*3.2. He Bubble Evolution as a Function of He*<sup>+</sup> *Fluence*

He bubble damage evolution in the grain matrices, as quantified by changes in the bubble number density, area, and the total change in volume (found using <sup>Δ</sup>*<sup>v</sup> <sup>v</sup>* <sup>=</sup> <sup>4</sup> <sup>3</sup><sup>π</sup> *<sup>r</sup>*<sup>3</sup> *cNv* where *Nv* is the bubble density in a 100 nm thick foils and *rc* is the average radius of the bubbles) due to bubble formation in the grain matrices for both tungsten grades, are shown in Figure 4 (data adjusted for the NCW for variations in focus, in different video segments, occurring during the dynamic in-situ experiments, as described in Figure S1 and the Supplemental Materials). The He bubble density in the NCW grade increased with He<sup>+</sup> fluence and reached a plateau (at a density of ~ 0.125 nm−2), while the average bubble size showed a decrease as a function of He<sup>+</sup> fluence and reached an average size of 3 nm2. The corresponding change in volume showed an increase with He<sup>+</sup> fluence at the start of implantation, peaked at ~0.7, and then decreased to ~0.4 at the last He<sup>+</sup> fluence (3.5 <sup>×</sup> <sup>10</sup><sup>16</sup> ion.cm<sup>−</sup>2). In the case of W-TiC (1.1%), the number density increased first (when the bubbles became visible) to 0.025 nm−<sup>2</sup> and then saturated throughout the implantation, while the average bubble area increased as a function of time (up to 8 nm2) in a similar trend to the corresponding change in volume, which reached a value of 0.6.

**Figure 2.** (**a**–**h**): Bright-field TEM micrographs of a small implanted region taken under Fresnel conditions (under-focused) showing He bubble formation and evolution as a function of He<sup>+</sup> fluence in the grain matrices and grain boundaries in equiaxial nanocrystalline tungsten (NCW) with an average grainsize of 85 nm implanted in-situ with 2 keV He<sup>+</sup> at 1223 K. Scale bar of (**b**–**h**) is the same and is shown in (**b**). Red box in (**a**) approximately represents a magnified region presented in (**b**) to (**h**).

**Figure 3.** (**a**–**h**): Bright-field TEM micrographs of a small implanted region taken under Fresnel conditions (under-focused) showing He bubble formation and evolution as a function of He<sup>+</sup> fluence in the grain matrices and grain boundaries in W-TiC (1.1%) implanted in-situ with 2 keV He<sup>+</sup> at 1223 K. Scale bar of (**b**–**h**) is the same and is shown in (**b**). Red box in (**a**) approximately represents a magnified region presented in (**b**) to (**h**).

**Figure 4.** (Color online) Helium bubble density, average area, and the total change in volume in the grain matrices of (**a**) W-TiC and (**b**) NCW as a function of He<sup>+</sup> implantation fluence. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

#### *3.3. Damage Evolution Comparison*

While the final total change in volume in the NCW is about half the change in volume in the W-TiC, indicating higher radiation tolerance of NCW to He<sup>+</sup> implantation, the damage evolution (and thus, the bubble density profile) is quite different. In the W-TiC, the change in volume followed the average bubble area profile (whereas the bubble number density was saturated). Therefore, bubble damage in this case is due to the growth of existing He bubbles through the absorption of vacancy, He, and/or He–vacancy complexes. The small bubble density in W-TiC compared to the NCW bubble density, in this case, indicates greater defect accumulation (vacancies and possible He–vacancy complexes with low migration energies) of the former. In addition to the role of high densities of grain boundaries in limiting irradiation or implantation damage, TiC dispersoids have been shown to assist in the defect recombination and absorption of excess vacancies acting as vacancy-biased sinks and to contribute to a higher radiation tolerance of the grain matrices to irradiation damage [12,14]. The same W-TiC material has previously been demonstrated, under heavy ion irradiation, to have higher irradiation resistance to cavity and loop formation (in the grain matrices) compared to commercial coarse-grained tungsten [14]. In previous work [15], nanocrystalline and ultrafine tungsten (where both elongated nanocrystalline and ultrafine grains coexist—SPD material) was implanted under the same conditions used in the current work. Bubble density in the grain matrices of the W-TiC was five times higher than in the SPD material but the average bubble area was ~2.5 times lower in the former. However, the total change in volume was the same. Relatively lower defect mobility, due to the impurities in the W-TiC, could potentially lead to more bubble nucleation sites compared to pure tungsten.

