Search

In this paper, we report on X-ray transition-edge sensor (TES) microcalorimeters optimized to have the best possible energy resolution for a limited energy range for the incoming X-rays, such as an energy resolution of 0.3 eV full width half maximum (FWHM) for energies up to $$\approx 0.8\,{\mathrm{keV}}$$ as is desirable for one of the Lynx X-ray Microcalorimeter subarrays. The test array we have fabricated has $$60\times 60$$ sensors on a pitch of $$50\,\upmu {\mathrm{m}}$$, and has $$46\times 46\,\upmu {\mathrm{m}}^2$$ absorbers that are one micrometer thick. We have measured a spectral energy resolution of the same device using 3 eV photons delivered through an optical fiber. For the one-photon 3 eV line, we have obtained an energy resolution of 0.25 eV FWHM, which is consistent with the estimated performance based on the signal size and noise. Further measurements will determine how the energy resolution degrades with energy. Based upon measurements of the TES transition characteristics, it appears that this level of energy resolution should be achievable up to 0.5 keV, and the performance will then gradually degrade to the measured energy resolution of around 2.3 eV at 1.5 keV. In this paper, we describe the full design and characterization of this detector, and discuss the performance limits of pixels designs like this.

The energy range of transition-edge sensor (TES) X-ray microcalorimeters with a multiplexed readout depends upon the width and shape of the TES superconducting transition, and also on the dynamic range of the readout. In many detector systems, the multiplexed readout slew rate capability will be the limiting factor for the energy range. In these cases, if we are willing to accept some energy resolution degradation, we can significantly extend the energy range by increasing the bath temperature of operation, essentially creating a second “extended energy range” mode of operation. For example, if we require the very highest energy resolution up to 7 keV, and wish to optimize the design up to this energy, for some measurements it could be very beneficial to have a mode where we can extend the energy range to 15–20 keV even if some energy resolution is sacrificed. In this paper, we explore the trade-off between dynamic range and energy resolution from changing the bath temperature of the TES. We present measurements of TES resolution and slew rate as a function of bath temperature and compare to numerical simulations.

We report on the development of multi-absorber transition edge sensors (TESs), referred to as hydras. A hydra consists of multiple x-ray absorbers each with a different thermal conductance to a TES. Position information is encoded in the pulse shape. With some trade-off in performance, hydras enable very large format arrays without the prohibitive increase in bias and read-out components associated with arrays of individual TESs. Hydras are under development for the next generation of space telescope such as Lynx. Lynx is a NASA concept under study that will combine a < 1 arcsecond angular resolution optic with 100,000-pixel microcalorimeter array with energy resolution of deltaE_FWHM ~ 3 eV in the soft x-ray energy range. We present first results from hydras with 25-pixels for Lynx. Designs with absorbers on a 25 micron and 50 micron pitch are studied. Arrays incorporate, for the first time, microstrip buried wiring layers of suitable pitch and density required to readout a full-scale Lynx array. The resolution from the coadded energy histogram including all 25-pixels was deltaE_FWHM = 1.66+/-0.02 eV and 3.34+/-0.06 eV at an energy of 1.5 keV for the 25 micron and 50 micron absorber designs respectively. Position discrimination is demonstrated from parameterization of the rise-time., 7 pages, 5 figures. Low Temperature Detectors 18 Conference, Milan Italy

The measured noise in Mo/Au transition-edge sensor (TES) microcalorimeters produced at NASA has recently been shown to be well described by a two-body electro-thermal model with a finite thermal conductance between the X-ray absorber and the TES. In this article, we present observations of a high-frequency peak in the measured current noise in some of these devices. The peak is associated with an oscillatory component of the TES response that is not predicted in a single-body model but can be qualitatively described by the two-body model.

