Evolution of hydrogen isotopes retention behavior of in-situ boronization films in EAST

Effective management of hydrogen isotopes retention in plasma-facing materials (PFMs) is crucial, particularly when utilizing tritium (T) as fuel, for the success of burning plasma operations. Boronization, a widely employed technique for controlling fuel recycling and mitigating impurity influx from plasma-surface interactions into the core of burning plasma, significantly influences hydrogen isotopes retention in PFMs. In this study, boronization films were generated by ion cyclotron range of frequency (ICRF) discharge assisted boronization with C2B10H12 as boron source on tungsten substrates at Experimental Advanced Superconducting Tokamak (EAST) which employing ITER-like water-cooled W monoblock PFMs and components (PFCs), followed by in-situ glow discharge (GD) cleaning for 2 h and 20 shots (180 s) edge-plasma exposure. Employing the Material and Plasma Evaluation System (MAPES), representative samples were analyzed after each process. The resultant carbon–boron films, dense and continuous, exhibited thickness up to 120 nm and were identified as amorphous in structure. It was observed that the D2-GD cleaning resulted in a significant isotopic exchange effect, effectively reducing the hydrogen (H) retention in the carbon–boron films. This hydrogen isotope replacement efficiency was found to be influenced by the thickness of the films. Notably, after boundary plasma exposure, samples with thicker films demonstrated an enhanced capacity to capture deuterium (D), adsorbing 10 times more D than bare tungsten (W). Our findings offer transformative insights for T recycling analysis and the plasma operation of devices like International Thermonuclear Experimental Reactor (ITER), highlighting the impact of boronization and subsequent treatments on hydrogen isotope retention behavior in PFMs.

Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

Introduction
Proper conditioning of plasma-facing components (PFCs) is crucial for the operation of current tokamak devices [1,2].Among various methods, 'boronization', or the deposition of boron coatings, is routinely utilized to manage fuel recycling and impurity influx from plasma-surface interactions into the core of the burning plasma [3].Effective control of hydrogen isotopes retention in PFCs, especially when using tritium (T) as fuel, is essential for successful burning plasma operations.Consequently, the retention behavior of hydrogen isotopes in boron (B) coatings has garnered significant attention [4][5][6].Characterization of hydrogen isotopes retention typically follows two strategies: The first involves in-situ measurement, such as gas balance.HT-7, which employing molybdenum (Mo) as PFCs, indicated that total deuterium (D) retention in the fusion device could be as high as ∼70%-80% postboronization [7].In the Alcator C-Mod, which employing Mo and tungsten (W) as PFCs, the study compared hydrogen/deuterium (H/D) retention rates under conditions of boronization and without boronization.It was observed that, in both scenarios, the retention rates ranged between 20% to 40% [8].The second in-situ method employs accelerator-based diagnostic techniques [9], enabling assessment of hydrogen isotope retention on PFC surfaces with boron films across extensive spatial ranges during various plasma discharge phases, thus revealing macroscopic evolutionary patterns.However, these methods have inherent limitations and environmental factors, leading to significant margins of error [10,11].An alternative is ex-situ analysis of boronization samples post-operation, typically involving thousands of tokamak pulses, using techniques like Thermal Desorption Spectroscopy (TDS) or Nuclear Reaction Analysis (NRA) [12,13].This strategy provides accurate data on hydrogen isotopes retention of boron films.Nonetheless, it offers limited insight into the dynamic evolution of the hydrogen isotopes retention.
The retention of T poses a significant challenge for International Thermonuclear Experimental Reactor (ITER), especially considering the projected injection of 50 g [6] of T per full-power D-T discharge and an in-vessel limit possibly as low as 700 g [14].Recently, ITER has considered a potential change in the first wall material from beryllium (Be) to W, with boronization suggested as an additional wall treatment.In the context of hydrogen isotopes, a key challenge is the predictive control of hydrogen isotopes recycling in the boronized tungsten divertor and first wall.Thus, a comprehensive understanding of the dynamic evolution of hydrogen isotopes retention is crucial for ensuring the reliable long-pulse operation of ITER.The EAST tokamak serves as an important experimental platform for conducting ITER-like steady-state advanced plasma science and technology research, featuring fully superconducting magnets and ITER-like water-cooled W monoblock PFCs [15,16].The boronization technique, employing carborane (C 2 B 10 H 12 ) as a boron source and assisted by ion cyclotron range of frequency discharge (ICRF), has been used in the HT-7 device and EAST device for many years [17][18][19].Consequently, EAST provides an excellent platform to study the hydrogen isotope retention behavior of in-situ boronization films.The Material and Plasma Evaluation System (MAPES) in the EAST tokamak [20] allows samples to be transferred to the main SOL region at the mid-plane, which can facilitate the acquisition of in-situ boronization samples and plasma exposure of the boronization samples.
In this work, we report the results of the hydrogen isotopes retention of in-situ boronization films obtained in 2023 summer EAST campaign employing MAPES.The evolution before and after deuterium glow discharge (D 2 -GD) exposure and boundary plasma exposure has been investigated.

