Session: Government Agency Student Posters

Paper Number: 173088

Investigating Interface-Driven Electronic Phenomena in Endotaxial (Sb₂ Te₃)ₘ (Sb₂)ₙ Heterostructures 

Endotaxial heterostructures are an emerging class of layered materials with tunable electronic properties and the potential to host topologically protected modes, enabling innovations in future device technologies. These characteristics establish them as a key materials system for driving advances in quantum information processing, spintronics, and low-power, fault-tolerant electronics, as well as energy conversion and sensing. The (Sb₂Te₃)_m(Sb₂)_n superlattice series, consisting of m quintuple layers of Sb₂Te₃ and n bilayers of Sb₂, provides a model system for exploring how engineered structural modulation governs electronic properties. These misfit-layered materials, which maintain coherent interfaces despite lattice mismatch, demonstrate how stacking and interface design can profoundly influence physical properties at the atomic scale. Techniques such as scanning tunneling microscopy (STM), scanning tunneling spectroscopy (STS), and quasiparticle interference (QPI) are well suited for investigating these phenomena due to their ability to resolve structural and electronic features with sub-nanometer precision. At the same time, their application is often challenged by thermal broadening and tunneling artifacts that can obscure subtle spectral features critical for understanding electronic behavior, necessitating advanced normalization and denoising strategies to recover meaningful signals.

To address these challenges, STM and STS were used to probe the atomic-scale structure and electronic properties of solid-state synthesized samples across the series at room temperature (~300 K) and low temperature (~80 K) in an ultrahigh vacuum (UHV) chamber. After collecting current and voltage data, a normalization procedure was applied to STS spectra, smoothing the current–voltage (I–V) curve through convolution with an exponential decay function. This step suppressed sharp fluctuations near band edges and stabilized the spectra in low-signal regimes where both current and conductance approached zero. The smoothed I–V curve was then used to rescale the conductance (dI/dV), yielding a stable representation of the local density of states (LDOS). Electronic band gaps were determined by extrapolating linear fits to the onsets of the valence and conduction band edges, enabling consistent gap extraction across compositions.

Processed spectra reveal clear and systematic trends in electronic structure. Sb₂Te₃​ exhibits a finite band gap consistent with its known semiconducting character, while the insertion of Sb₂ layers progressively widens the band gap, highlighting how endotaxial stacking influences electronic behavior in higher-n superlattices. These trends agree with density functional theory (DFT) calculations and prior reports, underscoring the role of structural modulation in tuning electronic states. The ability to directly resolve these variations via STS highlights its power in mapping electronic properties in layered materials.

As a next step, QPI analysis will be performed on samples in the series, paired with self-supervised denoising methods such as Noise2Self. These approaches are being developed to enhance STM dataset quality and enable high-fidelity QPI analysis. Neural networks trained with these methods learn directly from noisy experimental data, suppressing residual noise without requiring noisy-clean pairs or introducing the bias inherent in conventional filters. Together, advances in computational normalization, noise mitigation, and self-supervised denoising contribute to a deeper understanding of electronic structure in misfit-layered systems and support the broader goal of designing two-dimensional materials with tailored quantum properties for emerging technologies.

This work is supported by the National Science Foundation through the Materials Research Science and Engineering Center at the University of Michigan, Award No. DMR‑2309029.

Presenting Author: Katharine Moncrieffe Cooper Union

Presenting Author Biography: Katharine Moncrieffe is a 2025 CMI REU student in the Goldman group at the University of Michigan and an undergraduate electrical engineering student at The Cooper Union. Her research focuses on STM/STS studies of endotaxial misfit materials and computational analysis of their local electronic structure for next-generation device applications.

Authors:

Katharine Moncrieffe Cooper Union
Yi-Hsin Shen University of Michigan
Jakob Hammond-Renfro University of Michigan
P. F. P. Poudeu University of Michigan
Rachel Goldman University of Michigan