Deuterium retention in heavy-ion and helium-ion sequentially irradiated tungsten

Neutron irradiation as well as the presence of helium (He) significantly affects fuel inventory in plasma-facing materials. To investigate the synergistic effects of neutron and He irradiation on deuterium (D) retention behaviors in tungsten, heavy- and He-ion sequentially irradiation experiments were performed with various He fluence and/or heavy-ion damage levels, and then the samples were exposed to low-energy D plasmas at 450 K. It is shown that even a low He concentration of 0.5 atomic parts per million (appm) increases D concentration in the heavy-ion damaged region, which increases further with increasing He concentration under the parameters selected in this work (up to a maximum He concentration of 2.1 appm). The total D inventory in tungsten bulk also increases with He fluence due to the increase in D concentration both in the heavy-ion damaged region and the region irradiated by He-ion only. Furthermore, heavy-ion and He-ion successive irradiation slightly increases D retention in tungsten compared to the individual He ions irradiation. Similar to single heavy-ion damaged tungsten, the saturation of D retention is observed as heavy-ion irradiation damage above 0.2 dpa at a fixed He fluence.


Introduction
Plasma-facing materials (PFMs), which will withstand high dose of neutron irradiation, high particle flux and heat [1,2], are one of the biggest challenges standing in the way of commercializing fusion energy.Hence, plasma-wall interaction is one of the critical scientific topics in fusion energy research.
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.
As the most potential PFM in future fusion reactors, tungsten (W) has been chosen as the divertor material of the next fusion device ITER [3], and is considered the most promising PFM in DEMO [4] and CFETR [5] due to its high sputtering yield, high melting point and low fuel retention [3].Fuel inventory in tungsten, especially tritium (T) inventory, has been recognized as a safety and operational issue for future fusion reactors [6], and needs to be minimized.
In fusion devices, one of the main challenges for tungsten is to withstand irradiation by a large dose of 14 MeV neutrons produced by the D-T fusion reaction.For ITER, the neutron irradiation damage was estimated to be 0.6 dpa (displacements per atom, characterizing the damage level) and 1 dpa for the divertor and the first wall, respectively [6].The neutron damage in DEMO will reach values of 4-8 dpa in PFM at the end of 1.5-2 full-power years [2].The displacement damage in the tungsten lattice will enhance fuel retention due to the introduced additional trapping sites for hydrogen isotopes (HIs) [7].Numerous studies have shown that the total HIs retention in displacement-damaged tungsten is one to two orders of magnitude higher than in undamaged tungsten, and that the HIs concentration in the damaged region reaches 1.5-2.0at.% in the limit of high damage level [8][9][10][11][12][13].These studies also show that HIs saturation at high irradiation doses, which is caused by the generation and annihilation of radiation defects restrict each other under continual irradiation [13].Moreover, He is widely present in fusion devices and significantly affects HIs retention in tungsten.On the one hand, most of the He, which is a byproduct of D-T fusion, is loaded on the superficial layers of the PFMs with low energy [14] and significantly reduces D inventory [15].On the other hand, there is a non-negligible amount of He present in the tungsten bulk as summarized in our previous work [16], which is mainly contributed by He escaping from the magnetic confinement, and He produced by the negative beta decay of T and the transmutation of elements of PFMs via (n, α) nuclear reaction.These processes are accompanied or potentially accompanied by displacement damage.Hence, it is feasible to study the effect of He existence in tungsten bulk on D retention in tungsten using high-energy He ions irradiation.Several studies observed the increase in D retention in high-energy He ions irradiated tungsten, and found that this is caused by additional D trapping sites produced by displacement damage and He bubbles [16][17][18][19].Moreover, a few studies focused on the synergistic effects of displacement damage and bulk He existence on D retention in tungsten.It has been shown that He ion irradiation increased D inventory in heavy ions and He ions sequentially irradiated tungsten [20].In contrast, Markina et al showed that He ions irradiation, as well as He dose, had a negligible effect on D retention in tungsten [21].Other studies utilized simultaneous irradiation with dual beams of heavy ions and He ions and observed the lower D inventory in tungsten than in single heavy ion irradiated tungsten [22].
This work focuses on the synergistic effects of displacement damages and He accumulation on D retention behavior in the tungsten.Fe and Au ions were used to mimic the displacement damage that neutrons will cause in the tungsten.2.4 MeV He ions irradiation was performed to obtain bulk He distribution in tungsten.The main He accumulation region is away from the heavy-ion implanted region to obtain a lower He concentration and a more homogeneous He distribution in the heavy-ion damaged region.Two sets of experiments, Fe-He ions sequential irradiation and Au-He ions sequential irradiation, were performed.Among them, the Fe-He irradiation experiment investigates the dependence of D retention on He fluence, while the Au-He irradiation experiment is concerned with the dependence of D retention on damage level.D depth profiles were measured ex-situ by glow discharge optical emission spectroscopy (GD-OES), and D desorption behaviors were analyzed by thermal desorption spectroscopy (TDS).

