Chiu Fan Bowen "Leo" Lo

Under what conditions does a superconductor inherit topologically protected nodes from its parent normal state? In the context of inter-Fermi-surface pairing in three-dimensional Weyl semimetals with broken time-reversal symmetry, the pairing order parameter is classified by monopole harmonics and is necessarily nodal [Li and Haldane, Phys. Rev. Lett. 120, 067003 (2018)]. Here, we show that a similar conclusion could also be drawn for 2D Dirac semimetals, although the conditions for the existence of nodes are more complex, depending on the pairing matrix structure in the valley and sublattice space. We analytically and numerically analyze the Bogoliubov-de Gennes quasiparticle spectra for Dirac systems based on the monolayer as well as twisted bilayer graphene. We find that in the cases of intravalley intrasublattice and intervalley intersublattice pairings, the point nodes in the BdG spectra (which are inherited from the Dirac cone in the normal state) are protected by a 1D winding number. The nodal structure of the superconductivity is confirmed numerically using tight-binding models of monolayer and twisted bilayer graphene. Notably, the BdG spectrum is nodal even with a momentum-independent “bare” pairing, which, however, acquires a momentum dependence and point nodes upon projection to the Bloch states on the topologically nontrivial Fermi surface, similar in spirit to the Li-Haldane monopole superconductor and the Fu-Kane proximity-induced superconductor on the surface of a topological insulator.

Over the last decade, scattering-type scanning near-field optical microscopy and spectroscopy have been widely used in nano-photonics and material research due to their fine spatial resolution and broad spectral range. A number of simplified analytical models have been proposed to quantitatively understand the tip-scattered near-field signal. However, a rigorous interpretation of the experimental results is still lacking at this stage. Numerical modelings, on the other hand, are mostly done by simulating the local electric field slightly above the sample surface, which only qualitatively represents the near-field signal rendered by the tip-sample interaction. In this work, we performed a more comprehensive numerical simulation which is based on realistic experimental parameters and signal extraction procedures. By directly comparing to the experiments as well as other simulation efforts, our methods offer a more accurate quantitative description of the near-field signal, paving the way for future studies of complex systems at the nanoscale.