Method for releasing carbon foils from substrates

. The performance of a carbon foil (C-foil) as a charge stripper in an accelerator is directly affected by its condition and quality. C-foils are fabricated by evaporating carbon onto a substrate. Therefore, a suitable technique must be established for depositing and removing C-foils from substrates. The selection of appropriate releasing agents is crucial for the removal process. C-foils with thicknesses less than 0.1 mg/cm 2 were produced using the arc discharge method, wherein chloride materials were used as releasing agents. The magnetron sputtering method is suitable for manufacturing C-foils thicker than 0.1 mg/cm 2 using appropriate releasing agents, such as fluorine-based or silicon-based agents. The contact angle analysis demonstrates a negative relationship between the C-foil thickness and the wettability of the substrate treated with an appropriate releasing agent. The possibility of using ionic liquids as novel releasing agents is also investigated.


Introduction
The development of long-life carbon foils (C-foils) for charge strippers at RIKEN RI Beam Factory began in 1999. Because C-foils are manufactured via vapor deposition on a glass substrate or silicon wafer, they must be removed from the substrate or wafer prior to their installation in the accelerator beam line. The condition and quality of the C-foils directly affect the Cfoil lifetime and stability of the charge-stripped beam intensity. Therefore, a suitable technique is required for removing C-foils from the substrates.
The selection of an optimized releasing agent, which is applied on the substrate before the evaporation step, is crucial for the removal process. This agent aids the removal of the carbon layer, maintaining the C-foil condition as intact as possible [1]. Generally, the carbon layer is removed such that the foil floats easily on water. However, humidity entering carbon layer itself negatively affects the foil removal and causes damage resulting in a foil breakage. Therefore, it is necessary to devise a method to prevent humidity from entering the carbon layer while allowing water to permeate between the carbon layer and the substrate at an appropriate rate. This is the reason why we usually use hydrophilic releasing agents such as chloride, soap (a surfactant), and betaine saccharose. However, unless releasing agents with optimum amount and thickness are used, especially in the case of large thickness, water rapidly penetrates into the carbon layer and the C-foil breaks immediately. On the other hand, if the amount of the releasing agent is too small, the carbon foil would not be removed at all. * Corresponding author: hasebe@riken.jp 2 Releasing agents for the arc discharge method A thin C-foil with a thickness ranging from 10-80 μg/cm 2 was fabricated using the arc discharge method [2]. Before evaporating carbon, a chloride-based releasing agent was deposited onto the substrate using a resistance-heated source. Here, powder-type NiCl2 and LaCl3 were used as releasing agents because powdertype agents had smaller particles and were easier to deposit than granular type. The thickness of the releasing agent was measured using a crystal rate thickness monitor and was equivalent to 10 μg/cm 2 of carbon. The C-foil was removed from the substrate using the general method of floatation on water and was subsequently attached to a holder.

Releasing agents for the magnetron sputtering method
A C-foil thicker than 0.1 mg/cm 2 was fabricated using the magnetron sputtering method [3]. Similar to the case of the thin C-foils, the substrate surface was initially treated with a releasing agent. After testing various releasing agents manufactured by different companies, it was determined that using different releasing agents based on the thickness range is optimal. For carbon thicknesses of 0.1-0.3 mg/cm 2 , fluorine-based releasing agents, such as DAIFREE (GA-6310) [4] are suitable. For C-foil thicker than 0.3 mg/cm 2 , a silicon-based releasing agent RELEASE [5] is suitable. The substrate surface was treated as follows: EPJ Web of Conferences 285, 05001 (2023) https://doi.org/10.1051/epjconf/202328505001 INTDS2022 1. The releasing agent was uniformly sprayed onto the substrate using spraying cans, and thereafter, the substrate was dried. 2. The agent on the surface was wiped off with lowdust paper or cloth. 3. The agent layer was wiped thoroughly until it was not visible anymore to the naked eye. This is the general procedure for applying releasing agents. It takes several hours to overnight for a C-foil to float on the water while it takes only a few minutes in the case of the C-foil by the arc discharge method using the chloride-releasing agent. Most of the time is spent for water to penetrate between the carbon layer and the substrate. Our tests showed that this releasing agent can be applied to the fabrication of thin metal foils, such as palladium foils.

Method for measuring contact angles
To quantify the surface condition of the substrate before C-foil deposition, the contact angles of pure water were measured on various substrates using the sessile drop method. The apparatus used was the KYOWA Interface Measurement and Analysis System (FAMAS). Two microliters of pure water were dropped on the substrate, and after 5 seconds, the contact angle was measured using FAMAS.