Sample preparation
The materials used in this study are commercial high-purity (>99.9%)W obtained from ATTL Advanced Materials Co, Ltd.Samples were cut from the same hot-rolled polycrystalline W plate with dimensions of 15 mm × 12 mm × 1 mm using wire-cut.The grain sizes of the studied W were in the range of 2-10 µm, and the grains were elongated in the rolling direction and parallel to the sample surface.These samples were mechanically ground by SiC paper up to 2000 grit size, then polished to a mirror-like surface using diamond suspension.After that, polished samples were annealed at 1173 K for 1 h in a vacuum chamber (∼1 × 10 −5 Pa) and then naturally cooled to room temperature to eliminate the residual stress introduced by the polishing process and remove the impurity gases adsorbed on the surface.Sample surface cleaning using acetone and alcohol was performed before the irradiation experiment.

Ion irradiation experiments
W samples were pre-irradiated with heavy ions and then irradiated with high-energy He ions.Fe ion beam accelerated up to 2 MeV and Au ion beam accelerated up to 5 MeV were used for heavy ions irradiation.The irradiation of the Fe ions was performed on a 320 kV platform at the Institute of Modern Physics, Chinese Academy of Sciences.Au ions irradiation was carried out at a 2 × 1.7 MV tandem accelerator in the ion beam materials laboratory at Peking University.According to the results of [10], the selection of the two types of heavy ions in this experiment does not affect the D retention behavior.After heavy ion irradiation, the samples were irradiated by He ions with an energy of 2.4 MeV at a low energy intense-highly-charged ion accelerator facility at the Institute of Modern Physics, Chinese Academy of Sciences.All ion irradiation experiments were performed in the vacuum chamber with a background pressure below 5 × 10 -5 Pa.Besides, the sample temperature was maintained at room temperature with a minor fluctuation throughout the ion irradiation process as measured by a thermocouple attached to the sample holder.The implantation depth, ions distribution, and displacement damage were calculated by the Stopping and Range of Ions in Matter (SRIM-2008) code [23].The 'Ion Distribution and Quick Calculation of Damage' mode was selected for the simulation.The values of the 'Displacement Energy of atom' and the 'Lattice Binding Energy of atom in layer' were set to 90 eV [24] and 0 eV [25] respectively.For specific simulation and data processing refer to the study of Stoller et al [25].The detailed irradiation parameters of heavy ions irradiation and subsequent He ions irradiation are shown in table 1.The present study consists of two parts: (i) samples were irradiated by 2 MeV Fe 11+ with displacement damage level of 0.5 dpa and then subjected to 2.4 MeV He ions irradiation with different fluence (5.5 × 10 19 , 1.1 × 10 20 , and 2.2 × 10 20 He m −2 ).(ii) Samples were irradiated by 5 MeV Au 3+ with different  damage levels (0, 0.2, 0.4, 0.8, and 1.Moreover, the damage peak is located at around 3.4 µm.He is present at a maximum depth of around 3.9 µm, and the maximum He concentration is observed at around 3.5 µm.Details of the calculated maximum He concentrations and peak damage levels are given in table 1.

