Open Access
Add to BibSonomy
Abstract
The CaF2 window is the laser exit of the DUV lithography machine as well as its sealing component. Bearing the irradiation, high pressure, and discharging pollutants, the window is easy to damage and directly deteriorate the performance and reliability of the laser. In this paper, considering the effects of the above factors, two typical short-term damages to the CaF2 window - high energy induced damage and pollutant-induced damage are studied quantitatively. Using an experimental design, theoretical calculation, characterization analysis, and numerical simulation, we found that the damage induced by high-energy irradiation is dominated by defect propagation at the initial stage. At the later stage, it is dominated by heat and thermal stress with thermal melting and evenly distributed microcracks of 1∼10 µm in size. Low-energy irradiation only causes expansion and deformation of the window, but the highly absorbent electrode discharging particles with a diameter of 0.1∼1 µm strongly absorb the laser. Strong local heat is caused during the melting and gasification of the particles, which easily leads to ablation and cracks. The damaged area is proportional to the particle size, and the damaged rate is proportional to the average power density at a high repetition frequency. High energy density and the defects with high absorptivity, the electrode discharging particles, and the heat accumulation effect, are the main factors for the two short-term damage, respectively.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Due to its high transmittance, low birefringence and good stability, CaF2 has become the main optical component material for deep ultraviolet lithography (DUVL) [1,2]. With the continuous improvement of lithography energy, repetition frequency and other performance, the optical components damage caused by the laser of high single photon energy and high absorptivity [3], has become a weak link degrading the laser performance and restricting the development of DUVL [4].
The damage of CaF2 optical component is affected by laser parameters, impurities and defects, pollutants and so on [5,6]. They may cause different degrees of thermal effects, chemical effects, static and dynamic mechanical effects (thermal stress and impact), and lead to a variety of damage [7,8]. Lou’ experiments showed that the laser induced damage threshold (LIDT) of pure CaF2 crystal was 6 J/cm2 (0.1 Gw/cm2) at 193 nm wavelength and 60 ns pulse width [9]. But due to the defects [10,11], impurities [12,13], and stress during growth, cleaning, and polishing [14], the component was often damaged below this value [15]. Especially, contaminants with high absorption coefficients tended to be destructive and cause severe local temperature gradients [16]. Michael Bauer et al. found that the CaF2 was damaged at 80mJ/cm2 after millions of pulses, and the irradiated area heated up rapidly and was estimated to be over 150 °C [17]. The CaF2 decomposed and may react with air to form crystalline CaCO3. Daisuke Tei et al. believed the defect of CaF2 absorbed energy, locally broke crystal bonds and ionized [18]. The CaF2 window is the exit of the DUV laser as well as the sealing component of the discharge chamber. It suffers from the laser irradiation, high pressure in cavity and mechanical loading, discharging wastes of the discharge chamber. With the combined effect of these factors, its damage is more complicated and interesting than other optical components.
In this paper, two typical short-term damages of CaF2 window are studied: high energy induced damage and pollutant induced damage. By means of experimental design, theoretical calculation, characterization analysis and numerical simulation, the characteristics and damage mechanisms of the damage are cleared, the damage mechanism of DUV laser irradiation is revealed. This study will provide data and theoretical support for the damage and protection of DUV optical components.
2. Method and material
Uncoated excimer-grade CaF2 components (Nikon, 38.1 mm in diameter and 5 mm in thickness) were used as the output window of the self-developed ArF DUVL machine. The typical copper alloy was used as discharging electrodes. The working pressure was 0.4 Mpa, the working temperature was 40 °C, and the ambient temperature was 25 °C. F2 and Ar gas was used as the working gas, and Ne was used as the buffer gas. The laser discharge frequency was adjustable from 1 Hz to 6 kHz, the pulse width was 20 ns. A fiber optic thermometer probe was fixed on the outer surface of the window, and the steady state temperature during irradiation was recorded.
