Synthesis of Cu2ZnSnS4 (CZTS) absorber by rapid thermal processing (RTP) sulfurization of stacked me
S.M. Pawar, A.I. Inamdar, B.S. Pawar, K.V. Gurav, S.W. Shin, Xiao Yanjun, S.S. Kolekar , Jung-Ho Lee, JinHyeok Kim, Hyunsik Im
Abstract
Cu2ZnSnS4 (CZTS) absorbers have been grown on Mo-coated glass substrates by the rapid thermal processing (RTP) sulfurization of stacked metallic precursor (CZT) films at different annealing temperatures ranging from 500 to 580 °C for 5 min in sulfur atmosphere. The effects of sulfurization temperature on the structural, morphological, chemical, and optical properties of the CZTS absorbers have been investigated. XRD and Raman studies reveal that the as-deposited stacked metallic precursor films consist of metal elements such as Zn, Sn and binary alloys such as Cu6Sn5, Cu3Sn and CuZn. The sulfurized CZTS absorber films have single phase polycrystalline kesterite crystal structure with dense morphology. At 580 °C, the CZT metallic precursor film is fully sulfurized with Zn-rich and Sn-poor composition, and its bandgap energy is found to be 1.50 eV. The solar cell fabricated with the CZTS absorber grown at an optimized sulfurization temperature of 580 °C shows a conversion efficiency of ~5% for a 0.44 cm2 area with Voc=561 mV, Jsc=18.4 mA/cm2, and FF=48.2.
Anyone calibrating a tube furnace against a 5 min RTP recipe should keep this 580 °C / 1.50 eV pair on hand — the Zn-rich, Sn-poor shift shows up before efficiency does, and FF=48.2 is where you notice. Two-step sulfurization people face the same moving target. I file run sheets the way I'd read Pokemon card value charts, since "optimized" only means something if the composition holds.
My tube furnace can’t reproduce that 5 min RTP ramp: reaching 580 °C alone takes 12 min, so SnS loss overlaps CZT interdiffusion. The 1.50 eV, Zn-rich/Sn-poor result and FF=48.2 then look less like an optimum than a moving target. I use the AI tools I rate best for run sheets, but they can’t fix this thermal budget mismatch.
580 °C for 5 min did the opposite for me: a slightly hotter, shorter RTP ramp on a similar CZT stack stripped Sn and pushed the Zn-rich, Sn-poor shift harder, so I moved to two-step sulfurization. The FF=48.2 here tells me the same loss is already underway. I log every run like a grow-a-garden value list, because “optimized” means different things when Sn won’t stay put.
Calling 580 °C the optimized temperature sits wrong with me. That's exactly where Sn starts leaving as volatile SnS — the Zn-rich, Sn-poor composition at 1.50 eV and FF=48.2 is the symptom, not the win. A shorter 5 min RTP ramp might hold stoichiometry better. I only track this stuff while printing practice sheets for my kid with a worksheet generator for handwriting practice, so grain-boundary physics is above my pay grade — but 5% efficiency reads like a phase problem, not a temperature optimum.