Full-area Illumination Research Articles

Comparison of line-wise pl-imaging and area-wise pl-imaging

In this contribution we compare the results of a line-scanning camera and line-wise illumination based PL-imaging system (Line-PLI) to the results of a “classic” PL-imaging system featuring a matrix camera and a full area illumination of the wafer.We investigate the impact of the different injection conditions which result from the significantly higher irradiance user in the Line-PLI system. We derive a formula which can easily be used to predict how high the irradiance may be without causing the lifetime to be limited by intrinsic silicon recombination. We compare the image contrasts caused by regions of different recombination activity and we compare the robustness of the method, which calibrates the PL-images to local effective lifetime using photoconductance (QSSPC) measurements.As a result it can be said that contrasts between regions of high and low recombination activity are more pronounced using the investigated Line-PLI setup, which qualifies it excellently for the visualization of process inhomogeneities causing defects on solar cell precursor. We calculate that the irradiance of the investigated setup is still low enough to differentiate between regions of different Shockley-Read-Hall recombination if the effective sample lifetimes are roughly below 250 µs. Moreover we derive an uncertainty of 24 % on the absolute values of effective lifetime using the Area-PLI setup while we derive an uncertainty of only 8 % on the absolute values of effective lifetime using the Line-PLI setup for the investigated samples.

On the spectral response of PV modules

The spectral response of silicon wafer based and thin-film photovoltaic (PV) modules is studied using simulation and experimental methods. Circuit simulations show that the module spectral response (SR) depends on (1) the SR of the cells, (2) the shunt resistance Rshunt of the cells, and (3) the bypass diodes of the module. For realistic Rshunt values, the module SR is significantly higher than the minimal SR of the individual cells (which would be the module SR in the case of infinite Rshunt). Round-robin tests using different experimental methods (partial illumination and full-area illumination) to determine the SR of a wafer-based module and a thin-film silicon module were conducted. Both SR methods are found to agree reasonably well. However, circuit simulations indicate that if only one cell, or a few cells, within the module have significantly different characteristics but not known, the results may differ considerably. The partial illumination method can access the SR of the individual cells within a module, but it possibly requires a long measurement time in order to measure the SR and Rshunt of each cell for confirming the SR of the whole module. In contrast, full-area illumination methods measure the module SR directly, but they cannot access the cell SRs if problematic cells exist. An uncertainty analysis of the full-area illumination method is conducted, which reveals that—if the calibrated reference cell is chosen properly—the calibration uncertainty of the reference cell itself is the main source of uncertainty.

An inexpensive spectroscopic beam monitor for hard X-ray synchrotron applications

We describe an inexpensive beam monitor for hard X-raysynchrotron applications which has good spectroscopic abilities andcan operate without cooling. The device is centred on an inexpensive,commercial off-the-shelf, large area (1.2 × 1.2 mm2) Siphotodiode operated in single counting mode. Measurements carried outat the HASYLAB synchrotron research facility have shown that it isfully spectroscopic across the energy range 8 keV to 100 keV with ameasured energy resolution of ∼ 1.2 keV FWHM at roomtemperature. The measured resolutions were found to be the same underpencil-beam and full-area illumination, indicating uniformcrystallinity and stoichiometry of the bulk. The low cost, simplicityand performance of the detector make it suitable for a wider range ofapplications, e.g., in undergraduate laboratory experiments.

Hard X- and γ-ray measurements with a large volume coplanar grid CdZnTe detector

The recent introduction of coplanar grid techniques has led to resurgence in interest in developing large volume compound semiconductor detectors that have a reasonable γ-ray response but also good spectroscopic resolution. We report the results of a series of hard X- and γ-ray measurements on a large 15×15 mm2, 10 mm thick CdZnTe coplanar grid detector. The measurements were carried out at the HASYLAB and ESRF synchrotron radiation facilities using highly monochromatic pencil beams across the energy range 20–800 keV. Additional full-area measurements were carried out using radioactive sources. All measurements were carried out at room temperature. The measured energy resolution under full-area illumination was 8 and 12 keV FWHM at 59.95 and 662 keV, respectively. Under pencil beam illumination, the measured resolutions were essentially the same. The detector energy response function was found to be linear over the energy range 20–1400 keV with an average rms nonlinearity of 1.4%, consistent with statistics. The spatial uniformity of the detector was evaluated at 30, 60 and 180 keV by raster scanning a 20×20 μm2 monoenergetic X-ray beam across the active area. Apart from a few localized areas, the detector response was found to uniform at the few percent level, consistent with statistics. At 180 keV, the nonuniformity in the energy response due to the grids was estimated to be at the 0.7% level.

