Quantum computing and two dimensional van der Waals materials research have been two of the fastest growing fields of condensed matter physics research for the better part of the last two decades. In that time, advances in superconducting qubit design, materials and fabrication have improved their relaxation and coherence times by about 5 orders of magnitude. One of the key components that quantum devices such as qubits require are ultra low loss capacitance elements. Conventional parallel plate capacitors have been unable to fulfill this need due to bulk and inter-facial losses, necessitating the use of coplanar capacitors with extremely large footprints. In fact one of the driving forces behind increase coherence times has been the ever growing footprint of these coplanar capacitor pads, and the reduced electric field density and thus reduced surface losses that they provide. However, this style of capacitor creates a number of challenges when it comes to scaling the number of qubits in a system. First, the large geometric footprint of these pads limits the number of qubits that can be placed on a chip. Second, the dispersion of the electric field, above and below the plane of the capacitor pads can cause unwanted crosstalk between neighbouring qubits, again limiting the number of qubits that can be put on a chip without compromising coherence.
Since the isolation of a single atomic layer of graphene in 2004 and the ability to create heterostructures of a variety of two dimensional materials, the field of van der Waals materials research has exploded at a similar rate. Single crystals of van der Waals materials, can be grown with extremely low defect densities, and then be stacked to create heterostructures with ultra-clean laminated interfaces. This work explores how van der Waals materials may be used to create low loss parallel plate capacitors. The parallel plate geometry confines the electric field between the crystalline materials and low loss interfaces of a van der Waals heterostructure, limiting both losses at the surfaces as well as undesired cross talk between qubits. We begin by studying the microwave losses in hexagonal boron nitride (hBN). Next we report a method to make low loss microwave contacts to air sensitive superconducting van der Waals materials like niobium diselinde (NbSe₂). Finally, we demostrate coherence in a transmon where the primary shunt capacitor is an all van der Waals parallel plate capacitor, achieving a 1000× reduction in geometric footprint, when compared to a conventional coplanar capacitor.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/3cby-3v18 |
| Date | January 2022 |
| Creators | Antony, Abhinandan |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0119 seconds