E. Caroli(1), R. M. Curado da Silva(2), J.B. Stephen(1), F. Frontera(1,3), A. Pisa (3), S. Del Sordo (4) 1. INAF/IASF-Bologna, Italia 2. Departamento de Fisica, Univerisidade de Combra, Portugal 3. Dipartimento di Fisica, Università di Ferrara 4. INAF/IASF-Palermo, Italia 1
Why Hard-X ray polarimetry? Polarisation performance of CdTe pixel detectors Conventional analysis (imaging, spectroscopy and timing) of the high energy radiation from cosmic ray sources often provides two or more different models that successfully explain the observations; The combined measurement of polarization angle and degree of linear polarization can provide vital extra information to discriminate among the models. Solar Flares Solar flare emission is contaminated by thermal emission at lower energies and by lines above 1 MeV, the best energy range for polarimetry investigation from solar flares is from 0.1-1 MeV. Models for solar flares predict polarization levels as high as Interesting sources for focusing instruments 20 or 30% in this energy band. Gamma-ray Pulsars: hard Bursts X-ray. Since polarimetry the peak to energy understand of the GRB the spectrum extent is to narrowly which distributed gamma-ray at ~ photons 250 kev, are Compton related polarimetry to those is clearly at longer a requirement wavelengths for this topic. (e.g. Aa high polarisation level of polarization level >20% (>50%) is is expected from in GRB the CRAB if the prompt pulsar). emission, as recently suggested, is the result of a relativistic expansion of strongly magnetized electron-positron Soft Gamma-ray plasmas; Repeaters: in the one standard by-product (internal-external of magnetic shock) photon model, polarization splitting (50-500 levels of kev) about is 10-20% that are the expected reprocessed photons would exhibit a Other polarization possible sources level of Pulsars: ~25%. hard X-ray polarimetry to understand the extent to Other possible sources which gamma-ray photons are related to those at longer wavelengths (e.g. a polarisation Massive level >20% Black is expected Hole: from the the geometry CRAB pulsar). of the Soft the Gamma-ray accretion Repeaters: disk; For optical one byproduct disks of magnetic polarization photon levels splitting as (50-500 high kev) as is 30-60% that the are reprocessed possible photons while thin in would the exhibit optically a polarization thick regime level of lower ~25%. levels Massive (~5-10%) Black Hole: are the predicted. geometry of the the accretion disk; For optical thin disks polarization levels as high as 30-60% are possible while in the optically thick regime lower levels (~10%) are predicted. 2
The basic of Compton Scattering polarimetry Polarisation performance of CdTe pixel detectors! Q factor Klein-Nishina cross-section for linearly polarized photons: 2 1 2 2 d r0 - E' * & E' E 2 = + ( $ + ' 2sin 0 cos d. 2, E ) % E E'! Polarisation direction # /! " Q Q = = d# d# N N " " (" = 90)! d# (" = 0) (" = 90) + d# (" = 0)! + N N " ( ) Modulation Factor (Q) Q = 2 sin! E' E + " sin E E' 2! 3
2 mm Polarised Beam Energy range: 100 kev to ~1 MeV 2 mm 10 mm Advantages of a pixel detectors: Compton Scattering E E CdTe Pixel Dimensions!!Each element of the dectection plane is both a scatterer and a detector. Therefore all the sensitive area is used as polarimeter CdTe Pixel Matrix!The geometry of the detector select only events with a scattering angle close to 90º. For these scattering angles we have the best modulation factor. 4
