X-ray Microcalorimeter Spectrometer (XMS)

The operational principle of a microcalorimeter is depicted in the figure to the right. The general principle is to measure the temperature rise of the calorimeter when an x-ray photon is absorbed. When an x-ray is absorbed, the temperature in the absorber increases by an amount E/C, where E is the x-ray energy, and then recovers back to the steady state temperature. The thermal conductance of the weak link is engineered such that enough time is allowed for proper thermalization of the photon energy in the device, but fast enough such that the device recovers back to its base temperature in time to absorb the next x-ray.

A thermometer is integrated into the device to measure the temperature rise. To ensure high spectral resolution, the thermometer must be very well coupled to the absorber, and should be extremely sensitive to temperature changes. For this Gen-X, there are currently two candidate technologies for the thermometer: transition-edge sensors (TESs) and magnetic microcalorimeters (MMCs). In a TES the heat deposited by an individual X-ray photon changes the impedance of a superconducting bilayer held just below its transition temperature. In an MMC the heat from an X-ray photon changes the magnetization of a paramagnetic film. Both of these devices are read out using SQUID amplifiers. Currently TES microcalorimeters have provided the best spectral resolution that has been achieved (1.8 eV FWHM at 6 keV; Bandler et al., 2007), in arrays of pixels such as the one shown in the photograph to the right.

Microcalorimeters are usually designed to thermalize as quickly as possible to avoid degradation in energy resolution from position dependence to the pulse shapes. Each pixel consists of an absorber and thermometer, both decoupled from the cold bath through a weak thermal link. Each pixel requires a separate SQUID readout channel. For Gen-X, where we require millions of resolution elements, having an individual SQUID readout channel for each pixel becomes difficult. Furthermore, at a focal length of 50 m and an impressive 0.1" point-spread function, the Gen-X mirror requires small pixels at the focal plane. A Gen-X focal plane with 0.1" pixels would have pixels that are 30 um to a side. For comparison, the IXO X-ray Microcalorimeter Spectrometer uses 300 um pixels. Thus the area of Gen-X pixels is 100 times smaller than those for IXO. Fabricating microcalorimeter pixels that are two orders of magnitude smaller than the current state of the art is a technological challenge.

a) Schematic diagram of the 4-abosorber sensor design. b) Average pulse shapes for 5.89 keV photons absorbed in each of the four absorbers.

Both of these challenges (number of SQUID readout channels and pixel size), can in principle be reduced by an order of magnitude using position-sensitive microcalorimeters (Smith et al, 2008). These devices typically use one thermometer coupled to a number of different absorbers through thermal conductances that are engineered to give position dependence in the pulse shape. In the figure above, a single TES sensor is connected to four different absorbers. When x-rays are absorbed into different absorbers, the TES produces different pulse shapes depending on the absorption location. Thus each TES read-out element can be subdivided into several resolution elements. For Gen-X, we may have sensors with over ten absorbers attached to each sensor.

The most significant technical challenge in the development of microcalorimeters for Gen-X is to raise the pixel count from the current state of the art (a few dozen pixels), beyond the number needed for the International X-ray Observatory (a few thousand pixels) to a number in the range of a few million pixels. Very low-noise multiplexing is required if a reasonable number of signal paths are to carry the signals from every pixel in the cryogenic focal plane to the processing electronics. Currently the use of time-division multiplexing is a promising approach for arrays of medium size, such as IXO. The ability to scale this approach to much larger arrays, such as is needed for Gen-X, is limited by both power dissipation and the bandwidth of SQUID read-out approach being used. However, a read-out consisting of a microwave multiplexer offers a path to meeting the Gen-X requirements. Several different versions of this type of multiplexer exist, depending upon whether it is adapted for the read-out of TESs, MMCs or KIDs. For TESs and MMCs, unshunted, non-hysteretic rf SQUIDs are incorporated into the read-out which have negligible power dissipation even for extremely large arrays (Mates et al., 2008). Furthermore, rf SQUIDs can be coupled to high-Q microwave resonant circuits fabricated from superconducting coplanar waveguides with resonant frequencies of several gigahertz. In this approach, the bandwidth of each microcalorimeter is limited by the resonant circuit after amplification by the rf SQUID. A single high electron mobility transistor (HEMT) amplifier has the bandwidth and dynamic range to read out many hundreds of rf SQUIDs operated in superconducting microresonators tuned to different frequencies, all coupled to the same coplanar-waveguide feedline. This essential technology that must be developed in order to make this microcalorimeter feasible is microwave multiplexing.

The table below illustrates a plausible arrangement for the cryogenic microcalorimeter within the 50 m focal length telescope configuration.

References

"Performance of TES X-ray Microcalorimeters with a Novel Absorber" S.R. Bandler, R.P. Brekosky, A.-D. Brown, J. Chervenak, E. Figueroa-Feliciano, F. Finkbeiner, N. Iyomoto, R.L. Kelley, C.A. Kilbourne, F.S. Porter, J. Sadleir, S.J. Smith J. of Low Temp. Physics 151 (1/2), 400-405 (2008).

"Development of arrays of position-sensitive microcalorimeters for Constellation-X" S.J. Smith, S.R. Bandler, R.P. Brekosky, A.-D. Brown, J.A. Chervenak, M.E. Eckart, E. Figueroa-Feliciano, F.M. Finkbeiner, R.L. Kelley, C.A. Kilbourne, F.S. Porter, J. Sadlier Proc. SPIE 7011, 701126-1 - 701126-8, (2008).

"Demonstration of a Multiplexer of Dissipationless SQUIDs" J.A.B. Mates, G.C. Hilton, K.D. Irwin, L.R. Vale, K.W. Lehnert Appl. Phys. Lett. 92, 023514, (2008).