Steven W. Deiker

Characterization and reduction of unexplained noise in superconducting transition-edge sensors

The noise in superconducting transition-edge sensors (TESs) commonly exceeds simple theoretical predictions. The reason for this discrepancy is presently unexplained. We have measured the amplitude and frequency dependence of the noise in TES sensors with eight different geometries. In addition, we have measured the dependence of the noise on operating resistance, perpendicular magnetic field, and bath temperature. We find that the unexplained noise contribution is inversely correlated with the temperature width of the superconducting-to-normal transition and is reduced by a perpendicular field and in certain geometries. These results suggest paths to improved sensor performance.

Microfabricated transition-edge X-ray detectors

We are developing high performance X-ray detectors based on superconducting transition-edge sensors (TES) for application in materials analysis and astronomy. Using our recently developed fully lithographic TES fabrication process, we have made devices with an energy resolution of 4.5/spl plusmn/0.1 eV for 5.9 keV X-rays, the best reported energy resolution for any energy dispersive detectors in this energy range. These detectors utilize micromachined thermal isolation structures and transition-edge sensors fabricated from Mo/Cu bilayers with normal-metal boundary conditions. We have found the normal-metal boundary conditions to be critical to stable and reproducible low noise operation. In this paper we present details of fabrication and performance of these devices.

Time-Division SQUID Multiplexers

SQUID multiplexers make it possible to build arrays of thousands of low-temperature bolometers and microcalorimeters based on superconducting transition-edge sensors with a manageable number of readout channels. We discuss the technical tradeoffs between proposed time-division multiplexer and frequency-division multiplexer schemes and motivate our choice of time division. Our first-generation SQUID multiplexer is now in use in an astronomical instrument. We describe our second-generation SQUID multiplexer, which is based on a new architecture that significantly reduces the dissipation of power at the first stage, allowing thousands of SQUIDs to be operated at the base temperature of a cryostat.

Time-division multiplexing of high-resolution x-ray microcalorimeters: Four pixels and beyond

We present experimental results from a four-pixel array of transition-edge-sensor, x-ray microcalorimeters read out through a single amplifier channel via a time-division superconducting quantum interference device multiplexer. We map the dependence of the x-ray energy resolution of the microcalorimeters on multiplexer timing parameters. We achieve multiplexed, four-pixel resolution of 6.94±0.05eV full width at half maximum of the MnKα complex near 5.9keV, which is a degradation of only 0.44eV from nonmultiplexed operation. An analysis of straightforward upgrades to the multiplexer predicts that a linear array of 32 of these pixels could be multiplexed with a degradation in resolution of only 0.1eV. These results, the first demonstration of a time-division multiplexer for x-ray detectors, establish a clear path to the instrumentation of a kilopixel microcalorimeter array.

Superconducting transition edge sensor using dilute AlMn alloys

We have fabricated a bolometer using a transition-edge sensor (TES) made of Al doped with Mn to suppress the superconducting critical temperature (Tc) of Al from ∼1Kto∼100mK. The resulting detector exhibits low-frequency noise consistent with theory, with a noise-equivalent power of 7.5×10−18W∕√Hz. The addition of Mn impurities did not significantly increase the heat capacity of the TES. In addition, the detector is surprisingly insensitive to applied magnetic fields. The use of AlMn alloy films in arrays of TES detectors has advantages in simplicity of fabrication when compared to traditional bilayer fabrication techniques.

Dilute Al-Mn Alloys for Low-Temperature Device Applications*

We discuss results on the superconducting, electron-transport, and tunneling properties of Al doped with Mn in the range of 1000 to 3000 ppm. We demonstrate that the critical temperature of Al can be systematically reduced to below 50 mK. Films are prepared by sputter deposition, and show values of d(ln R)/d(ln T) of ∼500, indicating sharp superconducting transitions. Al-Mn/I/Al-Mn tunnel junctions show low sub-gap conductance and BCS-like characteristics. Our results in general suggest that the material is of interest for transition-edge sensors operating in the 100 mK regime and superconductor/insulator/superconductor (S/I/S) and normal/insulator/ superconductor (N/I/S) devices, in the latter case where heavily doped Al-Mn can replace the normal metal.

