Computerised Glow Curve Analysis Research Articles

UV effect on the cathodo- and thermoluminescence properties of a gem-quality Cr-rich diaspore (α-AlOOH)

This work reports on the cathodoluminescence (CL) and thermoluminescence (TL) properties of gem-quality diaspore samples from Milas/Muğla (Turkey) after 100 h of ultraviolet-C (UVC) exposure. The UVC exposure induces significant changes in the intensity of the CL emission in the range of 400–800 nm that would be mainly associated with photo-oxidation processes of the impurities (Cr3+, Ti3+, Fe2+) that substitute for Al3+ in the diaspore (α-AlOOH) lattice. The UVC effect on the 400 nm-TL behavior of beta irradiated samples in the range of 0.1–8 Gy modifies the TL glow curves probably due to both photo-transfer process where electrons release from deeper to shallower traps and redox reactions involving, also, breakages-linkages of chemical bonds. Meanwhile, the ‘as received’ samples consist of three maxima centered at about 120, 180, and 234 °C, the 100 h UVC-irradiated samples display three maxima at 122, 220 and 270 °C. The physical trapping parameters (intensity and peak position, trap depth and pre-exponential factor) for each TL curve were estimated by using a computerized glow curve analysis program.

Automatic detection of anomalous thermoluminescent dosimeter glow curves using machine learning

Computerized Glow Curve Analysis (CGCA) has been, and still is, an intensively-investigated subject for the past two decades. CGCA has applied different methods for various applications, from glow curve deconvolution into isolated peaks, through semi-automatic software tools for detecting outliers, to software that discovers exceptional curves by using predefined rules. The method presented herein addresses the subject using a new approach in which a completely automatic algorithm is used for the accurate detection of anomalies in thermoluminescent dosimeter (TLD) glow curves. A Support Vector Machines (SVM) technique, which is a machine learning classification algorithm, is used for the first time for radiation dosimetry applications. The algorithm classifies glow curves into two categories: an acceptable i.e. 'regular' curve, or a curve that exhibits any kind of anomaly i.e. an 'anomalous' curve. The classification method treats the glow curves raw data as a large ensemble of statistical data, and tries to identify exceptional glow curve shapes by statistical means. This classification method is performed in three steps. First, a library of glow curves is manually classified by a human user of the system into the above two classes. Then an iterative training algorithm is applied to these glow curves. The final stage applies a method of comparison between an unidentified glow curve and these two pre-classified sets, and assesses a classification probability to each of the two classes. The results show between 96.2% and 97.7% accuracy of the correct classification to either one of the classes, depending on the admissible false negatives rate.

Thermoluminescence dosimetry: State-of-the-art and frontiers of future research

The state-of-the-art in the use of thermoluminescence for the measurement of energy imparted by ionizing radiation is discussed. Emphasis is on the advantages obtainable by the use of computerized glow curve analysis in (i) quality control, (ii) low dose environmental dosimetry, (iii) medical applications (especially precision) and microdosimetric applications, and (iv) mixed field ionization-density–dosimetry. Possible frontiers of future research are highlighted: (i) vector representation in glow curve analysis, (ii) combined OSL/TL measurements, (iii) detection of sub-ionization electrons, (iv) requirements for new TL materials and (v) theoretical subjects involving kinetic modeling invoking localized/delocalized recombination applied to dose response and track structure theory including creation of defects.

Highlights and pitfalls of 20 years of application of computerised glow curve analysis to thermoluminescence research and dosimetry

The technical and dosimetric aspects of computerised glow curve analysis are described in detail including a review of the current 'state-of-the-achieved' in applications to environmental and personal dosimetry, clinical dosimetry, quality control, characterisation of new materials, continuing characterisation of 'old' materials, heavy charged particle dosimetry, mixed field n-gamma dosimetry, X-ray dosimetry and other aspects of thermoluminescence dosimetry. Fearless emphasis is placed on 'pitfalls' as well as successes.

Impact of reading pre-irradiation background signal on the post-irradiation glow curves of thermoluminescence dosimeters

Due to their sensitivity, signal reproducibility and stability, lithium fluoride thermoluminescence dosimeters are widely used in many applications. It was noticed that post-irradiation glow curves of LiF dosimeters that had undergone pre-irradiation background dose measurements dramatically differed from the curves of dosimeters irradiated without such preliminary measurements. Three types of dosimeters were studied in this respect, namely, TLD-700, TLD-600, and TLD-100. Impact of pre-irradiation background measurements on the glow curve areas, peak positions, and individual peak intensities were investigated using computerized glow curve analysis. Similarities in the behavior of the dosimeters of all three types were found. Areas of the glow peaks P 2a, P 3, P 4, P 5, and P 6 increased. Peak P 3 grew most significantly, followed by P 2, and P 5. Also, one of the common features was a shift of T max of P 5 to higher temperatures. Pre-irradiation background readings increase the total area of the glow curve in general and affect relative intensities and positions of specific peaks, which changes the shape of the glow curve. It is strongly recommended to perform pre-irradiation background measurements for all LiF-based dosimeters regardless of the values of the background doses relative to the radiation doses to be given.