Deuterium plasma exposure
After successive irradiation with heavy ions and He ions, samples were exposed to D plasmas.D plasma exposure experiments were performed on an experimental linear plasma device at the Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences.The ion fluxes measured by a single Langmuir probe (ESPION) in a radial position near the sample holder were around 1.0 × 10 21 D m −2 s −1 .All samples were exposed for 2.8 h corresponding to a total D fluence of 1.0 × 10 25 D m −2 .The incident ion energy was selected to be 38 eV D −1 , determined by the −100 V bias voltage applied to the sample holder and the 15 V plasma floating potential, as described in [26].During D plasma exposure, a sample temperature of 450 K was chosen to ensure that a sufficient amount of D could diffuse into the damaged region, as well as to avoid defect annealing such as vacancy migration [27].The sample temperature was measured by a thermocouple attached to the backside of the sample holder, and controlled by a thermostat device consisting of heating and alcohol cooling with a temperature fluctuation of ±10 K.

Characterization methods
After D plasma exposures, the surface morphology was analyzed by field emission scanning electron microscopy (JEM-6701F, JEOL).The focused ion beam (FIB, equipped on Helios Nanolab 600i, FEI) was used to observe the crosssectional microstructure.The FIB utilizes 10 kV Ga + ions under a beam current of 1 nA at a tilt angle of 55 • .The GD-OES and TDS measurements were performed to investigate the depth profiles and desorption behaviors of D. All D plasma exposed samples were stored in a desiccator for 15 d before GD-OES and TDS measurements to minimize the effect of storage time on D retention [28,29].
In the case of GD-OES analysis, a commercial device (GD PROFILER 2, Horiba Jobin Yvon) was employed.Samples were used as cathode and etched by argon plasma excited in radio frequency (RF) power mode maintained in a 4 mm diameter anode.The input power, frequency, and duty cycle of the RF source were set to 30 W, 3000 Hz, and 0.25 respectively, resulting in a final power of 7.5 W. For uniform etching, the plasma excitation/operating pressure was selected to be 650 Pa.Before measurement, the sample surface was purged with argon gas for 80 s using the 'flush' option to remove impurities such as dust.The data acquisition frequency was set to 10 ms.The total scan duration was 9 min to ensure a total measurement depth greater than 10 µm.The wavelength lines used were 121.534 nm for D and 429.461 nm for tungsten.He (wavelength of 587.562 nm) was also measured in the He-irradiated samples, but unfortunately, no He signal was detected due to He concentrations below the detection limit of the device.It should be noted that the GD-OES device used in this work has been calibrated for quantitative measurements of D [30] and He [31] in tungsten.
For TDS measurement, samples placed in the quartz tube were heated to 1173 K with an oven temperature ramp rate of 10 K min −1 and held at the maximum temperature for 10 min.The released D 2 (4 Da) and HD (3 Da) signals were detected by a quadrupole mass spectrometer (Pfeiffer QME220) and the total released D amounts were calculated from the integration of D 2 and HD signals.The signal of 4 Da was calibrated using a D 2 leak, and the 3 Da was calibrated using the average value of the signals of two calibrated H 2 and D 2 leaks.The D 2 and HD signals of an already out-gassed sample were measured as the background.All TDS results shown in this work are background-subtracted data.The actual temperature of the sample during the heating process was calibrated using a thermocouple in close contact with a tungsten sample of the same size as the present work.TDS measurement was performed for a Fe-He irradiated sample (He fluence of 2.2 × 10 20 He m −2 ) using the above parameters before D plasma exposure, and found that no peaks of mass 4 emerged, implying that He does not affect the signal of D under the parameters of TDS measurements in this work.

