A team of scientists at the U.S. Naval Research Laboratory (NRL) has uncovered a direct link between sample quality and the degree of valley polarization in monolayer transition metal dichalcogenides (TMDs). In contrast with graphene, many monolayer TMDs are semiconductors that show promise for future applications in electronics and optoelectronics technologies.
A ‘valley’ refers to the region in an electronic band structure where both electrons and holes are localized and ‘valley polarization’ refers to the ratio of valley populations — an important metric applied in valleytronics research.
"A high degree of valley polarization has been theoretically predicted in TMDs, yet experimental values are often low and vary widely," said Kathleen McCreary, Ph.D., lead author of the study. "It is extremely important to determine the origin of these variations in order to further our basic understanding of TMDs as well as advance the field of valleytronics."
Many of today’s technologies rely simply on the charge of the electron and how it moves through the material. However in certain materials, like the monolayer TMDs, electrons can be selectively placed into a chosen electronic valley using optical excitation.
"The development of TMD materials and hybrid 2D/3D heterostructures promises enhanced functionality relevant to future Department of Defense missions," said Berend Jonker, Ph.D., principal investigator of the program. "These include ultra-low-power electronics, non-volatile optical memory and quantum computation applications in information processing and sensing."
The growing fields of spintronics and valleytronics aim to use the spin or valley population, rather than the only charge, to store information and perform logic operations. Progress in these developing fields had attracted the attention of industry leaders and has already resulted in products such as magnetic random access memory that improve upon the existing charge-based technologies.
The team focused on TMD monolayers that have high optical responsivity and found that sample exhibiting low photoluminescence (PL) intensity exhibiting a high degree of valley polarization. The findings suggest a means to engineer valley polarization via the controlled introduction of defects and nonradiative recombination sites.
"Truly understanding the reason for the sample-to-sample variation is the first step towards valleytronic control," McCreary said. "In the near future, we may be able to accurately increase polarization by adding defect sites or reduce polarization by passivation of defects."
The results of this research were reported in the August 2017 edition of American Chemical Society’s Nano.