With the improving energy resolution of transitionedge sensor (TES) based microcalorimeters, performance verification and calibration of these detectors has become increasingly challenging, especially in the energy range below 1 keV where fluorescent atomic X-ray lines have linewidths that are wider than the detector energy resolution and require impractically high statistics to determine the gain and deconvolve the instrumental profile. Better behaved calibration sources such as grating monochromators are too cumbersome for space missions and are difficult to use in the lab. As an alternative, we are exploring the use of pulses of 3 eV optical photons delivered by an optical fiber to generate combs of known energies with known arrival times. Here, we discuss initial results of this technique obtained with 2 eV and 0.7 eV resolution X-ray microcalorimeters. With the 2 eV detector, we have achieved photon number resolution for pulses with mean photon number up to 133 (corresponding to 0.4 keV).

We are developing kilo-pixel arrays of transition-edge sensors (TESs) for the X-ray Integral Field Unit on ESA’s Athena observatory. Previous measurements of AC-biased Mo/Au TESs have highlighted a frequency-dependent loss mechanism that results in broader transitions and worse spectral performance compared to the same devices measured under DC bias. In order to better understand the nature of this loss, we are now studying TES pixels in different geometric configurations. We present measurements on devices of different sizes and with different metal features used for noise mitigation and X-ray absorption. Our results show how the loss mechanism is strongly dependent upon the amount of metal in close proximity to the sensor and can be attributed to induced eddy current coupling to these features. We present a finite element model that successfully reproduces the magnitude and geometry dependence of the losses. Using this model, we present mitigation strategies that should reduce the losses to an acceptable level.

We performed small-scale demonstrations at GSFC of high-resolution X-ray TES microcalorimeters read out using a microwave SQUID multiplexer. This work is part of our effort to develop detector and readout technologies for future space-based X-ray instruments such as the microcalorimeter spectrometer envisaged for Lynx, a large mission concept under development for the Astro 2020 Decadal Survey. In this paper we describe our experiment, including details of a recently designed, microwave-optimized low-temperature setup that is thermally anchored to the 55 mK stage of our laboratory ADR. Using a ROACH2 FPGA at room temperature, we read out pixels of a GSFC-built detector array via a NIST-built multiplexer chip with Nb coplanar waveguide resonators coupled to rf-SQUIDs. The resonators are spaced 6 MHz apart (at $$\sim $$ 5.9 GHz) and have quality factors of $$\sim $$ 15,000. In our initial demonstration, we used flux-ramp modulation frequencies of 125 kHz to read out 5 pixels simultaneously and achieved spectral resolutions of 2.8–3.1 eV FWHM at 5.9 keV. Our subsequent work is ongoing: to-date we have achieved a median spectral resolution of 3.4 eV FWHM at 5.9 keV while reading out 28 pixels simultaneously with flux-ramp frequencies of 160 kHz. We present the measured system-level noise and maximum slew rates and briefly describe our future development work.

In transition-edge sensors (TESs), the addition of normal metal stripes on top of the superconducting bilayer, perpendicular to the current direction, is known to globally alter the sensitivity of the resistance R to changes in temperature T and current I. Here, we describe measurements of the dependence of the TES current on magnetic field B, bath temperature and voltage bias in devices with various numbers of stripes. We show that the normal metal features have a profound effect on the appearance of localized regions of very large $$(T/R)\,\hbox {d}R/\hbox {d}T$$ . We associate this with changes in the current distribution and corresponding changes in the oscillatory pattern of I(B). $$140~\upmu \hbox {m}$$ TESs with no stripes are found to have a relatively smooth resistive transition and sufficiently low noise that the measured energy resolution is 1.6 eV for X-rays of 1.5 keV. The predicted energy resolution at 6 keV is better than 2 eV, once the heat capacity is optimized for these higher energies.

We have specialized astronomical applications for X-ray microcalorimeters with superconducting transition edge sensors (TESs) that require exceptionally good TES performance, but which operate in the small-signal regime. We have therefore begun a program to carefully characterize the entire transition surface of TESs with and without the usual zebra stripes to see if there are reproducible local “sweet spots” where the performance is much better than average. These measurements require precise knowledge of the circuit parameters. Here, we show how the Shapiro effect can be used to precisely calibrate the value of the shunt resistor. We are also investigating the effects of stress and external magnetic fields to better understand reproducibility problems.

