Measurement of detection efficiency and response functions for an Si(Li) x‐ray spectrometer in the range 0.1–5 keV

Abstract

Efficiency and response functions were determined for a modern commercial energy-dispersive x-ray spectrometer in the photon energy range 0.1-5 keV using dispersed and undispersed synchrotron radiation. The spectrometer was equipped with an Si(Li) detector crystal, a thin-film window and a digital pulse processor. Monochromatized synchrotron radiation from the PTB VUV radiometry beamline at the electron storage ring BESSY was used for the absolute determination of the detection efficiency with a typical relative uncertainty of 1-2% in the photon energy range 0.1-1.5 keV by direct comparison with calibrated photodiodes. At higher photon energies, the efficiency was found by comparison of the measured and calculated undispersed synchrotron radiation spectrum. The absolute intensity of the synchrotron radiation was known with a relative uncertainty of less than 1%. In the overlapping region, the results from the two independent experiments are in full agreement. The energy dependence of the measured efficiency can be explained only with a detector model assuming that there is no dead layer. A simple model for the effect of incomplete charge collection (ICC) was applied to describe the measured response functions. Consequences of the ICC such as broadening and shift of low-energy peaks and redistribution of counts around 1.84 keV are explained with the model, in accordance with the experimental results.

Modelling the response function of energy dispersive X‐ray spectrometers with silicon detectors

“…With monochromatic synchrotron radiation response functions could be measured for all photon energies of interest between 0.1 keV and several tens of keV. [6,10] Owing to the long calculation times needed and the statistical nature of the results, Monte Carlo simulations are not the way to go for practical applications where smooth response functions for large numbers of photon energies are needed to convolute calculated photon spectra for comparison with measured MCA pulse height spectra.…”

“…However, a also determines the depth z 0 where C approaches unity and is thus given by the total number of counts in the tail, see Eqn (3). In a previous article [10] we used a sum of two exponential functions; effectively, the first was used to adjust the slope at z = 0 and the second to adjust z 0 . In order to keep the number of free parameters as small as possible, the function…”

“…The response function of energy dispersive spectrometers (EDS) with Si(Li) detectors has been modelled by the so-called Hypermet function in different versions, [3] by an analytical physical model [9,10] or by means of Monte Carlo simulations. [6,8,11] The approach presented in this article is similar to that of Lowe [9] concerning the description of the shelf.…”

Quality Assurance Challenges in X-ray Emission Based Analyses, the Advantage of Digital Signal Processing

“…Our message differs from all previous investigators, as we declare that the signal processor DOES modify the spectra as confirmed by experiments. We recommend the reader look at the spectra presented in the literature [19][20][21][22][24][25][26] to see the large variations.…”

2 E 1 Ar17+ decay and conventional radioactive sources to determine efficiency of semiconductor detectors

“…Nowadays, the response functions are well understood and simulations, frequently based on Monte Carlo methods, reproduce well all the spectral features whatever the incident energy 4, 5,6,7,8 . Below a few keV, the low energy tail accompanying the full absorption energy peak is known to be mainly due to incomplete charge collection effects 5,6,7 that occur at low incident photon energies. For much higher incident photon energies (above a few tens of keV), a small fraction of photons interacts with the quasi-free electrons of the detector material and is scattered, losing part of their energy in the detector or escaping from the detector.…”