Preparation of High Purity CdTe for Nuclear Detector: Electrical and Nuclear Characterization

Available online at www.sciencedirect.com

ScienceDirect Physics Procedia 55 (2014) 476 – 484

Eighth International Conference on Material Sciences, CSM8-ISM5

Preparation of High Purity CdTe for Nuclear Detector: Electrical and Nuclear Characterization A. Zaioura*, M. Ayoubb, A. Hamiéc A. Fawaz d, M. Hage-alie a

Department of Physics and Electronics, Faculty of Sciences I, Lebanese University, Hadath, Lebanon b Kromek, Thomas Wright way, TS21 3FD, Sedgefield, UK c MCEMA, EDST, Faculty of Agriculture, Lebanese University, Beirut, Lebanon d Laboratoire 3S, Institut National Polytechnique de Grenoble, France e Centre des Recherches Nucléaires, Iness , 67037 Strasbourg, Cedex 2, France

Abstract High purity crystal with controllable electrical properties, however, control of the electrical properties of CdTe has not yet been fully achieved. Using the refined Cd and Te as starting materials, extremely high-purity CdTe single crystals were prepared by the traditional vertical THM. The nature of the defects involved in the transitions was studied by analyzing the position of the energy levels by TSC method. The resolution of 4.2 keV (FWHM) confirms the high quality and stability of the detectors: TSC spectrum was in coherence with detectors spectrum with a horizontal plate between 0.2 and 0.6 eV. The enhancement in resolution of detectors with a full width at halfmaximum (less than 0.31 meV), lead to confirm that the combination of vacuum distillation and zone refining was very effective to obtain more purified CdTe single crystals for photovoltaic or nuclear detectors with better physical properties. © 2014 Elsevier B.V. This is an open access article under the CC BY-NC-ND license © 2013 The Authors. Published by Elsevier B.V. (http://creativecommons.org/licenses/by-nc-nd/3.0/). Selection and/or peer-review under responsibility of CSM8-ISM5 Peer-review under responsibility of the Organizing Committee of CSM8-ISM5 Keywords: CdTe, THM, purification, defects, TSC,Ȗ-performance

1. Introduction

Cadmium Telluride (CdTe) solid-state detectors can provide high quantum efficiency with reasonably good energy resolution and can operate at near room temperature; its high stopping power and good energy resolution, are promising devices for the next generation of hard X-ray and gamma-ray astronomy applications. Unfortunately, such semiconductors are limited in their spectroscopic performance essentially by the poor whole transport properties. Thus, several techniques have been developed or are

Corresponding author. Tel.: (961) 3 768 878; fax: (961) 5 461 496. E-mail address: [email protected].

1875-3892 © 2014 Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/). Peer-review under responsibility of the Organizing Committee of CSM8-ISM5 doi:10.1016/j.phpro.2014.07.069

