528 IEEE TRANSACTIONS ON APPLIED SUPERCONDUCTIVITY, VOL. 19, NO. 3, JUNE 2009 Design of a Transition-Edge Hot-Electron Microbolometer for Millimeter-Wave Astrophysical Observations Emily M. Barrentine, Shafinaz Ali, Christine A. Allen, Ari D. Brown, Nga T. Cao, James A. Chervenak, Kevin L. Denis, Wen-Ting Hsieh, Timothy M. Miller, S. Harvey Moseley, Thomas R. Stevenson, Peter T. Timbie, Douglas E. Travers, Kongpop U-Yen, and Edward J. Wollack Abstract—We are developing a Transition-edge Hot-electron Microbolometer (THM) with the capacity to make sensitive and broadband astrophysical observations over frequencies ranging from 30–300 GHz (10-1 mm). This micron-sized bolometer con- sists of a superconducting bilayer Transition-Edge Sensor (TES) and a thin-film absorber. The THM employs the decoupling between electrons and phonons at low temperatures (below 300 mK) to provide thermal isolation. The devices are fabricated pho- tolithographically and read out with Superconducting Quantum Interference Devices (SQUIDs). We present the details of a thermal model for a THM detector and the design for new thermally opti- mized antenna-coupled THMs for illumination by a RF source at 40 and 100 GHz. Index Terms—Bolometers, hot-electron, millimeter wave detec- tors, superconducting sensors. I. INTRODUCTION A STROPHYSICAL measurements at millimeter wave- lengths now require arrays of photon-noise-limited detectors for both broad-band and narrow-band applications. In particular, future measurements of the Cosmic Microwave Background (CMB), including the observation of B-mode polarization to detect the imprint of gravitational waves from inflation, will require an order of magnitude increase in CMB detector sensitivity [1]. Presently, individual CMB detectors have reached photon-noise-limited levels, and further advances will require arrays of 1000 s of detectors. Low noise arrays are also required for millimeter-wave spectroscopy [2]. One of the most promising technologies for application to millimeter-wave detector arrays are cryogenic TES bolometric detectors read out by low power, low noise, multiplexed SQUIDs [3]. These TES detectors make use of the sharp transitional properties of Manuscript received August 25, 2008. First published June 30, 2009; current version published July 10, 2009. This work was supported in part by the NASA Graduate Student Researcher Program. E. M. Barrentine and P. T. Timbie are with the University of Wisconsin- Madison, Madison, WI 53706 USA (e-mail: barrentine@wisc.edu). S. Ali is with Merritt College, Oakland, CA 94619 USA. C. A. Allen, A. D. Brown, N. T. Cao, J. A. Chervenak, K. L. Denis, W. T. Hsieh, T. M. Miller, S. H. Moseley, T. R. Stevenson, D. E. Travers, K. U-Yen and E. J. Wollack are with the NASA-Goddard Space Flight Center, Greenbelt, MD 20771 USA. Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TASC.2009.2017956 superconductors to measure the temperature dependence of an RF absorbing bolometer. They can be voltage-biased to keep the TES in the sensitive transition region via electrothermal feedback due to Joule power dissipation in the TES [4], [5]. Bolometric detectors directly absorb incident radiation thermally. An important component to the operation of these bolometers is the thermal link between the detector and a cold bath. Common TES bolometers make use of micro-machined isolation structures, first introduced by Downey et al., [6], to precisely control the thermal conductance of the link. This thermal conductance affects the noise, time response, saturation level and bias point of the detector. These membrane structures are fragile and present both fabrication and design complexities, especially when the technique is extended to large arrays. The Transition-Edge Hot-Electron Microbolometer (THM) makes use of a different type of thermal isolation, one that is controlled by the weak coupling between electrons and phonons in the detector at low temperatures. This hot-electron design is similar to a design by Wei et al. [7], but contains a sep- arate absorbing structure. The advantages of this thermal de- sign are easy and robust fabrication, a small cross-sectional area for cosmic rays and close packing into the focal plane, a short thermal time constant, low thermal conductance, and separate impedance matching of the absorber to the transmission lines. II. THERMAL MODEL The basic design of the THM consists of a metal Bi absorber overlapping a superconducting bilayer Au/Mo TES. The ab- sorber terminates a Nb superconducting microstrip transmission line. The detector is operated at milliKelvin temperatures to in- crease sensitivity. The heat flow within the detector is controlled by elec- tron-electron scattering between the electrons in the TES and absorber and electron-phonon scattering between the electrons and the phonons in the detector. Andreev reflection of the electrons in the detector at the superconducting Nb transmis- sion lines keeps heat from dissipating out the leads [8]. The Kapitza boundary resistance between the detector and substrate phonons is minimal due to the thinness of the metal film relative to phonon wavelength [9]. A thermal model of the detector is show in Fig. 1. The thermal conductance for electron-electron scattering fol- lows [10]. Here R is the electrical resis- tance of the detector, is the Lorentz constant, where 1051-8223/$25.00 © 2009 IEEE Authorized licensed use limited to: University of Wisconsin. Downloaded on December 16, 2009 at 16:35 from IEEE Xplore. Restrictions apply.