Fe-He ions sequential irradiation
The first experiment studies D retention in tungsten samples irradiated sequentially by Fe and He ions with a damage level of 0.5 dpa for Fe ions irradiation and various fluence for He ions irradiation.The D depth profiles were measured ex-situ by GD-OES as shown in figure 2. D is measured within 6 µm in all samples and D depth profiles generally match the damage level.Specifically, for the reference sample which was only damaged by Fe 11+ and labeled 'He-free' in figure 2, the near-surface D concentration is ∼0.9 at.% and reaches a maximum at 0.32 µm, which is consistent with the location of the Fe ions damage peak calculated by SRIM.Besides, D is distributed within a depth of 4 µm and D concentration decreases rapidly outside the damaged region.In the case of Fe-He ions sequentially irradiated samples, the maximum depth of D distribution is 5.3-6 µm, which is greater than that of the reference sample.This is attributed to the larger irradiation depth of 2.4 MeV He ions.Furthermore, the additional He ions irradiation results in the enhancement of D inventory compared to the reference sample, and D concentration increases at all depths with increasing He fluence.The maximum D concentration is still located at the Fe ions damage peak.In addition, another minor D concentration peak located at ∼3.5 µm is observed in the He accumulation region which is more visible in the samples irradiated by He ions to a fluence of 1.1 × 10 20 and 2.2 × 10 20 He m −2 .As can be seen from the SRIM simulation (figure 1(c)), this minor D concentration peak is located at the maximum He concentration, which is consistent with our previous work [16].Figure 3 summarizes the maximum D concentrations in the Fe ions damaged region (∼0-0.9µm) and the so-called He accumulation region (∼2.5-4µm) based on the results of GD-OES measurements.According to the discussion above, these two maximum D concentrations are located at the damage peak (∼0.32 µm) and maximum He concentration (∼3.5 µm), respectively.One Figure 4 shows the D 2 release spectra for Fe-He ions sequentially irradiated samples.In general, D release begins at about 450 K and desorption is complete at 950 K.In the case of the single Fe ions irradiated reference sample, two well-distinguishable release peaks are visible: the first peak is located at about 590 K and the second one at about 730 K. Similar shapes and peak locations of desorption spectra in heavy-ion damaged tungsten were observed in our previous study, and the parameters selected for D plasma exposures and TDS measurements in the previous experiments are comparable to the present study [32].The shape of the desorption spectrum and the location of the peak are closely related to the types of D-trap defects, which can be dislocations, dislocation loops, vacancies, vacancy clusters, and voids (see e.g.[24,33]).For the reference sample, the peak at 590 K is attributed to D de-trap from the single-vacancies (trapping energy of 1.08-1.46eV) [34,35], and another peak at 730 K is ascribed  to D release from the vacancy clusters (trapping energy of 1.68-1.86eV) [35].In the case of Fe-He ions (He fluence of 5.5 × 10 19 He m −2 ) irradiated sample, compared with the reference sample, the peaks show a slight broadening which can be attributed to that D distributed in a larger depth range.Besides, the increase in the intensity of the 590 K peak is observed.There are two possible explanations: on the one hand, He irradiation introduces lattice defects [36].On the other hand, due to the strong attraction of defects to He, Hedefects clusters such as He-V complexes are readily formed, which is stabilized in tungsten bulk at lower temperatures [37,38].The formation of He-V complexes, which have almost the same D binding energy as the single vacancy [20], provides additional D trapping sites and exceeds the number of HIs atoms that a single vacancy can trap [39,40].We will discuss it in detail in the following.The increase in He fluence leads to a further increase in peak intensity, which means that higher doses of He ions irradiation produce more defects in tungsten.Moreover, as He fluence increases, the D release peaks shift to higher temperatures, which is caused by a larger depth of the center of weight of the D distribution due to the increase of D concentration in the He accumulation region (see figure 2) [41].
The D amounts measured by TDS and GD-OES as a function of He fluence are summarized in figure 5. D 2 (4 Da) and HD (3 Da) signals were considered in the total D inventory calculation measured by TDS, where the contribution of HD to the total amount of D retention was about 8%-11% and the contribution of D 2 to the total amount of D retention was about 89%-92%.The D amounts measured by GD-OES were calculated by integrating the D depth profiles.For all samples, the total D amount measured by GD-OES is 15%-20% lower than that measured by TDS.This is because D was not detected at low concentrations due to the detection limit of the GD-OES device, whereas it is only accessible by TDS.