A computerized glow curve analysis (GCA) method for WinREMS thermoluminescent dosimeter data using MATLAB

A computerized glow curve analysis (GCA) program for handling of thermoluminescence data originating from WinREMS is presented. The MATLAB program fits the glow peaks using the first-order kinetics model. Tested materials are LiF:Mg,Ti, CaF(2):Dy, CaF(2):Tm, CaF(2):Mn, LiF:Mg,Cu,P, and CaSO(4):Dy, with most having an average figure of merit (FOM) of 1.3% or less, with CaSO(4):Dy 2.2% or less. Output is a list of fit parameters, peak areas, and graphs for each fit, evaluating each glow curve in 1.5 s or less.

The use of computerised glow curve analysis will optimise personal thermoluminescence dosimetry measurements. Favouring the proposition.

Personal dosimetry is normally a regulated activity whose results must comply with established accuracy requirements. Although these requirements may not be extremely stringent, it is important to note that there are many factors influencing accuracy and therefore influencing results. These include the energy and angular dependence of dose equivalent response, signal reproducibility, stability and fading, along with other influence factors that tend to complicate the correct measurement of the low doses usually encountered in the monitoring of radiation workers. The statistical uncertainty associated with the measurement of dose equivalent in the workplace is dependent upon the reproducibility of the many conditions, and in the case of the readings produced by thermoluminescence (TL) dosemeters the analysis of the rather complex TL signals can be a source of uncertainty. Glow curve analysis may offer a method for reducing uncertainties, but it must be considered whether such an analysis can be accomplished in the context of a large scale worker dosimetry programme.

Pilot study on the application of computerised glow curve analysis in TL based personal dosimetry services.

Preliminary results of an inter-laboratory collaborative work on the application of computerised glow curve analysis to TL personal dosimetry are presented. Very simple analysis methods have proved to be useful for the evaluation of glow curves similar to those encountered in the dosimetric control of radiation workers. A first result obtained in the study has been the possibility of simplifying the TL working procedures by eliminating pre-annealing or pre-heating steps employed in conventional systems to avoid low temperature peaks. The presence of these unwanted peaks can be detected by the computerised evaluation methods, discriminating their contribution to the dosimetric TL data.

Computerised glow curve analysis: a tool for routine thermoluminescence dosimetry.

The basic features of computerised glow curve analysis methods are presented, together with the advantages derived from their use for dosimetry. The rapid development of personal computers with progressively greater computation capability has permitted the application of different analysis methods for practical dosimetry. A very simple glow curve analysis method, specifically developed for the measurement of low or very low doses usually measured in personal and environmental dosimetry, is also discussed.

Glow Curve Analysis of Long-term Stability of LiF:Mg,Cu,P as Compared to LiF:Mg,Ti

Over a six month period, the long-term stability of LiF:Mg,Cu,P (GR-200A, Beijing China) was compared with LiF:Mg,Ti (GR-100, Beijing China). This is of particular interest in applications where there can be considerable variations in the time intervals between TLD preparation, irradiation and readout, such as environmental, accident personal and mailed dosimetry. For each TLD material, slow and fast cool-down anneals, characteristic of those used in the above-mentioned applications, were employed to study the effect of annealing procedure on long term stability. An analytical approximation based on the Randall-Wilkins model employing first order kinetics was used to fit individual peaks. Glow curve features unique for each TLD material and annealing condition were quantified and compared. From computerised glow curve analysis, annealing at 220°C, instead of the pre-set temperature of 240°C, caused an increase in the contribution from peak 3 in LiF:Mg,Cu,P, causing the material to be more susceptible to fading and peak area interchanges.