Experimental setup
The experimental procedure at EAST consists of three distinct phases.The first phase involves in-situ wall treatment, introducing mainly boron coatings onto the samples.In the second phase, to manage the high H/(H + D) ratio in the film, a twohour D 2 -GD cleaning process is conducted, facilitating isotopic substitution of H with D. The final phase encompasses boundary plasma exposure.To investigate the retention behavior of hydrogen isotopes throughout these phases, up to nine samples were utilized, as detailed in table 1.

Substrate samples preparation
The substrate samples used in this study are ITER-grade rolled tungsten, provided by Advanced Technology & Materials Co.To reduce the influence of surface roughness.Each sample was mechanically polished to achieve a mirror-like surface, ensuring the polished face was perpendicular to the rolling direction.Following this, the samples were subject to ultrasonic cleaning for 15 min, first in high-purity ethanol and then in acetone.After cleaning, they were placed in a quartz furnace under high vacuum (10 −5 Pa), heated to 500 • C, and held at this temperature for 2 h.The surface morphology of the substrates is illustrated in figure 1.

Boronization
During the 2023 summer campaign of the EAST tokamak, the prepared tungsten substrate samples were positioned at a major radius of 2350 mm using the MAPES platform, as depicted in figure 2. The first boronization session utilized 30 kW ICRF, 12 g of C 2 B 10 H 12 , and D 2 as the auxiliary gas for sample 1-3.The second session involved 15 kW ICRF, 12 g of C 2 B 10 H 12 , employing helium (He) as the auxiliary gas for

Deuterium glow discharge and boundary plasma exposure
Two hours of D 2 -GDC were conducted using four fixed anodes positioned at the P-A, B-C, F-G, and J-K ports.During this process, each anode operated at an average power of 400 W (2 A, 200 V) under a working pressure of approximately 10 −1 Pa, with a magnetic field (B t ) of 0 T. The MAPES sample plate was positioned at a major radius of 2350 mm throughout the D 2 -GD exposure.In addition to the three sets of boronized samples, a group of bare tungsten was included for comparative analysis.Given the significant oxygen trapping capacity of carbon-boron films, as reported in the literature [21][22][23], it was crucial to minimize the air exposure time of the samples.The process involved extracting boronized samples from the MAPES sample exchange chamber for additional characterizations, while simultaneously introducing supplementary bare tungsten samples for D 2 -GD exposure along with the remaining boronized samples.The duration of air exposure for the samples was kept within ten minutes.Immediately after extraction, all samples designated for further analyses were placed in vacuum containers (pressure <10 −1 Pa).After D 2 -GD exposure, the MAPES sample exchange chamber was reopened to extract sample 2, sample 5 and sample 8, then quickly re-established vacuum.The remaining bare tungsten sample and the sample 3, sample 6, sample 9 were subsequently exposed to the EAST boundary D plasma.The leading boundary of the sample plate, or more precisely the sample surface, was advanced to the position at a major radius of 2355 mm, which is 5 mm behind the limiter.20 cycles of discharge with a single discharge configuration from #128758 to #128778 was followed.We use shot #128759 as an example to illustrate the discharge parameters, as presented in figure 3(a): B t = 2.4 T, I p = 400 kA, P total = 1.2 MW, employing exclusively the lower hybrid wave heating method, with an n e of 2.0-2.5 × 10 19 m −3 .Excluding the rising boundary and falling boundary, the effective discharge duration is approximately 9 s, that is, the total effective exposure duration is approximately 180 s.Parameters of the boundary plasma were estimated by Helium Beam Emission Spectroscopy (He-BES) [24].As shown in figure 3(b), T e and n e at the sample surface are estimated to be ∼13 eV and ∼3 × 10 18 m 3 , respectively.The temperature of the samples during plasma exposure was measured in real-time by a thermocouple mounted on the back of the sample holder, with the data visible in figure 3(c).The temperature of the samples gradually increases with the accumulation of discharge shots, reaching a maximum of 65 • C. Due to the thermocouple inability of the tip to withstand direct exposure to the plasma or to make direct contact with the sample surface, it is shielded at the rear of the sample holder, separated  by an approximate 5 mm-thick barrier.Consequently, there is an estimated deviation ranging from 1/3 to 1/5 between the measured and actual temperatures.This discrepancy suggests that the true maximum temperature of the sample surface is higher than the recorded.

