Skip to main content

NREL Reveals Benefits of O Contact with Defects in 2D Semiconductors

December 11, 2015

Researchers at the Energy Department's National Renewable Energy Laboratory (NREL) have explained the unusual behavior that stems from oxygen molecules interacting with defects in two-dimensional (2D) semiconductors.

Yuanyue Liu, Pauls Stradins, and Su-Huai Wei have published their findings, "Air Passivation of Chalcogen Vacancies in Two-Dimensional Semiconductors," in Angewandte Chemie International Edition. Their research, conducted within NREL's Materials Science Center and Materials Application and Performance Center, used computational resources at the High Performance Computing Center of NREL and the National Energy Research Scientific Computing Center.

2D semiconductors are promising candidates for next-generation electronics, optoelectronics, and energy conversion and storage devices. They have an extremely high surface-to-mass ratio because they are only one or two atomic layers thick. One result is that the atoms and atomic defects in these materials are sensitive to the environmental conditions surrounding them.

Defects are generally more reactive than the perfect atomic sites and have a stronger coupling with the environment. In 2D semiconductors, chalcogen vacancies are the most common defects, with chalcogens being elements such as sulfur, selenium, or tellurium.

The scientists studied the interaction between oxygen (O2) and the vacancies by performing first-principles calculations. At the time of the study, Liu was an NREL post-doctoral researcher and Wei was an NREL Research Fellow. Recently, they have both relocated—Liu to Caltech as a Resnick Prize Postdoc Fellow, and Wei to Beijing Computational Science Research Center as a professor and division head.

Together with Stradins, their research results at NREL have shown that chemisorbed oxygen molecules change the vacancies from carrier traps—usually thought to be harmful—into electronically benign sites. This unusual behavior originates from the isovalence between oxygen and chalcogen elements when bonded with metal.

Physisorbed molecules can modify the electronic structure of the defects, but this modification is not stable at room temperature. In contrast, chemisorbed oxygen cannot be released at ambient conditions because of the strongly exothermic adsorption. Therefore, it can irreversibly change the properties of the defects. However, chemisorption of oxygen does not occur spontaneously—an energy barrier must be overcome before reaching the chemisorbed state from the physisorbed state.

NREL's calculations show that chemisorbed oxygen molecules have a significant impact on the electronic structure of the chalcogen vacancies. These vacancies are typically harmful to the electronic performance of the semiconductor. But upon oxygen chemisorption, the deep gap states are removed completely, thus converting the vacancies to electronically inactive sites.

The researchers have proposed an approach based on their findings to improve the performance of 2D semiconductors by using oxygen to passivate the defects.

"Our results suggest that it is possible to improve the transport properties and the quantum efficiency of the 2D chalcogenide semiconductors by oxygen molecule treatment," Stradins said.

Although this research is focused on chalcogenide semiconductors, Liu, Stradins, and Wei believe that similar effects can also be found in oxide semiconductors. Detailed investigations are under way.

This work was supported by the Solar Energy Technologies Program at the DOE Office of Energy Efficiency and Renewable Energy.

Illustration of molybdenum sulfide with a chalcogen vacancy

Schematic of molybdenum sulfide (purple = molybdenum, yellow = chalcogen) with a chalcogen vacancy (missing atom in structure). Among the molecules in air, molecular oxygen (O2, represented by the two red spheres) can be chemisorbed at the chalcogen vacancy, converting the defect (which traps electrons as represented by e-) to an electronically benign site. This discovery can lead to an approach to improve the performance of 2D semiconductors.