The experiment of high energy induced damage requires high energy stability, so a specially constructed irradiation system (Fig. 1) was set up. The 193 nm laser (Mlase, energy stability >2%) was irradiated on the component after passing through the diaphragm and the focusing lens. The irradiation process was protected by N2 gas (4 L/min), and the damage monitoring was performed with an on-line microscope. The LIDT was tested by 10-on-1 method: the laser with the same energy density is irradiated at a certain position for 10 times at the same repetition frequency 10 Hz. Gradually increasing the energy density. The energy density corresponding to the occurrence of plasma flash is the plasma threshold. The zero probability LIDT can be obtained by fitting the probability of damage under different energy densities. The damage determination methods used in this paper include plasma method (plasma flash) and microscopic observation method (A microscope with a magnification of 100 to 150 times was used to observe the CaF2 surface, the morphological changes mean that the CaF2 is damaged.)
The pollutant induced damage was observed by the component irradiated at a repetition frequency of 4 kHz by the self-developed ArF DUVL machine, with the energy density of 58 mJ/cm2 and the spot area of 2.5×12.5 cm2. The discharging products that may be deposited on the component at any time make an online quantitative study impossible. So, the pollutants were collected and chartered, and then coated on the component surface to support the experiment of pollutant induced damage by the system in Fig. 1.
The absorptivity of the CaF2 material was measured by an ultraviolet-visible spectrophotometer (Lambda750). The defects and their evolution were measured by fluorescence spectrometer (FLS980). Particle size analyzer (Nano ZS) was used to determine the size of pollutants. The average damage depth was measured using a white light interferometer (WLI, NewView 8300). After sprayed the conductive carbon film, the cold field emission scanning electron microscope combined with an energy spectrum analysis (SEM, SU8010) was used to characterize the morphology and composition of the window. The mapping scanning was used to reduce errors caused by uneven distribution of elements. ANSYS software was used to simulate the effects of particle sizes and laser parameters on damage. The laser energy has a Gaussian distribution in the width direction, and flat-top distribution in the length direction, So the integral method is used to calculate the function formula of the body heat source. Body heat sources are applied through nodes on each layer mesh of the element. The main parameters were shown in Table 1. Since the particle influence zone and the heating effect of laser on CaF2 irradiation are limited, a range of ten times the radius of the particle was selected for the pollutant induced research, and the laser heat source was applied to the upper surface of the particle. The heating rate of particles is in the order of milliseconds, so the heat exchange between the elements and the environment is ignored.
Table 1. Main parameters of the simulation
3. Results and discussion
3.1 Short-term damage of the CaF2 window
The damage of components is mainly considered as three types: threshold damage (high energy induced damage), pollutant-induced damage, and long-term damage during operation. In this paper, we focus on the first two types (short term), which damage under one to tens of thousands of pulses, instead of the case of long-term with millions of pulses or more.
3.1.1 High energy-induced damage and its evolution
As showed in Fig. 2, the average absorptivity of CaF2 is 3.9%/5 mm at 193 nm. The content of impurity elements and defects with high absorptivity is very low, since no significant absorption peak in the curve is observed. Figure 3 shows the zero-probability LIDT is 2.62 J/cm2 with laser spot area 3000µm2. With the increases of energy density, the damage points on the surface increase, and the damage area gradually expands into damage pits. The damage threshold of plasma flash was observed to be 3.18 J/cm2.
At present, studies have shown that surface machining defects, crystal structure defects, impurities, etc. can lead to fluorescence peaks [19]. Figure 4 shows fluorescence spectrum results of the damages. The fluorescence defects already exist on the surface and subsurface of the unirrigated material. After irradiation, defects show different degrees of growth. In most cases, the peak value of surface microdefects (380 nm) and irradiation structure defects (450∼600 nm) [19] increased significantly. The surface defects (600-740 nm) are generally considered to be defects that cause surface absorption or scattering. Since its higher absorptivity, the defect tends to heat up and generates thermoelastic stress, which further causes its expansion. The power density corresponding to the zero-probability LIDT is 1.7×108W/cm2, which is enough to make the extended defect further absorb energy to gasify and even generate plasma. The randomness of defect distribution is an important factor to the fluctuation of LIDT within a certain range.
3.1.2 Characteristics of the pollutant induced damage
CaF2 window is exposed to the energy density of 58 ∼ 80 mJ/cm2 in the operation, which could not induce a short-term damage in theory. However, the operation of laser produces discharging particle products. Unavoidably, a small amount of the particle would deposit on the surface of the window and induce damage under irradiation.
