Open Access iconOpen Access

ARTICLE

Active Learning-Driven Optimization of Sulfurization–Selenization Processes in Sb2(S,Se)3 Thin Films for Enhanced Photovoltaic Efficiency

Yunpeng Wen1,*, Bingyang Ke2, Junrong Ding3

1 College of Engineering, Oregon State University, Corvallis, OR 97330, USA
2 College of Engineering, Northeastern University, Brighton, MA 02135, USA
3 College of Business, Oregon State University, Corvallis, OR 97330, USA

* Corresponding Author: Yunpeng Wen. Email: email

Chalcogenide Letters 2026, 23(1), 4 https://doi.org/10.32604/cl.2026.076587

Abstract

This study reports an active learning (AL)-guided strategy to optimize the sulfurization–selenization processing conditions of Sb2(S,Se)3 thin-film photovoltaic absorbers for enhanced power conversion efficiency (PCE). By coupling Gaussian process modeling with iterative experimental feedback, we explored 20 targeted annealing conditions across the full compositional spectrum (x = 0–1) and identified an optimal S/(S + Se) ratio of 0.40 (x = 0.60), which yielded a band gap (Eg) of ~1.34 eV, close to the theoretical Shockley–Queisser optimum. The optimized process employed a controlled two-step 420°C anneal with sequential H2Se→H2S exposure, which produced large plate-like grains (300–500 nm) with columnar cross-sectional morphology and eliminated the ribbon-like anisotropy typical of pure Sb2Se3. XRD confirmed a single-phase orthorhombic structure with no secondary phases, while EDS mapping showed homogeneous S/Se distribution and a mild back-surface S gradient. Optical absorption coefficients exceeded 105 cm1 (hν > 1.7 eV), and photoluminescence revealed a reduced Stokes shift (~40 meV) and low Urbach energy (~20 meV), indicating suppressed defect states. Devices fabricated with a Zn(O,S) buffer achieved average metrics of Voc = 0.58 V, Jsc = 25.0 mA cm2, FF = 78%, and PCE = 12.2%, representing a ~62% relative improvement over our baseline Sb2Se3 cells (7.5%). The EQE exceeded 90% across 400–900 nm, and the integrated Jsc matched measured values within 5%. Preliminary stability tests retained 95% of initial PCE after 100 h under 1-sun illumination at 45°C. These findings establish AL-driven exploration as a powerful method for rapidly converging on high-performance chalcogenide processing regimes.

Keywords

Grain coarsening; defect suppression; optical absorption; compositional gradient; photovoltaic stability

Cite This Article

APA Style
Wen, Y., Ke, B., Ding, J. (2026). Active Learning-Driven Optimization of Sulfurization–Selenization Processes in Sb2(S,Se)3 Thin Films for Enhanced Photovoltaic Efficiency. Chalcogenide Letters, 23(1), 4. https://doi.org/10.32604/cl.2026.076587
Vancouver Style
Wen Y, Ke B, Ding J. Active Learning-Driven Optimization of Sulfurization–Selenization Processes in Sb2(S,Se)3 Thin Films for Enhanced Photovoltaic Efficiency. Chalcogenide Letters. 2026;23(1):4. https://doi.org/10.32604/cl.2026.076587
IEEE Style
Y. Wen, B. Ke, and J. Ding, “Active Learning-Driven Optimization of Sulfurization–Selenization Processes in Sb2(S,Se)3 Thin Films for Enhanced Photovoltaic Efficiency,” Chalcogenide Letters, vol. 23, no. 1, pp. 4, 2026. https://doi.org/10.32604/cl.2026.076587



cc Copyright © 2026 The Author(s). Published by Tech Science Press.
This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 135

    View

  • 34

    Download

  • 0

    Like

Share Link