrossRef]
3. J. Han, M. C. Heaven, P. J. Moran, G. A. Pitz, E. M. Guild, C. R. Sanderson, and B. Hokr, “Demonstration of a CW diode-pumped Ar metastable laser operating at 4 W,” Opt. Lett. 42(22), 4627–4630 (2017). [CrossRef]
4. D. J. Emmons, D. E. Weeks, B. Eshel, and G. P. Perram, “Metastable Ar(1s5) density dependence on pressure and argon-helium mixture in a high pressure radio frequency dielectric barrier discharge,” J. Appl. Phys. 123(4), 043304 (2018). [CrossRef]
5. P. A. Mikheyev, J. Han, A. Clark, C. Sanderson, and M. C. Heaven, “Production of Ar and Xe metastables in rare gas mixtures in a dielectric barrier discharge,” J. Phys. D: Appl. Phys. 50(48), 485203 (2017). [CrossRef]
6. B. Eshel and G. P. Perram, “Five-level argon–helium discharge model for characterization of a diode-pumped rare-gas laser,” J. Opt. Soc. Am. B 35(1), 164–173 (2018). [CrossRef]
7. R. Wang, Z. Yang, Q. Liu, K. Han, H. Wang, and X. Xu, “Demonstration of a diode-pumped plasma jet-type rare gas laser,” Opt. Lett. 47(13), 3279–3282 (2022). [CrossRef]
8. X. Xu, R. Wang, and Z. Yang, “The second fusion of laser and aerospace - an inspiration for high energy lasers,” Opto-Electronic Advances, 220113 (2023).
9. J. Han and M. C. Heaven, “Kinetics of optically pumped Ar metastables,” Opt. Lett. 39(22), 6541–6544 (2014). [CrossRef]
10. D. J. Emmons and D. E. Weeks, “Effect of Ar(3p(5)4p; 2p)+M –> Ar(3p(5)4s; 1s)+M branching ratio on optically pumped rare gas laser performance,” Opt. Express 27(24), 35689–35699 (2019). [CrossRef]
11. Z. Yang, G. Yu, H. Wang, Q. Lu, and X. Xu, “Modeling of diode pumped metastable rare gas lasers,” Opt. Express 23(11), 13823–13832 (2015). [CrossRef]
12. R. Wang, Q. Liu, Z. Yang, H. Wang, and X. Xu, “Revealing kinetics of a diode-pumped metastable Ar laser in pulsed and CW lasing,” Opt. Lett. 47(21), 5477–5480 (2022). [CrossRef]
13. P. Sun, D. Zuo, X. Wang, J. Han, and M. C. Heaven, “Investigation of dual-wavelength pump schemes for optically pumped rare gas lasers,” Opt. Express 28(10), 14580–14589 (2020). [CrossRef]
14. D. J. Emmons and D. E. Weeks, “Kinetics of high pressure argon-helium pulsed gas discharge,” J. Appl. Phys. 121(20), 203301 (2017). [CrossRef]
15. R. Wang, Z. Yang, K. Li, H. Wang, and X. Xu, “Experiment and modeling of the pulsed lasing in a diode-pumped argon metastable laser,” J. Appl. Phys. 131(2), 023104 (2022). [CrossRef]
16. R. Wang, Z. Yang, H. Tang, L. Li, H. Zhao, H. Wang, and X. Xu, “A linewidth narrowed diode laser for metastable Argon atom pumping,” Opt. Commun. 502, 127398 (2022). [CrossRef]
17. X. M. Zhu, Y. K. Pu, N. Balcon, and R. Boswell, “Measurement of the electron density in atmospheric-pressure low-temperature argon discharges by line-ratio method of optical emission spectroscopy,” J. Phys. D: Appl. Phys. 42(14), 142003 (2009). [CrossRef]
18. Y.-K. Lee, S.-Y. Moon, S.-J. Oh, and C.-W. Chung, “Determination of metastable level densities in a low-pressure inductively coupled argon plasma by the line-ratio method of optical emission spectroscopy,” J. Phys. D: Appl. Phys. 44(28), 285203 (2011). [CrossRef]
19. M. H. Kabir and M. C. Heaven, “Energy transfer kinetics of the np5(n + 1)p excited states of Ne and Kr,” J. Phys. Chem. A 115(34), 9724–9730 (2011). [CrossRef]
20. A. Kramida, Y. Ralchenko, J. Reader, and NIST ASD Team (2022). NIST Atomic Spectra Database. Available: https://physics.nist.gov/asd.
21. K. Waichman, B. D. Barmashenko, and S. Rosenwaks, “Dependence of static K DPAL performance on addition of methane to He buffer gas: 3D CFD modeling and comparison with experimental results,” J. Opt. Soc. Am. B 36(12), 3464–3470 (2019). [CrossRef]
22. M. Endo, R. Nagaoka, H. Nagaoka, T. Nagai, and F. Wani, “Modeling of diode-pumped cesium vapor laser by combination of computational fluid dynamics and wave-optics,” Jpn. J. Appl. Phys. 59(2), 022002 (2020). [CrossRef]
23. R. J. Knize, B. V. Zhdanov, and M. K. Shaffer, “Photoionization in alkali lasers,” Opt. Express 19(8), 7894–7902 (2011). [CrossRef]