A. Zaiour et al. / Physics Procedia 55 (2014) 476 – 484

being developed in order to improve this performance. Nevertheless, some points remain difficult to solve in this situation; behavior of native defects and impurities, their interaction has not been fully clarified. Different growth techniques were developed in the past for the preparation of CdTe crystals. All the growth techniques using quartz sealed-tubes (THM, standard Bridgman) suffers from the nonhomogeneity of the material due on the one hand, to the thermal traveling heater method (THM) growth conditions, which induces the formation of Te precipitates owing to the retrograde CdTe phase diagram [1] and, on the other hand, to the impurities introduced during the growth. Materials grown by this growth technique are reasonably limited in dimension to diameter less or equal to about 100mm, it means that large dimensions, which are a demand for the future, are not attainable by these growth methods. The primary goal on which every growth technique is focused is the production of defect-free, high-resistivity material, in sufficient quantities to ensure high yield and low cost. The measurements on both low and high resistivity samples suggest that shallow levels and deep levels are related to intrinsic defects, so confirming their crucial role in the compensation mechanism providing the high resistivity in CdTe crystals [2]. Room temperature semiconductor detectors, particularly CdTe are confirmed as the best choice for imaging applications, their stability is better than expected; the most important requirements for this application are high resistivity material with high photosensitivity, very short response time and large carrier lifetime [3]. More laterally-uniform crystal can be grown with a transverse magnetic field because the high-velocity parallel layer located adjacent to the crystal-melt interface reduces the lateral segregation. A more axially-uniform crystal can be grown with an axial field with a strong magnetic field [4].Technical problems posed by the High Pressure Bridgman technique were avoided by grown by encapsulated (B2O3) obtaining high-purity semiinsulating CdTe crystals [5]. A new method to grow large dimension CdTe based on solvent evaporation from Te-rich solution made of cadmium and tellurium in open tube and proceeds in a crucible maintained at a constant temperature; resistivity was about 1010 Ƿcm [6]. It was shown that the electrical properties of the grown crystals are very sensitive to the procedure employed for adjusting the stoichiometry deviations: the native defects are strongly correlated with material stoichiometry deviation and their electron levels and densities in the forbidden gap play a crucial role in deep-shallow level compensation. This hypothesis seems to be confirmed by the results of annealing treatments at 600°C for 2 h on CdTe crystals, the high resistivity failed to its low values [7]. Any quality improvement is especially made difficult because impurity effects on the electrical properties are not easily separated from the effects of crystalline defects and off-stoichiometry. This unsatisfactory quality is further worsened when dopants in high concentration are added induces the formation of electrically active defects and complexes which tend to compensate the material, thus allowing the attainment of high electrical resistivity (ȡ§ 107 to 1010 Ƿcm). In the cases tested, the BiParametric Spectrum BPS improved detection efficiency (75%) without degrading energy resolution (± 6.5% even for monolithic detectors) [8]. The large atomic number of the CdTe promotes a high registration efficiency of J-radiation (compared, e.g., with silicon). At the same time, with the increase in the atomic number the registration efficiency for a given energy range of gamma-radiation varies over a wider range of values. However, this energy dependence can be successfully leveled out (with an accuracy of 71%) both by electronic signal processing and by application of absorbing filters of sophisticated designs [9]. The characterization of high purity CdTe single crystals, grown by the improved PVT growth method with a Cd reservoir, has been done using photoluminescence, photo excited cyclotron resonance (CR) and infrared transmission (IR) methods. The results demonstrated that the crystals are not only extremely chemically pure but also structurally pure [10]. In the present paper, extremely three high – purity CdTe single crystals with high resistivity were prepared: ABl : Te1 was purified by horizontal zone refining AB2 : Te2 was purified by vacuum distillation AB3 : Te3 was purified by combination of both processes Growth of single crystals was achieved in same conditions of THM (Traveling Heater Method). Quality of three types of detectors and measurements of TSC was detailed and discussed, in order to

477

478

A. Zaiour et al. / Physics Procedia 55 (2014) 476 – 484

clarify the purification effect on the electrical properties of the samples and to choose the best purification process leading to CdTe crystals with good efficiency. 2. Experimental procedures 2.1. Growth material Table 1: Mean concentration of 21 impurities measured in each ingot

Elements Ag Al As B Ca Cr Cu Fe In Mg Mn Na Ni Pb Pd Se Si Sn Ti V Zn

AB1 Distilled Te d0.01 d0.03 d0.07 d0.1 d0.15 d0.03 d0.02 d0.03 d0.1 d0.1 d0.06 d0.1 d0.05 d0.02 d0.01 d0.1 d0.1 0.1 d0.1 0.1 d0.03

AB2 Zone refined Te d0.02 d0.1 d0.1 d0.1 d0.02 d0.04 d0.02 d0.1 d0.1 d0.05 d0.07 d0.2 d0.05 d0.02 d0.1 0.5 0.2 0.08 0.2 0.1 0.06

AB3 Distilled + Zone refined Te d0.01 d0.01 d0.05 d0.2 d0.08 d0.01 d0.05 d0.1 d0.4 d0.1 d0.06 d0.1 d0.1 d0.09 d0.03 0.2 d0.1 0.1 0.01 0.1 d0.03

Cd source material of high purity (6N cominco) was used with the three samples of purified tellurium, in required proportions. The THM using tellurium as solvent has been successfully applied to the growth of single crystals CdTe. In addition to the specific advantages inherent the low temperature growth, this method leads to an effective purification. Chlorine doped crystals grow from tellurium solution. High quality crystals and detectors have been made in this way. However, the low growth temperature limits the growth rate to typically few millimeters a day. The slow growth rate is the main disadvantage of this method. Three CdTe ingots are grown by THM (traveling heater method) in Te – rich CdTe liquid solution with one zone heating. All stoichiometric CdTe single crystals were grown with the same apparatus. The growth rate was 3mm/h and the post – growth cooling rates were about 4oC/h. The temperature was controlled with in ± 0.2 K throughout the growth process. According to the poor heat conductivity of CdTe, the low applied temperature profile was made to have a steep temperature gradient in the crystallization end, handled precaution to prevent the thermal perturbation due to the high melting point of CdTe. Mean concentration of 21 impurities are measured in each ingot as shown the table 1.