Hard X- and γ-ray measurements with a 3×3×2mm3 CdZnTe detector

Abstract We report the results of a series of X- and γ-ray measurements on a 3 × 3 mm 2 , 2 mm thick CdZnTe detector carried out at the HASYLAB and ESRF synchrotron radiation facilities. The detector energy response function was found to be linear over the energy range 10–100 keV with an average rms non-linearity of 0.6%, consistent with statistics. Under full area illumination, the FWHM energy resolution was 270 eV at 5.9 keV and rises to 930 eV at 59.54 keV. Under pencil beam illumination, the measured energy resolution at 10 keV was 310 eV FWHM and rises to ∼1000 eV at 100 keV. At 60 keV the resolution was found to be ∼30% lower than that measured under uniform illumination, indicating a degree of non-uniform crystallinity and stoichiometry in the bulk. For energies

The X-ray response of TlBr

We present the results of a series of X-ray measurements on several prototype TlBr detectors. The devices were fabricated from mono-crystalline material and were typically of size 2.7×2.7×0.8 mm3. The material is extremely pure, having impurity concentrations <100 ppm. The measured electron and hole mobility–lifetime products were found to be 3×10−4 and 1×10−5 cm2 V−1, respectively, which are about an order of magnitude higher than previously reported values. Three detectors were fabricated and extensively tested over the energy range 2.3–100 keV at three synchrotron radiation facilities: the Physikalisch-Technische Bundesanstalt (PTB) laboratory at the Berliner Elektronenspeicherring fur Synchrotronstrahlung (BESSY II), the European Synchrotron Radiation Research Facility (ESRF) and the Hamburger Synchrotron-strahlungslabor (HASYLAB) radiation facility. Room temperature energy resolutions under full-area illumination of 1.8 and 3.3 keV FWHM have been achieved at 5.9 and 59.95 keV, respectively. At reduced detector temperatures of −30°C, these fall to 800 eV and 2.6 keV FWHM, respectively. Under monochromatic pencil beam illumination, the measured energy resolutions at 6 and 60 keV were 664 eV and 3 keV FWHM at the same temperature. For energies <20 keV, the measured spectra display symmetric photopeaks. However, the peaks become increasingly tailed at higher energies. At the highest energies, the energy-losses due to the electrons and holes are clearly separated. Whilst the detectors gave reproducible results over 12 months of operation, it was observed that for synchrotron beam measurements above 45 keV, they were unstable, showing rate dependent gain shifts and polarization effects. These were not observed at lower energies. The spatial uniformity of the detectors was measured using a 50×50 μm2, 12 keV mono-energetic X-ray beam, raster scanned over the forward active area. Whilst two detectors were spatially uniform to a level commensurate with statistics, the third was not. In all cases, evidence was found for charge collection problems caused by field fringing.

Hard X-ray spectroscopy using a small-format TlBr array

We report X-ray measurements on a prototype 3 � 3 TlBr pixel array, produced to assess the technological feasibility of making a Fano limited imager, which can operate near room temperature. The device was fabricated on monocrystalline material of size 2.7 � 2.7 � 1.0 mm 3 . It has a pixel size of 350 � 350mm 2 and pitch of 450mm. Measurements were carried out on all pixels over the energy range 5.9–662 keV using radioactive sources. The leakage currents were found to be low enough to allow room-temperature operation, with typical energy resolutions of B20 keV FWHM at 59.95 keV under full-area illumination. At a reduced detector temperature of –301C, these fell to B4 keV FWHM. Although the spectral performance of the present array is currently impaired by material limitations, its spectral acuity was found to be greatly enhanced by the small pixel effect. Additional photon metrology was carried out at the Hamburger Synchrotron-strahlungslabor radiation facility. Under monochromatic pencil beam illumination, the measured energy resolutions at 20 keV were B3 keV FWHM at � 301C. The spatial uniformity of the array was measured using a 50 � 50mm 2 , 20 keV monoenergetic X-ray beam, raster scanned over the entire active area. The response, in terms of count rate, gain and energy resolution was found to be uniform at the few percent level, consistent with statistics. It was observed during these measurements, that the X-ray response of the pixels was unstable, showing time-dependent gain shifts indicative of polarization effects. The magnitude of the effect was proportional to the total energy deposition per unit time. Lastly, the use of TlBr arrays in nuclear medicine applications is discussed with particular emphasis on radio-guided surgical probes. Recommendations for an optimized design are given. r 2002 Elsevier Science B.V. All rights reserved.

