Two-dimensional semiconductors such as molybdenum disulfide have a band gap that makes them promising for future transistors, but growing them on wafers has been held back by a fundamental problem: crystals start forming at random locations, and when adjacent crystals meet they create boundaries that degrade electrical performance. A team from KAIST and South Korean startup TDS Innovation reports in Nature a method to control where nucleation begins, using an etching flux to suppress crystal formation everywhere except the geometric center of each patterned growth region.
The researchers built working field-effect transistors with the resulting material and measured higher charge-carrier mobility than previously reported for selectively grown MoS₂ transistors. In a separate scaling test across a two-centimeter substrate, the process produced single crystals at 397 of 400 patterned sites, a 99.3% yield. That suggests the approach can be made uniform, though it stops short of demonstrating full commercial chip production.
The significance lies in what controlled growth enables: stacking 2D semiconductor devices directly on silicon could put logic and memory closer together, reducing the energy spent moving data between them. The paper does not specify a production timeline, but TDS says it is aiming for commercial use around 2030. The company notes that accelerating demand from AI, including physical AI and robotics, could help bring adoption forward, subject to further validation.