Since 18 of December 2019 conferences.iaea.org uses Nucleus credentials. Visit our help pages for information on how to Register and Sign-in using Nucleus.

# 28th IAEA Fusion Energy Conference (FEC 2020)

10-15 May 2021
Virtual Event
Europe/Vienna timezone
The Conference will be held virtually from 10-15 May 2021

## Ammonia production, isotopic exchange and sticking on materials relevant to fusion reactors: tungsten and 316L stainless steel

14 May 2021, 14:00
4h 45m
Virtual Event

#### Virtual Event

Regular Poster Fusion Energy Technology

### Speaker

Dr Régis Bisson (Aix-Marseille Université, CNRS, PIIM UMR 7345)

### Description

For high-power operations in ITER, it is foreseen to inject extrinsic impurities into the edge plasma to dissipate part of the plasma exhaust power through radiation and maintain the power fluxes to the plasma-facing components within tolerable limits. To date, the best compromise between radiative efficiency$^{1}$ and hot plasma performance$^{2,3}$ has been achieved with the injection of molecular nitrogen (N) in the edge plasma region. However, injection of nitrogen in a divertor plasma environment next to metallic components leads to the conversion of ~10% of the injected nitrogen into NH$_3$, as it was observed in ASDEX-Upgrade and JET tokamaks during N-seeded discharge$^{4,5}$. In ITER, ammonia production will be concurrent with a deuterium/tritium plasma thus radioactive tritiated ammonia is expected. The formation of large quantities of tritiated ammonia has consequences for several aspects of the ITER plant operation in terms of tritium retention, gas reprocessing and duty cycle. It is well known that ammonia is a polar molecule and could stick on (shadowed) metallic in-vessel components as well as on pumping ducts. This represents an issue for the tritium recycling plant operated to retrieve tritium from cryopumps and exhaust gases. It is currently unclear how and where ammonia formation predominantly occurs in fusion devices, which makes it difficult to predict the ammonia formation rate in ITER as well as to determine the areas in the vacuum vessel where ammonia might accumulate.

In this contribution, we present new results that will help to model the tritiated ammonia inventory in ITER. In particular, we address the following questions:
- What is the dominant ammonia surface formation mechanism and how does the formation rate depend on the surface material?
- What is the surface assisted isotope exchange efficiency of ammonia?
- What is the sticking probability of NH$_3$ molecules on ITER-relevant material (tungsten and 316L stainless steel)?
Our studies are performed in an ultra-high-vacuum environment$^{6}$ using ion beam and molecular beam exposures of polycrystalline tungsten (W) and 316L stainless steel (316L-SS) samples probed by various mass spectrometry and spectroscopic methods.

To understand the mechanism of ND$_3$ formation from elements implanted in the material bulk, sequential implantation of N$_2$$^+ and D_2$$^+$ ions at 300 K is performed and Temperature Programmed Desorption (TPD) is used to quantify HD, D$_2$, N$_2$ and ND$_3$ production rates. On the one hand, we show that deuterated ammonia (ND$_3$) is produced on both metals when bulk deuterium (D) diffusion is activated. On the other hand, the absolute quantity of ND$_3$ produced is found to be strongly dependent on the sample material. This difference in ND$_3$ production rate is related to dissimilarities in the formation process. On polycrystalline tungsten, the formation mechanism involves the implanted N atoms that are present at the surface. Nitrogen implanted deeper in tungsten is unable to participate in the production of ND$_3$. In stark contrast, on 316L-SS, N atoms naturally contained in the bulk of the alloy participate significantly in the formation of ND$_3$ through diffusion to the surface, which occurs in the temperature range investigated here (300 – 800 K).

The surface assisted production of ammonia isotopologues NH$_{3-x}$D$_x$ has been studied with two complementary approaches. To quantify the efficiency of isotope exchange between neutral species, a molecular beam of NH$_3$ molecules impinging on a deuterated tungsten surface has been analyzed by on-line mass spectrometry. To evaluate which edge plasma species is most efficient for surface assisted isotope exchange, NH$_x$ surface precursors have been exposed to either, a neutral D$_2$ flux, an ionic D$_2$$^+$ flux or an atomic D flux and the resulting surface precursors have been probed by High Resolution Electron Energy Loss Spectroscopy. It is found that D radicals are most efficient while D$_2$ neutral fluxes are least efficient.

Finally, the absolute sticking probabilities of NH$_3$ molecules on W and 316L-SS surfaces have been measured with the molecular beam reflectivity method of King & Wells$^{7}$. For each material, three parameters have been varied independently: the material surface temperature, the NH$_3$ kinetic energy and the NH$_3$ coverage on the material surface. Firstly, the sticking probability of NH$_3$ decreases exponentially from near unity down to a few percent when increasing the material surface temperature from 125 to 425 K. Secondly, ammonia sticking probability is divided by a factor of $\sim$3 when increasing the NH$_3$ kinetic energy from 55 meV up to 255 meV. Thirdly, NH$_3$ sticking probability exhibits a complex behavior as a function of NH$_3$ coverage on the surface: the sticking probability initially increases and then decreases. Finally, NH$_3$ dissociation is found to be negligible below 450 K as seen by in situ Auger Electron Spectroscopy. These experimental observations reveal a molecular adsorption of ammonia on both W and 316L-SS that is assisted by two surface precursors, a NH$_3$ molecule bouncing on the bare surface (intrinsic precursor) and a NH$_3$ molecule bouncing atop a NH$_3$ covered area of the surface (extrinsic precursor).

A Generalized and Separable Kisliuk (GSK) kinetic model describing this precursors-mediated adsorption has been developed and it is able to reproduce quantitatively the evolution of NH$_3$ sticking probabilities with NH$_3$ surface coverage and the material surface temperature. The departure of the experimental data to the GSK model at the highest coverage (Figure 1) is explained by a long living ($\sim$ second lifetime) extrinsic precursor which manifests itself by a NH$_3$ partial pressure spike when the molecular beam exposure is stopped.

$^{1}$ A. Kallenbach et al., Plasma Phys. Control. Fusion 52, 055002 (2010)
$^{2}$ J. Schweinzer et al., Nucl. Fus. 51, 113003 (2011)
$^{3}$ C. Giroud et al., Nucl. Fus. 53, 113025 (2013)
$^{4}$ D. Neuwirth et al., Plasma Phys. Control. Fusion 54, 085008 (2012)
$^{5}$ M. Oberkofler et al., J. Nucl. Mater. 438, S258 (2013)
$^{6}$ R. Bisson et al., J. Nucl. Mater. 467, 432 (2015)
$^{7}$ King and Wells, Surf. Sci. 29, 454 (1972)

Country or International Organization France Aix-Marseille Université

### Primary authors

Dr Régis Bisson (Aix-Marseille Université, CNRS, PIIM UMR 7345) Dr Marco Minissale (Aix-Marseille Université, CNRS, PIIM UMR 7345) Mr Axel Dunand (Aix-Marseille Université, CNRS, PIIM UMR 7345) Dr Florin Ghiorghiu (Aix-Marseille Université, CNRS, PIIM UMR 7345) Mr Taki Aissou (Aix-Marseille Université, CNRS, PIIM UMR 7345) Prof. Thierry Angot (Aix-Marseille Université, CNRS, PIIM UMR 7345) Dr Gregory De Temmerman (ITER Organization)