Transition-Edge Superconducting Antenna-Coupled Bolometer

Citation

Hunt, Cynthia Lee (2004) Transition-Edge Superconducting Antenna-Coupled Bolometer. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/RG6K-C898. https://resolver.caltech.edu/CaltechETD:etd-05282004-131005

Abstract

The temperature anisotropy of the cosmic microwave background (CMB) is now being probed with unprecedented accuracy and sky coverage by the Wilkinson Microwave Anisotropy Probe (WMAP), and will be definitively mapped by the Planck Surveyor after its launch in 2007. However, the polarization of the CMB will not be mapped with sufficient accuracy. In particular, the measurement of the curl-polarization, which may be used to probe the energy scale of the inflationary epoch, requires a large advance in the format of millimeter-wave bolometer arrays. SAMBA (Superconducting Antenna-coupled Multi frequency Bolometric Array) is being developed to address these needs for the next generation of submillimeter astronomical detectors. SAMBA consists of a focal plane populated with microstrip-coupled slot antennas, whose signals are coherently added and sent to transition-edge superconducting (TES) bolometers via microstrip lines. SAMBA eliminates the need for the feedhorns and optical filters currently used on CMB observational instruments, such as Planck and Boomerang. The SAMBA architecture allows for a high density of pixels in the focal plane with minimal sub-Kelvin mass. As a precursor to a full monolithic high-density antenna array, we are developing a single-band antenna-coupled bolometric detector.

In this thesis, I report test results for a single-pixel antenna-coupled bolometric detector. Our device consists of a dual slot microstrip-coupled slot antenna coupled to an Al/Ti/Au voltage-biased TES. The coupling architecture involves propagating the signal along superconducting microstrip lines and terminating the lines at a normal metal resistor collocated with a TES on a thermally isolated island. The device, which is inherently polarization sensitive, is optimized for 140 GHz measurements. In the thermal bandwidth of the TES, we measure a noise equivalent power (NEP) of 2.0x10^-17 W/√ Hz in dark tests which agrees with the calculated NEP including only contributions from phonon, Johnson and amplifier noise. We do not measure any excess noise above this expectation at frequencies between 1 and 200 Hz. We measure a thermal conductance G=55 pW/K. We measure a thermal time constant as low as 437 µs at 3 µV bias when stimulating the TES directly using a light emitting diode.

Thesis Files