The traditional method of applying a digital optimal filter to measure X-ray pulses from transition-edge sensor (TES) devices does not achieve the best energy resolution when the signals have a highly non-linear response to energy, or the noise is non-stationary during the pulse. We present an implementation of a method to analyze X-ray data from TESs, which is based upon principal component analysis (PCA). Our method separates the X-ray signal pulse into orthogonal components that have the largest variance. We typically recover pulse height, arrival time, differences in pulse shape, and the variation of pulse height with detector temperature. These components can then be combined to form a representation of pulse energy. An added value of this method is that by reporting information on more descriptive parameters (as opposed to a single number representing energy), we generate a much more complete picture of the pulse received. Here we report on progress in developing this technique for future implementation on X-ray telescopes. We used an 55Fe source to characterize Mo/Au TESs. On the same dataset, the PCA method recovers a spectral resolution that is better by a factor of two than achievable with digital optimal filters.

Arrays of transition-edge sensors (TESs) X-ray microcalorimeters can provide high energy resolution and a large area necessary for the future X-ray mission Athena (2028~). An array with 4000 TESs will be employed on the X-ray satellite for the first time. The detector can achieve energy resolution of 2.5-3.0 eV and a high quantum efficiency in the energy range of 0.3-12 keV. Multiplexing the readout is necessary to minimize the number of readout amplifiers, the amount of wiring in the cryostat, and the cooling power required at the base-temperature. We are developing frequency-domain multiplexing (FDM) readout of TES microcalorimeters. In the FDM configuration, the TES is ac voltage biased at well-defined frequencies (between 2 and 6 MHz). For the development of the readout of the Athena instrument, we are using a nearly quantum-limited high-dynamic-range two-stage superconducting quantum interference device amplifier from VTT and high-Q lithographical LC resonators. In this paper, we will present the current status of the development of FDM readout of TES microcalorimeters. We will report on the results obtained using arrays fabricated at the NASA/Goddard Space Flight Center group.

We are developing kilo-pixel arrays of small-pitch transition-edge sensors for high spectral-resolving, high count-rate applications in astrophysics and solar physics measurements. We have fabricated and tested pixels that are $$65\,\upmu $$ m in size on a silicon substrate with an X-ray flux of $${\sim }100$$ counts per second (cps) per pixel. The X-ray pulses were recorded and analyzed in various ways to obtain high throughput with good energy resolution. We have demonstrated 2.3 eV FWHM resolution with 99.6 % throughput for a 6-keV X-ray flux of 100 cps.

Mo/Au transition-edge sensors exhibit weak-link behavior in the measured temperature, $$T$$ and field, $$B$$ dependence of the critical current $$I_\mathrm{{C}}(T,B)$$ . This is a consequence of the longitudinal proximitization between the Nb electrical bias contacts and the bilayer. Understanding how weak-link superconductivity impacts the resistive transition and the detector energy resolution is of great interest. In this contribution we present studies of $$I_\mathrm{{C}}(T,B)$$ for three devices that have different geometries of metallic depositions on top of the sensor used for noise mitigation and X-ray absorption. Results show that these features change the measured $$I_\mathrm{{C}}(T,B)$$ compared to the previously seen measurements on devices without additional deposition layers. Measurements of the small signal transition parameters $$\alpha $$ and $$\beta $$ also reveal differences between designs that impact the measured response to X-rays and energy resolution.

