Ion irradiation study of lithium silicates for fusion blanket applications

https://doi.org/10.1016/j.jnucmat.2023.154281Get rights and content

Abstract

This study reports on the ion-irradiation induced microstructural and compositional evolutions, behavior of deuterium release, and lithium loss from lithium silicates. Pellets of biphasic Li4SiO4 and Li2SiO3 were fabricated using hot pressing of milled powders. Sequential irradiation with Si+, He+ and D+ ions was performed up to 773 K to emulate one-year 6Li burnup in 6Li4SiO4 inside the SlimCS DEMO. Crystalline Li4SiO4 phase was not observed in the irradiated depth region in this study. In spite of full amorphization in the near-surface region, Li2SiO3 irradiated to a high dose at 773 K remained crystalline in the damage peak region, suggesting that Li2SiO3 is more irradiation resistant to amorphization than Li4SiO4. Radiolysis may be primarily responsible for the silicate decomposition and amorphization. Data from this study also show that D release from the pellet is very efficient during ion irradiation at 773 K, which is accompanied with a significant Li loss. Thermal annealing for 10 min at 773 K for the room-temperature irradiated pellet leads to a nearly complete D release without an observable Li loss. A concept of Ni coating on tritium breeder materials is discussed with supporting data to minimize Li loss without a significant impact on tritium diffusion and release.

Introduction

Tritium (T) is subject to β decay 3T3He+e+νe¯+18.6keV with a half-life of 12.3 years. It must be produced on site as a fuel to operate thermonuclear fusion reactors for power generation based on 2D+3T3.5MeV4He+14.1MeVn. Tritium production is accomplished by transmutation of isotope 6Li that captures a neutron, i.e., n+6Li2.75MeV3H+2.05MeV4He. A concept of tritium breeding blanket containing a 6Li-enriched compound (tritium breeder material) behind the first wall of a fusion reactor has been proposed [1], [2], [3]. In order to achieve optimum performance, the tritium breeder material should have a good structural stability, rapid tritium release and low lithium volatility under the fusion extreme conditions, including high neutron flux and high temperature.

A density functional theory (DFT) study [4] indicates that T diffusion via the first nearest-neighbor Li vacancies in γ-LiAlO2 is the most energetically preferred mode. A separate DFT calculation of the breeder material [5] suggests that tritium trapping in Li vacancies with O-T bonding is the most energetically preferred configuration. Nuclear magnetic resonance (NMR) measurements [6] appear to suggest that tritium occupies Li vacancies with O-T bonding in γ-LiAlO2. An atom probe tomography (APT) study [7] confirms the heterogeneous distributions of both T and OT in low-Li regions of neutron irradiated γ-LiAlO2. These results generally suggest that maximizing the Li/O ratio in Li ceramics can enhance T diffusion, while minimizing the O/T ratio can reduce the likelihood of T trapping in hydroxyls or other compounds. Both maximizing the Li/O ratio and minimizing the O/T ratio can optimize tritium production that originates from 6Li. Thus, ceramics with a high Li atomic density are generally preferred for fusion blanket applications. While ceramic materials with extremely high Li densities are being explored [8], lithium orthosilicate (Li4SiO4) [9], [10], [11], [12], [13], [14], lithium titanate (Li2TiO3) [15], [16], [17], and lithium zirconate (Li2ZrO3) [18], [19], [20] have been the primary ceramic materials that have attracted most extensive research as candidates for solid-state tritium breeders. Li4SiO4 has the highest Li/O ratio of the three and is chosen for this study. It should be pointed out that Li-richer compounds may have lower melting points and higher Li volatilities. In fact, Li4SiO4 powders are commonly mixed with lithium metasilicate phase (Li2SiO3), which is consistent with the phase diagram of the Li2O-SiO2 system [21].

Previous investigations [22] show that Li4SiO4 phase decomposes and transforms to Li2SiO3 during long-time heating at 1073 K in air. Tritium release from neutron irradiated Li4SiO4 pebbles contains mainly HT (72%) and T2 (19%) in addition to a small fraction of HTO [23]. Similar release behavior from bi-phases of Li4SiO4 and Li2TiO3 was recently reported with little tritium retention [24]. The modified Li4SiO4 pebbles with addition of Li2TiO3 were first reported by Knitter et al., which show improved mechanical, thermal and chemical properties compared to pure Li4SiO4 [25]. The kinetics of tritium release from the bi-phases under neutron irradiation has been investigated as a function of the phase ratio [26]. Amorphization and He bubble evolution have also been reported in Li4SiO4 during sequential irradiation with He+ and H+ ions at nominally room temperature (RT) and post-irradiation thermal annealing at 723 K [27]. This study reports on the microstructural evolution, compositional change and deuterium behavior in the bi-phases of Li4SiO4 and Li2SiO3 during ion irradiation at various temperatures.

Section snippets

Pellet fabrication

Lithium silicate pellets are not commercially available. We have explored pellet fabrication by hot-pressing of powders. Powders of bi-phases of lithium orthosilicate Li4SiO4 and lithium metasilicate Li2SiO3 with a nominal purity of 99.9% (metals basis) and particle sizes of −100 meshes (<150 μm) were obtained from American Elements (Los Angeles, CA). Our analysis shows that the as-received powder contains 64.2 wt.% Li4SiO4 and 35.8 wt.% Li2SiO3. High-energy ball milling was performed for the

Crystalline phase

GIXRD at a fixed incident angle of 1° was performed for irradiated (and unirradiated) lithium silicate pellets to enhance the surface detection sensitivity. Estimation shows that an intensity of 10% reflection for incoming angle of 1° and outgoing angle of 30° corresponds to a probing depth (perpendicular to the surface) of ∼4 μm in Li4SiO4 and ∼6 μm in Li2SiO3 as compared to the depth of 1.5 μm for the damage produced by Si+ ion irradiation (see Fig. 1). Fig. 2 shows the GIXRD data with 3

Conclusion

Hot pressing of powders was applied to fabricate pellets of biphasic Li4SiO4 and Li2SiO3. The pellet samples were irradiated sequentially with Si+, He+ and D+ ions at 300, 573 and 773 K to 30.6 dpa and 2.2 at.% He and D at the depth of ∼500 nm, respectively, corresponding to one-year 6Li burnup in 6Li4SiO4 inside the SlimCS DEMO. One additional irradiation was also performed at 773 K to a double dose and gas species concentrations. Li2CO3 phase near the surface formed due to graphite tooling

CRediT authorship contribution statement

Weilin Jiang: Conceptualization, Methodology, Formal analysis, Data curation, Writing – original draft, Supervision. Libor Kovarik: Investigation, Formal analysis. Mark G. Wirth: Investigation. Zihua Zhu: Investigation. Nathan L. Canfield: Investigation. Lorraine M. Seymour: Investigation. Larry M. Bagaasen: Methodology, Supervision. Mark E. Bowden: Investigation, Formal analysis. Tamas Varga: Investigation, Formal analysis. Nicole R. Overman: Investigation. Zhihan Hu: Investigation. Lin Shao:

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This research was supported by both the Office of Fusion Energy Sciences (for the Lithium silicate study) and the National Nuclear Security Administration (for the Ni study), U.S. Department of Energy (DOE), and performed at the Pacific Northwest National Laboratory under Contract DE-AC05-76RL01830. The authors are grateful to Prof. Steven Zinkle at the University of Tennessee Knoxville for his helpful suggestions on the design of the ion irradiation experiment. Ion irradiation was performed at

References (38)

Cited by (0)

View full text