Approaching the Intrinsic Properties of Moiré Structures Using Atomic Force Microscopy Ironing

Nano-spherical tip-based smoothing with minimal damage for 2D van der Waals heterostructures

Two-dimensional materials and their heterostructures have significant potential for future developments in materials science and optoelectronics due to their unique properties. However, their fabrication and transfer process often introduce impurities and contaminants that degrade their intrinsic qualities. To address this issue, current atomic force microscopy (AFM) probe contact mode methods provide a solution by allowing in situ cleaning and real-time observation of the nanoscale cleaning process. Nevertheless, existing pyramidal probes may scratch surfaces and damage heterostructures during force application. Therefore, we proposed a method based on the nano-spherical probe contact mode to clean residual and polymer contamination for minimum damage cleaning of MoS2/hBN substrates. Comparative experiments with pyramidal probes in 2DM morphology and photoluminescence (PL) have shown that nano-spherical probes are exceptionally effective in cleaning bubbles of various sizes, compared to uncleaned MoS2, where PL full width at half maximum (FWHM) averages 0.115 eV, nano-spherical probes reduce it by 30% to 0.08 eV. Pyramidal probes, however, only clean smaller bubbles and leave residuals in larger ones, resulting in less optimal PL mapping data with values in both the 0.09 eV and 0.0115 eV regions. We also collected the standard deviation of the FWHM data points for the uncleaned region and the regions cleaned by the pyramidal and nano-spherical probes, which were 0.02773, 0.01895, and 0.00531, respectively. Notably, the standard deviation of the FWHM in the nano-spherical probe-cleaned region is only 28% of that in the pyramidal probe-cleaned region. Then, increasing the applied force leads to damage in the crystal structure, resulting in potential inconsistencies across different areas, as evidenced by KPFM and SEM observations. In contrast, nano-spherical probes demonstrate a uniform potential in KPFM and consistently maintain a smooth surface morphology in SEM throughout the process. This approach highlights the potential of nano-spherical probes to advance minimum-damage cleaning techniques in 2D material research and applications.

Nanoironing van der Waals Heterostructures toward Electrically Controlled Quantum Dots

Assembling two-dimensional van der Waals (vdW)-layered materials into heterostructures is an exciting development that sparked the discovery of rich correlated electronic phenomena. vdW heterostructures also offer possibilities for designer device applications in areas such as optoelectronics, valley- and spintronics, and quantum technology. However, realizing the full potential of these heterostructures requires interfaces with exceptionally low disorder which is challenging to engineer. Here, we show that thermal scanning probes can be used to create pristine interfaces in vdW heterostructures. Our approach is compatible at both the material- and device levels, and monolayer WS2 transistors show up to an order of magnitude improvement in electrical performance from this technique. We also demonstrate vdW heterostructures with low interface disorder enabling the electrical formation and control of quantum dots that can be tuned from macroscopic current flow to the single-electron tunneling regime.

Identification and Structural Characterization of Twisted Atomically Thin Bilayer Materials by Deep Learning

Two-dimensional materials are expected to play an important role in next-generation electronics and optoelectronic devices. Recently, twisted bilayer graphene and transition metal dichalcogenides have attracted significant attention due to their unique physical properties and potential applications. In this study, we describe the use of optical microscopy to collect the color space of chemical vapor deposition (CVD) of molybdenum disulfide (MoS2) and the application of a semantic segmentation convolutional neural network (CNN) to accurately and rapidly identify thicknesses of MoS2 flakes. A second CNN model is trained to provide precise predictions on the twist angle of CVD-grown bilayer flakes. This model harnessed a data set comprising over 10,000 synthetic images, encompassing geometries spanning from hexagonal to triangular shapes. Subsequent validation of the deep learning predictions on twist angles was executed through the second harmonic generation and Raman spectroscopy. Our results introduce a scalable methodology for automated inspection of twisted atomically thin CVD-grown bilayers.