The focal plane of the X-ray integral field unit (X-IFU) for ESA's Athena X-ray observatory will consist of approximately 4000 transition edge sensor (TES) x-ray microcalorimeters optimized for the energy range of 0.2 to 12 kiloelectronvolts. The instrument will provide unprecedented spectral resolution of approximately 2.5 electronvolts at energies of up to 7 kiloelectronvolts and will accommodate photon fluxes of 1 milliCrab (90 counts per second) for point source observations. The baseline configuration is a uniform large pixel array (LPA) of 4.28 arcseconds pixels that is read out using frequency domain multiplexing (FDM). However, an alternative configuration under study incorporates an 18 by × 18 small pixel array (SPA) of 2 arcseconds pixels in the central approximately 36 arcseconds region. This hybrid array configuration could be designed to accommodate higher fluxes of up to 10 milliCrabs (900 counts per second) or alternately for improved spectral performance (less than 1.5 electronvolts) at low count-rates. In this paper we report on the TES pixel designs that are being optimized to meet these proposed LPA and SPA configurations. In particular we describe details of how important TES parameters are chosen to meet the specific mission criteria such as energy resolution, count-rate and quantum efficiency, and highlight performance trade-offs between designs. The basis of the pixel parameter selection is discussed in the context of existing TES arrays that are being developed for solar and x-ray astronomy applications. We describe the latest results on DC biased diagnostic arrays as well as large format kilo-pixel arrays and discuss the technical challenges associated with integrating different array types on to a single detector die.

Athena is a European X-ray observatory, scheduled for launch in $$\sim $$ 2028. Athena will employ a high-resolution imaging spectrometer called X-ray integral field unit (X-IFU), consisting of an array of $$\sim $$ 4000 transition edge sensor (TES) microcalorimeter pixels. For the readout of X-IFU, we are developing frequency domain multiplexing, which is the baseline readout system. In this paper, we report on the performance of a TES X-ray calorimeter array fabricated at Goddard Space Flight Center (GSFC) at MHz frequencies for the baseline of X-IFU detector. During single-pixel AC bias characterization, we measured X-ray energy resolutions (at 6 keV) of about 2.9 eV at both 2.3 and 3.7 MHz. Furthermore, in the multiplexing mode, we measured X-ray energy resolutions of about 2.9 eV at 1.3 and 1.7 MHz.

Weak link behavior in transition-edge sensor (TES) microcalorimeters creates the need for a more careful characterization of a device’s thermal characteristics through its transition. This is particularly true for small TESs where a small change in the bias current results in large changes in effective transition temperature. To correctly interpret measurements, especially complex impedance, it is crucial to know the temperature-dependent thermal conductance, G(T), and heat capacity, C(T), at each point through the transition. We present data illustrating these effects and discuss how we overcome the challenges that are present in accurately determining G and T from I–V curves. We also show how these weak link effects vary with TES size. Additionally, we use this improved understanding of G(T) to determine that, for these TES microcalorimeters, Kaptiza boundary resistance dominates the G of devices with absorbers while the electron-phonon coupling also needs to be considered when determining G for devices without absorbers

Recent measurements of micron-sized Mo/Au bilayer TESs have demonstrated that the TES can behave like an S-S′-S weak link due to the lateral proximity effect from superconducting leads. In this regime the Tc is a function of bias current, and the effective Tc shifts from the bilayer Tc towards the lead Tc. We explore the idea that a micron-sized S-N-S weak link could provide a new method to engineer the TES Tc. This method would be particularly useful when small size requirements for a bilayer TES (such as for a hot-electron microbolometer) lead to undesirable shifts in the bilayer Tc. We present measurements of a variety of micron-sized normal Au ‘TES’ devices with Nb leads. We find no evidence of a superconducting transition in the Au film of these devices, in dramatic contrast to the strong lateral proximity effect seen in micron-sized Mo/Au bilayer devices. The absence of a transition in these devices is also in disagreement with theoretical predictions for S-N-S weak links. We hypothesize that a finite contact resistance between the Nb and Au may be weakening the effect. We conclude that the use of the lateral proximity effect to create a superconducting transition will be difficult given current fabrication procedures.

We are developing small pitch transition-edge sensor (TES) X-ray detectors optimized for solar astronomy. These devices are fabricated on thick Si substrates with embedded Cu heat-sink layer. We use 35 x 35 square micrometers Mo/Au TESs with 4.5 micrometer thick Au absorbers. We have tested devices with different geometric absorber stem contact areas with the TES and surrounding substrate area. This allows us to investigate the loss of athermal phonons to the substrate. Results show a correlation between thc stem contact area and a broadening in the spectral line shape indicative of athermal phonon loss. When the contact area is minimized we have obtained exceptional broadband spectral resolution of 1.28 plus or minus 0.03 eV at an energy of 1.5 keV, 1.58 plus or minus 0.07 eV at 5.9 keV and 1.96 plus or minus 0.08 eV at 8 keV. The linearity in the measured gain scale is understood in the context of the longitudinal proximity effect from the electrical bias leads resulting in transition characteristics that are strongly dependent upon TES size.

