Understanding W Transport from the Closed DIII-D SASVW Divertor via Experimentation and Interpretive Modeling
Date Issued
August 2026
Author(s)
Messer, Seth H
The University of Tennessee
Advisor(s)
David Donovan
Additional Advisor(s)
Lawrence Heilbronn
Greg Sinclair
Jake Nichols
Abstract
High-Z impurity transport from a closed, upper slot divertor is examined across the detachment threshold in the DIII-D tokamak using experimental data and assistive code analysis. An experiment is conducted to induce divertor detachment by increasing the main ion density via D2 gas puffing in the DIII-D SAS-VW divertor. SAS-VW is a closed, W-coated divertor in an otherwise carbon device designed to more easily facilitate divertor detachment while reducing impurity leakage. Experimental measurements of tungsten (W) accumulation in the core and far-Scrape-Off Layer (SOL) indicate increased W leakage with the onset of divertor detachment despite the reduction of target gross erosion predicted by modeling. Two separate, distinct routes are explored in an attempt to explain this phenomenon: (1) diagnostic measurements from the experiment are analyzed and interpreted, and (2) SOLPS-DIVIMP simulations for comparable discharges are conducted to examine the parallel-to-B impurity momentum balance.
Experimental measurements indicate W from SAS-VW travels through the private flux region (PFR) to the inner wall in a series of re-erosion/re-deposition steps. A series of USN L-mode discharges in each toroidal magnetic field (BT) direction crossed the divertor detachment threshold by scanning the line-averaged density from 3.15×1019 to 4.35×1019 m-3. W deposition on far-SOL collector probes (CPs) is compared to campaign-integrated W concentration measurements at the wall measured with X-Ray Fluorescence (XRF). Results suggest W accumulates on the high-field side wall and acts as a secondary W source. Lastly, W deposition is 2.5× larger on the side of the probe that connects to the inner target only in detached conditions (ion B×∇B drift directed towards the X-point). This may be due to broadening of the far-SOL on the high-field side or enhanced E×B flow from the inner target with the onset of detachment.
SOLPS-DIVIMP simulations of W transport highlight the need to use kinetic models over fluid models for high-Z impurity leakage predictions in the SOL, and particularly in the far-SOL. A quantifiable metric is introduced and indicates that for low-charge states of Tungsten (e.g., W1+), non-Maxwellian effects are significant, indicating a clear failure of the Maxwellian fluid approximation in certain transient regimes.
Subjects
Degree
Doctor of Philosophy
Major
Nuclear Engineering
File(s)![Thumbnail Image]()
Name
MESSER-PRIMARY-2026.pdf
Size
5.48 MB
Format
Adobe PDF
Checksum (MD5)
c154fa9b768d94749e420e9cf7e8b612