Microwave Kinetic Inductance Detectors

Citation

Mazin, Benjamin A. (2005) Microwave Kinetic Inductance Detectors. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/GZ72-V784. https://resolver.caltech.edu/CaltechETD:etd-10042004-120707

Abstract

Low temperature detectors have been a subject of intense interest to the scientific community over the last decade. These detectors work at very low temperatures, often well below 1 Kelvin, to minimize the noise in the measurement of photons. This leads to very powerful detectors applicable to a broad wavelength range.

Since these detectors are so sensitive even single pixels and small arrays (up to several hundred pixels) enable deeper explorations of the cosmos than ever before. Instruments based on these technologies have been used at submillimeter, optical, and X-ray wavelengths. The scientific prospects for these detectors increase as they grow in pixel count. For some applications, especially for Cosmic Microwave Background (CMB) polarization work, a large focal plane will not only increase efficiency but will also enable new and vital science.

Current superconducting technologies, such as Transition Edge Sensors (TESs), can currently deliver extremely high sensitivity in the submillimeter and read-noise free imaging spectroscopy at Optical/UV and X-ray wavelengths, but the largest arrays contain less that 100 pixels. In order to make real progress these arrays must contain many thousands of pixels. This is a formidable technical challenge.

This thesis will explore a promising emerging technology called Microwave Kinetic Inductance Detectors (MKIDs). MKIDs make use of the change in the surface impedance of a superconductor as incoming photons break up Cooper pairs. This is accomplished by making the strip of superconductor part of a microwave resonant circuit, and monitoring the phase of a signal transmitted through (or past) the resonator. The primary advantage of this technology is that by using resonant circuits with high quality factors, passive frequency domain multiplexing will allow up to thousands of resonators to be read out through a single coaxial cable and a single HEMT amplifier. This eliminates the cryogenic electronics (SQUIDS) and wiring problems associated with current superconducting devices. Inexpensive and powerful room-temperature readout electronics can leverage the microwave integrated circuits developed for wireless communications.

When this project started four years ago MKIDs were just a concept. In this thesis I will recount the progress we have made in taking this concept and turning it into a detector technology, building on the work we published in Nature in 2003. I will demonstrate that we have overcome the major technical obstacles and shown conclusively that MKIDs work, and should provide much larger arrays than are possible with other technologies. Due to our work at Caltech and JPL, MKIDs are considered one of the leading technologies to reach the ambitious goals set for future ground and space missions. The noise performance is already good enough for sky-limited, ground-based, submillimeter astronomy.

Despite MKIDs acceptable noise performance for ground-based astronomy, there are many applications which require lower detector noise. The simple MKIDs we measure in this thesis still exhibit a noise higher than theory would predict which could limit their usefulness for some applications.

In Chapters 7-9 we perform several experiments which identify the source of the excess noise as the substrate.

In the course of these experiments to characterize and identify the noise, we successfully demonstrate two distinct approaches which dramatically reduce the noise excess.

Thesis Files