Microcalorimeters onboard future x-ray observatories require an anti-coincidence detector to remove environmental backgrounds. In order to most effectively integrate this anti-coincidence detector with the main microcalorimeter array, both instruments should use similar read-out technology. The detectors used in the Cryogenic Dark Matter Search (CDMS) use a phonon measurement technique that is well suited for an anti-coincidence detector with a microcalorimeter array using SQUID readout. This technique works by using a transition-edge sensor (TES) connected to superconducting collection fins to measure the athermal phonon signal produced when an event occurs in the substrate crystal. Energy from the event propagates through the crystal to the superconducting collection fins, creating quasiparticles, which are then trapped as they enter the TES where they produce a signal. We are currently developing a prototype anti-coincidence detector for future x-ray missions and have recently fabricated test devices with Mo/Au TESs and Al collection fins. We present results from the first tests of these devices which indicate a proof of concept that quasiparticle trapping is occurring in these materials.

We have developed a process to integrate microstripline wiring with transition edge sensors (TES). The process includes additional layers for metal-etch stop and dielectric adhesion to enable recovery of parameters achieved in non-microstrip pixel designs. We report on device parameters in close-packed TES arrays achieved with the microstrip process including R(sub n), G, and T(sub c) uniformity. Further, we investigate limits of this method of producing high-density, microstrip wiring including critical current to determine the ultimate scalability of TES arrays with two layers of wiring.

We are developing kilo-pixel arrays of TES microcalorimeters to enable high-resolution X-ray imaging spectrometers for future X-ray observatories and laboratory astrophysics experiments. Our current array design was targeted as a prototype for the X-ray Microcalorimeter Spectrometer proposed for the International X-ray Observatory, which calls for a 40x40-pixel core array of 300 micron devices with 2.5 e V energy resolution (at 6 keV). Here we present device characterization of our 32x32 arrays, including X-ray spectral performance of individual pixels within the array. We present our results in light of the understanding that our Mo/Au TESs act as weak superconducting links, causing the TES critical current (Ic) and transition shape to oscillate with applied magnetic field (B). We show Ic(B) measurements and discuss the uniformity of these measurements across the array, as well as implications regarding the uniformity of device noise and response. In addition, we are working to reduce pixel-to-pixel electrical and thermal crosstalk; we present recent test results from an array that has microstrip wiring and an angle-evaporated Cu backside heatsinking layer, which provides Cu coverage on the four sidewalls of the silicon wells beneath each pixel.

Transition-edge sensor microcalorimeter arrays in compact geometries and large formats experience local heating from bias power and x-ray hits that must be dissipated in the frame. For devices on solid, non-perforated silicon substrates, we have introduced an underlying embedded copper heatsinking layer to enhance the ability of the frame to remove this heat. In particular, such a layer can mitigate thermal crosstalk between nearby pixels within the array. Further improvements in array performance, such as decreased magnetic field sensitivity and stray inductance, are possible by turning the heatsinking layer into a superconducting ground plane. In this presentation, we report on the development of heatsinking layers consisting of a 1-2 μm thick high-quality copper layer which is sandwiched between two thin refractory metal-based diffusion barriers. These diffusion barriers are designed to avoid copper migration into the surrounding material over time, especially during our high temperature TES fabrication process which takes place in excess of 400°C . A 0.3-0.5 μm thick PECVD SiO2 cover layer isolates the heatsinking layer from the detector circuit. We present first results on our attempt to tailor the materials forming the diffusion barrier to fabricate both well defined superconducting ground planes and non-superconducting layers with the desired barrier characteristics.