A. Zaiour et al. / Physics Procedia 55 (2014) 476 – 484

2.2. Detectors The characterization of the detectors involved the measurement of some key parameters: the leakage current, the energy resolution, the photopeak efficiency, the stability of gain and the energy distribution of trapping levels in the band gap. All these parameters were measured prior to and again after each irradiation. Indeed, the performance of the elaborated detectors depends not only on the presence of impurities but also on the concentration of the structural defects and complexes with impurities. To reduce the structural defects, such as large angle grain boundaries and the densities of dislocation and low angle grain boundary, high resistive CdTe may be produced by further annealing at 600oC in a closed quartz ampoule, with an argon gas atmosphere. However, pure CdTe is not gamma active. Addition of a donor dopant activates the material. Detectors with 3 ×5mm2 electrodes and inter-electrode distance ranging from 1.0 to 2.5 mm, in steps of 0.6 mm, were fabricated . Two platinum electrodes (50-80nm thick) were deposited, by means of the electroless technique, on the two opposite 3×5 mm2 faces. Bias voltages were selected through preliminary tests and correspond to an average electric of 500 V/cm. All the measurements were performed at room temperature. Anode pulses were derived, through a decoupling capacitor, via commercial charge-sensitive preamplifiers; the opposite side was connected to ground. The pulses were shaped and recorded by a standard nuclear electronic chain.

Figure 1:Standard measurement scheme for gamma detector characterization

A further improvement could be made by modifying the detector size; the present electrode area makes the edge effects to be significant; enlarging the electrode area, probably by the use of a support to avoid problems of crystal breaking, could improve the observed performance. The quality of detectors was tested with 57Co source. Electrical characterizations were achieved by TSC in the aim to determine the best way to obtain high purity CdTe. Standard measurement scheme for gamma detector characterization was showed in Fig 1. The experimental measurement of this spectroscopic characteristic was performed by using low energy y ray sources such as 241Am and 57Co. 3. Results and discussion:

479

480

A. Zaiour et al. / Physics Procedia 55 (2014) 476 – 484

3.1. Electrical measurements: Electrical measurements appear more efficient to characterize CdTe ingots. In fact, the performance of elaborated detectors depends not only on the presence of impurities, but on the concentration of the structural defects and complexes with impurities. Helpfully electrical spectrometric methods like TSC (transient stimulated current), allow to distinguish the role of each type of defects and complexes with sensitivity up to 105. A possibility for obtaining CdTe material with different properties is to employ an amphoteric dopant, which for low concentrations acts as a donor impurity, compensating the Cd vacancies and leading to high resistivity crystals suitable for X- and gamma-ray detectors. For high dopant concentrations, due to the amphoteric behavior, the impurity acts as an acceptor, changing the conductivity to p-type and diminishing the resistivity. Among the possible dopants, recently Bi has been reported as useful for obtaining semi-insulating CdTe. Mostly, it could be expected that Bi would occupy a Te position, but high resistivity CdTe crystals have been reported for low Bi concentrations, suggesting an amphoteric behavior of Bi, occupying Cd sites leading to compensation of Cd vacancies [11]. If this is true, Bi could be a very interesting dopant, with a strong effect on the physical properties of CdTe. Finally, when quantitatively discussing the distribution of impurity elements (Fluor, Bromine, Sodium) within the CdTe active layer, one has to bear in mind that SIMS analysis on its own is not able to confirm whether the impurities recorded have any electrical activity in the CdTe layer with a direct impact on the performance of the final devices based on these structures [12]. As stated above, all the structures reported in this study had their CdTe layers grown using (•6 N) purity CdTe source material, meaning that the total amount of impurity elements to be anticipated from a starting material of this grade is about 2×1016 cm-3. This finding emphasizes the fact that even before undergoing any processing step, the structures have already some Si, O, Br and Cl coming from impurity sources other than the CdTe starting material. 3.2. TSC: Thermally stimulated current (TSC) method has been used to identify the electrical defects into the band gap. The principle of this method is based on the illumination of the sample by 0.63 Pm wave length of laser diode following by a cooling down the temperature to 76K. A bias voltage is applied on the sample during both the illumination and the decrease in temperature. The sample is then heated from 76K to 350K with a slow heating ramp. The leakage current is recorded and then plotted with the temperature variation to give finally the TSC spectrum. Three samples have been chosen from the bottom, middle and top of the ingots and for each purification in order to carry out the TSC measurements. The figures 2(a) (AB1), 2(b) (AB2), and figure 3 (AB3) show the TSC spectrum for the three purification methods. Different defect bands have been identified for these three methods. These defect bands have been located at 0.15eV, 0.3eV, 0.5eV and 0.75eV or midgap. - 0.15eV band: This band is present for the three purification methods. Two defects are identified for the AB1 and AB2 methods with energy 0.16eV and 0.19eV. While for AB3 only one defect is identified at 0.12eV. This band of defects is always present in CdTe:Cl and it is attributed to an acceptor complex. Meyer et al. [13] have identified the acceptor complex (VCd-ClTe ) at 0.12 eV using photoluminescence and optically detected magnetic resonance (ODMR) techniques. Fiederle et al. [14] found the same complex in the 0.14–0.17eV band. In fact, this defect level, usually called the A center, plays a significant role in the compensation process of CdTe:Cl as it seems to be the shadow of a deep donor level, which neutralizes the native VCd acceptor defects. - 0.3eV band: Four defects are identified with respective energies 0.23, 0.3, 0.36 and 0.38eV. Three of these defects except 0.36eV are present for the AB1 samples while two defects are identified for AB2 ones and two with different defect energy for the AB3 samples. One can notice that the defect 0.3eV is present for all AB3 samples while the 0.38eV is present only for the AB3 top sample and it disappears for the others middle and bottom samples to be replaced by a lowest energy defect located at 0.23eV as shown the figure AB3. Two defects at 0.3eV and 0.36eV have been only identified for