The NCW demonstrated a high density of very small bubbles in the grain matrices which was unexpected. It also showed a decrease in the average bubble size as a function of He<sup>+</sup> fluence. The decrease in average size could be due to the nucleation of new small bubbles, which can decrease the average bubble size. However, from the histograms of bubble sizes at four different He<sup>+</sup> fluences shown in Figure 5, the average bubble size decrease was shown to be due to a decrease in the sizes of the individual bubbles. Moreover, the average bubble size continued to decrease even after the bubble density plateaued. Such a decrease in bubble size has to be associated with the balance of interstitial and vacancy fluxes against the bubbles in addition to any possible thermal emissions of vacancies at 1223 K (the latter can be difficult due to the high binding energies of vacancies and He) [24] and internal pressure changes due to the He concentration change in the bubble. Both a higher level of interstitial absorption and thermal vacancy emission (the latter of which being suppressed as mentioned above) can lead to the shrinkage of the bubbles. Moreover, nanoporosity from magnetron deposited films can be responsible for higher densities of bubble nucleation sites but this porosity can reduce with irradiation [29] due to adatom accumulation occurring before reaching a stable bubble

size. Nanoporous materials were shown to enhance irradiation tolerance in different materials [30–32]. Nanoporosity and possible impurities can also limit the mean free path of defect mobility and thus explain the high density of He bubbles in the NCW. Dislocation loops formed in the NCW were shown to be highly mobile [13]. The interaction of these loops with He bubbles may cause them to shrink. There can also be an effect of residual stress on defect absorption, which is yet to be understood in nanocrystalline materials. Under heavy ion irradiation with no gas introduction, the NCW grades demonstrated higher irradiation resistance to void formation (but with higher void density) compared to the SPD material and commercial coarse-grained tungsten [13]. It was shown in the same study that interstitial and vacancy defects have a higher recombination probability at smaller grain sizes, meaning that NCW would have less cavity damage. In the current work, the bubble sizes at the grain boundaries in the NCW (~8–10 nm as shown in inset in Figure 2h) were smaller than those in the W-TiC grade (~20–25 nm as shown in the inset in Figure 3h). They were also smaller than the bubbles (~10 nm) at the grain boundaries in the SPD material at the same He<sup>+</sup> fluence [15]. This represents lower overall bubble damage, which is indicative of a lower He–vacancy complex (with low migration energies) and vacancy transport to the grain boundaries (presumably due to higher levels of recombination). Moreover, the high density of grain boundaries can lead to lower defect transport (to grain boundaries) per boundary. It should be mentioned that the very large bubble sizes at the grain boundaries in the W-TiC material could be due to the TiC impurities at the grain boundaries, which can change the grain boundary sink efficiency and have been shown to be vacancy-biased, as previously discussed. The SPD material in reference [33] showed one order of magnitude higher He<sup>+</sup> fluence threshold for fuzz formation compared to commercial tungsten. The NCW in the current work (with its lower total bubble damage in the grain matrices and grain boundaries) is similarly expected to demonstrate an improved fuzz resistance. Such a material needs to be examined under plasma conditions to assess its mechanical property behavior to evaluate its suitability to fusion conditions.

**Figure 5.** (Color online) normalized bar graphs of bubble size distributions in the grain matrices in (**a**) NCW and (**b**) W-TiC as a function of implantation He<sup>+</sup> fluence. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

#### **4. Conclusions**

In summary, He bubble damage evolution was studied using in-situ TEM as a function of He<sup>+</sup> implantation fluence for two strong fusion material candidates (NCW and ultrafine W-TiC alloy). The results have been compared to previously studied SPD nanocrystalline and ultrafine tungsten [15]. Grain boundaries were shown to be decorated with large bubbles, indicating efficient He trapping. He bubble evolution, profiles, and ov]erall bubble damage (total change in volume in grain matrices and grain boundaries) were found to be different in both grades. The NCW showed a higher density of

small bubbles but a lower overall change in volume than the W-TiC grade and the previously published SPD tungsten. The high density of small bubbles in the NCW was unexpected but several phenomena can govern this observation. Since SPD tungsten possessed a He<sup>+</sup> fluence threshold for fuzz formation that was one of order of magnitude higher than coarse-grained tungsten, the NCW is then expected to have an even higher He<sup>+</sup> fluence threshold (for fuzz formation). This is true if the fuzz formation mechanism depends on high bubble densities near the surface, as illustrated in previous literature works. These results indicate that while these materials may have improved fuzz resistance under plasma exposures and an enhanced tolerance to He damage due to neutron transmutation reactions, further studies of these materials under fusion conditions need to be performed to develop materials that can withstand these extreme conditions.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1996-1944/13/3/794/s1, Figure S1: (Color online) Helium bubble density, average size, and total change in volume in the grain matrices of NCW and W-TiC as functions of He<sup>+</sup> fluence. (a) shows the original and the adjustment data.

**Author Contributions:** Conceptualization, O.E.A. and S.M.; Methodology, J.H., G.G. and O.E.A.; Formal analysis, K.U., W.S.C. and S.F.; Writing—original draft preparation, O.E.A.; Writing—review and editing, O.E.A., K.U., W.S.C., S.F, J.H., G.G. and S.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** Research presented in this article was supported by the Laboratory Directed Research and Development program of Los Alamos National Laboratory under project number 20160674PRD3. We gratefully acknowledge the support of the U.S. Department of Energy through the LANL/LDRD Program and the G. T. Seaborg Institute for this work. The authors would also like to thank EPSRC for the funding of the MIAMI facility through grant EP/M028283/1.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