We report on the development of position-sensitive transition-edge sensors (PoST's) for future X-ray astronomy missions such as the International X-ray Observatory (IXO), under study by NASA and ESA. PoST's consist of multiple absorbers each with a different thermal coupling to one or more transition-edge sensors (TESs). This results in a characteristic pulse shape for each absorber element and allows position discrimination. PoST development is motivated by a desire to achieve maximum focal-plane area with the fewest number of readout channels. We report detailed characterization of our single TES PoST's or Hydras, which consist of four electroplated Au/Bi absorbers coupled to a low noise Mo/Au TES. Using a numerical model of the Hydra we fit to measured complex impedance curves and determine device parameters that allow us to accurately reproduce the measured pulse shapes and noise spectra. Results from Hydras with different internal thermal conductances reveal the trade-offs in optimizing for energy resolution or position-sensitivity. We report a best achievable energy resolution of < 6.0 eV across all pixels for a device with transition temperature of 86 mK, coupled with straightforward position discrimination by rise-time.

In a large close-packed array of x-ray microcalorimeters, sufficient heat sinking is important to minimize thermal crosstalk between pixels and to make the bath temperature of all the pixels uniform. We have measured crosstalk in our 8 times 8 pixel arrays. The shapes of the thermal crosstalk pulses are reproduced well as a convolution of heat input from the source pixel and the thermal decay in the receiver pixel. The amount of the thermal crosstalk is clearly dependent on the degree of electrothermal feedback. We have compared the magnitude of thermal crosstalk with and without a heat-sinking copper layer on the backside of the silicon frame as a function of distance between the source and receiver pixels. Using the results obtained, we have estimated the degradation of energy resolution that is expected as a function of count rate. We have also studied the temperature distribution within an array due to continuous heating from the TES bias to estimate impacts on the uniformity of the pixel performance.

We have successfully modeled our new TES (transition-edge sensor) X-ray microcalorimeters with a novel X-ray absorber design that is suitable for close-packed array with high quantum efficiency. We have determined device parameters that reproduce complex impedance curves and noise spectra throughout the transition. Observed pulse height, decay time and baseline energy resolution were in good agreement with simulated results using the same parameters.

We have studied crosstalk on our 8×8 transition-edge sensor (TES) x-ray microcalorimeter arrays. The shapes of thermal crosstalk pulses are well reproduced as a convolution of heat input from the source pixel and decay in the receiver pixel. We have measured the amount of thermal crosstalk as a function of distance between the source and receiver pixels. Using the results, we have estimated the degradation of energy resolution due to the thermal crosstalk as a function of count rate to determine the level of thermal crosstalk that is needed to satisfy the requirements of NASA’s Constellation-X mission.

We describe our current fabrication process for arrays of superconducting transition edge sensor microcalorimeters, which incorporates superconducting Mo/Au bilayers and micromachined silicon structures. We focus on materials and integration methods for array heatsinking with our bilayer and micromachining processes. The thin superconducting molybdenum bottom layer strongly influences the superconducting behavior and overall film characteristics of our molybdenum/gold transition-edge sensors (TES). Concurrent with our successful TES microcalorimeter array development, we have started to investigate the thin film properties of molybdenum monolayers within a given phase space of several important process parameters. The monolayers are sputtered or electron-beam deposited exclusively on LPCVD silicon nitride coated silicon wafers. In our current bilayer process, molybdenum is electron-beam deposited at high wafer temperatures in excess of 500 degrees C. Identifying process parameters that yield high quality bilayers at a significantly lower temperature will increase options for incorporating process-sensitive auxiliary array components (AAC) such as array heat sinking and electrical interconnects into our overall device process. We are currently developing two competing technical approaches for heat sinking large compact TES microcalorimeter arrays. Our efforts to improve array heat sinking and mitigate thermal cross-talk between pixels include copper backside deposition on completed device chips and copper-filled micro-trenches surface-machined into wafers. In addition, we fabricated prototypes of copper through-wafer microvias as a potential way to read out the arrays. We present an overview on the results of our molybdenum monolayer study and its implications concerning our device fabrication. We discuss the design, fabrication process, and recent test results of our AAC development.