Au-He ions sequential irradiation
Another concern is whether the presence of He in tungsten bulk affects the damage saturation in tungsten, i.e. whether the presence of He increases D retention above the damage saturation threshold of 0.2 dpa.Thus, to investigate the effect of increasing damage level on D retention in the case of a fixed single He concentration, Au-ion irradiation with varying damage levels and subsequent He-ion irradiation with the same fluence were performed.Damage levels of 0, 0.2, 0.4, 0.8, and 1.7 dpa were selected for Au ions irradiation, and the He fluence was 1.1 × 10 20 He m −2 .Figure 6 shows D depth profiles in Au-He ions irradiated tungsten samples.For the individual He ions irradiated sample, D is distributed within 4.5 µm.D concentration reaches 1.5 at.% in the surface layer and decreases with depth until a minor peak appears in the socalled He accumulation region at 3-4 µm.In the case of Au-He irradiated tungsten, D concentrations show a maximum at a depth of around 0.21 µm, corresponding to the damage peak of the Au ions.This is similar to the case of the Fe-He ions irradiated samples as shown in figure 2. The value of this maximum D concentration remains almost constant at around 2.2-2.3 at.% as the damage dose increases.A slight increase in D concentration at depths of 1-1.5 µm is observed with increasing damage levels.In the He accumulation region at a depth of 3-4 µm, the minor peak of D concentration also appears, as in the individual He ions irradiated sample.
Blistering was observed in the individual He ions irradiated tungsten after the given D plasma exposure as shown in figure 7.There are two types of blisters on the sample surface: large blisters with a size of 5-10 µm (figure 7(a)) and small blisters with a size of less than 1 µm (figure 7(b)).Large blisters are randomly distributed on the surface of the sample, covering several grains.While the small blisters are distributed individually or in clusters inside the grains.The cross-sectional morphologies of the two types of blisters prepared by FIB are also shown in figure 7. The actual depth of the cross-section is calculated according to h = h ′ /sin 55 • , where h represents the actual depth, h ′ is the measured depth, and 55 • is the tilt angle of FIB.For the large blister, an inter-granular crack at a depth of 1.95 µm parallel to the sample surface can be observed, which nucleates and propagates along the grain boundary.In the case of the small blister, the depth of the crack underneath the blister is much smaller than that of the large blister (∼0.28 µm), and the crack propagates inside the grain.The mechanism of blister formation has been much reported (see e.g.[42][43][44][45]).Briefly, supersaturation of D at grain boundaries or dislocations causes plastic deformation of the tungsten, resulting in the formation of blisters on the surface.Manhard et al directly observed a significant increase in dislocation density in blister caps due to strong deformation [46].These dislocations act as trap sites for D, leading to a high concentration of D within 1 µm in the individual He ions irradiated tungsten.
TDS spectra of D trapped in Au-He ions sequentially irradiated samples are shown in figure 8.For the individual He ions irradiated sample, the sharp bursts are observed at low temperatures, meaning that D is released from the burst D blisters.Moreover, two desorption peaks which are located at about 580 K and 700 K can be clearly distinguished.In the case of Au-He ions irradiated samples, these two peaks can still be identified, but the intensity and location of the peaks have changed slightly.Specifically, the intensity of the peaks of the 0.2 dpa Au + He ions irradiated sample increases compared with the individual He ions irradiated sample.As the damage level of Au ions irradiation increases, the intensity of the 580 K peak is slightly diminished and the peak at 700 K is shifted to higher temperatures.As discussed above, the shift of the peaks implies the change of the center of weight of the D distribution [41], which corroborates the increase in D concentration with dpa at the depth of 1-1.5 µm as shown in figure 6.
The amount of D retention was calculated by integrating the D depth profile and desorption spectrum, as shown in figure 9.As mentioned above, the D inventory measured by GD-OES is lower than that of TDS due to the weaker detectability of GD-OES for low concentrations of D. For the individual He ions irradiated sample, the amount of D retention is about 1.41 × 10 21 D m −2 , which is higher than that in only Fe ions irradiated sample (∼1.1 × 10 21 D m −2 , see figure 5).Although the damage level for He ions irradiation is smaller than that for individual Fe ions irradiation, the deeper damage range of He ions irradiation and the plasma-induced defects (e.g.blisters) in the He-irradiated sample contribute to this result.The total D retention increases by 16% to 1.64 × 10 21 D m −2 with the increase of damage level of Au ions irradiation to 0.2 dpa.A further increase in the damage level has little effect on D retention, which indicates that D is saturated in the sample.