Precision for Low Dose Assessment Using TLD-100 Chips and Computerised Glow Curve Analysis

The precision for low dose assessment using LiF:Mg,Ti (TLD-100) thermoluminescent detectors and computerised glow curve analysis was investigated. The detectors were irradiated at low doses, from 0.01 to 3 mGy. Two experiments were carried out, one with detectors annealed at 400(C for 15 min and another where the detectors were submitted to no regeneration annealing at all. The software utilised was the so-called TLD GLOW, developed at GSF (Germany). No stabilisation annealing before readout was utilised. The thermoluminescent system used was a Harshaw 3500. Both experiments show that doses as low as 0.05 mGy can be evaluated with a precision better than 6% (1 SD), and for doses higher than 0.1 mGy, a precision better than 1.5% (1 SD) can be achieved.

Estimation of the Time Elapsed Between Exposure and Readout Using Peak Ratios of LiF:Mg,Cu,P (MCP-N,GR200A)

The applicability of peak area ratios (peak2/peak4 and peak3/peak4) in the glow curves of LiF:Mg,Cu,P for the estimation of time between exposure and readout, t ER , was studied. MCP-N and GR-200 LiF:Mg,Cu,P detectors were exposed to simulate an 'accidental' dose at times before readout ranging from 24 h to 3 months. Two groups of detectors were exposed simultaneously to determine the fading of irradiated detectors and the sensitivity variations of stored, non-irradiated detectors. Peak areas were evaluated using a computerised glow curve analysis code. The peak 2/peak 4 ratio is well suited for evaluating t ER below 1 week, while the peak 3/peak 4 ratio serves for t ER up to 3 months. The method can be applied in personal and environmental dosimetry to evaluate the date of a high dose rate, accidental exposure.

TL Measurements with Computerised Glow Curve Analysis in Clinical Dosimetry

A MS Windows software has been developed that abandons low temperature annealing in TL dosimetry with LiF:Mg,Ti (TLD-100). This is achieved by a glow curve analysis based on the first order kinetics model of Randall and Wilkins. Different from commercial deconvolution, and aimed to process exact trap parameters, the new software is optimised for straightforward assessment of peak areas as a measure of dose. The program allows manual or automatical analysis of glow curves (even of different TL materials). Typical for LiF:Mg,Ti is a processing time of less than 1 s per glow curve. Under laboratory conditions coefficients of variation, V, were found within 0.2 < V < 0.6% for 50 non-selected LiF:Mg,Ti samples re-used 10 times at 1 Gy. No thermal fading could be found for one week storage at temperatures up to 40 °C. The flexible software is user-friendly and may serve for routine and quality assurance programmes in clinical dosimetry.

Effects of Annealing on Glow Peak Parameters of LiF:Mg,Ti (TLD-100) Dosimetry Material

Computerised glow curve analysis has been used to examine the dependence of the peak parameters of LiF:Mg,Ti (TLD-100) on the pre-irradiation annealing procedure in the temperature range 353 K to 673 K. The intensities of the main peaks 1, 2, 3, 4 and 5 as well as the activation energies and frequency factors were found to vary strongly with the pre-irradiation low temperature annealing. The trends in the changes of the glow curves with respect to peaks 2 and 5 could be explained by the defect reaction equation MgV r (MgV)2 r (MgV) 3 r precipitates, and an additional assumption that the traps responsible for peak 2 react during read out with MgV and (MgV)2. The 'precipitation' process was found to be frozen in at temperatures lower than 373 K. The influences of variations in the cooling rate in the annealing procedure on the glow curve shape are discussed. It was found that for accurate dosimetry the cooling applied in the annealing procedure is not a very critical factor for high cooling rates (> 1 K.s -1 ) but may be a very critical factor for low cooling rates (<0.08 K.s -1 ).

On the peculiarities of peak 4 in LiF TLD-100

A detailed study of the different glow curves that can be obtained from LiF TLD-100 is presented. This study has been made measuring glow curves after different preparation and readout conditions, and using computerized glow curve analysis to resolve individual glow peaks. In particular, the evolution of peak 4 has been carefully followed. This peak is detected with very different intensity, shape and temperature position depending on the characteristics of the cooling stage of the preparation thermal treatment, and also on the readout heating rate. After very fast cooling and also for fast readout, peak 4 is the most intense peak in the TLD-100 glow curve. The unique behaviour of peak 4 is discussed in the light of the current models for thermoluminescent processes.

Eight Years Application of Computerised Glow Curve Analysis at the Rossendorf Nuclear Research Centre

Abstract Since 1983 computerised glow curve analysis has been carried out at the Rossendorf Nuclear Research Centre. The application fields, the methods developed and the results obtained are: (1) Skin dose determination in beta/gamma fields. The glow curve of a fast, directionally heated thick TL detector is deconvoluted into contributions of standard glow curves for beta and gamma irradiation of known depth dose distribution. Doses in tissue depths from 0.05 mm to 1.6 mm are obtainable. (2) Fading reduction. The evaluation of only the separated main peak yields the fading-independent dose and, additionally, the evaluation of the separated low temperature peak gives an indication of the time elapsed since a single high exposure. (3) High range exposure. For exposures of CaSO4:Dy detectors to doses from 0.1 kGy to 5 kGy the saturation effects can be reduced by deconvolution of the glow curve into up to five Gaussian peaks. The dose is obtained from two or more peak areas as well as from the ratio of peak height and location.

