Design and Manufacturing of Low Relative Humidity Chamber for Laser Processing of Lithium Metal

In order to overcome the energy density limitations of lithium-ion batteries consisting of graphite anodes, studies on lithium metal batteries (LMBs) have been actively conducted. However, most LMBs related studies focus on suppressing the growth of unpredictable dendrites. Research for production processes has been rarely conducted. In the paper, laser processing is introduced to improve the drawback of conventional processing for Li metal. Moreover, a low humidity maintenance chamber is manufactured to prevent oxidation of Li metal during laser processing because Li metal easily reacts with moisture. The chamber has a closed space that does not allow outside air to enter, and a glass that allows the laser beam to pass through is installed at the upper part. In addition, silica gel is installed to maintain low relative humidity. The dew point inside the chamber drops to -17.4 ℃ in 1 hour. This result implies that the chamber can prevent Li metal oxidation. Next, we analyze the effect of transparent plate glass on laser beam with Gaussian distribution. Finally, it is confirmed through experiments that lithium metal is not contaminated by moisture during laser processing using a manufactured chamber.


Introduction
Over the past 30 years, the energy density, power density, and lifespan of lithium-ion batteries (LIBs) have been steadily developed and have good performance compared to lead-acid batteries, Ni-Cd batteries, and Ni-MH batteries [1] [2]. The development of LIBs has improved the performance of electric applications such as smartphones, laptops, tablets, and electric vehicles. Currently, the energy density of the commercialized LIBs is approximately 250 ℎ/ , reaching almost theoretical capacity [3], [4]. Conventional LIBs have limited effective capacity enhancements due to the limitation of the theoretical capacity of graphite anodes (370 ℎ/ ) [5]. However, customers want to drive more distance by electric vehicles and use a wireless electronic device for a long time. For this reason, the development of new batteries that exceed the performance of LIBs consisting of graphite anode is actively underway.
Lithium (Li) metal is being actively discussed as an anode for the next generation batteries.
Because the theoretical capacity of the Li metal has 3860 ℎ/ , which is about 10 times higher than that of graphite, and has the lowest reduction potential (3.04 V vs. the standard hydrogen electrode) and density (0.534 / 3 ). Batteries consisting of Li metal anode are so-called Li metal batteries (LMBs) LMBs were developed before LIBs but were rarely used as secondary batteries [6]- [8]. LMBs have unpredictable Li dendrites growing on Li metal surface during operation. Li dendrites result in a reduction of lifespan, explosion, and low coulombic efficiency of batteries [9]. Currently, many studies are being conducted to suppress Li dendrite growth. Li et al. [10] produced a 3D porous Cu current collector/Li metal composite anode using copper mesh. The manufactured anode induced low local current density due to its large surface area and reduced interfacial resistance between electrolytes and electrodes. As a result, the formation of dendrites was suppressed and a stable solid electrolyte interphase layer was formed, so that the stability and lifespan of LMBs increased. Yang et al. [11] synthesized a mesoporous silica thin film (MSTF) on an anodic aluminum oxide (AAO) membrane through the polymer transfer method. When MSTF⊥AAO separator was applied to Li-Li cells, It showed stable cycling performance at ultrahigh current density (10 / 2 ) by inducing a uniform Li + flux. Park et al. [12] produced patterns on Li metal surface through stamping techniques. During Li plating/stripping processes, Li + was induced and drained in fine patterns to suppress dendrite formation, enabling a stable cycle. However, mechanical processing methods such as stamping technique may easily contaminate tools when processing Li metal which has low Mohs hardness (0.6). Contaminated tools cause secondary contamination of the workpiece, which requires continuous cleaning. This can cause productivity to be reduced in the production process [13]. Therefore, a method to achieve high repeatability, fast production speed, and high quality in Li metal processing is required. Laser processing has the advantages of contactless processing, high precision, fast process speed and easy automation, and has already successfully replaced conventional process methods in various fields [14]- [19].
However, few studies have been conducted on the interaction characteristic between laser and Li metal to apply laser processing to Li metal. This is because, in order to handle Li metal, a very low relative humidity (RH) environment such as a dry room or glove box should be required. Figure 1 shows the Li metal exposed to air at a temperature of 25℃ and RH of 40.0% over time. Silver-white surface can be observed on initial Li metal, and a black coating layer is gradually formed on Li metal over time.
According to the literature review, this black coated layer is formed by the reaction of Li metal with H2O in the air to form Li hydroxide (LiOH), Li nitride(Li3N), and Li carbonate (Li2CO3) [20], [21]. A day later, the Li metal turned completely white. When the Li metal turned black, no significant difference is observed from the surface of the initial Li metal in the SEM images, but after one day, numerous cracks are observed on the surface. Such contamination of the Li metal surface reduces the performance of the Li metal battery by causing the non-uniform current density when the battery is operated [22].