We are developing arrays of superconducting transition-edge sensors (TES) for imaging X-ray spectroscopy telescopes such as the XMS on Constellation-X. While our primary focus has been on arrays that meet the XMS requirements (of which, foremost, are an energy resolution of 2.5 eV at 6 keV and a band pass from ∼0.3 keV to 12 keV), we are also beginning to investigate other optimizations that might be used to extend the XMS capabilities. In one of these optimizations, improved resolution below 1 keV is achieved by reducing the heat capacity. These low-energy pixels can be added to an array with broadband response either as a separate array or interspersed, depending on other factors that include telescope design and science requirements. To explore optimizations for higher count rates, we are also optimizing the design and operating temperature of pixels that are coupled to a solid substrate. We present analysis of the preliminary performance of such variations.

We are developing position-sensitive transition-edge sensors (PoSTs) for future X-ray astronomy missions such as NASAs Constellation-X. The PoST consists of 1 or more transitions-edge sensors (TESs) thermally connected to a large X-ray absorber, which through heat diffusion, gives rise to position dependence. The development of PoSTs is motivated by the desire to achieve the largest focal-plane coverage with the fewest number of readout channels. In this paper, we investigate optimized signal processing algorithms for single channel PoSTs or ‘Hydras’ (consisting of 4-absorbers connected to 1 TES). Using simulated data, we investigate the impact different parameters such as the thermal conductances and the inductance of the bias circuit have on device performance.

We discuss the development of an array of superconducting Transition-Edge Sensor (TES) bolometers intended for use at millimeter wavelengths. Each bolometer in the array uses a proximity-effect TES sensing element and each has integrated Normal-Insulator-Superconductor (NIS) refrigerators to cool the bolometer below the reference bath temperature. The NIS refrigerators and bolometers are fabricated on a single Silicon substrate. Four short, thin and narrow legs suspend the millimeter wavelength radiation-absorbing element. The leg geometry is used to control and optimize the thermal conductance of the bolometer. Using the technology developed at NIST, we fabricated NIS refrigerators at the base of each of the suspension legs. These NIS refrigerators remove hot electrons by quantum-mechanical tunneling and should be capable of reducing the temperature of the biased (∼10 pW) bolometers by more than 100 mK when operating from a local thermal reservoir provided by a 3He system at ∼300 mK. This lower temperature at the bolometer will allow the detectors to approach background-limited performance despite the simple cryogenic system. In this paper we describe the progress toward the development of these detectors designed for use in observations at the 100 m Green Bank Telescope at 3 mm wavelength and the results of laboratory tests of a prototype detector.

We are developing arrays of Mo/Au bilayer transition-edge sensors (TES’s) for future X-ray astronomy missions such as NASA’s Constellation-X. The physical properties of the superconducting-to-normal transition in our TES bilayers, while often reproducible and characterized, are not well understood. The addition of normal metal features on top of the bilayer are found to change the shape and temperature of the transition, and they typically reduce the unexplained ‘excess’ noise. In order to understand and potentially optimize the properties of the transition, we have been studying the temperature, widths and current dependence of these transitions. We report on the characterization of devices both deposited on silicon substrates and suspended on thin silicon nitride membranes. This includes key device parameters such as the logarithmic resistance sensitivity with temperature α, and the logarithmic resistance sensitivity with current β, and their correlation with excess noise.

Arrays of superconducting transition-edge sensors (TES) can provide high spatial and energy resolution necessary for x-ray astronomy. High quantum efficiency and uniformity of response can be achieved with a suitable absorber material, in which absorber x-ray stopping power, heat capacity, and thermal conductivity are relevant parameters. Here we compare these parameters for bismuth and gold. We have fabricated electroplated gold, electroplated gold/electroplated bismuth, and evaporated gold/evaporated bismuth 8x8 absorber arrays and find that a correlation exists between the residual resistance ratio (RRR) and thin film microstructure. This finding indicates that we can tailor absorber material conductivity via microstructure alteration, so as to permit absorber thermalization on timescales suitable for high energy resolution x-ray microcalorimetry. We show that by incorporating absorbers possessing large grain size, including electroplated gold and electroplated gold/electroplated bismuth, into our current Mo/Au TES, devices with tunable heat capacity and energy resolution of 2.3 eV (gold) and 2.1 eV (gold/bismuth) FWHM at 6 keV have been fabricated.