As showed in Fig. 5, the original surface of the window is smooth and flat, but local cracks and pits with sizes ranging from micron to tens of microns exist on the damaged surface. The residual particles showed a slight melting phenomenon. The components of the particles were Cu, Zn and a small amount of F element, no other substances were detected. Combined with the previous work [20], the particles are the discharging products of the electrodes of the laser.
Figure 6 shows the size distribution and morphology of the particles collected in the discharge chamber. The diameter of most particles is within 0.8µm, only a small number of it is above 1µm. The particles were coated on the surface of a clean CaF2 component and irradiated with 1000 pulses at 58 mJ/cm2 and 4 kHz. Damage pits with a diameter of about 1∼3 µm were produced in the light spot area, as showed in Fig. 7(a). Most of the damage pits only contained Ca and F elements, but some Cu and Zn elements still existed locally. A small amount of exfoliated CaF2 particles were found on the surface. As showed in Fig. 7(b), the particles at the edge of the light spot area showed a melting morphology, and the size is about 0.5-2 µm. No plasma flash was observed in the experiment at 1000 Hz, and no damage was observed with the same number of pulses irradiated at 10 Hz. The energy density in this experiment is lower than the damage threshold of CaF2 and copper materials, but the peak power is about 2.5×106 W/cm2.
The continuously accumulated energy is enough to support the melt and vaporize of the particles at high repetition rate, and may cause surface damage of the window in the form of local high-temperature heat source.
3.2 Influencing factors and mechanism of short-term damage
Laser irradiation is the key reason of the two damages. Once the laser energy of the pulse width on the order of 10 ns was absorbed by the CaF2 material, it would be conducted within the component in the form of heat. Considering the laser energy evenly distributed, since the spot size is much larger than the propagation depth of the heat in the pulse width, the spot area can be regarded as a semi-infinite object whose surface is uniformly heated. The surface temperature of the component can be approximately expressed by the formula 1:
Wherein, ${a_t}$ is the thermal diffusivity of CaF2, t is the heating time, T is the temperature, Ps is the laser power density, ${\lambda _t}$ is the thermal conductivity, ${a_A}$ is the absorptivity of the surface. So the surface temperature of CaF2 is proportional to the absorptivity and the square root of the heating time. With the increase of irradiation time, the surface absorbs heat to form a melting layer, and then the temperature continues to rise until evaporates. Considering the critical state that causes evaporation, but the vapor cannot cause a strong absorption of the laser, the damage rate of the material can be expressed as formula 2:The damage depth caused by a single pulse is very low, so the heat generated by irradiation mainly propagates from the bottom of the spot area. The characteristic time of heat propagation can be expressed as formula 3:
Wherein, S is the bottom area of the light spot, ${\alpha _t}$ is the thermal diffusivity. As the spot area used is 3000µm2, it takes 0.84 ms for the heat of a single pulse to transfer to the depth for the next pulse. Therefore, at low repetition frequency, such as 1 Hz and 10 Hz, the heat of a single pulse will transfer exhausted in the pulse interval. But with the increase of repetition frequency, the thermal accumulation effect will become obvious, and the damage will gradually serious.3.2.1 Heat accumulation effect of laser irradiation
To verify the above analysis, measure the temperature at the edge of the spot area (Te) on the outer surface of the CaF2 window (Fig. 8). When the laser power density is 35 W/cm2 and the repetition frequency is 100 Hz and the irradiation reaches steady state, Te is 42 °C, which is only slightly higher than the temperature of the discharge chamber. In most cases, the surface temperature increases with the average power. When the repetition frequency is 1000 Hz, with the power density increasing from 0.138 kW/cm2 to 0.15 kW/cm2 and then to 0.225 kW/cm2, Te rises from 44 °C to 49 °C and then to 57 °C. In the case of similar average power, the higher the repetition frequency, the more obvious the thermal accumulation effect, and the higher the surface temperature. When the repetition frequency increasing from 1kHz to 4kHz, and the laser power density decreasing from 264 kW/cm2 to 240 kW/cm2, the Te has increasing from 80°C to 93°C. Therefore, the average power and repetition frequency have direct influence on component temperature.