Discussion
Table 2 lists the SRIM-calculated He concentrations and dpa of He ions at the Fe ions damage peak (0.33 µm) in the Fe-He ions sequentially irradiated samples.The displacement damage caused by He ions irradiation with different fluences at this depth is 0.0013, 0.0026, and 0.0053 dpa, corresponding to He fluences of 5.5 × 10 19 , 1.1 × 10 20 , and 2.2 × 10 20 He m −2 , respectively.In this work, the Fe ions damage level of 0.5 dpa is higher than the known displacement damage defects saturation dose value of 0.2 dpa in tungsten [13].Hence we determine that such low levels of displacement damage caused by He ions irradiation would essentially not result in obtaining more D trapping sites at this depth to increase the D concentration.Moreover, the formation of blisters, which allows strong deformation of the near-surface region leading to a high concentration of dislocations and providing additional D trapping sites [47], was not observed in this part of the work.Therefore, the increase in D concentration at this depth cannot be ascribed to the formation of blisters.Hence, it is most likely that the presence of He dominates the increase in D retention at the Fe ions damage peak.According to SRIM simulations, He concentrations corresponding to the three He fluences (5.5 × 10 19 , 1.1 × 10 20 , and 2.2 × 10 20 He m −2 ) at this depth are 0.5, 1.0, and 2.1 atomic parts per million (appm), respectively (see table 2).The increase in D retention due to such a low He concentration is surprising: the selected maximum He fluence in this work corresponds to a He concentration of only 2.1 appm in this depth, which results in a 61% increase in D concentration, as discussed above in figure 3. Due to the low migration energy of He in tungsten [48] and its strong attraction to defects [49,50], He atoms are easily captured by defects.Markelj et al inferred that the implanted He in the damaged region is able to rapidly form He-V or He-V clusters with defects introduced by irradiation or with its intrinsic defects [20].These He-V complexes strongly attract D and provide more D trapping sites surrounding the He-V complex compared to the single vacancy [40].We speculate that the interaction of He with displacement damage defects, which provides additional D capture sites, is an important factor in the increase in D concentration described above.One could argue that He provides the pathway for D at the deeper region to diffuse into the damaged region, as revealed in previous studies [51][52][53][54].However, this possibility can be easily ruled out under the experimental parameters of the present study.Firstly, the porous structure with open pathways formed by bubbles occurs at the He bubble volume fraction of more than 16% [52,54,55], and it is clear that this phenomenon does not occur in the case of the He fluence chosen in this work.Secondly, it was shown in [56] that the mobility of He is very low at temperatures below 993 K. Hence, at the temperatures chosen in this experiment, it was not possible to obtain the porous structure by the migration of the He-V complex.Thirdly and more importantly, the defects are saturated in the damaged region, and even if D enters the damaged region via the porous structure formed by the He bubbles, its probability of further D trapping is low.This further confirms that the presence of He provides additional D capture sites and has a strong influence on the increase in D concentration at this depth.He concentration in tungsten under fusion environment is expected higher than that in the damaged region of this work.Gilbert et al calculated that He concentration of around 30 appm will be generated in tungsten during five full power years under first wall fusion power-plant conditions via (n, α) nuclear reaction [57].In addition, according to the model of Shimada and Merrill, the concentration of He in tungsten produced by tritium decay may reach 600 appm over a period of 3 years [58].Hence the effect of bulk He on D retention cannot be ignored in fusion devices as well as in future power plants.
In the Au-He ions irradiated tungsten, it is found that the maximum D concentration in the Au-ion damaged region as well as the total amount of D bulk retention are saturated at the Au-ion damage level above 0.2 dpa.This indicates that D saturation persists in the Au-He ions irradiated tungsten at a fixed He fluence, as in the individual heavy-ion damaged tungsten.In addition, it can be seen that the total D retention is around 16% higher in the Au-He ions irradiated samples compared to the individual He ions irradiated samples, which to some extent indicates that the D retention is dominated by He.The mechanism of D saturation in the heavy-ion and Heion sequentially irradiated tungsten is currently not clear.On the one hand, He irradiation creates additional damage like interstitial-type defects (dislocations) [59] and vacancy-type defects [36] in tungsten.Although the evolution of defects in heavy-ion damaged tungsten, e.g. the evolution of dislocations and vacancies as a function of damage dose, has been relatively well studied [13], the additional He ions irradiation complicates the evolution of the defects: for example, collisional cascades as well as He occupation of vacancies change the defect density, type, mobility, and annihilation mechanisms.Previous works have investigated the evolution of defects in tungsten sequentially irradiated by heavy-ion and He-ion.El-Atwani et al studied the evolution of defects in Kr + He sequential experiment via in-situ TEM [60].It was found that the dislocation loops induced by Kr ions irradiation were annihilated at the beginning of He implantation, and subsequently, the nucleation and growth of dislocation loops due to He implantation were observed.It should be noted that this experiment was performed at a temperature of 1223 K, which may be somewhat different from the room temperature experiment.Hou et al found that the addition of He inhibited the migration of vacancies and hence kinetically suppressed the formation of large vacancy clusters [61].Qin et al observed that pre-irradiation with Kr ions inhibits the growth of large He bubbles and increases the density of He clusters [62].On the other hand, the He-defect interaction provides new types of D trapping sites while consuming defects such as vacancies, further complicating the D retention mechanism.We can confirm that D saturation persists in tungsten irradiated by additional He ions, although the determination of the specific threshold needs to be experimentally investigated by designing a larger range of parameters.This is considered beneficial for the control of D retention in the fusion device.