481

A. Zaiour et al. / Physics Procedia 55 (2014) 476 – 484

the AB2 samples. s The in ntensity of thiss defect band increases i from m the bottom too the top sampples for the AB2 and AB3 meethods. In facct, the 0.3eV defect band is i attributed to t a metal im mpurity Cu, Ag, Au, Pbb, Bi) [15]. Othhers authors atttribute this deefect band to thhe TeCd contaminaation such as (C anti-site level [16] possibly introducced during thee growth proccess. It seems that with thee AB3 purificatioon method we reduce r the mettallic impurity contaminationn and so the ellectrical defectts. This can influennce the electriccal properties of o the material and also the gaamma ray perfformance. - 0.5eV bannd: Four defects have been identified i at 0.52, 0.56. 0.622 and 0.62eV. All these defeects are present forr the AB1 sam mples except thhe 0.62eV defect for the midddle and the topp ones while foor AB2 and AB3 only three deffects are identiified. The 0.566eV and 0.65eeV defects are present for thhe AB3 3 middle samplle where the 0.52eV 0 is identtified in place of 0.56eV, whhile for samples exxpect the AB3 the AB2 samples two deefects are foundd at 0.52eV and 0.62eV. Thiss defect band is generally attrributed V [17] or a doubble charge donnor Cdi with ennergy 0.64eV. Höschl H to a Cdi doonor defect loccated at 0.56eV et al. havee found a defecct at 0.6eV and assigned it to a VCd [18]. (a)

(b)

Fig 2: the TSC sppectrum for the thrree purification meethods. (a) TSC Spectrum S for the samples s taken from m bottom, middle, and top of ingots for the horizontal zone refining purrification (AB1)). (b) TSC Spectrum S for the samples taken froom bottom, middlle, and top of inggots for the vacuuum distillation purrification (AB2)).

482

A. Zaiour et al. / Physics Procedia 55 (2014) 476 – 484

Fig. 3: TSC Speectrum for the saamples taken from m bottom, middle, and top of ingoots for the combiination of AB1 and a AB2 purification proceesses (AB3).

