10.1117/2.1201002.002620
Magnetic calorimeter arrays for
x-ray astronomy
Simon Bandler
A new class of detectors has great potential as x-ray cameras for future
astrophysical observatories, with unprecedented energy resolution and
array sizes greater than one megapixel.
Ever since the early 1980s, microcalorimeters were thought to
have the potential to determine the energy of an x-ray photon
with an accuracy of 1eV, a goal that promises to greatly enhance
our understanding of astrophysics.1 A microcalorimeter detector
at the focus of a high-angular-resolution x-ray telescope is a
unique instrument because of its ability to simultaneously determine the energy of each x-ray photon extremely precisely while
also producing x-ray images of various astronomical sources
with very high detection efficiency. By studying the precisely
measured distribution of x-ray photon energies emitted by
various astronomical objects, we learn important information
about them such as their composition, temperature, density,
and dynamics. We can study a broad spectrum of astrophysical
sources, such as the matter-accretion process near black holes,
supernovae remnants, and the growth and evolution of galaxies
and galaxy clusters. Since the invention of microcalorimeters,
scientists around the world have developed ever more sensitive
and innovative techniques to detect photons.2 At present,
microcalorimeters with superconducting transition-edge sensors (TESs) hold the record for the best energy resolution (1.8eV)
for detecting a 6keV x-ray.3 Metallic magnetic microcalorimeters
(MMCs) are now on the verge of demonstrating the first ever
‘sub-eV’ energy-resolution x-ray detector.
MMCs use magnetism to produce a high-precision temperature sensor. The MMCs that have been developed use the paramagnetic susceptibility of gold doped with a low concentration
of erbium ions (Au:Er) when placed in a dc magnetic field.4
In a paramagnet, the magnetization is inversely proportional
to temperature, making it very sensitive to small changes at
low temperatures. The paramagnet is attached to an x-ray absorber as depicted in Figure 1. When an incoming x-ray hits the
microcalorimeter’s absorber, its energy is converted into heat,
which a thermometer then measures. The temperature rise is
Figure 1. Schematic representation of a magnetic calorimeter. SQUID:
Superconducting quantum interference device.
directly proportional to the x-ray’s energy and is approximately
0.0005 Kelvin for a 6keV photon in a calorimeter operated at a
temperature of 0.04 Kelvin.
During the development of MMCs it became clear that the
full potential of high-resolution detectors would only be realized
by microfabrication techniques. Groups at Heidelberg University (Germany) and my group at NASA’s Goddard Space Flight
Center (GSFC) in collaboration with George Seidel at Brown
University have led efforts to develop fully microfabricated
detectors and detector arrays. Techniques have been devised to
make the best geometries for sensing the change in magnetic
susceptibility using a meander-shaped pickup coil connected to
a current sensor that is a specially designed state-of-the-art
superconducting quantum interference device (SQUID).5 On top
of these coils we sputter-deposit 1m-thick films of Au:Er to
produce the MMC sensor, and then fabricate cantilevered, highquantum-efficiency x-ray absorbers consisting of a bismuth-gold
bilayer (see Figure 2).6
Continued on next page
10.1117/2.1201002.002620 Page 2/3
Figure 2. (left) Scanning-electron-microscope image showing how the
absorbers of a 55 array of pixels are cantilevered above the substrate.
(right) Photograph of an individual pixel. The pattern of the underlying meander-shaped pickup is visible through the Au:Er film and
the absorber. The central pickup is also the region on a silicon nitride
membrane that supports the cantilevered absorber.
We have tested the thermodynamic properties of the MMC,
in particular the magnetic susceptibility and the heat capacity,
and these properties are now approaching those that have been
determined for an MMC made using bulk Au:Er. The presently
best-achieved energy resolution at 6keV is EFWHM (full width
at half maximum) D 2:8eV in Heidelberg and EFWHM D 3:3eV
at GSFC7 (as shown in Figure 3). These arrays of MMCs have
some flaws in their fabrication (GSFC) and testing (Heidelberg)
that are now well characterized and understood at both institutions, and can be overcome. In the next generation, implementing a modified fabrication technique, we anticipate that the
energy resolution will break the 1eV barrier.
To make some of the more advanced future concepts for astrophysics possible, such as Generation-X,8 it is desirable to have
both sub-eV energy resolution and a dramatic increase in the size
of arrays. Ideally, we will develop megapixel arrays of tiny pixels
just as in everyday digital cameras. One of the key advantages
of using MMC-based microcalorimeters is that no heat is dissipated within the pixels associated with the process to measure
the change in magnetic susceptibility. The significance of this is
that MMCs can be more easily scaled up to megapixel-sized
arrays.