The X-ray response of InP: Part B, synchrotron radiation measurements

In this, the second part of a detailed study into the X-ray response of InP, we present results of a series of X-ray measurements on a 3.142 mm 2×180 μm thick semi-insulating InP detector at the BESSY II and HASYLAB synchrotron radiation research facilities. Photon metrology was carried out at energies ranging from 8 to 100 keV. Additional measurements were made using radioactive and fluorescent target sources. At −60°C, under full-area illumination, the FWHM energy resolution was 2.4 keV at 5.9 keV rising to 8.5 keV at 59.54 keV. Under pencil-beam illumination, the measured resolutions were generally less, being 2 keV FWHM at 8 keV rising approximately linearly to 5 keV at 100 keV. Analysis of the energy resolution function indicates that poor charge transport presently limit the performance of InP detectors and specifically hole trapping. This is borne out by the observed low-energy tailing of the pulse height spectra at intermediate and high energies. At very low count rates, it was found that the device was linear to within 1%. However, for count rates above ∼100 s −1 and energies above ∼50 keV, the detector began to display time variable gain shifts indicative of polarization effects. The spatial uniformity of the detector was investigated by raster scanning a 50×50 μm 2, 15 keV monoenergetic X-ray beam across the active area. Apart from a few localized areas, the detector response was found to be reasonably uniform at the 10% level, although inconsistent with that expected from counting statistics alone (2%).

The hard X-ray response of a large area HgI 2 detector

We summarise the results of a number of X-ray measurements on a 1.06 cm 2, 0.18 cm thick HgI 2 detector carried out at the European Synchrotron Research Facility at Grenoble, France. The detector energy response function was found to be linear over the energy range 8–27 keV with an average rms non-linearity of 0.6%, consistent with statistics. At room temperature, under full area illumination, the FWHM energy resolution was 1.4 keV at 5.9 keV rising to 3.9 keV at 59.54 keV. Under pencil beam illumination, the measured energy resolution at 8 keV was 1.4 keV FWHM at a detector temperature of 9°C. Using a 50×50 μm 2 10 keV monoenergetic X-ray beam, the spatial uniformity of the detector response was determined by a raster scan technique to be at the ±5% level.

The hard X-ray response of HgI 2

We summarise the results of a number of X-ray measurements on a 7 mm 2, 0.5 mm thick HgI 2 detector carried out both in our laboratory and at two synchrotron radiation facilities in Germany. The detector energy response function was found to be linear over the energy range 2.3– 100 keV with an average rms non-linearity of 0.2%. Using a 50×50 μm 2 15 keV monoenergetic X-ray beam, the spatial uniformity of the detector response was determined by a raster scan technique to be better than 1%. At room temperature, the FWHM energy resolution was 600 eV at 5.9 keV rising to 6 keV at 100 keV with no statistically significant difference between pencil beam (50×50 μm 2) and full-area illumination. It was found that the FWHM energy resolution decreased with decreasing temperature reaching a minimum of 400 eV for 5.9 keV X-ray illumination at −19.2°C. Below this temperature, the FWHM begins to rise slowly. By considering the various contributions to the FWHM energy resolution, we estimate the charge collection efficiency to be (95.2±0.7)% at room temperature and (98±0.2)% at −19°C with a Fano factor of 0.30±0.24 at room temperature.

The X-ray response of InP

We present the results of X-ray measurements on a prototype InP detector. The device was fabricated from Fe-doped bulk material of size 3×3×0.18 mm 3. X-ray measurements have been carried out using a number of radioactive and fluorescent target sources. The detector energy response function was found to be linear over the energy range 5.9–88 keV with an average rms non-linearity of 0.7%, consistent with statistics. At a detector temperature of −60°C, the FWHM energy resolution under full-area illumination was 2.5 at 5.9 keV rising to 12 at 88 keV. Analysis of the energy resolution function indicates that poor charge transport presently limits the performance of InP detectors.

The X-ray response of CdZnTe

We report the results of a series of X-ray measurements on a 3.1 mm 2, 2.5 mm thick CdZnTe detector carried out at the BESSY II and HASYLAB synchrotron radiation facilities. The detector energy response function was found to be linear over the energy range 2.3–100 keV with an average rms non-linearity of 0.6%, consistent with statistics. At room temperature, under full-area illumination, the FWHM energy resolution was 1.6 keV at 5.9 keV rising to 2.9 keV at 59.54 keV. At a reduced detector temperature of −20°C, these fall to 380 and 818 eV FWHM, respectively. Under pencil beam illumination, the measured energy resolution at 2.5 keV was 360 eV FWHM rising to 1558 eV at 100 keV. By best fitting the expected resolution function to the experimental data, we derive a value for the Fano factor of (0.099±0.02). Cooling the detector to −50°C, resulted in no noticeable difference in Δ E above 60 keV, but a 40% reduction at energies <10 keV. At 5.9 keV, the measured resolution under full-area illumination using an 55Fe source was found to be ∼10% broader than that measured at an equivalent energy using a 50 μm diameter pencil beam of synchrotron radiation. However, at 59.54 keV there was no statistical difference between pencil beam and full-area illumination, indicating uniform cystallinity and stoichiometry in the bulk. For energies <50 keV, the measured energy-loss spectra display symmetric photopeaks, becoming increasing tailed at higher energies due to hole trapping. The spatial uniformity of the detector was investigated by raster scanning a 50×50 μm 2, 10 keV monoenergetic X-ray beam across the active area. Apart from a few localized areas, the detector response was found to uniform at the few percent level, consistent with statistics