Simulation of the Collection Properties of CdTe Strip Detectors A.Cola' , F.Quaranta', E.Caroli2, W.Dusi2 and E.Perillo3 'CNR-IME, Via Arnesano, 73100, Lecce, Italy 2CNR-TESRE. Via Gobetti 101,40129 Bologna, Italy 3Dip. di Fisica. Universita' Federico IT and sezione INFM, Napoli, Italy Abstract CdTe is an attractive. material for X and Gamma-Ray detectors but the poor transport properties of holes affect the performance by introducing a low energy tailing in the observed spectra. .A possible solutjon to thls problem is to optimize the electrode- geometry, for example by reducing the dimension of the anode.,with respect to., the cathode. In this way the charge signal. in the external circuit is mostly due to the electrons moving' ,towards the, anode that is where the , weighing field becomes localized. The optimization of the electrode geometry requires a ,numerical -, analysis as .the problem is complicated by stochastic trapping and detrapping processes, which are difficult to be treated analytically. In this 'work we present a numerical simulator based on a finite difference numerical method. (which- follows the weighing field approach) and on a MonteXarlo procedure which is able to analyze, in two -dimensions, the effect of different electrode configurations: single strip, multiple strip, a single ' strip with a lateral extended cathode 'and, for comparison;.ttie uniform geometry. 'The ,results are analyzed in terms of the maps of the local charge collection efficiency and their histograms,. equivalent to the spectra due to high energy x-rays. ,(. ,. :'. I. INTRODUCTION ~ , I , . , .. , . . . . . . '. 1 I . . . . The scope of our-collaGoration is td develop an innovative position. "sensitive spectrometer -for , hjgh . 'energy' x-ray, astronomy[l]. -It shod$ .be ',of highLefficiency and". good' spectral respcmse"over' a' wide energy-range (-10 'keV 1 MeV).' Vertically oriented detectors should' be placed. close each"other'.in 'oider to cover an 'appieciable'(>lOO' cm') area. stkting from detectors with'sectidns of few mm2. ' ' For this application, Cadmium Telluride (CdTej is..weli , suitable . as . detecting material :to, high ,efficiency. and spectroscopic properties. However, it is known that trapping, of holes seriously limits the spectra energy resolution and,the overall detector performance: 'Different methods. have -.been proposed and..demonstrated to'., mitigate .. and, eventually, overcame this limit: including i. 'electronic pulse discrimination[2-43 and unipolar 'devices [5-8]. ' The 'last methods are based on a proper geometry of the detector and of the 'electrical contacts, optimized to. minimize the effect of hole trapping. If the two contacts, entirely .cover the two opposite surfaces.of ,the detector,,the electric field is uniform, and .'the .relative- collection '.of each tfpe of carriers .only depends on the mean free path, which, for electrons;,is much ! longer than for holes. Reducing the dimension of the anode c 4 , . t : , I ! ,- . increases the weighing field beneath this contact and it favors the electron collection. The critical dimension at which this sinall pixel effect starts is the thickness of the detector. The. true unipolar 3D device., as first realized by Malm [S],'consists of an hemispherically shaped detector where'one contait is the outer surfaci of the hemisphere and the other, the anode, located at the center of the flat surface, has a much smaller radius. In a 2D2detectorgeometry, the corresponding improvement of electron collection is achieved with the thin strip; moreover, the 2D version of the true unipolar detector is represented by the hemicylindrical geometry. With detectors of parallelepiped shape this geometry can be approximated by extending the cathode on the lateral sides [9]. However, the realization of these single-strip devices present some technical disadvantages, especially if imaging performance are required and a large number of detectors should '.be independently realized and assembled together. In this case, a multi-element detector can be preferred. cqllection properties of CdTe strip detectors with different contact configurations. To this .scope -we have ,developed a numerical simulator . based on finite difference numerical method and. on Monte-Carlo procedure. After :a. brief description of the principles of the ,method, we will present, and discuss the results of the simu1,ations .in: two: sections. The first section refers .to .different contact cpnfigurations: one dimensional, single strip, single. strip with extended.,cathode, and multiple strips. The other section investigates the case of multiple strips by varying the relevant geometrical parameters. The aim of this work is to study and to compare the charge ' 11. PRINCIPLES OF THE METHOD The model is two dimensional: the, Laplace equation is solved in a x-y section of the detector over a numerical grid with a constant-mesh, lOpm thick, finite-difference iterative method. In this way the potential P(x,y) and the electric field E(x,y) distributions are found; the weighing field Ew(x,y) is calculated by grounding all the electrodes with. the exception of the investigated anode which is raised to unit potential. If only one anode exists, E(x,y) is !proportional to E,(x,y); otherwise, as in the case of multi-strips, these electric fields differ each other. Hence, .we can evaluate the elemental charge induced in the external ,circuit, by a fharge q moving through a small displacement ds, as: , , , (1) d9(X,Y) = - 9 ds@,y).Ew(x,y) 4-97 0-7803-6503-X/01/$10.00 2001 IEEE