TES-based microcalorimeter for future X-ray astronomy missionsSoftware development for instrument calibration

The Instrument

The XMS microcalorimeter can gauge the energy of absorbed radiation by converting it into heat. The energy of an incoming X-ray photon can be determined by measuring the rise in temperature of the absorber. This temperature increase is directly proportional to the energy of the photon.

A Transition Edge Sensor (TES) is used because it acts as an incredibly sensitive “thermometer” due to the abrupt phase transition from the superconducting state to the normal state, where the resistance changes from 0 to a finite value within a temperature range of very few mK. The TES is cooled to <100 mK and biased in its transition.

The TES is coupled to a super-conducting quantum interference device (SQUID) in order to read out the signal (ie. pulses in current).

The Software

Consists of a series of processing chains developed to carry out the detection of pulses and to characterize & calibrate the instrument, more specifically aiming to:

• Calculate the energy resolution of the instrument by analyzing the pulses resulting from X-ray photons hitting the detector.

• Obtain characteristics of the IV-curve (current vs. voltage).

• Calculate the complex impedance of the TES.

• Analyze the noise spectrum of the TES.

Pulse Detection

• The input data consists of values of current I(t) vs. time (t).

• The data is low-pass filtered (using a box-car function) to eliminate noise.

• Using the 1st derivative of the filtered data provides a much more sensitive way to handle pulses that are piled-up.

• An initial threshold value is set to look for single pulses.

• An adjusted 1st derivative and a new threshold level are used to search for secondary pulses.

• Finally, the pulse detection task calculates the initial & end times of ALL pulses in an event.

Science with the XMS

The XMS instrument, because of its excellent spectral resolution (2.5-3 eV at 6 keV) has the potential to carry out breakthrough science in the field of astrophysics. Those contributions as described in past proposed space missions (IXO/ATHENA) with this instrument as payload would further our understanding of:

• Cosmic Feedback processes involving the energy outflow from supermassive black holes and of the surrounding hot medium in galaxy bulges, groups and clusters.

• Missing baryons & the detecting WHIM lines both in emission, and in absorp-tion against a bright background source.

• Cluster physics & evolution.

• Chemical evolution through cosmic time by measuring the abundances of all elements from C to Zn in the hot intracluster medium of clusters.

• Map the abundance patterns of Supernova Remnants.

• Reveal the physical nature of gas and dust in the diffuse interstellar medium of our Galaxy, via absorption features imprinted on the X-ray spectrum.

• Interactions of space plasmas and magnetic fields by giving us deeper insights into the complex workings of planetary magneto spheres and exospheres.

Absorber

Substrate

Heat Sink < 0.1 K

Normal metalSuperconductor

slow thermallink

X-rayphotons

(Cu/Bi)

(Ti/Au)

Fraga-Encinas, R. (1) ; Cobo, B. (1) ; Ceballos, M. (1) ; Schuurmans, J. (2) ; van der Kuur, J. (2) ; Carrera, F. (1) ; Barcons, X. (1)

(1) Instituto de Física de Cantabria (CSIC-UC), 39005, Santander, Spain. (2) Netherlands Institute for Space Research (SRON), Utrecht, Netherlands.

The XMS (X-ray Microcalorimeter Spectrometer) is an instrument prototype with imaging capability in X-rays and high-spectral resolution. This instrument is a microcalorimeter based on transition edge sensors (TES). As part of the Spanish contribution to the advancement of the XMS, we present the work carried out by the X-ray astronomy group at the Instituto de Física de Cantabria in collaboration with The Netherlands Institute for Space Research. Our main task involves the development and testing of software for this prototype with the purpose of instru-ment calibration, X-ray pulse detection and energy resolution calculations.

2.2. Phase transition and critical temperature 21

0.080 0.090 0.100 0.110 0.120Temperature (K)

-0.050.00

0.05

0.10

0.15

0.200.25

Res

ista

nce

(Ω)

Figure 2.2: Example of the superconducting to normal transition of a Ti/Au bilayer. Inbulk form, Ti has a critical temperature Tc of 0.40 K. By using a bilayer of 18 nm Ti and30 nm Au, the Tc was suppressed to 0.097 K.

called the proximity effect. By changing the film thicknesses of the superconducting andthe normal metal, the strength of this effect can be adjusted and the Tc of the bilayer canbe tuned. For the Cooper pairs, a so-called intrinsic coherence length can be defined:

ξ0 =hvFπ∆

(2.1)

≈ 0.18hvFkBTc

(2.2)

where vF is the Fermi velocity. For a pure superconductor, there is almost no suppres-sion of Tc when the thickness exceeds ξ0. For thinner films, Tc is suppressed furtherwith increasing normal metal thickness, until the coherence length of the normal metal isreached:

ξn =

√

hD2πkBT

(2.3)

where D = vFnln/3 is the diffusion constant of the normal metal, with vFn its Fermivelocity and ln its electron mean free path [10]. For greater normal metal thicknesses,Tc stays constant. This is schematically illustrated in figure 2.3. In thin films at lowtemperature, the mean free path ln will be limited by the film thickness. Therefore, normalmetal films with thicknesses below (hvF/(6πkBT ))1/3, will always be thinner than theircoherence length.

These coherence lengths give an upper limit as to the required thicknesses for thedesired suppression of Tc. In practice however, there appears to be quite a large margin

PHASE TRANSITION

superconducting R = 0

normal state

Tcritical

ECT

∆T =

X-ray photon

AbsorberFast thermal link

SensorSlow thermal link

Heat sink

-0.02 -0.01 0 0.01

Zreal (Ohm)

0.02 0.03

-0.02

-0.01

0

0.01

Zimag (Ohm)Complex Impedance

FREQUENCY (Hz)10 1e3 1e5

3e-11

1e-10

3e-10

1e-9

3e-9

1e-8

NO

ISE

SP

EC

TRA

L D

EN

SIT

Y (A

mp/

Hz1/

2 )

Noise spectrum

Non-filtered pulse Filtered pulse

400 500 600 700 800 900

7e-7

9e-7

1e-6

3e-6

5e-6

Time

Cur

rent

0 2000 4000 6000 8000 10000

0 2000 4000 6000 8000 10000

1

0.8

0.6

0.4

0.2

0.0

-0.2

-0.4

1

0.8

0.6

0.4

0.2

0.0

-0.2

-0.4

PulseFirst derivative

Threshold

PulseAdjusted derivative

New Threshold

missedpulse

recovered

5x5 pixel array with Ti/Au TES of 150x150 μm2

5860 5880 5900 5920 Energy (eV)

0

100

200

Cou

nts

ΔE = 2.5 ev @ 5.9 keV

9.0

x 10-7

8.0

7.0

6.0

5.0

Time for datapoint

Cur

rent

(A)

Pulse height ∝ E

Read-out signal from the TES

Transition from superconducting to normal state

Transition from normal to superconducting state

Normal state (conducting)

power plateau

Voltage (V) Vmax

Cur

rent

(A)

Ic+

Ic-

Supe

rcon

duct

ing

IV - curve

TES

absorber

References

Bergmann Tiest, W. 2004, Tekst. Proefschrift Universiteit Utrecht.

Boyce, K. et al. 1999, Proc SPIE 3765 (XRS).

Den Herder, J-W. et al. 2011, SRON-XMS-RP-2011-033.

ATHENA assessesment study report, ESA/SRE(2011)17.