Samples characterization 2.4.1. Bonding information characterization.
The elemental bonding information in the film layer was determined through Raman spectroscopy (Lab-Ram HR).

Cross-sectional morphology and film thickness characterization.
A dual-beam focused ion beam (FIB, FEI-Strata 400S) was utilized to prepare transmission electron microscope (TEM) specimens from the cross sections of samples 1 and 3.The microstructure of these samples was examined using a TEM (Talos-F200S) equipped with an energy-dispersive x-ray spectrometer (EDS) detector.Phase identification was performed using selected-area electron diffraction (SAED).The precise thickness of the carbon-boron film in samples 1 and 3 was determined using TEM.The film thickness for samples 4 and 7 was inferred by comparing the film thickness ratios obtained from glow discharge optical emission spectroscopy (GDOES, Spectrum GDA 750) for samples 1,4,7 with the accurately measured thickness of sample 1 via TEM.

Hydrogen retention behavior characterization.
Hydrogen retention behavior in the samples was quantified using TDS.During the TDS measurements, the samples were heated in a quartz tube under a typical vacuum background pressure of approximately 1 × 10 −5 Pa, from room temperature to 1000 • C, with a ramping rate of 10 • C min −1 .The temperature of the samples was monitored using a thermocouple, which was inserted into the vacuum chamber and attached to the samples during heating.The mass signal was detected by a quadrupole mass spectrometer.