Conclusion
D retention behaviors in the heavy-ion and He-ion sequentially irradiated tungsten have been investigated.Here, 2 MeV Fe ions irradiation with a damage peak of 0. In the case of Fe-He ions sequentially irradiated tungsten, the results show that even a low He concentration of 0.5 appm increases D concentration in the heavy ion damaged region, which increases further with He fluence under the parameters selected in this work (up to a maximum He concentration of 2.1 appm).Furthermore, an increase in D concentration outside the heavy ion damaged region was observed in He ions irradiated samples with higher fluence.The total D inventory in tungsten bulk follows the same trend, which is contributed by the increase in D concentration both in the heavy ion damaged region and outside this region caused by He irradiation only.For the future fusion reactor, the increase in D retention due to such a low concentration of He is not negligible.For Au-He ions irradiated tungsten, Au-ion and Heion successive irradiation slightly increases D retention compared to individual He ions irradiation.Moreover, similar to single heavy-ion damaged tungsten, D saturation is observed at damage level of Au ions above 0.2 dpa.This contributes to the control of HIs retention in the future fusion reactor.

Figure 1 .
Figure 1.SRIM simulated results of damage distributions for (a) Fe-He ions sequential irradiation, (b) Au-He ions sequential irradiation, and (c) He ions irradiation (bottom).He distribution profiles (top) are also plotted in (c).
7 dpa) followed by the same fluence of 2.4 MeV He ions (1.1 × 10 20 He m −2).The values of the above damage levels are the peak damage simulated by SRIM code.In the following, we refer to these two parts of the experiment as Fe-He ions sequential irradiation and Au-He ions sequential irradiation, respectively.Figure1shows the SRIM calculated results of Fe-He ions sequential irradiation, Au-He ions sequential irradiation, and He ions irradiation.The damage profile of tungsten by 2 MeV Fe ions has a damage peak at around 0.32 µm, and the maximum damage depth is located at around 1 µm (figure1(a)).In the case of 5 MeV Au ions irradiation, the damage peak and the maximum damage depth are located at around 0.21 and 0.7 µm (figure 1(b)), respectively.The 2.4 MeV He ions irradiation introduces a displacement damage range from the surface to 3.8 µm, and the damage profiles show a gentle increase followed by a rapid increase at around 2.5 µm (see figure1(c)).

