On the energy response function of a CdTe Medipix2 Hexa detector

Abstract

X-ray imaging based on photon counting pixel detectors has received increased interest during the past years. Attached to a semiconductor of choice, some of these devices enable to resolve the spectral components of an image. This work presents the results from measuring the energy response function of a Medipix2 MXR Hexa detector, where six individual Medipix detectors were bump bonded to a 1 mm thick cadmium telluride sensor in order to form a 3×2 array of 4.2×2.8 cm2 size. The average FWHM of the photo peak of an 241Am source was found to be 2.2 and 2.1 keV for single pixels and bias voltages of 200 and 350 V, respectively, across the whole Hexa detector. This corresponds to a relative energy resolution of less than 4%. Adding up all pixel spectra of individual chips lead to an only small deterioration of energy resolution, with line widths of 2.7 and 2.5 keV. In general, a lower detection efficiency was observed for the lower voltage setting, along with a shift of the peak position towards lower energies.

Full text

Page 1

On the energy response function of a CdTe Medipix2 Hexa detector
Thomas Koeniga,?, Andreas Zwergerb, Marcus Zubera, Patrick Schuenkea, Simeon Nilla, Ewald Gunic,
Alex Faulerb, Michael Fiederleb, Uwe Oelfkea
aGerman Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120 Heidelberg, Germany
bFreiburg Materials Research Center (FMF), Stefan-Meier-Strasse 21, 79104 Freiburg, Germany
cErlangen Centre for Astroparticle Physics (ECAP), Erwin-Rommel-Strasse 1, 91058 Erlangen, Germany
a r t i c l e i n f o
Available online 21 November 2010
Keywords:
Semiconductor detector
Cadmium telluride
Photon counting
Energy resolution
a b s t r a c t
X-ray imaging based on photon counting pixel detectors has received increased interest during the past
years. Attached to a semiconductor of choice, some of these devices enable to resolve the spectral
componentsofanimage.Thisworkpresentstheresultsfrommeasuringtheenergyresponsefunctionofa
Medipix2MXRHexadetector,wheresixindividualMedipixdetectorswerebumpbondedtoa1 mmthick
cadmium telluride sensor in order to form a 3?2 array of 4.2?2.8 cm2size. The average FWHM of the
photo peak of an241Am source was found to be 2.2 and 2.1 keV for single pixels and bias voltages of 200
and 350 V, respectively, across the whole Hexa detector. This corresponds to a relative energy resolution
oflessthan4%.Addingupallpixelspectraofindividualchips leadtoanonly smalldeteriorationofenergy
resolution, with line widths of 2.7 and 2.5 keV. In general, a lower detection efficiency was observed for
the lower voltage setting, along with a shift of the peak position towards lower energies.
& 2010 Elsevier B.V. All rights reserved.
1. Introduction
Conventional digital X-ray imaging relies on measuring a signal
proportional to the integrated amount of energy deposited in a
pixel during exposure, which normally follows the conversion
of X-rayphotons intovisible lightin a scintillatingmaterial. During
the last decade, this approach has been supplemented by direct
conversiontechniquesbasedonpixelizedsemiconductordetectors
[1]. While many of these devices are capable of counting individual
photons and thus providing spectral resolution, there is ongoing
debate as to which combinations of pixel sizes, sensor materials
and thicknesses to choose in order to achieve both high spatial
and spectral resolution in combination with a high quantum
efficiency. Depending on detector properties and sensor material,
the recorded spectrum can significantly differ from the actual one.
In order to interpret and reconstruct these spectra, a careful
measurement of the so-called energy response function is neces-
sary.Thisworkpresentstheresultsfrommeasuringthisfunctionat
a single photon energy of about 60 keV for a Medipix2 MXR [2]
Hexa detector attached to cadmium telluride (CdTe) single crystal.
With a size of 4.2?2.8 cm2, this device represents the largest
Medipix detector assembled to a single sensor so far, and is
intended to be used for small animal imaging at DKFZ.
In the following section, we will briefly describe the detector
andtheprocessingstepsweemployedinordertoobtaintheresults
presented in Section 3. Section 4 will then give a summary as well
as an outlook on future experiments.
2. Materials and methods
The energy response function of pixelized semiconductors can
be severely biased by electric charges shared among neighboring
pixels.Inordertomitigatetheseeffects,apixelpitchof165 mmwas
chosen rather than the 55 mm which is provided by the Medipix
architecture. This was achieved by connecting only every ninth
pixeltothesensor(3?3)bymeansofintentionallyohmiccontacts,
resulting in 258?172 pixels (44 376 in total). The sensor was a
1 mm thick CdTe single-crystal manufactured by Acrorad. These
single-crystals represent the state-of-the-art of high resistivity,
detectorgrade material and are commercially available with a
diameter of 75 mm and thicknesses of 1 and 2 mm. The crystals
were grown by the travelling heater method, which offers the
advantage of a low growth temperature that reduces the concen-
tration of defects and increases homogeneity. Characterization of
thesewafersobtainedahomogeneousdistributionoftheresistivity
with an average value of 5 ? 109O cm and a variation of less than
10%.Telluriuminclusions(secondphasedefects)wereidentifiedby
infrared microscopy with an average diameter of less than 10 mm
and a concentration of about 103cm?2. Hybridization was per-
formed at FMF with low temperature solder bumps. The overall
process temperature was kept below 130 1C, which is of crucial
importance in order to maintain high level sensor properties. All
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/nima
Nuclear Instruments and Methods in
Physics Research A
0168-9002/$-see front matter & 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.nima.2010.11.071
?Corresponding author.
E-mail address: t.koenig@dkfz.de (T. Koenig).
Nuclear Instruments and Methods in Physics Research A 648 (2011) S265–S268