Characterization of in-situ boronization films
TEM was employed to examine the cross-sectional morphology and compositional distribution of the samples.Figure 5(a) displays the TEM micrograph of the entire specimen.The average thickness of the carbon-boron film on Sample 1, formed under the conditions of 30 kW ICRF, 12 g C 2 B 10 H 12 , and D 2 as the auxiliary gas, is approximately 120 nm.The corresponding SAED pattern, shown in figure 4(b), indicates that the structure of the carbon-boron film is amorphous.This finding is consistent with previous research conducted at Surface modification Testsland, which reported that the structure of the boron film was confirmed as amorphous [3].Moreover, TEM bright-field micrographs of the carbon-boron film, as seen in figure 4(c), reveal a large number of nano-sized black particles dispersed throughout the film.These particles tend to exhibit a certain orientation, vertically aligned, extending in straight lines either towards the upper surface of the film or towards the upper surface of the tungsten substrate.The primary components of the film layer, C, B, and O, were revealed by EDS.It is a pity that the chemical composition of the black particles cannot be confirmed owing to the limitations of the equipment.
Figure 5(c) reveals the calibration results of the SAED diffraction pattern near the film area of the tungsten substrate, indicating that the analyzed area corresponds to the (110) crystallographic plane of the tungsten substrate lattice.To further examine the interface between the tungsten substrate and the carbon-boron film, a clear lattice view was obtained by applying fast Fourier transform (FFT) and inverse FFT (IFFT) to the TEM image.As depicted in figure 5(b), the examined region, encompassing a few nanometers of the tungsten substrate and the carbon-boron layer, reveals that within the carbon-boron film, the area in contact with the tungsten substrate features a 1-2 nm layer that does not exhibit a completely amorphous structure.This layer shows certain preferential orientations corresponding to the lattice arrangement of the tungsten substrate, with the carbon-boron film atoms closer to the substrate surface displaying more pronounced orientation.Beyond 2 nm  from the substrate, the film transitions to a fully amorphous structure.
The thicknesses of samples 4 and 7 were estimated based on the TEM findings from sample 1, in conjunction with the GDOES results.Figure 6 shows the depth distribution of B and W elements obtained by GDOES for sample 1, sample 4, and sample 7. Notable peaks in the B signal and ascending steps in the W signal were observed.Given that the fabrication process of these three samples is consistent, differing only in parameters, it is feasible to infer that their etching rates are similar.Consequently, the ratios of the peak widths in the B signals and the timings of the rising steps in the W signals can be approximated to the thickness ratios of the films.Thus, the thickness ratios for sample 1, sample 4, and sample 7 are roughly 6:1:1.5.Based on these ratios derived from GDOES, the thickness of the carbon-boron film for sample 4, formed under the conditions of 15 kW ICRF, 12 g C 2 B 10 H 12 , and He as the auxiliary gas, is estimated to be about 20 nm.Similarly, the thickness of the carbon-boron film for sample 7, prepared using 30 kW ICRF, 6 g C 2 B 10 H 12 , and He as the auxiliary gas, is approximately 30 nm.
To gain a deeper understanding of the chemical composition of the carbon-boron film, Raman spectroscopy was utilized.The thin films on the surfaces of sample 4 and 7 are beyond the detection limit of the Raman spectroscopy used, so the compositional analysis primarily focuses on findings from Sample 1, as illustrated in figure 7. Notable peaks are observed in the ranges of 700 cm −1 -900 cm −1 and 2400 cm −1 -3200 cm −1 .Drawing on existing literature, peaks at 3068 cm −1 , 2929 cm −1 , 2597 cm −1 , and 772 cm −1 are identified as stretching vibrations of C-H/C-D, C-H/C-D, H-B/D-B, and B-H/B-D [25,26], respectively.Considering the sample's complex composition, the dominant Raman peaks might overshadow those from minor constituents.To address this point, deconvolution analysis was conducted to discern the minor components in the sample.Peaks at 3068 cm −1 , 2950 cm −1 , and 2887 cm −1 are all associated with C-H/C-D stretching vibrations, while those at 2604 cm −1 , 2535 cm −1 , and 777 cm −1 corresponded to B-H/B-D stretching vibrations.Additionally, weaker peaks at 865 cm −1 and 815 cm −1 , identified as closely related to B(OH) 3 /B(OD) 3 [27], suggest the potential presence of trace amounts of these compounds in the sample.
TDS was employed to analyze the retention behavior of hydrogen isotopes in samples following in-situ ICRF boronization.The results, presented in figure 8, show that the total release amount of H 2 from ITER-grade rolled tungsten substrate after preparation is 8.36 × 10 19 m −2 .For the boronized samples, prominent hydrogen isotope release peaks in sample 1 and sample 7 are observed, with the main desorption peaks of H 2 , HD, and D 2 located at 350 • C and 630 • C, as illustrated in figures 8(a)-(c).When comparing sample 1 and sample 7, the HD desorption peak in Sample 4 is not as evident, as shown in figure 8(b), and no D 2 desorption peak is observed in figure 8(c).Raman spectroscopy results indicate that hydrogen in the sample mainly exists in the form of compounds formed with C and B. Therefore, the release peaks at around 500 • C and 750 • C could be speculated to correspond to hydrogen isotopes bound to B and C, respectively [28][29][30].
The total amount of D desorbed from sample 1 is measured at 5.52 × 10 19 m −2 , which is ten times the 5.25 × 10 18 m −2 from sample 4 and 2.5 times the 2.25 × 10 19 m −2 from sample 7, as indicated in table 2. After accounting for the total H release from the tungsten substrate, the H/(H + D) ratio in the samples was calculated.No significant difference was observed in the H/(H + D) ratio among the three samples.Specifically, the ratios of H/(H + D) were 0.956 for sample 1, 0.977 for sample 4, and 0.995 for sample 7. The D retention in sample 1 results from the use of D 2 as the auxiliary gas and D desorption from the EAST tokamak's entire inner surface during ICRF discharge.Similarly, the D retention in samples 4 and 7 primarily stems from D desorption across the same surface area during ICRF discharge.These findings suggest that the H/(H + D) ratio in the in-situ boronized carbon-boron films is unaffected by the choice of D 2 as the auxiliary gas.
After accounting for the total amount of H released from the tungsten substrate, the combined amount of H and D released from sample 1 is calculated to be 1.24 × 10 21 m −2 .This amount is significantly higher than that from sample 4, which is 2.49 × 10 20 m −2 (approximately 4.98 times lower) and from sample 7, which is 4.37 × 10 20 m −2 (approximately 2.84 times lower).These results suggest that as the thickness of the carbon-boron film increases, there is a corresponding increase in the total retention of hydrogen isotopes in whole sample including the tungsten substrate and the carbonboron coating.This indicates a direct relationship between film thickness and isotope retention capacity in these coated substrates.