The POLCA pixellated CdTe detectors Experiment at the beam line ID15 of the ESRF (July 2002) Thickness (mm) Pixel (mm 2 ) Bias (V/mm) Resistivity ($.cm) D.C. (na) 3.4 5 7.5 2.5 # 2.5 ~100 1 5 x10 9 20-40 Custom Multi-parametric system with16 shaping amplificator, coincidence logic, autocalibration system, ADC (12 bit) and parallel data output: " Coincidence time: 5!s " Trigger threshold: 25-30 kev " Dynamic range: 50 (20-1000 kev) 5
POLCA: Experiment at the beam line ID15 of the ESRF (July 2002) Corner pixel irradiaton Rotation of the 4x4 matrix 90 double events distribution symmetry Allows extrapolation to a 7x7 matrix 1 16 1 16 180 90 1 1 2 3 4 5 6 7 8 270 9 10 11 12 16 13 14 15 16 6
POLCA: Experiment at the beam line ID15 of the ESRF (July 2002) The single events are used to correct the double counts maps for pixel response non uniformity in order to reduce sistematics effects on the evaluation of the Q factor: Single events Double events N x/y = M x/y x N t /N sxy M = detected double events N t = Total events N s = Single events The red arrow give the polarisation plane direction 7
POLCA vs MonteCarlo POLCA experiment Matrix number (thickness) 1167/11 (3.4 mm) 1283/26 (5.0 mm) 1186/49 (7.5 mm) Polarimetric Q factor 100 kev 300 kev 400 kev 0.15 ± 0.051 0.46 ± 0.036 0.36 ± 0.091 _ 0.40 ± 0.12 0.33 ± 0.13 _ 0.39 ± 0.060 0.31 ± 0.065 POLCA 7.5 mm 300 kev Monte Carlo Matrix number (thickness) 1167/11 (3.4 mm) 1283/26 (5.0 mm) 1186/49 (7.5 mm) Polarimetric Q factor 100 kev 300 kev 400 kev 0.49 ± 0.005 0.47 ± 0.003 0.43 ± 0.003 _ 0.43 ± 0.002 0.39 ± 0.003 _ 0.40 ± 0.002 0.35 ± 0.002 MC 7.5 mm 300 kev 8
0,60 0,50 Central pixels suppression First order suppression Second order suppression Symmetric module Q Factor 0,40 0,30 0,20 0,10 0,00 Q Factor vs energy for a CdTe detector with 2x2 mm 2 pixels and 10 mm thickness 100 200 300 400 500 600 700 800 900 1000 Energy (kev) 9
The effect of photon incidence angle (i.e. focussing photons) on the scatter distribution E=200 kev Polarisation Direction! = 0 = 0 %% = = 10 5 %% = = 10 5 Non polarised 100 % polarized First Order Pixels Central Pixels First Order Pixels 10
0,60 0,50 0,40 0 and 2 inclined beams Q Factor vs energy for a CdTe detector with 2x2 mm 2 pixels and 10 mm thickness 0,30 Q Factor 0,20 0,10 0,00 100-0,10 200 300 400 500 600 700 800 900 1000-0,20-0,30-0,40 Energy (kev) Y polarised beam 0 Y polarised beam 2 Unpolarised beam 2 X polarised beam 2 In the proposed configuration the inclination of photon coming out from the laue lens is <2º 11
CZT pixel detector with Laue wide band lens Monte Carlo Parameters:! PSF: 2 cm Ø (FWZR) independent from Energy (first order approximation)! Energy band: 2 band for LE lens (60-120 kev, 120-200 kev) and 3 energy band for the HE lens (150-250 kev, 250-400 kev, 400-600 kev)!spectra: Crab like spectra (E -2, ph/cm 2 /s/kev)!background: LEO, High inclination (ISS), conservative hypothesis! Detector: CZT (planar square pixel array)! LE A eff HE A eff 12
LE lens scattered events maps Detector parameters: pixel = 1x1 mm 2, thickness = 10 mm _E = 60-120 kev _E = 120-200 kev 13
HE lens scattered events maps Detector parameters: pixel = 1x1 mm 2, thickness = 10 mm _E = 150-250 kev _E = 400-600 kev 14
LE lens scattered events maps Detector pixel = 1x1 mm 2 ; _E = 120-200 kev Thickness = 3 mm Thickness = 10 mm 15
Polarimetric Perfomance Minimum Detectable Polarisation MDP ( ) 100% = n A.!. S " F. Q Polarimetric Sensitivity 100 A.!. S T F + B n$ B 2 F min = ph /( s! cm! kev ) #! Q AT" E 100 A = geometric sensitive area (cm 2 ) _ = double events efficiency S F = source flux (ph/s/cm 2 ) B = background (c/s) Q 100 = modulation factor for 100% polarised photons T = observation time (s) _E = energy band width (kev) 16