We report the latest results from our development of Position-Sensitive Transition-edge sensors (PoSTs), which are one-dimensional imaging spectrometers. In PoSTs with segmented Au absorbers, we obtained 8 eV energy resolution on K K α lines, which is consistent to the baseline energy resolution and the design values, on all of the nine pixels, by choosing the best combination of the thermal conductance in absorbers and in links that connects the absorbers. The pulse decay time of 193 μ s is fast enough for our purpose. In a PoST with a continuous Bi/Cu absorber, by dividing the events into 63 effective pixels, we obtained energy resolutions of 16 eV at the center “pixel”, which is comparable to the baseline energy resolution, and 33 eV at the outer “pixel”. The degradation of the energy resolution in the outer “pixel” is due to position dependence, which we can cancel out by dividing the events into smaller “pixels” when we have sufficient X-ray events.

Bismuth's large atomic number and low carrier density makes it an attractive X-ray absorber material for microcalorimeters. Bismuth's long fermi wavelength and long mean free paths have motivated much interest in the fabrication of high quality bismuth films to study quantum size effects. Despite such incentives, fabrication of high quality bismuth films has proven difficult, and measured properties of such films are highly variable in the literature. Implementing a bismuth deposition process for TES (superconducting Transition Edge Sensor) device fabrication presents additional challenges particularly at interfaces due to the inherent granularity and surface roughness of its films, its low melting point, and its tendency to diffuse and form undesired intermetallic phases. We report observations of Bi–Cu and Bi–Au diffusion in our devices correlating with large shifts in T c (superconducting transition temperature). Using SEM and in situ R vs T annealing experiments we have been able to study these diffusion processes and identify their activation temperatures.

An ideal microcalorimeter is characterized by a constant energy resolution across the sensor's dynamic range. Any dependence of pulse shape on the position within the absorber where an event occurs leads to a degradation in resolution that is linear with event's energy (excess broadening). In this paper we present a numerical simulation that was developed to model the variation in pulse shape with position based on the thermal conductivity within the absorber and between the absorber, sensor, and heat bath, for arbitrarily shaped absorbers and sensors. All the parameters required for the simulation can be measured from actual devices. We describe how the thermal conductivity of the absorber material is determined by comparing the results of this model with data taken from a position sensitive detector in which any position dependent effect is purposely emphasized by constructing a long, narrow absorber that is readout by sensors on both ends. Finally, we present the implications for excess broadening given the measured parameters of our X-ray microcalorimeters.

We present a geometrical model to describe excess electrical noise in transition-edge sensors (TESs). In this model, a network of fluctuating resistors represents the complex dynamics inside a TES. The fluctuations can cause several resistors in series to become superconducting. Such events short out part of the TES and generate noise because much of the current percolates through low resistance paths. The model predicts that excess white noise increases with decreasing TES bias resistance (R/RN) and that perpendicular zebra stripes reduce noise and alpha of the TES by reducing percolation.

We describe progress in the development of a close-packed array of bolometers designed for use in photometric applications at millimeter wavelengths from ground-based telescopes. Each bolometer in the array will use a proximity-effect Transition Edge Sensor (TES) sensing element and each will have integrated Normal-Insulator-Superconductor (NIS) refrigerators to cool the bolometer below the thermal reservoir temperature. The NIS refrigerators and acoustic-phonon-mode-isolated bolometers are fabricated on silicon. The radiation absorbing element is mechanically suspended by four legs, whose dimensions are used to control and optimize the thermal conductance of the bolometer. Using the technology developed at NIST, we fabricate NIS refrigerators at the base of each of the suspension legs. The NIS refrigerators remove hot electrons by quantum-mechanical tunneling and are expected to cool the biased (10 pW) bolometers to