Hydrogen isotopes retention behavior after D 2 -GD
The analysis results of TDS regarding the retention of hydrogen isotopes and the thermal release behavior of boron isotopes in the sample group after D 2 -GD exposure are presented in figure 9 and table 3. It is observed that there is no significant change in the main temperature range of the primary release peaks of the sample group before and after D 2 -GD exposure (as shown in figures 9(a)-(d)).This indicates that the principal trapping sites for hydrogen isotopes within the film remained unchanged before and after the D 2 -GD exposure.
Regarding the total desorption of hydrogen isotopes, as detailed in table 3: In sample 2, compared to sample 1, the total H desorption decreases by 25.6%, while D desorption increases by 21.2%.In Sample 5, relative to sample 4, the total H desorption decreases by 52.8%, and D desorption increases by 434.3%.In sample 8, compared to sample 7, the total H desorption decreases by 36.7%, and D desorption increases by 73%.After subtracting the total amount of H inherent in the tungsten substrate, the H/(H + D) ratios in Sample 2, 5, and 8 were calculated, yielding results of 0.927 for Sample    Table 3.Total release amounts of hydrogen isotopes from tungsten and tungsten with different thickness of carbon-boron film after D 2 -GD exposure.
Tungsten 1.12 × 10 20 2, 0.545 for Sample 5, and 0.842 for Sample 8.These results indicate that D 2 -GD exposure markedly reducing H adsorption and enhancing D adsorption.Notably, there exists a negative correlation between the hydrogen isotope replacement effect and the carbon-boron film's thickness.This suggests that the optimal duration of D 2 -GD cleaning, aimed at facilitating hydrogen isotope exchange in the carbon-boron film, should be tailored according to the film's actual thickness.
Nonetheless, given the current state of research, it cannot be ruled out that during D 2 -GD (low energy, low beam current D plasma) cleaning, D predominantly accumulates in the film's shallow surface region, complicating the replacement of internally adsorbed H. Therefore, advancing research to elucidate the mechanisms underlying this phenomenon is crucial for refining the treatment process to achieve the desired isotopic compositions within the film.

