Probing the Quality of Bi₂Se₃ Films Grown by Molecular Beam Epitaxy In-Situ with Spectroscopic Ellipsometry

Quantum materials such as topological insulators (TI) like Bi₂Se₃ feature massless Dirac-like states due to their spin-momentum coupling and time reversal invariance. Those massless topological surface states (TSS) enable the study of new quantum physics and with that allow access to a plethora of technological advancements and the realization of applications with new functionalities for more energy efficient and faster computation schemes. The key to exploiting the unique properties of TIs lies in high-quality materials synthesis. However, Bi₂Se₃ synthesis is plagued with a large number of unintentional carriers, shifting the Fermi level up into the conduction band and masking the TSS. Methods like compensation doping, or cation/anion mixing are being pursuit successfully to prevent the Fermi level drift out of the band gap, but the price paid is a substantial carrier mobility reduction in the film limiting TSS observation to very low temperatures. To date, best results for achieving low bulk carrier concentrations and high mobilities are reported for a meticulously designed buffer layer (BL) growth approach consisting of Bi₂Se₃, In₂Se₃, and (Bi_xIn_1-x)₂Se₃ layers. This approach, on the other hand, is slow and thus expensive due to the complexity of the BL-Bi₂Se₃ synthesis engineering and subsequent TSS probing iteration. Developing a fast, ideally <i>in-situ</i> feedback loop between growth and electronic properties is highly advantageous to speed up this approach and realize meaningful progress towards highest-quality Bi₂Se₃ thin film fabrication.<br/>In this talk, we elucidate how <i>in-situ</i> spectroscopic ellipsometry (SE) can be employed as a non-invasive, non-contact, real-time, ultra-fast probe of the electronic properties of Bi₂Se₃ during growth. Evaluating the <i>in-situ</i> SE data obtained before and after growth of each layer from differently designed Bi₂Se₃, In₂Se₃, and (Bi_xIn_1-x)₂Se₃ heterostructures, we found that the observed SE measurables for Bi₂Se₃ depend sensitively on the sample heterostructure design. To gain deeper insights about the observed differences, we developed oscillator models for the dielectric functions of Bi₂Se₃, In₂Se₃, and (Bi_xIn_1-x)₂Se₃, and compared the weighted average of the broadening parameter of the oscillators that make-up the dielectric function. As a result, we found that the weighted average broadening parameters for the Bi₂Se₃ oscillators of films grown on a In₂Se₃, and (Bi_0.7In_0.3)₂Se₃ heterostructure were 20 % smaller than those of films grown directly on Al₂O₃. Smaller broadening parameters are generally associated with the degree of momentum-conservation between electronic transitions due to less pronounced electron scattering. Our finding thus indicates that the Bi₂Se₃ grown on the In₂Se₃, (Bi_0.7In_0.3)₂Se₃ heterostructure has superior quality compared to films grown on bare Al₂O₃. This conclusion was corroborated by Hall mobility measurements conducted at room-temperature, which ranged highest for Bi₂Se₃ films grown on In₂Se₃, (Bi_0.7In_0.3)₂Se₃ heterostructures. Based on the presented data, <i>in-situ</i> SE has the potential to serve as an ultra-fast, real-time probe of the electronic properties of Bi₂Se₃, and is thus highly advantageous for speeding up the engineering process of high-quality TIs growth.