Scanning tunneling microscope study of the superconductor-insulator transition in molybdenum nitride

Abstract

The superconductor-insulator transition (SIT) in films provides a versatile platform to explore critical behavior in quantum phase transitions, thus, it has been an active topic for decades. Large amount of research captured this transition in a wealth of material systems and concomitant novel phenomena. Simultaneously, different theoretical pictures are employed to explain these experimental observations. Regrettably, consensus about the underlying mechanism is not yet reached. Regarding the disputable nature of the insulating state in the superconductor-insulator transition, insights with the microscopic processes of the transition is demanded. Thus we attempt to adopt scanning tunneling spectroscopy combined with shot noise and scanning tunneling microscopy measurements to locally study the superconductor-insulator transition at microscopic scale.This work starts with the summary of the research status of SIT which involves representative experimental observation including the spatial inhomogeneity of super- conducting state revealed by STS and latest intermediate metallic state, together with theoretical scenarios that trigger our research interest. The following chapter 2, presents superconductor-insulator transition related necessary background including BCS theory, Josephson effects, magnetic field related BKT transition. Beyond that, fundamental thermal noise, flicker noise and shot noise widely existing in mesoscopic systems are presented.Then chapter 3 presents working principles of our experimental techniques, especially the design of electronics which provide the possibility to detect differential conductance and shot noise simultaneously. Chapter 4 firstly gives general properties of molybdenum nitride (MoN) films we are going to study. Previous research mainly focused on MoN’s excellent mechanical and catalytic properties thus the superconductivity is not widely studied. All films are fabricated by atomic layer deposition with accurate control of thickness achieved by cycle numbers during growth. Preliminary characterization reveal superconductor-insulator transition by evolution of resistance v.s. temperature and find the critical thickness of MoN film fabricated at 400◦C is between 5.5 nm and 7 nm. Before cooling down, three-step ultrasonic cleaning is performed to remove the protecting layer of the samples.Finally in chapter 5, we address typical results and corresponding discussion of three aspects of measurements. A good deal of topography images with different scanning ranges visualize sample surface in nanoscale. Disregarding different thickness and fabrication temperature, all samples are polycrystalline with grains size in tens of nanometers and surface roughness about 3-7 nm. And the local tunneling density of states (LDOS) obtained at 300 mK demonstrates disorder-induced spatial inhomogeneities of the superconducting gap in every single sample. An overview of experimental and BCS gap versus film thickness shows the suppression of the superconducting gap upon de- 84creasing film thickness. At last, shot noise results of four samples are demonstrated. The Fano factor varies from 1 to 3 at different detecting spots. In context with STS spectra at 4.2 K, the variation of Fano factor is probably related to the spatial inhomogeneity of LDOS. In areas still maintain superconductivity at 4.2 K, the electrons correlate and tunnel as Cooper pairs that induce the twofold shot noise compared to Poisson noise. Further experiment is needed to verify this possibility.