m band: two t defects havve been identiffied at 0.72eV and 0.78eV. These T two defeects are - 0.7eV or midgap present foor the AB1 sam mples. For AB B2 the 0.7eV defect d is well present p for thee bottom sampple and should be present for th he others two ones o but the measurement m has been done and stopped att room me case for thhe AB3 samplees. This band of o defect is gennerally assigneed to a temperaturre. It is the sam complex defect d (VCd – metallic m or others impurities). In fact, the prresence of a deeep level locateed near the mid gap g necessarilly involves thhe Fermi-level process whicch takes placee in semi-connductor materials such as CdT Te:Cl. This defect d band plays p an impoortant role onn the shallow w level nd thus on the electrical propperties of the materials m in parrticularly on thhe bulk compensattion process an resistivity [17] and [19]. n of deffects are foundd for the AB1 samples s and leess for both AB B2 and We can nootice a higher number AB3 ones. The T intensity of o the defects varies betweeen the differennt purification methods as well w as between the chosen c samplees along the poosition in ingoot. The best gaamma ray perrformance in teerm of energy resolution and efficieency was obtaiined for the puurification methhod AB3. Thiss can be explainned by od to reduce or o remove thee metallic impuurities defectss, to neutralizee some the efficiencyy of the metho electrical defeects, to minimize the number of the defects and thus to haave a good com mpensation proccess of the materials. 3.3. Detectorss quality: The high performance of o the detectors demands coonsisting in: (i) ( high resistiivity of the material m g signal-to--noise ratio andd (ii) high carrrier mobility-liifetime producct (μIJ) allowingg good leading to a good collection of photo-generatted charge. Coontrary to the first case, whhere the comppensation of shallow 1 cm-3 is presuumed, the secoond task requirres the defects by deep defects witth a high enough density 1015 ping centers, thhe density of thhose should deecreases below 1013 cm -3 at leeast. significant redduction of trapp Room tem mperature semiconductor detectors are chharacterized byy a noticeablee trapping of charge carriers. Thiss makes the photopeak mucch more asym mmetric than thhe scintillator and the high-purity semiconductoor detectors. Fo or this reason the experimenntal data have been fitted byy using two diifferent analytic functtions. The firstt serie of data has been obtaiined by fitting the entire phootopeak with a GMG curve (that is a Gaussian fun nction modifieed by means off a distortion paarameter), whiile the second serie s is o by using a purely Gausssian function. the best fit of the right side only

A. Zaiour et al. / Physics Procedia 55 (2014) 476 – 484

The large atomic number of the CdTe constituents promotes a high registration efficiency of Ȗradiation (compared, e.g., with silicon). At the same time, with the increase in the atomic number the registration efficiency for a given energy range of Ȗ-radiation varies over a wider range of values. In other side, undoped material is not gamma sensitive than active and only n-type doped material is active. Under-doped detectors operate reasonably at high bias voltage, which is about three times as much as optimized detectors. Properly doped detectors have good spectral resolution. However, the resistivity of the n-type material is lower than the p-type detectors made by other growth methods, and the spectral resolution is current limited. Over-doped detectors have very high mobility-lifetime product (ȝIJ), but the performance is very poor since the electrical resistance is very low. For these detectors an appropriate selection of the width of the irradiated region can give the possibility to increase the active detection area with respect to the best charge collection region, still maintaining prefixed values for the energy resolution, in order to match the requirements of particular applications. Dependence of the parity of starting materials and the physical properties of grown CdTe was investigated by performing three types of CdTe detectors: their spectrometric performance at room temperature was shown in figures 3a, 3b and 3c, after exposing under 57Co radiation source. The measured leakage current is ~10-8A under 100V of bias voltage. In all samples, a light decreasing of detectors quality appears with position from the head to the bottom of each ingot. As expected, the detectors showed more efficient counting in 122 keV and 136 keV, however their count rate was found dropping slowly over a period of hours. This kind of behavior was attributed to the polarization (under bias) causing a reduction thickness of the detector sensitive volume [21]. Samples prepared from AB1 (figure 3a) produced a resolution of 5.6 keV (FWHM); the apparent peak confirmed its lowest value by comparison with AB3 detectors samples (Fig 3c).

Figure 3: 57Co J-ray performance for the samples taken from the three different purifications methods: a) AB1, b) AB2 and c) AB3

AB2 samples (Fig 3b) showed an improvement in resolution of 4.8 keV, trapping effects were most important. Contribution of impurities in AB1 was due to high values of segregation coefficients of impurities in CdTe (K • 1). Indeed, it comes from elements having the same vapor tension of tellurium for AB2, with more stability in comparison with AB1 [22]. The high quality and stability was shown in AB3 detectors (Fig 3c), with a resolution of 4.2 keV (FWHM) evident from the separation of 122 and 136 keV peaks. Deceased trapping was clear in the side of low energies, TSC spectrum was in coherence with detector spectrum with a horizontal plate between 0.2 and 0.6 eV. Taking into account the presence of the confusion between impurities and structural defects, the decreasing of concentration levels in TSC and the high quality of nuclear detectors, consist a significant purity of CdTe ingot which mean the successful purification of the starting materials.