3.2.2 Influencing factors and mechanism of high energy induced damage
The morphology of the irradiation damage pits under high energy irradiation was showed in Fig. 9. When the pulses were limited and energy densities near the threshold, the surface of the material did not melt significantly, but mainly fragmented and peeled off. As the pulses increase, the molten morphology appears, but the wall around the damage pit still has an irregular fracture morphology. When the energy density increase, not only the melting morphology appears, the smooth melting morphology appears on the wall of the damage pit. The bottom of the pit distributed uniformly microcrack with a size of 2∼3 microns caused by thermal stress, and sharp corners and fragmentation morphology of the wall basically was replaced by molten morphology.
Fig. 9. Morphology of Damage Pits with Fth, 5 pulses (a), 1.2Fth, 20 pulses (b), 1.3Fth, 5 pulses (c), 1.4Fth, 20 pulses (d).
Damage depth of CaF2 increases approximately exponentially with the energy density (Fig. 10(a)). This is due to the peak and average power increase correspondingly with the laser energy density. As showed in fEqs. (1) and (2), the total absorption and damage rate of the material increase, the thermoelastic stress and the plasma impact effect intensifies, resulting in a more severe damage.
With the repetition frequency increase from 1 Hz to 500 Hz, the damage depth of CaF2 increases rapidly at first and then slow down under the same irradiation energy (Fig.10b). This is due to the peak power of the laser is constant, but the average power is proportional to the repetition frequency, the thermal accumulation effect is increasing with the repetition frequency. The damage depth is proportional to the repetition frequency, and the total irradiation time is inversely proportional to the repetition frequency, so the damage rate (damage depth/time) is proportional to the repetition frequency (average power density). So, the primary factor of high energy induced damage is energy density, followed by heat accumulation effect.
3.2.3 Influencing factors and mechanism of pollutant-induced damage
The steady-state temperature and deformation, the temperature distribution at the outer edge of the spot area under the condition of 58 mJ/cm2 and 4 kHz is simulated by Ansys. The simulated Te is 88 °C, basically consistent with the measured result of around 90 °C in Fig. 8. The thermoelastic stress on the surface of the window is higher than the working gas pressure, the expansion of the inner and outer surface is 2 µm and 8 µm, as showed in Fig. 11. So, irradiation under operating energy density does not directly cause the damage of the window, but causes thermoelastic tensile stress on the surface.
Consider the particles are in contact with the window in the form of a hemisphere (Fig. 7). Due to the limited temperature transmission range, the substrate with a radius of ten times the particle size is selected for the calculation. The thermal diffusivity of copper is 2.95×10−5 m2/s, much higher than that of CaF2 (3.56×10−6 m2/s), so the laser energy will be quickly absorbed by the particles and transferred to the inside.
The effects of particle size on damage were studied. Figure 12 shows the temperature and deformation of particles with a radius of 1 µm and 0.1 µm under 80 mJ/cm2 and 4 kHz conditions. For a particle with a radius of 1 µm, it takes 16 ms for the surface to boil, but for the particle with a radius of 0.1 µm, it takes only 8 ms. The time consumption is much higher than the pulse width of the laser, and the heat accumulation effect cannot be ignored at high repetition rate. The boiling points of Copper Alloy and CaF2 are 2595 °C and 2500 °C, while the melting points of them are 1083 °C and 1423 °C. The high temperature copper-zinc particles can easily cause strong local heat and stress gradient on the CaF2 surface, resulting in ablation damage and cracking until the particle until the particle vaporized completely. The influence area of the CaF2 is about 1.3 times of the particle size, and the expansion deformation of the interface between the particle and the window is about 0.75% of the particle size. Both of them increase almost linearly with the particle size.
The effects of different irradiation parameters on particles with a radius of 1 µm were studied, as showed in Table 2. The surface temperature is proportional to the irradiation energy density, and when the repetition frequency is above kHz, the pulse interval is close to or even lower than the characteristic time of thermal diffusion, the thermal accumulation effect will remarkably raise the surface temperature. So, the temperature rising speed of the particle surface is proportional to the average power, not the peak power. As can be seen from Fig. 12, the surface of the particle is constantly heating up by absorbing the laser energy, and heats the interior of the particle and the CaF2 in the form of a heat source. The ablation damage starts from the contact interface which exceeds the melting boiling point of CaF2. This result can be confirmed by the experimental results of pollutant-induced damage in Fig. 13. When the pollutant is not completely ablated (the red part), the ablation of the CaF2 is around the outer surface of the particle (the blue part). Therefore, the damage begins from there. The damage rate of pollutant induced damage is inversely proportional to the time when the particle surface reaches the melting point of CaF2. So the damage rate is proportional to the average laser power laser until the particle is completely ablated, forming a damage pit. Figure 14 shows the formation and its influence of the pollutant particle: Under the irradiation of laser, the surface temperature of the particles produced by the electrode reaction increases and affects the CaF2 material, causing annular ablation and gradually expanding to the pit. In severe cases, the thermal stress may even cause cracks.
