Substrate-dependent deposition efficiency on mirrors exposed in the TCV divertor

https://doi.org/10.1016/j.fusengdes.2007.04.044Get rights and content

Abstract

Given the planned extensive use of metallic mirrors in the optical diagnostics systems of ITER, the study of reflectivity changes induced by erosion and/or redeposition of impurities on the mirror surfaces is of primary importance for the reliability of the diagnostics signals. This contribution will demonstrate that the mirror material choice can exert a significant influence on the relative importance of erosion/redeposition affecting the mirror reflectivity. A dedicated manipulator has been designed to allow exposure of mirror samples in the divertor region of the Tokamak à Configuration Variable (TCV) tokamak. Mirrors from different materials have been exposed both during short experimental campaigns and boronisation procedures. Before and after exposures the mirrors were characterized with different surface analysis techniques (XPS, SIMS, SEM, EDX, ellipsometry). Under identical exposure conditions, the mirror material can strongly influence the deposit thickness found on the sample: the carbon layer thickness on a Si sample is much higher than on a Mo sample. These results have potentially important consequences for the first mirror material choice in ITER.

Introduction

Reliable optical plasma diagnostic systems will be mandatory tools for the success of the International Thermonuclear Experimental Reactor (ITER) [1]. They will be used to characterise, control and understand the burning plasmas that ITER will produce and will be essential for machine protection. In contrast to conditions in today’s tokamaks, high radiation levels and neutron fluxes are expected in ITER. To minimize the possible neutron leakage, the plasma light/radiation will be transmitted by first mirrors to diagnostics through a labyrinth embedded in the shielding material [2]. This arrangement has the advantage of recessing the vacuum windows or the optical fibres located at the end of the labyrinth some metres away from the zone of intense radiation and to shield them by substantially reducing the neutron fluxes they will face. Indeed, refractive components (windows, fibres) suffer from radiation-related effects (luminescence, absorbance) which obviate their use in the vicinity of high neutron or radiation fluxes [3]. The possible deterioration of the first mirror reflectivity as a result of erosion by charge-exchange neutrals (CXN) and re-deposition of impurities on the mirror surface represents a serious concern for the reliability of spectroscopic signals.

The deterioration of the mirror reflectivity induced by these different damaging effects has been previously studied in various laboratory experiments [4], [5], [6]. Dedicated experimental programmes have also been initiated in a number of tokamaks to further address this issue [7], [8], [9], [10]. In these experiments, although erosion and deposition effects were investigated, the mirror material was thought to be of importance only under erosion conditions, where a high-Z material (like molybdenum) in the form of a single-crystal has demonstrated its ability to withstand erosion conditions with a sufficient lifetime [11]. Deposition of impurities was, however, thought to be independent of the substrate material. It has recently been shown [12] that under similar conditions the erosion/deposition patterns of various materials can be very different. This contribution provides a great deal more details on the experiment and results briefly mentioned in ref. [12] and will show clearly how the choice of substrate material can have a significant impact on the longevity of mirror surfaces in the presence of erosion/redeposition processes, such as those likely to be found in the ITER tokamak environment. These results thus have potentially important consequences for ITER first mirror material choices.

Section snippets

Design and location of the sample manipulator

The Tokamak à Configuration Variable (TCV) is a medium sized tokamak (major radius 0.89 m, minor radius 0.25 m, vacuum toroidal field 1.43 T) designed with 16 individually powered poloidal field shaping coils, allowing a huge variety of magnetic equilibria to be created in its novel, rectangular shaped vacuum vessel [13]. The latter is completely open, with no dedicated baffling structures for specific diverted plasmas. Both limiter and divertor configurations are regularly produced for

Surface analysis of the deposited material

Table 1 summarises the different materials tested, the experimental conditions (recess distance, number of tokamak shots, glow discharge conditioning time, etc.) and estimation of the deposited thickness. Evidently, only very thin layers have been found on the different samples, especially when high-Z materials (Mo, W) have been exposed. No difference in the carbon layer thickness was found when Mo and W were exposed simultaneously. The very low values found after experiment numbers 2 and 3 are

Conclusion

This paper has described the results of experiments designed to study the deposition of carbon layers formed on samples exposed in the divertor SOL of the TCV tokamak, remote from the plasma. The results demonstrate that the first mirror material choice should take into account the fact the substrate material itself plays an important role in determining the eventual thickness of any deposited layer. Indeed at the beginning of the carbon film growth, the deposition rate on a high-Z material is

Acknowledgements

The authors are grateful to P. Conti, J.M. Mayor and B. Joye from CRPP Lausann, for assistance in the conception, manufacture and installation of the sample manipulator. The SIMS measurements were done by U. Breuer and A. Scholl from Zentralabteilung für Chemische Analysen, Forschungszentrum Jülich, Germany.