Page 2

measurementsdescribedbelowwereperformedusingtheMedipix
USB interface [5] and the Pixelman software [6].
As every otherMedipix2 detector,the Hexadetector introduced
here offers two modes of operation: The single threshold mode,
where every event abovea predefined energy threshold is counted,
and the dual threshold or window mode, which introduces an
additional threshold in order to obtain an energy window.
The response of every pixel can be calibrated with two sets of
three adjustment bits to bring each of the two thresholds more in
line with their neighbors. The process determining the optimal
values of these bits is called threshold equalization and was
performed according to standard procedures. In detail, the Ka
fluorescence of a silver foil was used to equalize the lower
thresholds, while the upper thresholds were equalized employing
the switching point between single and dual threshold modes as
described by Tlustos et al. [3].
Energy calibration was performed using the Kalines of molyb-
denum (17.4 keV) and silver (22.2 keV) aswell asthe photo peak of
an241Am source (59.6 keV), which is described below. Here, one
has to keep in mind that charge sharing shifts the maximum
position of the peak to slightly lower energies. This shift was
determined using a Monte Carlo simulation as developed by Durst
et al. [4]. The corrected energy values used for the three sources
were 16.4, 21.4 and 59.0 keV for a bias voltage of 350 V.
In principle, the window mode can be used to obtain spectro-
scopic information about the incident X-rays. However, the energy
window must not be chosen too small as manufacturingtolerances
lead to a slight variation of each pixel’s response to radiation even
after threshold equalization, also called residual threshold disper-
sion. The remaining small differences become more and more
dominant as the energy window gets thinner, and so the effective
windowsizecannotbesetbelowafewkeV,whichinturnleadstoa
blurring of the spectra recorded. While this behavior does not
represent a drawback in most imaging applications, it prevents the
precise characterization of the detector under study. For this
reason, the energy response function is usually measured differ-
ently byscanning insingle threshold modeacross the energy range
of interest, yielding an integrated spectrum. This spectrum is then
differentiated to obtain the actual photon energy spectrum.
In order to determine the energy response function, i.e. the
detector’s response to monochromatic radiation, we employed the
same241Am source that we already used for energy calibration. It
emits X-ray photons with an energy of about 59.6 keV and is
therefore well suited for the investigation of the energy response
function at an X-ray energy typical for medical diagnostics, when
other monochromatic sources such as synchrotron light are not
available.
The241Amusedherehasanominalactivityofabout1.1 GBqand
was placed in front of the detector at a distance of about 5 cm. The
measurement was performed at a bias voltage of 350 V for an
energy range of 10–60 keV and took about 18 h in order to acquire
enough photonsto perform the peakfittingwithoutany significant
low-pass filtering, which would have biased the width of the photo
peak. Another measurement of just the photo peak was performed
at 200 V to investigate the dependance on the bias voltage.
The full width at half maximum (FWHM) of the photo peak was
determined by fitting a Gaussian to individual pixels as well as the
