Recently, Professor Jinlan Wang’s team from the School of Physics made significant progress in the kinetic regulation mechanism of epitaxial growth for 2D semiconductors. Collaborating with Professor Xinran Wang’s team from the Suzhou Laboratory, they utilized Metal-Organic Chemical Vapor Deposition (MOCVD) technology with an oxygen-assisted strategy to precisely regulate growth kinetics.This collaboration successfully overcame challenges inherent in traditional MOCVD techniques—such as carbon contamination, small grain domain sizes, and low mobility. They achieved a core mass production technology for 6-inch (150 mm) transition metal dichalcogenide (TMD) single crystals, marking a crucial step in transitioning 2D semiconductors from the laboratory to industrial application. This work, titled "Kinetic control of transition-metal dichalcogenide growth by oxygen-assisted MOCVD," was published online in the journal Science on January 29.
Figure 1. A group photo of Professor Jinlan Wang 's team (left 3rd from left) and Professor Xinran Wang 's team (right 3rd from right).
2D semiconductors like molybdenum disulfide (MoS2) are considered ideal materials for extending Moore’s Law due to their atomic-scale thickness, high carrier mobility at the ultimate scale, and low power consumption. However, their industrial-scale fabrication faces two major bottlenecks: while lab-level chemical vapor deposition (CVD) can produce high-quality single crystals, it suffers from limited size, poor uniformity, and low reproducibility. Meanwhile, traditional MOCVD, despite its scalability, typically yields low-quality polycrystalline materials with high defect density and low electron mobility due to reaction kinetics limitations. The lack of a universal mass-production technology for large-area single-crystal 2D semiconductors has been a key obstacle hindering progress in this field.
Based on first-principles computational simulations, Professor Jinlan Wang’s team systematically revealed the reaction kinetics limitations of traditional MOCVD. Calculation results showed that the sulfidation reaction energy barrier for Mo(CO)6 in conventional MOCVD is as high as 2.02 eV, severely restricting the growth rate and resulting in nanoscale crystal domains along with carbon contamination. Further investigation revealed that introducing oxygen fundamentally alters the reaction pathway: Mo(CO)6 and CS2 undergo pre-reactions with oxygen, forming molybdenum trioxide (MoO3) and elemental sulfur as active intermediates. This strategy achieves precise control over growth kinetics—compared to the conventional pathway, it increases the absolute enthalpy change of the exothermic reaction by 15-fold and significantly reduces the reaction energy barrier from 2.02 eV to 1.15 eV, thereby enhancing precursor reaction rates by approximately three orders of magnitude. Additionally, this approach suppresses the formation of carbon-containing intermediates at the source, effectively eliminating carbon contamination and ensuring ultra-high material purity. This provides fundamental theoretical support for the large-area, high-quality epitaxial growth of MoS2 single crystals and offers clear guidance for subsequent process design.
Figure 2. (a) Kinetic process of MoS2 growth by traditional MOCVD and oxy-MOCVD, and (b-e) characterization and electrical properties.
Building on this theoretical mechanism, Professor Jinlan Wang collaborated with Xinran Wang’s team to translate the core concept of "oxygen-assisted reconstruction of reaction pathways" into an industrially viable oxy-MOCVD technical solution. The team innovatively designed a pre-reaction chamber structure to achieve precise mixing and pre-oxidation of oxygen with precursors (Mo(CO)6, CS2), efficiently generating the theoretically predicted active intermediates MoO3 and elemental sulfur. This process replicates the key mechanism of "no carbon intermediate formation" predicted by theory, eliminating carbon contamination at the source. Experimental results demonstrated that the new approach significantly enhances MoS2 crystal domain growth rates compared to traditional MOCVD, with individual domain sizes increasing from hundreds of nanometers to hundreds of micrometers (up to 260 μm) and achieving unidirectional ordered alignment along specific crystal orientations, thereby solving the challenge of large-area uniform growth for mass production. Electrical performance tests further confirmed the theoretical advantages: field-effect transistor arrays fabricated from 6-inch MoS2 single crystals grown via oxy-MOCVD exhibited a maximum electron mobility of 123 cm2·V-1·s-1 (more than 10 times higher than traditional MOCVD) and an on/off ratio of 109. This achievement not only validates the theoretical prediction that "kinetic regulation improves material quality" but also marks a substantive breakthrough in core technologies for mass-producing 2D semiconductor single crystals, laying a material foundation for their large-scale applications in integrated circuits, flexible electronics, and sensors.
The paper's co-first author is Ruikang Dong (a recent PhD graduate from Southeast University, now a postdoc at Suzhou Laboratory). Professor Jinlan Wang from the School of Physics is a co-corresponding author. The work conducted at Southeast University was supported by the National Natural Science Foundation of China (Innovative Research Group Project, Key Program, Excellent Young Scientists Program) and the Jiangsu Frontier Leading Technology Basic Research Project.
Link: https://www.science.org/doi/10.1126/science.aec7259
