Unraveling the plasma-material interface with real time diagnosis of dynamic boron conditioning in extreme tokamak plasmas
Summary¶
First in-chamber XPS (via MAPP) on samples between deuterium plasma exposures probes boronized ATJ graphite after boronization. Reactive molecular dynamics explains how B, C, O, and D interplay controls retention; QCMD checks trends. Simulations decode oxygen and boron roles and predict D uptake on boronized carbon similar in magnitude to prior predictions for lithiated, oxidized carbon. The paper positions boronization as a common wall conditioning strategy in tokamaks and stresses that fuel retention on PFM materials couples plasma exposure to evolving surface chemistry—motivating in situ spectroscopy rather than post-mortem coupons alone (introduction themes; extract).
Methods¶
Experiment (tokamak PMI). The Material Analysis Particle Probe (MAPP) on NSTX-U performs XPS on samples retracted to an in-vessel analysis chamber between deuterium plasma exposures—Mg Kα excitation (hν = 1253.4 eV) with a hemispherical analyzer configuration described in Nucl. Fusion §2.1. MAPP positions ATJ graphite coupons flush with lower divertor tiles during the 2015–2016 campaign window reported here, focusing on a boronized surface state ~12 days after boron deposition; an Au reference sets the binding-energy scale (4f\(_{5/2}\) at 88 eV as stated).
MD application (ReaxFF + QCMD). Reactive MD with ReaxFF treats B/C/O/D chemistry on boronized graphite models meant to mirror XPS-resolved surface speciation under D exposure. Quantum–classical MD (QCMD) is used as a cross-check on selected retention trends (coupling philosophy and software names appear in the Methods section). Supercell construction, PBC, temperature, timestep, thermostat, ensemble, and trajectory length are specified in Nucl. Fusion §2 onward; this wiki page does not transcribe those tables from the short front-matter extract.
Force-field training is N/A (published reactive parameters as cited). Bulk standalone DFT is N/A as the headline method (QCMD supplies QM-level forces in a hybrid sense where used).
MD blueprint honesty. Reactive molecular dynamics with ReaxFF and QCMD checks use PBC graphite/boronized surface cells as described in Nucl. Fusion. LAMMPS/other engine strings, explicit NVT/NPT/NVE labels, timestep, thermostat, barostat/pressure, and equilibration/production durations (ps/ns) are N/A on this page—copy from the PDF Methods.
Findings¶
- Time-resolved (between-exposure) XPS resolves B–C–O–D chemistry not accessible with ex situ post-campaign analysis alone.
- Reactive MD reproduces the subtle B/O/C/D interplay governing D retention on boronized surfaces.
- Predicted D uptake for boronized carbon is close to prior lithiated + oxidized carbon predictions—highlighting oxygen-mediated retention pathways.
- Integrated picture: boron and oxygen co-species on graphite are argued to set effective trapping sites that compete with bare C–D chemisorption scenarios emphasized in older models (discussion as summarized in abstract/extract).
- Community context: linking boronized PFMs to fuel retention benchmarks matters for ITER-era material choices and divertor conditioning strategies discussed in the fusion literature (introduction framing).
Limitations¶
Single campaign window and specific sample history; MD models specific surface models and may not span all divertor conditions.
Relevance to group¶
Shows ReaxFF-style reactive MD integrated with tokamak PMI diagnostics—adjacent to plasma–surface and carbon hydrogen retention modeling in the broader ReaxFF literature.