Contact angle on the glass slide
The contact angles of pure water on a glass slide were measured with and without the application of a releasing agent. Figure 1 shows photographs of the contact angle measurements using glass slides treated with chloridebased agents: (a) NiCl2 and (b) LaCl3. The contact angle on the untreated glass slide was 20.2°. The angles measured for the samples treated with NiCl2 and LaCl3 were 31.5° and 36.5°, respectively. The contact angles measured on the slides treated with chloride agents were larger than those on the untreated slides, indicating a decrease in wettability. The contact angle was larger for substrate treated with LaCl3 than that with NiCl2.

Contact angles on the Si wafer
The contact angles on a Si wafer were measured after treating it with fluorine-and silicon-based releasing agents. Figure 2 shows photographs of the measurements obtained using the Si wafer. Figure 2 (a) and (b) show the contact angles of the droplets on wafers treated with the fluorine-based agent DAIFREE and the silicon-based agent RELEASE, respectively. The contact angles on the untreated Si wafers and Si wafers treated with DAIFREE and RELEASE are 30.2°, 57.6°, and 103.5°, respectively. Similar to the case of the glass substrate, the contact angles were larger for the treated wafers than those for the untreated ones. Furthermore, the application of silicon-based RELEASE led to significantly larger contact angles than that of fluorinebased DAIFREE.
(a) DAIFREE 57.6° (b) RELEASE 103.5°  Table 1 lists the measured contact angles for each combination of the substrate and releasing agent along with the C-foil thickness ranges and deposition methods. Release agents giving higher contact angles are needed for foils with increased thickness. In other words, a negative relationship exists between the C-foil thickness and the required wettability of the substrate.

Other releasing agents
Furthermore, potential candidates for application as efficient releasing agents have also been identified. Ionic liquids (ILs), which negligibly evaporate at room temperature, or even in vacuum, are potentially promising materials as releasing agents. Since ILs do not easily change state in any environment, it is thought that the effect on the C-foil is small as well even if it is taken out of the chamber after vapor deposition. Moreover, if IL is deposited on a substrate while maintaining its properties as a liquid, an extremely thin and uniform layer of release agent can be formed. This time, a water-soluble IL (ILA48-32) [6] was tested to determine the feasibility of using it as a releasing agent. This IL had a decomposition temperature of 170 °C. However, most ILs are toxic and require careful handling. The IL was tested as a releasing agent as follows: 1. The IL was placed in a deep Ta boat and covered with a perforated Ta lid (Fig. 3 (a)). 2. The IL was evaporated with a resistance-heating evaporation source. 3. Carbon was deposited via the arc discharge method. 4. The carbon-deposited substrate was removed from the device and allowed to float on water. The carbon layer was then peeled off.
The peeled C-foil condition is poor, as shown in Fig.  3 (b). This might be because the IL amount was not optimized, and the IL did not spread uniformly on the substrate. The optimization of the IL amount is difficult because the thickness of the IL layer cannot be measured using a crystal rate thickness monitor. Herein, the entire releasing agent is heated simultaneously, and the IL is immediately boiled, forming various sizes of molecular clusters of the IL as a flux to the substrate. As a result, the IL does not adhere uniformly to the substrate because of the various sizes of the molecular clusters in the deposits.
It is thought that when the IL amount is controlled and a small amount of IL is evaporated many times in a short period to result in a flux with a uniform molecular density, the IL will adhere uniformly to the substrate. However, the evaporation source with a Ta boat used in this study is not suitable for this purpose. The laser evaporation of IL would be a better method for C-foil fabrication compared to that used in this study.

Summary
Chloride-based release agents are suitable for C-foils with thicknesses of 10-80 μg/cm 2 , which are fabricated using the arc discharge method. Among the chloride materials, NiCl2 and LaCl3 are suitable for C-foil thicknesses lower and higher than 50 μg/cm 2 , respectively. Fluorine-and silicon-based releasing agents are appropriate for C-foil thicknesses ranging from 0.1-0.3 and 0.3-10 mg/cm 2 , respectively. The damage caused by humidity is significantly reduced when appropriate releasing agents are applied. Additionally, we found that there is a relationship between the C-foil thicknesses and contact angles of pure water droplets on the substrate treated with an adequate releasing agent. ILs are potentially promising candidates for application as releasing agents when used in conjunction with an appropriate method.