Study of ageing effects in LiF:Mg, Ti by analysis of thermoluminescence glow curves

The effects of ageing on the thermoluminescence of LiF(TLD-100) crytals have been investigated by computerized glow curve analysis. After an annealing of 1 h at 400°C and quick cooling the crystals were aged at 80°C for 0, 6, 20, 60, 200, 800 and 1200 min and readout at 6°C s −1 or 0.24°C s −1. The intensities of peaks 2, 3 and 4 decrease with increasing ageing time. Peak 5 initially increases (up to 20%) due to clustering of dipoles and decreases at longer ageing times due to precipitation. Dramatic changes in the values of the activation energies of peaks 2 (from 1.35 to 0.72 eV), 3 (from 1.44 to 1.17 eV) and 4 (from 1.55 to 2.32 eV), have been observed for ageing times longer than 20 min and a readout heating rate of 6°C s −1. These changes are smoothed out at when the readoutheating rate is slow. The strong changes of E values indicate a trap modification process which may be caused by precipitation of defect clusters. The relevance for practical dosimetry is discussed.

The Analysis of Thermoluminescent Glow Peaks in CaF&lt;sub&gt;2&lt;/sub&gt;:Tm (TLD-300)

Computerised glow curve fitting, the variable heating rate method, Tm-Tstop procedure and the dose variation method are used to determine the number of peaks, the order of kinetics, the activation energy E and the attempt-to-escape frequency s associated with the glow peaks in single crystals of CaF2:Tm (TLD-300). Dose variation and the Tm-Tstop procedure indicate that there are only first-order kinetic TL processes in the temperature range from 300 K to 600 K. Glow curve analysis indicates that the glow curve in this temperature range can be best described as a superposition of six peaks. Trapping parameters found with glow curve fitting are found to be constant for glow curves measured at heating rates varying from 0.24 K.s-1 to 3.0 K.s-1. Glow curve fitting and the variable heating rate method yield trapping parameters for main peak 3 which agree reasonably well (E = 1.15 ± 0.02 eV and 1.22 ± 0.04 eV respectively). For all other peaks there is no, or doubtful, agreement. The discrepancy may be caused by the presence of several emission bands at different wavelengths. The possibility of using computerised glow curve analysis in obtaining more information on the radiation quality in mixed radiation fields is discussed.

Comparative Study of Trapping Parameters of LiF (TLD-100) from Different Production Batches

Computerised glow curve analysis has been used to determine the trapping parameters of the main peaks of the thermoluminescent (TL) material LiF(TLD-100). The TL material (solid state chips) originated from six different production batches with at least 19 chips per batch. The maxima of glow peaks 2 to 5 are found at the same temperature within very small limits. The activation energy and frequency factor of the main glow peak (peak 5) of TLD-100 originating from two batches differ significantly from those of the other four investigated batches. Nevertheless, the sensitivity of glow peak 5 is more or less the same for all batches. The trapping parameters of glow peaks 2 to 4 of TLD-100 vary little from batch to batch. The measured half-life of peak 2 differed strongly from batch to batch. For all investigated peaks no correlation has been found between glow peak sensitivity and trapping parameters. The results of this study suggest that both defect concentration and nature of the trapping centres vary from batch to batch. It would appear that as a consequence of selection by the manufacturer, the differences between the batches in terms of total light output are small.

Computerised Glow Curve Analysis of CaSO&lt;sub&gt;4&lt;/sub&gt;:Dy Teflon Dosemeters for Routine Environmental Monitoring

A computerised glow curve analysis of CaSO4:Dy Teflon environmental dosemeters was applied for fading independent dose evaluation and for the assessment of time elapsed since a single high exposure. The method is based on unfolding the glow curve into a main peak and a sub-peak using Gaussian functions, the least squares technique and a stripping procedure. If only the main peak area is used for dose evaluation the error due to fading can be reduced from 40% to about 5%. Using the fading of the low temperature peak an estimate of the time elapsed since a single high exposure can be made for known and constant storage temperature, but effects from storage temperature and storage time, which both influence the fading, could not be separated from each other.