Hydrogen isotopes retention behavior after D 2 -GD and boundary plasma exposure
TEM was utilized to examine the cross-sectional morphology of sample 3, which had been exposed to both D 2 -GD and boundary plasma.As indicated in figure 10(a), following the boundary plasma exposure, the remaining thickness of the carbon-boron film is approximately 75 nm.This suggests that around 45 nm of the film is eroded by the boundary plasma during exposure, with an estimated erosion rate of about 2.25 nm per shot.After exposure to boundary plasma, samples 6 and 9 still retain a carbon-boron film on their surfaces indicated by characterization of surface morphology, hinting at a potential role played by the recycling, recombination, and redeposition of scraped B and C on the surface in the etching process by edge plasma on the carbon-boron film.Figure 10(b) presents the cross-sectional morphology of the carbon-boron film, showing no significant differences when compared to the carbon-boron film's cross-sectional morphology depicted in figure 4(c).Figure 10(c) displays the interface between the carbon-boron film and the tungsten substrate.Furthermore, figure 10(d) illustrates the results of applying FFT followed by IFFT to a localized area of figure 4(c).This result is almost identical to what is seen in figure 5(b).In the region of the carbon-boron film that is adjacent to the tungsten substrate surface, a 1-2 nm area exhibits a lattice arrangement that correlates with the lattice arrangement of the tungsten substrate.This observation suggests that the D 2 -GD exposure and boundary plasma exposure processes do not have a structural impact on the interface between the carbon-boron film and the tungsten substrate.The stability of this interface under such conditions is noteworthy, indicating the resilience of the carbon-boron film's structural integrity in response to these experimental treatments.
The TDS results of the sample group following exposure to the EAST boundary plasma are depicted in figure 11 and summarized in table 4. The temperature range for the main release peaks does not show significant movement, indicating that the boundary plasma exposure does not substantially affect the primary trapping sites for hydrogen isotopes within the film.Table 4 summarizes the total amounts of H and D desorption in both bare tungsten and the sample group after exposure to the boundary plasma.For bare tungsten, compared to the state after two hours of D 2 -GD exposure, H desorption decreases by 21.5%, whereas D desorption increases by 10.8%.For sample 3, compared to sample 2, H desorption drops by 1.9%, but D desorption surges by 364.1%.In sample 6, relative to sample 5, H desorption increases by 41.2%, while D desorption decreases by 50.3%.For sample 9, compared to sample 8, there is a decline of 5.7% in H desorption and a reduction of 57.7% in D desorption.In summary, after 20 shots (180 s) exposure under the boundary plasma conditions of T e ∼ 13 eV and n e ∼ 3 × 10 18 m −3 , the samples exhibit the following two phenomena: 1. Regarding total H desorption, bare tungsten, sample 3, and sample 9 remained relatively stable with a potential slight decrease, whereas sample 6 exhibited an increase instead of a decrease.2. As for total D desorption, bare tungsten shows roughly the same levels with a potential slight increase, while samples 6 and 9 experiences decrease, and sample 3 demonstrates a dramatic surge.
The observed decrease in D desorption in samples 6 and 9 can be attributed to the thinning of the carbon-boron film, which is caused by the etching by the boundary plasma.This reduction in film thickness leads to less retention of D in these samples.On the other hand, the increased D retention seen in Sample 3 highlights the carbon-boron film's superior capability to capture D atoms.Notably, for Sample 3, the boundary plasma scraped off 45 nm of the carbonboron film.This implies that under the experimental conditions set, when a tungsten substrate surface is coated with a 120 nm thick carbon-boron film, the sample's capacity to capture D increases to 20 times that of bare tungsten.This finding is significant from a materials science perspective, as it demonstrates that the presence of the carbon-boron film is a key factor in the enhanced total D retention observed after boronization.