Fig. 12. The temperature and deformation of particles under 80 mJ/cm2 and 4 kHz conditions: Particle Temperature (a), size effect (b), Tb-time to reach boiling temperature, Re-radius of the affected area, δmax-the max deformation of the particle, δc- the deformation of the contact place of the particle and the window, and deformation distribution of the particle (c) and the window (d).
Fig. 14. The process of particle generation (a), heating (b), melting (c), and making the CaF2 local ablation (d) and final damage (e).
Table 2. The effects of irradiation parameters on particle temperature
In summary, under the low energy irradiation, electrode discharging particles deposit on the window, absorb the laser energy, and damage the window during melting and gasification. The damage area is proportional to the particle size, and the damage rate is proportional to the average laser power density at high repetition frequency. We only preliminarily revealed the thermal state and its facts of the particles under laser irradiation in this paper, the local thermal conditions caused by particles, the influence of the multiple particle clusters and differential response of the crystal phase will be worth our further efforts.
4. Conclusion
In this paper, considering the operating conditions of the window, including mechanical loading, laser irradiation and cavity pressure, two typical short-term damage of CaF2 window - high energy induced damage and pollutant induced damage are studied. Both damages are related to photothermal absorption. Under irradiation with energy density above LIDT of CaF2, the material, especially the defects absorb the laser and expands gradually. With the increase of the absorption, the damage is dominated by heat and thermal stress- thermal melting and uniformly distributed microcracks with the size of 2∼3 µm appear in the spot area. Therefore, the high absorptivity defects are the direct factors in the high energy induced damage. The operating power density of the DUVL could not directly cause damage, but make an expansion of 2 µm on the internal surface and 8 µm on the external surfaces. The electrode discharging particles with high absorptivity and a size of 0.1∼1 µm, will absorb the laser and cause strong local heat and thermal stress on the surface during melting and gasifying, resulting in ablation and crack damage of the CaF2. The damage area is proportional to the size of the particles and the damage rate is proportional to the average power density at high repetition rates.
Funding
National Natural Science Foundation of China (No. 61705235); Natural Science Foundation of Shaanxi Province (No. 2022JQ-032); National Science and Technology Major Project (No. 2016ZX02201).
Acknowledgments
X.G. thanks the National Natural Science Foundation of China, Shaanxi Natural Science Foundation and National Key Technical Projects of China for supporting this work. The authors acknowledge the support of State Key Laboratory for Strength and Vibration of Mechanical Structures of Xi’an Jiaotong University, Institute of Microelectronics of the Chinese Academy of Sciences, and Beijing RSLaser Opto-Electronics Technology Co.
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. Wang and G. P. Cox, “ArF laser-induced damage of calcium fluoride windows with protected anti-reflection coatings,” Opt. Eng. 60(03), 1 (2021). [CrossRef]
2. C. Görling, U. Leinhos, and K. Mann, “Surface and bulk absorption in CaF2 at 193 and 157 nm,” Opt. Commun. 249(1-3), 319–328 (2005). [CrossRef]
3. W. Liu, C. Wei, K. Yi, and J. Shao, “Postprocessing treatments to improve the laser damage resistance of fused silica optical surfaces and SiO2 coatings,” Chin. Opt. Lett. 13(4), 041407 (2015). [CrossRef]
4. C. Li, X. Kang, W. Han, W. Zheng, and L. Su, “Nanosecond laser-induced surface damage and material failure mechanism of single crystal CaF2 (111) at 355 nm,” Appl. Surf. Sci. 480, 1070–1077 (2019). [CrossRef]
5. H. Guo and Q. Liu, “Evolution mechanism of fused silica damage by repetitive laser pulses,” Laser & Infrared 48, 958–964 (2018).
