1. Investigation of diamond sensors for calorimetry
K. Afanaciev 1, I. Emeliantchik 1, A. Ignatenko 1,
E. Kouznetsova 2, W. Lohmann 2, C. Grah 2
1 NCPHEP, Minsk; 2 DESY, Zeuthen,
ECFA ILC Workshop, Vienna, 2005

2. Application ILC Detector forward calorimeter BeamCal: Sampling
diamond-tungsten calorimeter
Expected dose from beamstrahlung
is about 10 MGy per year
Large area detector is needed

3. Detection mechanism Diamond works as a solid state ionization chamber
Charge carriers, generated by ionizing particle move in an external electric field and induce charge on the surface electrodes
This charge is detected by a charge-sensitive preamplifier

4. The samples Chemical vapor deposited (CVD) polycrystalline samples:
A large number of samples from Fraunhofer IAF, Freiburg
12x12 mm plates with thickness m
Metallization: 10 nm Ti nm Au
1 pad or 4 pad segmentation
samples from GPI, Moscow
samples from Element Six
HPHT synthetic monocrystalline samples:
From ADAMAS BSU
Typical size 4x4 mm, thickness m
Metallization: Fe-Ni with B ion implantation
1 sample of natural monocrystalline diamond

5. IV Measurement Most of the samples show ohmic
behavior and very high resistance
in the order of ohm
Settling time of the current is in the
order of 100 s. Hysteresis is due to
long settling time of the signal.

6. Charge collection distance
Performance of diamond as particle detector is usually characterized by charge collection efficiency  or distance (CCD)
Efficiency is given by
 = Qmeas/Qind
where Qmeas is measured charge and Qind - total charge generated
Charge collection distance equals
CCD = *d
where d is the thickness of the sample
CCD could be considered as the drift distance between the charge carrier pair before trapping

7. CCD Measurement
To measure CCD we need to know the amount of charge carriers generated
But we know that a MIP
generates about 36 e--h pairs
per 1 m pathlength in diamond
Sr90
&
Gate PA
discr
delay
ADC
In this case CCD = Qmeas /36
diamond
Scintillator allows triggering
only from particles which could
be considered MIPS
Scint.
PM1
PM2

8. CCD Measurement Pedestal (random trigger) Signal (MIP trigger)
Signal charge value could be calculated from this pair of spectra
by subtracting pedestal and multiplying by charge/ADC channel
value from calibration.
The signal is fitted to the Landau distribution

9. CCD Results

10. CCD vs HV and time Charge collection distance depends on bias voltage
applied to the sample
Saturation is observed at
field strength of about 1V/m
There is also dependence
of CCD on time with HV on
About 25% signal decrease

11. CCD vs Dose
Diamond is stable under irradiation with doses up to 10 MGy
But in the process of irradiation the signal properties could change
Most of the samples show similar behavior
CDD is rising with the dose by approx %
and then stabilizes
But some samples show complete
degradation of signal and sharp rise
of the current under irradiation

12. Trapping
Trapping of the charge carriers is the main cause of low CCD.
Traps could be impurities, defects and other irregularities of crystalline structure.
Filling of traps produces space charge in the crystal and could give rise to polarisation effects
If we have several trapping levels with different activation energy, density and cross section, the dynamics of trapping will be really weird.
For now we have three methods of investigation of trapping levels: TSC, PL and Raman

13. TSC measurements TSC means thermally stimulated currents
The idea is that if we have charge carrier trapping in the diamond at room temperature, then heating the sample up could free trapped charge carriers and produce detectable current. The current peak position vs temperature gives the idea about energy levels of traps.
Setup consists of heating plate
electronic thermometer with
thermopair, HV supply and
picoammeter

14. TSC measurements Peak position corresponds to energy of about 0,8-1 eV
FAP21, 100 V
TSC “resets” sample, but the shape is similar for different samples

15. Linearity
A linear behavior over a large dynamic range is required for the calorimetry application.
We have checked this with a beamtest at CERN
Setup consists of a box
with diamond and preamplifier and
scintillator trigger box
Incident hadronic beam
with approx. 4 GeV energy
and intensity
from 105 / s
up to / 10ns

16. Linearity Signal at high intensity clearly visible w/o preamplifier
Good linearity for two samples over
the dynamic range of 102
(flux approx. 2x104-3x106 MIP/cm2
per 10 ns)
MIP on the same scale
FAP2.2
E 6-4

17. Linearity
Particle flux
Sample response
particle flux
E 6-4

18. PL Spectra Photoluminescence spectra of different samplesSi
There seems to be a correlation between detector properties and PL

19. Raman Spectra For diamond the Raman line is 1332.5 cm-1 at 300 K.
The presence of graphite in a diamond causes a strong line at 1580cm-1, microcrystalline graphite provides also a 1355cm-1 line. Amorphous carbon gives a broad peak between 1100 and 1700cm-1 E6
150 m
FAP5
40 m
FAP23
10 m

20. Conclusion Future work
Diamond sensors work and we have results both from CVD
and synthetic monocrystalline diamonds.
The sensors have good radiation hardness and linear response.
The CCD vs time behavior and dependence of CCD on dose
needs further investigation. .
Future work
Need to establish close contact with manufacturer, find
a method to characterize samples (PL, Raman) and get
interpretation from the material scientists
A study of radiation hardness with large doses is planned
A new batch of samples is ready for tests.

21. Diamond as detector Diamond have very high radiation hardness
(doses up to 10 MGy)
This makes diamond very promising material for the next generation colliders
It also have unique properties, such as low conductivity, high thermoconductivity, wide bandgap, chemical inertness

22. Testbeam pad1 pad3 pad2 pad4 diamond 10x10 mm scint. 5x 5 mm
Misalignment or change
in the beam profile which
wasn’t seen by scint.