Niobium nitride in bulk form exhibits relatively high superconducting transition temperature (Tc), around 16 K. Therefore, it is intensively studied for applications such as qubits in quantum computing, infrared light detectors or superconducting radio frequency cavities in particle accelerators. The application fields can be further broadened by making the preparation of thin NbN films compatible with the CMOS process technology and using directly silicon wafers as a substrate; note that silicon is not inherently suitable for the growth of high-quality NbN due to the large lattice mismatch.
Our study aims to deposit thin films of NbN with sufficiently high Tc with a technique that is scalable to semiconductor industrial demands and has low thermal stress on the sample. In the first part of our study, we have identified stoichiometry and porosity as the most essential characteristics of the NbN thin films. They can be tuned by varying deposition parameters during the film preparation. Specifically, magnetron sputtering along with its ionized variant is shown to be a highly promising technical approach. In order to find the best deposition conditions, we have focused on increasing the density and conductivity of our 100-nm thick NbN films. These are found to be correlated to each other and significantly improved by using high-power impulse magnetron sputtering (HiPIMS) with the disadvantage of a somewhat lowered deposition rate.
In the second part of the study, the microstructure of the HiPIMS samples is characterized in detail and its influence on the superconducting transition temperature and width is analysed. The stoichiometry of the films, determined by the elastic recoil detection analysis in combination with X-ray reflectivity and X-ray diffraction, seems to be crucial for reaching high Tc. Also, decreasing internal stress and microstrain helps to increase Tc. On the contrary, the transition temperature is found not much dependent on the grain size.
Finally, our approach is found suitable for creating test circuits and device structures of dimensions on the order of hundreds of nanometers. These structures have been used in our critical current density measurements.
This project was funded within the QuantERA II Program through the Swedish Research Council (2021-06025) and the French National Research Agency (SUNISIDEuP – ANR-19-CE47-0010).