6. G. J. Exarhos, M. A. Norton, A. H. Guenther, L. W. Hrubesh, Z. Wu, M. R. Kozlowski, K. L. Lewis, E. E. Donohue, M. D. Feit, M. J. Soileau, M. R. Kozlowski, D. Milam, K. P. Neeb, W. A. Molander, A. M. Rubenchik, W. D. Sell, and P. J. Wegner, “Growth of laser-initiated damage in fused silica at 351 nm,” in Laser-Induced Damage in Optical Materials (2001).
7. H. Song and C. Huang, “Progress in thermal-mechantical effects induced by laser,” Adv. in Mech. 46, 435–477 (2016). [CrossRef]
8. Z. Jia, T. Zhang, H. Zhu, Z. Li, Z. Shen, J. Lu, and X. Ni, “Stress damage process of silicon wafer under millisecond laser irradiation,” Chinese Opt. Lett. 16, 1 (2018).
9. Q. Lou, J. Xu, and S. Luo, “Laser resistance of the optical materials and films for ArF and KrF excimer laser,” Laser & optoelectronics progress, 9–15 (1997).
10. H. Jin, “Defects induced damage behavior and mechanism of fused silica and KDP crystal with high power laser radiation,” Ph.D. dissertation (University of science and technology of China, 2018).
11. W. Deng and C. Jin, “Laser induced damage threshold testing of DUV optical substrates,” Chin. Opt. Lett. 11, S10702 (2013).
12. Y. Liang, “Reseach on the Influence of Optical Components Surface Contamination on the Damage Resistance Capability,” Ph.D. dissertation (University of Electronic Science and Technology of China, 2016).
13. X. Yin, Y. Zhuang, Y. Zhang, G. Zhang, X. Lu, S. Zhou, W. Ma, and J. Zhu, “Experimental study on characteristics of nanosecond laser-induced damage on optical elements,” Chin. Opt. Lett. 12(s2), S21409 (2014). [CrossRef]
14. X. Li, X. A. Dou, H. Zhu, Y. Hu, and X. Wang, “Nanosecond laser-induced surface damage and its mechanism of CaF2 optical window at 248 nm KrF excimer laser,” Sci. Rep. 10(1), 5550 (2020). [CrossRef]
15. C. W. Carr, G. J. Exarhos, M. J. Soileau, D. Ristau, V. E. Gruzdev, J. Melchior, E. Mason, and S. Das, “Accelerated testing of high fluence protective coated optics,” in Laser-Induced Damage in Optical Materials 2018: 50th Anniversary Conference, (2018).
16. L. Wang, W. Ma, and L. Ji, “Influence of metal particles on damage threshold of fused silica at 3ω,” Chinese Journal of Lasers 39, 22–26 (2012).
17. M. Bauer, M. Bischoff, S. Jukresch, T. Hülsenbusch, A. Matern, A. Görtler, R. W. Stark, A. Chuvilin, and U. Kaiser, “Exterior surface damage of calcium fluoride outcoupling mirrors for DUV lasers,” Opt. Express 17(10), 8253–8263 (2009). [CrossRef]
18. C. W. Carr, G. J. Exarhos, M. J. Soileau, D. Ristau, V. E. Gruzdev, H. Mizoguchi, Y. Usami, J. Fujimoto, M. Hattori, D. Tei, and T. Kumazaki, “Laser-induced damage and defect analysis of calcium fluoride window caused by the high pulse repetition rate of ArF excimer laser radiation,” in Laser-Induced Damage in Optical Materials 2018: 50th Anniversary Conference, (2018).
19. C. Muehlig, G. J. Exarhos, W. Triebel, A. H. Guenther, N. Kaiser, G. Toepfer, A. Jordanov, K. L. Lewis, M. J. Soileau, C. J. Stolz, A. Giesen, and H. Weber, “Calcium fluoride for ArF laser lithography: characterization by in-situ transmission and LIF measurements,” in Laser-Induced Damage in Optical Materials: 2002 and 7th International Workshop on Laser Beam and Optics Characterization, (2003).
20. X. Guo, J. Ding, Y. Zhou, and Y. Wang, “Corrosion behaviors of the copper alloy electrodes in ArF excimer laser operation process,” High Power Laser Sci. Eng. 6, e1 (2018). [CrossRef]
