Research
2026년 10월 21일 수요일
New Study Pinpoints Where Semiconductor Crystals Begin to Grow
Abstract Nucleation control is fundamental to semiconductor deposition, governing when, where and how crystalline materials form. Conventional strategies can regulate nucleation density and, through area-selective growth, confine deposition to designated regions. However, they generally do not determine where an individual nucleus forms within the growth regions, leaving even single-nucleation events spatially stochastic and limiting deterministic construction of crystalline materials. Here we report etching-flux-mediated single-centred nucleation of two-dimensional (2D) semiconductors to deterministically localize a single nucleation event. The etching flux released from the barrier suppresses nuclei near the pattern boundary, leaving a single surviving nucleus at the pattern center, and systematic experiments elucidate the mechanism and establish nucleation design rules for zero, single and multiple nucleation regimes. Etching-flux-mediated single-centred nucleation enables single-crystal molybdenum disulfide growth at the 10-μm scale, field-effect mobilities of up to 117 cm2 V−1 s−1 and large-area uniformity with process compatibility. This in-plane chemical-flux strategy realizes spatially programmed growth, demonstrating nucleation–growth decoupling for line-shaped single crystals, multiple transistors integrated within a large common crystal, and aligned 2D lateral heterostructures for self-aligned contacts. These capabilities open a path towards advanced 2D electronic integration and expand semiconductor deposition from controlling where materials grow to controlling where and how crystals can begin to form. Controlling where a semiconductor grows is one thing. Determining exactly where its first crystal forms has been far more difficult. A joint research team from UNIST, KAIST, Hanyang University, and TDS Innovation Inc. has developed a technique that precisely controls where crystal growth begins, addressing a longstanding challenge in semiconductor fabrication. The technique enables a single crystal of molybdenum disulfide (MoS2), an atomically thin semiconductor, to form at the center of a designated growth area. Professor Byungjo Kim of the Graduate School of Semiconductor Materials and Devices Engineering at UNIST UNIST led the simulations that helped explain how the mechanism behind this precise control. Conventional deposition techniques can restrict semiconductor growth to selected regions, but they offer little control over where individual crystals first emerge. When multiple crystals grow and merge, boundaries from between them, disrupting the orderly arrangement of atoms and potentially impeding electron transport. To overcome this limitation, the researchers developed a method called etching-flux-mediated single-centered nucleation (EF-SCN). The approach relies on chemical species released from oxide barriers surrounding the growth area. These species inhibit crystal formation near the edges, where their influence is strongest, while allowing a single crystals to emerge at the center. As that crystal expands, the surrounding chemical environment continues to suppress competing nuclei. The results were striking. In experiments, involving 400 patterned regions, more than 99% produced a single MoS2 crystals at the intended central location. The researchers also demonstrated field-effect mobilities of up to 117 cm²/V·s, indicating promising electronic performance. Simulations led by Professor Kim's team at UNIST revealed why this approach works. By modeling the earliest stages of crystal formation, the researchers showed how the chemical environment varies across the growth area, creating conditions that favor nucleation at its center. Their findings provided a quantitative explanation for the experimental results. Beyond individual crystals, the team demonstrated line-shaped single crystals, multiple transistors integrated within a larger shared crystal, and aligned lateral heterostructures that could support self-aligned electrical contacts. These demonstrations suggest that controlling nucleation could offer greater flexibility in designing semiconductor structures. The findings also have implications for vertically stacked chips, an approach being explored to improve computing efficiency. Placing computing and memory components closer together could reduce the time and energy needed to transfer data between them. Because 2D semiconductors are only a few atoms thick, they are promising candidates for these densely integrated devices. “Crystal nucleation is inherently probabilistic, but this work shows that we can control where it begins,” said Professor Kibeom Kang of KAIST. “The ability to grow high-quality 2D semiconductors directly at designated locations could provide a foundation for vertically stacked semiconductor devices.” “By reproducing the earliest stages of crystal formation through simulations, we were able to quantitatively explain why nucleation occurs at a single location near the center of the growth region,” said Professor Byungjo Kim of UNIST. The study, co-first authored by Dr. Jeongwon Park, Jiyun Kim, and Sumin Kang of KAIST, as well as Jieun Oh of Hanyang University and Youngmin Sunwoo of UNIST, was published in Nature on October 8, 2026. Journal Reference Jeongwon Park, Jiyun Kim, Sumin Kang, et al. , “Spatially deterministic nucleation of 2D semiconductors by etching flux,” Nature , (2026).
소개
연구시설
교수진
입학