Figure 1 Chemical change of Li metal in air
In this paper, we discuss laser processing on Li metal using chamber, which has rarely been reported. Firstly, a chamber is designed and manufactured to maintain low RH. The chamber has three main parts: container lid with transparent plate glass, blocking frame, and specimen stage. The internal temperature, RH, and dew point are measured to analyze the performance of the chamber, and the validation of the chamber is verified. In addition, when a Li metal is processed using a chamber, the laser beam interacts with the Li metal after passing the transparent glass plate. Therefore, the laser beam characteristics after passing the transparent glass plate are analyzed. Finally, laser ablation experiments of Li metal are carried out using the manufactured chamber to verify the system.  The boundary conditions required for chamber manufacturing are defined as follows.

Design and manufacturing chamber
A. The dew point in the chamber should be below -15 ℃ B. Fluid entering or leaving the chamber should be blocked.
C. Laser beam irradiated from outside should be transferable to Li metal inside the chamber.
To meet the above conditions, container lid, transparent plate glass, blocking frame, and silica gel are used in manufacturing the chamber. Figure 2 shows a schematic of chamber, 3D modeling, and finished products.
For the A condition, as mentioned above, Li metal is oxidized by rapidly reacting with oxygen and nitrogen in the air. In other words, maintaining the low RH in the air can slow down the reaction between oxygen and nitrogen as much as possible. Here, the criteria are defined as dew point because RH varies with temperature and pressure. The dew point means the amount of moisture in the air regardless of the temperature and pressure. Moreover, the reason for setting the minimum dew point of -15 is that Li metal can have been treated at -15 ℃ in previous studies [23]. To keep the internal dew point of the chamber, silica gel is used. As shown in Figure 2, the position of the silica gel is located at the bottom of the specimen stage and inside the blocking frame. A-type and B-type of silica gel are utilized, and present physicochemical property of each silica gel in Table 1. A-type silica gel has a surface area of 773.60 2 / and a pore diameter of 22 Å and is effective in relatively low RH environments. On the other hand, B-type silica gel has a surface area of 574.30 2 / , and a pore diameter of 66 Å and is effective at relatively high RH [24]. For this reason, two different silica gels are used. The two silica gels are mixed at a volume ratio of 1:1 to fill about 80% of the chamber volume.
In the case of condition B, because the dew point is not kept stable when high RH air flows into the chamber, the container lid and blocking frame block the entering fluid from the outside.
In the C condition, transparent plate glass made of silicon dioxide (SiO2) is used because laser beam externally irradiated must be reliably transferred to the surface of Li metal. Transparent plate glass is bound to container lid as shown in Figure 2. 3. Characteristic analysis of low relative humidity maintenance Therefore, since the manufactured chamber can maintain a dew point of below -15 ℃, it is implied that laser processing is possible by minimizing Li metal contamination caused by moisture.