sum over each of the six chips’ pixel spectra. A binomial filter of
size three was applied to the spectra prior to fitting in order to
make the results more stable (which broadens the peak by less
than 0.05 keV). The medians and the standard deviations obtained
from the fits on the single pixel level can then be considered a
measure for the inter-pixel dissimilarity, while the fits on the sum
spectra give an impression of how much the residual dispersion of
the lower thresholds’ responses affect the energy resolution of a
whole chip.
In order to demonstrate how the energy response function
affects the acquisition of broad band X-ray spectra, the tube
spectrum from a medical Siemens Powerphos X-ray tube was
measured at a voltage of 110 kVp and a current of 1 mA. The tube
features a built-in aluminium filter, which is normally used to
reduce the skin dose in patients and blocks almost every photon
below 20 keV from exiting. The detector was placed about 1.2 m
from the tube to prevent both signal pile-up in the detector
electronics as well as polarization of the CdTe sensor caused by
the low hole mobility. Again, the spectrum was recorded by a scan
insinglethresholdmodebetweenenergiesof10and110 keVusing
exposure times of 4.5 s per value of the internal Medipix channel
(about 0.2 keV).
To finally give an impression of the overall quality of our Hexa
detector, we acquired two images of a plastic lighter in window
mode at two different photon energies (30–35 and 65–70 keV).
Furthermore, we discuss the influence of the two bias voltages on
the resulting image quality using the example of the241Am scan.
3. Results and discussion
The energy response function of nine pixels measured at 350 V
is depicted in Fig. 1. The spectra in this figure have been aligned by
meansofacross-correlationtoenhancethevisibilityofthepeaks.It
demonstrates the occurrence of not only the charge sharing back-
ground, but also of the Ka lines of cadmium (23.2 keV) and
tellurium (27.5 keV) plus their associated escape peaks (36.3 and
32.0 keV). It follows that the pixel pitch of 165 mm is not large
enoughtoreducethenumberoffluorescencephotonsthatescapea
pixel down to a level where their influence would be neglectable.
The medians and the standard deviations of the FWHMs for the
single pixel spectra are shown in Table 1 for the two voltages
investigated. It can be seen that increasing the bias voltage from
200 to 350 V only amounts to a decrease of the peak width by
0.1 keV. However, this goes along with a further movement of the
peakpositionto lowerenergiesby about0.6 keV onaverage, which
is due to charge sharing (data not shown).
Thesame conclusioncanbe drawnfor thespectrasummedover
thewholechip,aslistedinTable2.Furthermore,theincreasedueto
the summation is found to be only about 20%, a result that is
certainly depending on the quality of the adjustment bits obtained
from the lower threshold equalization procedure. Fig. 2 shows the
samespectraasFig.1withoutalignmenttogiveavisualimpression
of the residual variations in the threshold equalization which are
responsible for this behavior.
20 304050 60
0
500
1000
1500
2000
2500
3000
3500
Energy [keV]
Differential Counts
Fig.1. Singlepixelspectrafoundina3?3neighborhoodofarandomlychosenpixel
(Vbias¼350 V). Note that these spectra were low-pass filtered using a binomial
kernel of size n¼25 (for visualization only).
T. Koenig et al. / Nuclear Instruments and Methods in Physics Research A 648 (2011) S265–S268
S266