Conclusions and outlook
During the EAST 2023 summer campaign, we utilized the MAPES experimental platform to produce carbon-boron films via in-situ boronization under three different conditions.Following this, bare tungsten and a set of samples underwent a 2 h, 1.6 kW D 2 -GD discharge (sample 2, 5, and 8), succeeded by exposure to 20 cycles (180 s) of EAST's 1.2 MW single configuration discharge (sample 3, 6, and 9).The evolution of the hydrogen isotopes retention behavior before and after deuterium glow discharge exposure and boundary plasma exposure has been investigated mainly by TDS, which are summarized and illustrated in figure 12.
The main conclusions of this work are listed below: (1) Amorphous carbon-boron films with a thickness of up to 120 nm were successfully deposited in-situ on the tungsten substrate.Following exposure to D 2 -GD and 1.2 MW Lmode discharge boundary plasma, there was a noticeable reduction in the thickness of the films.However, the adhesion of the carbon-boron film to the tungsten substrate remained stable throughout these processes.cleaning time under the tokamak's actual operating conditions, is crucial.After exposure to boundary plasma, the sample with a remaining film thickness of 76 nm demonstrated a remarkable ability to adsorb D, capturing ten times more D than bare tungsten under similar plasma exposure conditions.This observation underscores the potential role of the carbon-boron film in enhancing total D retention in tokamak devices, especially following wall treatment processes.

Figure 2 .
Figure 2. Material and plasma evaluation system on EAST.

Figure 3 .
Figure 3. (a) Discharge parameters of shot #128759 of EAST.(b) Parameters of boundary plasma of shot #128759 estimated by HeBES.(c) Samples temperature during boundary plasma exposure estimated by armed thermocouple.

Figure 4 .
Figure 4. TEM images of sample 1: (a) HADDF image of the overall appearance for the FIB specimen, (b) the SAED pattern of the substrate tungsten and the carbon-boron film, (c) TEM bright-field micrographs of the carbon-boron layer.

Figure 5 .
Figure 5. (a) Image of the interface between tungsten substrate and carbon-boron layer.(b) The fast Fourier transform (FFT) and an inverse fast Fourier transform (IFFT) on the interface between tungsten substrate and carbon-boron film.(c) The corresponding SAED pattern of the tungsten substrate.

Figure 6 .
Figure 6.B and W intensity of samples as a function of sputtering time measured by GDOES depth profiling.

Figure 10 .
Figure 10.TEM image of sample 3 after D 2 -GD exposure and Boundary plasma exposure: (a) TEM bright-field micrographs of the cross-sectional morphology (b) TEM bright-field micrographs of the carbon-boron layer, (c) image of the interface between tungsten substrate and carbon-boron layer.(d) The fast Fourier transform (FFT) and an inverse fast Fourier transform (IFFT) on the interface between tungsten substrate and carbon-boron film.

( 2 )
Following the boronization process, which utilized C 2 B 10 H 12 as the boron source, a substantial amount of H was adsorbed onto the carbon-boron films.The D 2 -GD cleaning can effectively reduce the total retention of H + D in the film layer.The D 2 -GD cleaning effectively facilitates the replacement of H with D, consequently reducing H adsorption and enhancing D adsorption within the films.The efficiency of this hydrogen isotope replacement may be impacted by the thickness of the carbon-boron film.Investigating the exchange mechanism between D in the plasma and H trapped in the film during D 2 -GD cleaning, and elucidating the correlation between film thickness and

Figure 12 .
Figure 12.The evolution of the total desorption amount in bare tungsten and carbon-boron film samples: (a) H, (b) D, (c) H + D.

Table 1 .
Summary of the samples employed in this work.

The third boronization session comprised 30 kW ICRF and 6 g of C 2 B 10 H 12 , with He as the auxiliary gas for sample 7-9, as detailed in table 1. Wall baking has been kept in the off state for the whole boronization processes, ensur- ing that the first wall was cooled to room temperature before proceeding with the boronization. The boronization process of
EAST begins with a 3 h ICRF cleaning, followed by introducing a crucible of solid C 2 B 10 H 12 into the device.The process ensures pressure stability at 0.05 Pa through feedback control while heating the crucible to 80 • C, enabling C 2 B 10 H 12 volatilization and initiating the coating.Pressure adjustments are made as needed to maintain this stability.The process concludes once the C 2 B 10 H 12 has fully evaporated, marking the completion of the boronization process.

Table 2 .
Total release amounts of hydrogen isotopes from ITER grade rolled tungsten substrate and hydrogen isotopes from tungsten boronized samples.