To build such arrays, one of the techniques being pursued is
the use of position-sensitive MMCs. In this device multiple absorbers, each with a different thermal conductance, are coupled
to one magnetic sensor. This results in different pulse shapes
(see Figure 4), which enables position discrimination. We have
fabricated and tested a position-sensitive magnetic calorimeter
in which a single sensor was used to read out four absorbers.
An energy resolution of less than 4.7eV was observed for 6keV
Figure 3. Spectrum of manganese K˛ x-rays from an iron-55 source.
The light blue line shows the intrinsic line shape, and the broadening
of this shape is consistent with a metallic magnetic microcalorimeter
(MMC) energy resolution of 3.28eV. FWHM: Full width at half
maximum.
x-rays in each of the absorbers, which is the present record
for any position-sensitive microcalorimeter at this energy.9 Improved energy resolution is also expected for the next generation
of designs for these devices.
In summary, MMC-based microcalorimeter arrays hold great
potential for meeting the requirements of advanced mission concepts such as Generation-X, which has a goal of a focal-plane
detector with less than 1eV energy resolution over a million
pixels. The energy resolution that has been achieved in small arrays of single pixels and position-sensitive detectors is already
competitive with other state-of-the-art technologies, and further
progress is likely in the very near future from microfabricating arrays of pixels that have the same magnetic properties as
has been seen in previous detectors with bulk magnetic samples. Our next steps are to build and test MMC arrays that will
attempt to demonstrate the further energy-resolution potential,
before moving on to increase the size of arrays to over a thousand pixels.
The X-ray Microcalorimeter Group at Goddard is led by Richard
Kelley, Caroline Kilbourne, and F. Scott Porter, and includes Joseph
Adams, Catherine Bailey, Simon Bandler, Meng Chiao, Megan Eckart,
Fred Finkbeiner, Nikhil Jetvala, Jan-Patrick Porst, Jack Sadleir, and
Stephen Smith. Wen-Ting Hsieh was the lead engineer responsible for
fabricating the devices described.
Continued on next page
10.1117/2.1201002.002620 Page 3/3
References
1. S. H. Moseley, J. C. Mather, and D. McCammo, Thermal detectors as x-ray spectrometers, J. Appl. Phys. 56 (5), pp. 1257–1262, 1984. http://link.aip.org/link/?JAP/56/
1257/1, doi:10.1063/1.334129
2. C. Enss ed., Cryogenic Particle Detection, Topics Appl. Phys. 99, Springer,
Berlin, 2005.
3. S. R. Bandler et al., Performance of TES x-ray microcalorimeters with a novel absorber
design, J. Low-Temp. Phys. 151 (1-2), pp. 400–405, 2008. doi:10.1007/s10909-0079673-6
4. A. Fleischmann, C. Enss, and G. Seidel, in C. Enss ed., Cryogenic Particle Detection, Topics Appl. Phys. 99, ch. Metallic magnetic calorimeters, pp. 151–216,
Springer, Berlin, 2005.
5. D. Drung, C. Assmann, J. Beyer, A. Kirste, M. Peters, F. Ruede, and T. Schurig,
Highly sensitive and easy-to-use SQUID sensors, IEEE Trans. Appl. Supercond. 17,
pp. 699–704, 2007.
6. W.-T. Hsieh et al., Microfabrication of high resolution, micro-fabricated, x-ray magnetic
calorimeters, AIP Conf. Proc. 1185, pp. 591–594, 2009.
7. S. R. Bandler et al., Performance of high resolution, micro-fabricated, x-ray magnetic
calorimeters, AIP Conf. Proc. 1185, pp. 579–582, 2009.
8. http://www.cfa.harvard.edu/hea/genx.html Generation-X Vision Mission,
Harvard-Smithsonian Center for Astrophysics. Accessed 30 January 2010.
9. J.-P. Porst et al., Development of position-sensitive magnetic calorimeter x-ray detectors,
AIP Conf. Proc. 1185, pp. 579–582, 2009.
Figure 4. (top left) Schematic drawing showing the layout of a positionsensitive MMC. (top right) Photograph of such a device. (bottom) Measured average pulse shapes for x-ray events at the different absorbers.
After the initial equilibration signal, the pulses decay with the same
exponential time constant. The differences in the rise times and pulse
shapes allow us to determine which element the x-ray was absorbed by.
Author Information
Simon Bandler
NASA/GSFC
Greenbelt, MD
Simon Bandler is an associate research scientist at the University
of Maryland who works in the X-ray Astrophysics Laboratory
at NASA’s GSFC. He received his PhD in physics from Brown
University in May 1995. He has also worked at Heidelberg
University (Germany) and the Smithsonian Astrophysical
Observatory.
c 2010 SPIE