EUVIMEDEUVIMED entdecken
Crossref

Condition-gated graph neural networks with autoregressive training for plasma-etching profile prediction

JungHo Song, Ki-Hwan Cho, Hyunsik Kim

Journal of Vacuum Science & Technology A · 01.12.2026 · 44 (6)

DOI: 10.1116/6.0005847 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Vacuum Science & Technology A, online erschienen am 03.11.2026. Verfasst von JungHo Song und 2 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Gefördert von: National Research Council of Science and Technology, National Fusion Research Institute.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate prediction of plasma-etching profile evolution can support process exploration, but recursive surrogate predictions may accumulate shape errors. We present a condition-gated graph neural network (GNN) trained with dataset aggregation on a benchmark of 200 simulated Si etching processes in Cl2 plasma. Using the benchmark’s existing polar-coordinate representation, we treat each angular sample as a graph node and connect neighboring samples. Process conditions gate the features at each node, while a gated recurrent unit predicts profile evolution over time. On 29 common validation records, GNN-full achieves a mean intersection over union of 0.9904 for the final single step and 0.9561 for a 39-step rollout, compared with 0.9880 and 0.9487 for a retrained cascade recurrent neural network (CRNN). Corresponding mean radial-profile errors are 0.43 and 2.10 nm for GNN-full vs 0.59 and 2.78 nm for CRNN. Three-seed ablations identify nodewise conditioning and autoregressive training as the main contributors to long-horizon stability. A parameter-matched transformer retains similar performance, indicating that the angular-mixing mechanism is less decisive. Training on 50 selected records reduces training time approximately threefold while retaining most autoregressive accuracy. Both methods reproduce major process-condition trend signs, but CRNN has a lower vertical-depth error in the one-factor-at-a-time sweeps. GNN inference takes approximately 1.4 ms per step, with greater training and repeated-inference cost than CRNN. The results demonstrate a benchmark-specific accuracy–cost trade-off; manufacturing significance and prospective generalization require further validation.

Förderung

  • National Research Council of Science and Technology
  • National Fusion Research Institute

Zitieren

APA

Song, J., Cho, K., & Kim, H. (2026). Condition-gated graph neural networks with autoregressive training for plasma-etching profile prediction. Journal of Vacuum Science & Technology A, 44(6). https://doi.org/10.1116/6.0005847

Vancouver

Song J, Cho K, Kim H. Condition-gated graph neural networks with autoregressive training for plasma-etching profile prediction. Journal of Vacuum Science & Technology A. 2026;44(6). doi:10.1116/6.0005847

BibTeX
@article{Song2026,
  author = {Song, JungHo and Cho, Ki-Hwan and Kim, Hyunsik},
  title = {Condition-gated graph neural networks with autoregressive training for plasma-etching profile prediction},
  journal = {Journal of Vacuum Science & Technology A},
  year = {2026},
  volume = {44},
  number = {6},
  doi = {10.1116/6.0005847}
}
Kostenlos starten →