483

484

A. Zaiour et al. / Physics Procedia 55 (2014) 476 – 484

4. Conclusion: Three purification processes for tellurium were used to grow and characterize high purity CdTe single crystals. The constant proof of resistivity along the position in ingot, the enhancement in resolution of detectors, especially for AB3 ingot lead to confirm that one purification process, especially the zone refining process, with was not sufficient to improve purity and quality of CdTe detectors; the combination of vacuum distillation and zone refining was very effective to obtain more purified CdTe single crystals for photovoltaic or nuclear detectors with better physical properties. Despite the high quality achieved in CdTe single crystal, a further improvement of the purity of starting materials was still necessary toward an interesting CdTe material with electronic grade for nuclear and electronic applications with good efficiency. Further studies are also needed to attribute an origin to these defects and to fully understand their role on the resulting electrical properties of the material. More statistics and improvements in the crystal production and treatment and changes in the detector size are required to understand the device behavior better and to select an optimum combination of crystal characteristics, size and applied electric field. References: [1] J. H. Greenberg, Journal of Crystal Growth, 1996; 161: 1–11,. [2] N. Armani , C. Ferrari, G. Salviati, F. Bissoli, M. Zha, L. Zanotti Materials Science and Engineering 2002; B91–92: 353–357 [3] Y. Eisen, Nucl. Nuclear Instruments and Methods (A) 1992; 322: 596. [4] Martin V. Farrell, Nancy Ma, International Journal of Heat and Mass Transfer 2004; 47: 3047–3055 [5] M. Zha, A. Zappettini,, F. Bissoli , L. Zanotti , V. Corregidor , E. Dieguez Journal of Crystal Growth 2004; 260: 291–297 [6] B. Pelliciari, F. Dierre, D. Brelliera, B. Schaub; Journal of Crystal Growth 2005; 275: 99–105 [7] M. Zha, F. Bissoli, L. Zanotti, C. Paorici; Journal of Materials Processing Technology 2003; 143–144:425–429 [8] J. Rustique, G. Sanchez . Nuclear Instruments and Methods in Physics Research A 2001; 458: 297-309 [9] A.V. Rybkaa,*, L.N. Davydova, I.N. Shlyakhova, V.E. Kutnya, I.M. Prokhoretza, D.V. Kutnya, A.N. Orobinskyb ,Nuclear Instruments and Methods in Physics Research A 2004; 531: 147–15 [10] B. Yang, Y. Ishikawa, Y. Doumae, T. Miki, T. Ohyama, M. Isshiki, J. Crystal Growth 1997; 172: 370 [11] M.Isshiki, M. Sato andK.Masumoto, J . of Crystal Growth 1986; 78: 58. [12] M. Emziane , K. Durose , D.P. Halliday , N. Romeo , A. Bosio, Thin Solid Films 2006; 511 – 512: 66 – 70 [13] B. K. Meyer,W. Stadler, and D. M. Hofmann, Journal of Crystal Growth , 1992; 117: 656–659 [14] M. Fiederle, D. Ebling, C. Eiche, D. M. Hofmann, M. Salk, W. Stadler, K. W. Benz, and B. K. Meyer, Journal of Crystal Growth, 1994; 138: 529–533 [15] M. Hage Ali and P. Siffert, Nucl. Instrum. Nuclear Instruments and Methods in Physics Research A, 1992; A322: 313–323, [16] M. Samimi, B. Biglari, M. Hage Ali, J. M. Koebel, and P. Siffert, Phys. Status Solidi, 1987; A100: 251–258, [17] M. Ayoub, M. Hage-Ali, J. M. Koebel, F. Klotz, C. Rit, R. Regal, P. Fougeres, and P. Siffert, IEEE Trans. Nucl. Sci., 2003; 50: 229–237, [18] P. Höschl, R. Grill, J. Franc, P. Moravec, and E. Belas, Materials Science and Engineering B, 1993; 16: 215–218, [19] A. Zumbiehl, S. Mergui, M. Ayoub, M. Hage Ali, A. Zearrai, K. Cherkaoui, G. Marrakchi, and Y. Daric Materials Science and Engineering, 2000; B71: 297–300, [20] V. Nagarkar, M. Squillante, G. Entine, I. Stern, D. Sharif, Nuclear Instruments and Methods in Physics Research A 1992; 322: 623 [21] M. Samimi, B. Biglari, M. Hage – Ali, J.M. Koebel and P. siffert, Nuclear Instruments and Methods A 1989; 283: 243 [22]A. Zaiour, K. Zahraman, M. Roumie, J. Charara, A. Fawaz, F. Lmai, M. Hage-Ali, Materials Science and Engineering B 2006; 131: 54–61