Toward a 2-eV microcalorimeter x-ray spectrometer for Constellation-X

COnstellation-X is a cluster of identical observatories that together constitute a promising concept for a next- generation, high-throughput, high-resolution, astrophysical x-ray spectroscopy mission. The heart of the Constellation-X mission concept is a high-quantum-efficiency imaging spectrometer with 2 eV resolution at 6 keV. Collectively across the cluster, this imaging spectrometer will have twenty times the collecting efficiency of XRS on Astro-E and better than 0.25 arc minute imaging resolution. The spectrometer on each satellite will be able to handle count rates of up to 1000 counts per second per imaging pixel for a point source and 30 counts per second per pixel for an extended source filling the array. Focal plane coverage of at least 2.5 arc minutes X arc minutes, comparable to XRS but with a factor of thirty more pixels, is required. This paper will present the technologies that have the potential to meet al these requirements. It will identify the ones chosen for development for Constellation-X and explain why those were considered closer to realization, and it will summarize the results of the development work thus far.

TES detector noise limited readout using SQUID multiplexers

The availability of superconducting Transition Edge Sensors (TES) with large numbers of individual detector pixels requires multiplexers for efficient readout. The use of multiplexers reduces the number of wires needed between the cryogenic electronics and the room temperature electronics and cuts the number of required cryogenic amplifiers. We are using an 8 channel SQUID multiplexer to read out one-dimensional TES arrays which are used for submillimeter astronomical observations. We present results from test measurements which show that the low noise level of the SQUID multiplexers allows accurate measurements of the TES Johnson noise, and that in operation, the readout noise is dominated by the detector noise. Multiplexers for large number of channels require a large bandwidth for the multiplexed readout signal. We discuss the resulting implications for the noise performance of these multiplexers which will be used for the readout of two dimensional TES arrays in next generation instruments.

Surface micromachining for transition-edge detectors

We are developing arrays of high-performance detectors based on superconducting transition-edge sensors (TES) for application in x-ray materials analysis as well as x-ray and sub-millimeter astronomy. In order to obtain the desired thermal time constants, as well as to provide thermal isolation from adjacent pixels, these arrays utilize micromachined thermal-isolation structures. Until recently, we have achieved thermal isolation of single-pixel devices by anisotropic wet etching of the entire Si wafer behind the pixel, leaving the detector supported by a thin Si/sub 3/N/sub 4/ membrane. Limitations of this technique make it undesirable for the fabrication of close-packed arrays. One possible means to achieve thermal isolation of close-packed arrays is surface micromachining. Here, a TES is fabricated on top of a Si/sub 3/N/sub 4/ membrane that is held above the substrate by a small number of support legs. Because the underlying wafer is not thinned or removed, the resulting detector chip is strong and requires no special handling. In this paper we describe the fabrication processes and present preliminary data on the properties of 64-pixel arrays of surface-micromachined TES x-ray detectors.

Progress on the Micro-X rocket payload

The Micro-X High Resolution Microcalorimeter X-ray Imaging Rocket is sounding rocket experiment that will combine a transition-edge-sensor X-ray-microcalorimeter array with a conical imaging mirror to obtain high-spectral-resolution images of extended and point X-ray sources. Our first target is the Puppis A supernova remnant, which will be observed in January 2011. The Micro-X observation of the bright eastern knot of Puppis A will obtain a line-dominated spectrum with up to 90,000 counts collected in 300 seconds at 2 eV resolution across the 0.3-2.5 keV band. Micro-X will utilize plasma diagnostics to determine the thermodynamic and ionization state of the plasma, to search for line shifts and broadening associated with dynamical processes, and seek evidence of ejecta enhancement. We describe the progress made in developing this payload, including the detector, cryogenics, and electronics assemblies. A detailed modeling effort has been undertaken to design a rocket-bourne adiabatic demagnetization refrigerator with sufficient magnetic shielding to allow stable operation of transition edge sensors, and the associated rocket electronics have been prototyped and tested.