Page 3

It is worthy to note that the energy resolution does not vary
significantly across the sensor for the 350 V setting, with 2.1 keV
obtained for the best chips and 2.2 keV for the worst, which is well
below one standard deviation.
The situation changes when one considers the chip spectra
where all the contributions from individual pixels have been
summed up and peak widths exceeding 3 keV can be observed.
Whether this behavior can be changed with a different threshold
equalization or not will have to be investigated in further studies.
The tube spectrum is shown in Fig. 3 and has been obtained by
summing up all pixel spectra for chip 4 in order to increase the
signal-to-noise ratio. It clearly shows that the charge sharing
backgroundandtheCdTefluorescencesdistorttheactualspectrum
predominantly at lower energies, where almost no X-rays happen
topasstheX-rayfilter.ThetungstenKalinecanbefoundjustbelow
60 keV, whereas the Kbline is barely visible at about 67 keV. The
escape peaks encountered in Fig. 1 are now a continuum due to the
broad X-ray spectrum.
Fig. 4 shows the two images of the plastic lighter acquired in
windowmode.Asexpected,thecontrastbetweenmetalandplastic
parts diminishes at higher energies. The images also show some
defectivebumpbondsatthebottomandontheleft.Apartfromthat
the occurrence of assembly defects seems to be well controlled.
The influence of the bias voltage on the resulting images is
shown in Fig. 5. Striking is the generally higher number of counts
observed for the higher bias voltage. This behavior is probably due
Table 1
Medians and standard deviations of the FWHMs obtained for the single pixel
spectra, averaged over the whole sensor or individual chips.
FWHM (200 V) (keV)FWHM (350 V) (keV)
All chips
Chip 0
Chip 1
Chip 2
Chip 3
Chip 4
Chip 5
2.270.4
2.270.6
2.570.4
2.470.3
2.170.3
2.170.2
2.370.3
2.170.3
2.270.3
2.270.3
2.270.3
2.170.3
2.170.2
2.170.2
Table 2
Medians and standard deviations of the FWHMs obtained when summing up all
single pixel spectra over entire chips. The first quantity (‘‘all chips’’) is obtained as
the median and the standard deviation of the other measurements.
FWHM (200 V) (keV)FWHM (350 V) (keV)
All chips
Chip 0
Chip 1
Chip 2
Chip 3
Chip 4
Chip 5
2.770.3
NAa
2.7
3.2
2.5
2.3
2.8
2.570.3
3.0
2.6
2.9
2.4
2.1
2.3
aUnreliable fit, as the spectra added up to two distinct Gaussians.
20 304050 60
0
500
1000
1500
2000
2500
3000
3500
Energy [keV]
Differential Counts
Fig.2. AsinFig.2,butwithoutalignmentofthespectrainordertoshowtheresidual
threshold dispersion.
204060 80100
0
0.5·106
1.0·106
1.5·106
Energy [keV]
Differential Counts
Fig.3. RecordedspectrumofanX-raytubeoperatedat110 kVp(noisyline:original
data after differentiation, smooth line: low-pass filtered version (n¼101);
Vbias¼350 V).
Fig. 4. Two flat-field corrected images of a plastic lighter acquired in window mode
at different X-ray energies (left: 30–35 keV, right: 65–70 keV). The salt-and-pepper
noise is despite the flat-field correction and occurs because the bias voltage (200 V)
had notbeenswitched onlong enough before themeasurements took placeð ? 1 hÞ.
Fig. 5. Example of how assembly defects can influence the resulting images at bias
voltages of 200 V (left) and 350 V (right). Blue: low number, red: high number of
counts.
T. Koenig et al. / Nuclear Instruments and Methods in Physics Research A 648 (2011) S265–S268
S267