Figure 2 .
Figure 2. D depth profiles in the Fe-He ions sequentially irradiated samples measured by GD-OES.'He-free' means that the sample has only been damaged by Fe ions.The depth profile of the damage level calculated by SRIM is shown as a dashed line.

Figure 3 .
Figure 3. Maximum D concentrations in SRIM-calculated Fe ions damaged region (at ∼0.33 µm) and He accumulation region (at ∼3.5 µm) as a function of He fluence from GD-OES measurements.

Figure 4 .
Figure 4. TDS spectra of D trapped in Fe-He ions sequentially irradiated samples.

Figure 5 .
Figure 5.The integrated total amount of D derived from TDS (figure 4) and GD-OES (figure 3) measurements.
Our discussion of D retention is hence based on the TDS results.One can see that the amount of D retention increases significantly with He fluence.Specifically, for the reference sample, the amount of D retention determined by TDS is about 1.1 × 10 21 D m −2 , while it increases by about 36% to 1.5 × 10 21 D m −2 in the 5.5 × 10 19 He m −2 sample.Further, the amount of D retention reaches 2.0 × 10 21 and 2.4 × 10 21 D m −2 when He fluence increases to 1.1 × 10 20 and 2.2 × 10 20 He m −2 , respectively, which are ∼83% and ∼110% higher than that in the reference sample.No significant saturation of D retention is found at these selected He fluences.As discussed above, He ions irradiation introduces the defects that provides additional D trapping sites in both the damaged region as well as the He accumulation region.

Figure 6 .
Figure 6.D depth profiles in Au-He ions irradiated tungsten with various damage levels of Au ions (0, 0.2, 0.4, 08, 1.7 dpa) and fixed He fluence (1.1 × 10 20 He m −2 ) measured by GD-OES.The depth profile of the damage level calculated by SRIM is shown as a dashed line.

Figure 7 .
Figure 7. SEM images of surface (top) and cross-sections (bottom) morphologies of only He ions irradiated tungsten.(a) Large blister, (b) small blisters.

Figure 8 .
Figure 8. Thermal desorption behaviors of D trapped in Au-He ions sequentially irradiated samples.

Figure 9 .
Figure 9. Plot of D retention in tungsten as a function of damage levels of Au ions irradiation.
5 dpa and 2.4 MeV He ions irradiation with various fluence (5.5 × 10 19 , 1.1 × 10 20 , and 2.2 × 10 20 He m −2 ) were performed successively to study the effect of He fluence on D retention.5 MeV Au and 2.4 MeV He ions irradiation were carried out with various damage levels of Au ions irradiation (0, 0.2, 0.4, 0.8, 1.7 dpa) and fixed He fluence (1.1 × 10 20 He m −2 ) to study the effect of damage level on D retention.The main He accumulation region is away from the heavy-ion implanted region to obtain a lower concentration and a more homogeneous He distribution in the heavyion damaged region.After ion irradiation, all samples were exposed to D plasma with an incident energy of 38 eV D −1 and fluence of 1.0 × 10 25 D m −2 at a temperature of 450 K. D retention behaviors were measured ex-situ by GD-OES and TDS.

Table 1 .
Detailed irradiation parameters of heavy-ion irradiation and subsequent He-ion irradiation.Dpa is the peak value of the damage level simulated by SRIM code.Cmax represents the maximum He concentration calculated by SRIM code.

Table 2 .
SRIM-calculated He concentrations and He ions damage levels at the damage peak of Fe ions (0.33 µm).Concentrations displayed in both appm and at.%.