100 10mm 3mm MDP (%) 10 LE Laue Lens (120-200 kev) 1 1,0E+03 1,0E+04 1,0E+05 1,0E+06 Observation Time (s) 17
100 150 to 250 kev 250 to 400 kev 400 to 600 kev MDP (%) 10 1 HE Laue Lens 10 mm thick CZT pixel detector 1,0E+03 1,0E+04 1,0E+05 1,0E+06 Observation Time (s) 18
Modulation Factor vs Pixel Size Polarisation performance of CdTe pixel detectors 0,330 2,5 0,325 2 Q factor 0,320 0,315 0,310 1,5 1 MDP (%) 0,305 _E= 120-200 kev 0,5 0,300 0 0,0 0,5 1,0 1,5 2,0 2,5 Lateral Pixel Size (mm) 19
Q factor optimisation I Polarisation performance of CdTe pixel detectors E = 150-250 kev Pixel = 1x1 mm 2 Thickness = 10 mm First estimation value The structure is due to the large scale pixellisation and to the criteria for acceptance (inside a) of any pixel inside the integration over the wide detector sector. The second plot is the effective Q factor that take into account the fraction of double events that are used in the calculation. 20
Background reduction methods: Polarisation performance of CdTe pixel detectors Q factor optimisation II " Accept only events that have one hit inside the PSF of the pointed source. " Reject event in which the two energy deposits are not compatible with the Compton kinematics. " Reconstruction of the photons incidence cone using Compton kinematics in order to accept (or reject) only photons for which the incidence cone is compatible with the instrument aperture (lens). The efficiency of these methods is very sensitive to the performance and the characteristics of the focal plane detector: energy resolution, spatial resolution (pixel size in XY), thickness (Z axis). The Doppler broadening should be taken into account for the photon incidence cone direction and aperture uncertainties evaluation (between 1 and 10 in CZT detectors with the considered geometry depending on energy. 21
Expected Polarimetric performance Source Polarisation Level 50% 10% 2% Sensitivity for a polarized source (10 6 s, 3_) Ph/cm 2 /s/kev _E = 60-600 kev 5#10-8 ( ~0.5 mcrab) 2.5#10-7 ( ~2 mcrab) 1.3#10-6 ( ~10 mcrab) MDP for a 100 mcrab source (3_) Time 10 4 s 10 5 s 10 6 s _E = 60-600 kev 10% 3% 1% 22
The study is in progress... Polarisation performance of CdTe pixel detectors POLCA II Experiment 40x40 mm 2 CZT detector 2.5x2.5 mm 2 pixel, 16x16 pixels 5 mm thick ASIC readout electronics Monte Carlo Development Implementation of a more symmetric configuration (e.g. Hexagonal) Polarimetric response on the range 100-700 kev Q factor vs the angle between polarisation vector and detector SNR improvement using background reduction methods based on Compton kinematics Better modelling of the Laue lens PSF (vs E and off-axis sources) 12-15 axisseptember 2005 23
Conclusion... The implementation of the Laue lens technique can provide an improvement in sensitivity of about 2 orders of magnitude with respect to current hard X and soft gamma ray telescopes. A detector that can exploit high polarimetric performance together with good imaging and fine spectroscopy in the focal plane of this kind of telescope can attain unprecedented performance...why not add this challenging feature to the same detector we will use for spectroscopy, imaging and timing? 24
Detection efficiency vs energy for a CdTe detector with 2x2 mm 2 pixels and 10 mm thickness 30 90 Double events efficiencies (%) 25 20 15 10 5 Double events/detected photons Double events/sent photons Total detection efficiency 80 70 60 50 40 30 20 10 Total detection efficiency (%) 0 0 100 200 300 400 500 600 700 800 900 1000 Energy (kev) 25