Page 4

tochargesharing, butmay evenbea resultof aninsufficientcharge
collection efficiency. Further investigations of this behavior are
planned.
It can also be observed that a higher bias voltage affects the size
of the non-working detector areas negatively. Furthermore, the
pixels in the periphery of these defects show higher counts in this
single threshold measurement than the areas free of (visible)
defects, which indicates that charges generated near the defects
get deflected towards the sides. Note that while the border pixels
with higher counts can be dealt with by applying a flat-field
correction, those in the interior show zero counts and require an
interpolation during image reconstruction.
Although the reduction of dead pixels is generally favorable
especially when it comes to computed tomography, the lower
detection efficiency along with the movement of the peak position
is probably more serious, and using the higher voltage setting is
advised.
4. Conclusions and outlook
The results presented in the previous section show that the
quality of CdTe single-crystals available on the market is now good
enoughtofacilitatetheproductionandtheuseoflargermonolithic
sensors ð4:2 ? 2:8 cm2Þ than previously possible. We have demon-
strated that Medipix2 detectors with a pixel pitch of 165 mm and a
1 mm thick CdTe sensor can reach energy resolutions of about
2.1 keV under near optimal conditions, and that an average of
2.2 keV is a realistic value for assemblies like the Hexa detector
characterized in this study. In conjunction with bigger pixels
this leaves room for increasing the sensor thickness to improve
absorption.
We have also shown that a value of 200 V for the bias voltage is
notappropriatetooperatethiscombinationofdetectorandsensor.
Although the decrease in spectral resolution amounts to only
0.1 keV and the influence of assembly defects on the images gets
significantly smaller, a lower number of counts was observed for
the lower bias voltage, and so the use of this setting is discouraged.
It will be the subject of further studies to investigate the optimal
trade-off between spectral resolution, dead areas and photon
detection efficiency for imaging applications.
Acknowledgements
The authors would like to thank Gisela Anton and Thilo Michel
(both ECAP Erlangen) for helpful discussions and the provision of
the Am source at their institute.
References
[1] J. Schlomka, E. Roessl, R. Dorscheid, S. Dill, G. Martens, T. Istel, C. Bumer,
C.Herrmann,R.Steadman,G.Zeitler,A.Livne,R.Proksa,Experimentalfeasibility
of multi-energy photon-counting k-edge imaging in pre-clinical computed
tomography,Phys.Med.Biol.53(15)(2008)4031doi:S0031-9155(08)75296-5.
[2] X. Llopart,M.Campbell,R. Dinapoli,D.S. Segundo,E. Pernigotti,IEEETrans. Nucl.
Sci. NS-49 (5) (2002) 2279.
[3] L. Tlustos, R. Ballabriga, M. Campbell, E. Heijne,K. Kincade,X. Llopart, P. Stejskal,
IEEE Trans. Nucl. Sci. NS-53 (1–2) (2006) 367.
[4] J. Durst, P. Bartl, B. Kreisler, T. Michel, G. Anton, Monte Carlo simulation of
pixelated photon counting X-ray detectors like the Medipix2 and the Medipix3
using high-Z sensor materials, in: IEEE Nuclear Science Symposium Conference
Record, 2008, 2008, pp. 4990–4994.
[5] Z. Vykydal, J. Jakubek, S. Pospisil, Nucl. Instr. and Meth. 563 (1) (2006) 112.
[6] T.Holy,J.Jakubek,S.Pospisil,J.Uher,D.Vavrik,Z.Vykydal,Nucl.Instr.andMeth.
A 563 (1) (2006) 254.
T. Koenig et al. / Nuclear Instruments and Methods in Physics Research A 648 (2011) S265–S268
S268