Analysis of gaussian beam characteristics
We analyze the effect of transparent glass plate on Gaussian beam characteristics. Gaussian beam distribution can be represented as . 2 [26], [27].
where 0 is peak intensity, is laser power, is the distance to the center of the beam, and 0 is beam waist. Therefore, analysis of , and 0 is required to understand the Gaussian beam that changed after passing the transparent glass plate.
In Figure 5 (a), the Gaussian beam propagating along the z-axis (optical axis) on a uniform medium can be represented . 3 [28]- [30]. Where, is complex beam parameter, is radius of the phase front of the beam, is beam waist, ( 2 ) is wavenumber, is wavelength ,and is refractive index. From . 3 − 5, . 6 can be obtained.
( ) = + 2 ( = 1,2,3,4 … ) When the Gaussian beam passes through the optical system, the relationship between the beam before passing ( 1 ) and the beam after passing ( 2 ) can be obtained through ABCD matrix. ABCD matrix is 2 × 2 matrix that allows a beam to be tracked according to elements of the optical system when a beam has paraxial ray properties, such as a Gaussian beam. Elements A, B, C and D of Matrix are determined by the optical system, and the relational expression can write it down as shown in . 7 [30].
In Figure 5(b), can be represented by a column vector of height , angle from the optical axis.
So, we can rewrite it as . 8.
The matrix above is developed as follows: Here, the following conditions are established due to the characteristics of the paraxial ray.
We can estimate the Gaussian beam change by . 14 , when the Gaussian beam passes through a medium with a refractive index of 1 to 2 .
Let`s estimate the change in Gaussian beam passing through two interfaces, such as glass.
When the Gaussian beam proceeds to a flat dielectric interface such as Figure 5 If the Gaussian beam moves by on the same medium, each element of ABCD matrix is = 1, = , = 0, = 1.
Next, if the Gaussian beam progresses from 2 to 3 , it can be expressed in the same way as . 16.
. 20 can be obtained from . 15 − 19. In this case, negative values must be substituted because the slope of the incident beam is less than zero.
Δ can be calculated as follows: For this study, since 1 = 3 , it can be simply calculated as follows: . 20 indicates that if the Gaussian beam passes through transparent plate glass, the focal length is determined by the refractive index and thickness of the transparent plate glass. Therefore, the glass used in the chamber is approximately 1200 of thickness and the refractive index is 1.42, so that it is located approximately 353 away from the original focal length.

Experiments and results
Section 4 conducts experiments to verify the theoretical approach in Section 3.2. Nanosecond   the DOF of laser beam passed through the glass moves as much as the focus has far away. Moreover, a crater diameter formed by the laser beam passed through glass is produced smaller than that of the original beam. This is because the power loss is caused by the glass. Hereby, the laser beam change after passing through the glass is experimentally verified.    [32]. In addition, the crater diameter increases with increasing pulse energy. When pulse energy above 15 is applied, burr begins to form. In particular, no contamination was observed on the surface of Li metal by moisture, and it was confirmed that Li metal laser processing was possible through manufactured chamber.

Conclusion
In this study, a low humidity maintenance chamber was manufactured to apply the laser process to Li metal. The performance of the manufactured chamber was evaluated and the effect of plate glass on Gaussian beam was analyzed. Finally, the possibility of laser processing of Li metal was confirmed through experiments. The main results of this study are as follows: 1. The dew point inside the manufactured chamber fell to -17.4 ℃ in 60 minutes. After 240 minutes, the dew point was maintained at -23.8 ℃~-26.4 ℃. As a result, contamination of Li metal by H2O could be prevented through the manufactured chamber.

2.
When the Gaussian beam passed through the plate glass, the laser power was lost due to absorption and reflection of the glass, and the focal length was determined depending on the thickness and reflective index of the plate glass.
3. Laser ablation of Li metal was conducted using a chamber. Like other metal laser ablation, traces of melting, evaporation, and melt expulsion were observed. In addition, no contamination of Li metal was observed after experiments. Therefore, it is believed that Li metal laser processing is possible through the manufactured chamber.
Future plans will study the interaction characteristics between lithium metal and nanosecond laser using the manufactured chamber. In particular, laser ablation thresholds and incubation coefficients are studied.