Financial support of the Swiss Federal Office for Education and Science, and of the Federal Office of energy is gratefully acknowledged. This work was partly funded by EURATOM and the Swiss

References (32)

  • A.E. Costley et al.

    Fusion Eng. Des.

    (2001)
  • V.S. Voitsenya et al.

    J. Nucl. Mater.

    (2001)
  • T. Sugie et al.

    J. Nucl. Mater.

    (2004)
  • P. Wienhold et al.

    J. Nucl. Mater.

    (2005)
  • M. Lipa et al.

    Fusion Eng. Des.

    (2006)
  • R.A. Pitts et al.

    J. Nucl. Mater.

    (1997)
  • J. Marki et al.

    J. Nucl. Mater.

    (2007)
  • M. Mayer et al.

    J. Nucl. Mater.

    (2003)
  • K.G. Tschersich

    J. Nucl. Mater.

    (1989)
  • C. Arnas et al.

    J. Nucl. Mater.

    (2005)
  • J. Winter et al.

    J. Nucl. Mater.

    (1999)
  • W. Möller et al.

    Comput. Phys. Commun.

    (1988)
  • ITER physics basis, Nucl. Fusion 39-12 (1999)...
  • T. Sugie et al.

    J. Nucl. Mater.

    (2002)
  • V. Voitsenya et al.

    Rev. Sci. Instrum.

    (2001)
  • G. De Temmerman, M.J. Rubel, J.P. Coad, R.A. Pitts, J.R. Drake, and JET-EFDA contributors, Proc. of the 32nd EPS conf....
  • Cited by (12)

    • TCV mirrors cleaned by plasma

      2017, Nuclear Materials and Energy
      Citation Excerpt :

      The n (at 632.5 nm) varied from 1.48 to 1.56 for all points and k was calculated to be zero. This revealed a soft hydrocarbon film typical for mirrors exposed in TCV [12] and resembled T-10 tokamak (Kurchatov Institute) cleaning discharge films on mirrors [13]. The reflectivity calculated using the model: air, our Cauchy film, SiO film (10 nm) and an Au layer using the Palik constant is plotted in Fig. 2.

    • Experimental studies of first mirror exposure and surface recovery on EAST

      2015, Journal of Nuclear Materials
      Citation Excerpt :

      The research on FM in a deposition-dominated condition at existing tokamaks is very important for the next-step burning plasma devices, such as ITER. Various experiments of FM exposure, deposition mitigation and surface recovery have been carried out in many tokamaks, such as JET [3], Tore Supra [7], HL-2A [8], TCV [9], TEXTOR [10], ASDEX-Upgrade [11] and DIII-D [12]. The authors of [1] have summarized the FM experiments carried out in 10 tokamaks in Ref. [1] and concluded that the deposition of contaminants cannot be avoided even if the FMs are located in the diagnostic ports or protected with a diaphragm.

    • Experimental study of contamination and cleaning of in-vessel mirrors for ITER optical diagnostics on T-10 and QSPA-T facilities

      2013, Journal of Nuclear Materials
      Citation Excerpt :

      For example, thin (≈10 nm) layer of carbon contamination on the surface of the first metallic mirror for ITER optical diagnostics changes significantly the reflectivity owing to interference effect [1]. For reliable protection and cleaning of optical elements, it is important to know deposition rates, chemical composition and structure of ITER-relevant deposits as well as to possess sensitive spectroscopic methods for their characterization [2,3]. In this work, co-deposition of hydrogen, carbon and some metallic impurities on metallic (Mo and stainless steel) mirrors and Si probes have been studied on T-10 tokamak (“Kurchatov Institute”) and QSPA-T plasma gun (TRINITI) facilities under controlled ITER-relevant conditions.

    View all citing articles on Scopus
    View full text