**Welcome to my personal website!**

Dr. Vasilis Pagonis

Professor of Physics Emeritus (2020), McDaniel College, Westminster, MD 21157, USA

*Areas of interest:* Experimental and modelling work in Thermoluminescence and Optically stimulated Luminescence

Email: vpagonis@mcdaniel.edu

Click **HERE** for my CV (updated March 2021)

**Click for information on the new R-book from SPRINGER (May 2021)**

**Click for information on the new CLASSICAL MECHANICS undergraduate book! **

**Click for information on my new TL/OSL Book (2019)**

**Click HERE for information on my two older TL/OSL Books (2006, 2012)**

**Below you will find links to PDF FILES for my publications in TL and OSL.**

**LINKS TO PUBLICATION****S
**

**142. On the deconvolution of promptly measured luminescence signals in feldspars**

Vasilis Pagonis, Nathan D. Brown, Jun Peng, George Kitis, George S. Polymeris

Journal of Luminescence 239 (2021) 118334

**Abstract**

Thermally and optically stimulated luminescence signals from feldspars have been the subject of numerous experimental and modeling studies, due to the importance of these signals in luminescence dosimetry and luminescence dating. Despite these extensive previous studies, there are no standardized computerized deconvolution methods for analyzing these signals in the literature. In this paper we present a simple and consistent method which can be used to perform computerized deconvolution of thermoluminescence (TL), continuous wave infrared stimulated luminescence (CW-IRSL) and linearly-modulated (LM-IRSL) signals from feldspars. The method can be used with promptly measured luminescence signals from laboratory irradiated samples. R scripts are used in several examples to carry out the deconvolution of TL, CW-IRSL and LM-IRSL luminescence signals from different types of feldspars, including K feldspars extracted from bedrock and from a suite of geological museum samples. The results of the analysis show that the TL glow curves can be described in a uniform manner, as the superposition of several components corresponding to different activation energies. All TL, CW-IRSL and LM-IRSL signals can be described by a narrow range of the dimensionless acceptor density parameter, consistent with previous analysis of these luminescence signals in the literature, which have concluded that these luminescence signals are likely to share a common recombination center.

**141. RLumCarlo: Simulating Cold Light using Monte Carlo Methods**

Sebastian Kreutzer, Johannes Friedrich, Vasilis Pagonis, Christian Laag, Ena Rajovic, Christoph Schmidt

Accepted at *The R Journal* Vol. XX/YY, 2021

**Abstract**

Luminescence phenomena of insulators and semiconductors (e.g., natural minerals such as quartz) have various application domains. For instance, Earth Sciences and archaeology exploit luminescence as a dating method. Herein, we present the R package *RLumCarlo* implementing sets of luminescence models to be simulated with Monte Carlo (MC) methods. MC methods make a powerful ally to all kind of simulation attempts involving stochastic processes. Luminescence production is such a stochastic process in the form of charge (electron-hole pairs) interaction within insulators and semiconductors. To simulate luminescence-signal curves, we distribute single and independent MC processes to virtual MC clusters. RLumCarlo comes with a modularised design and consistent user interface: (1) C++ functions represent the modelling core and implement models for specific stimulations modes. (2) R functions give access to combinations of models and stimulation modes, start the simulation and render terminal and graphical feedback. The combination of MC clusters supports the simulation of complex luminescence phenomena.

**140.Quantitative analysis of thermoluminescence signals of glass displays from mobile phones**

Vasilis Pagonis, Clemens Woda, Michael Discher

Radiation Measurements (2021)

https://doi.org/10.1016/j.radmeas.2021.106614

**ABSTRACT**

Glass displays from mobile phones show a characteristic thermoluminescence (TL) signal after the exposure to ionizing radiation. Therefore they can be used as accident and emergency dosimeters. This paper carries out quantitative analysis of such TL signals by using several standard methods of analyzing luminescence signals. Application of the initial rise method to 120 TL glow curves obtained by the Tmax-Tstop method, shows that the TL signals can be described by a quasi-continuous distribution of energies in the range of E=0.8-1.5 eV. The shape of the TL signals was simulated by using five Gaussian trap energy distributions centered at five discrete activation energy values and a single frequency factor s. The result of the simulations were in a good quantitative agreement with all the experimental TL glow curves. Two additional experiments investigating the variation of the TL signal with the heating rate, and studying the fading of the TL signal at room temperature over long periods of time, showed that this material exhibits the well known anomalous heating rate effect, and anomalous fading of the TL signal. The results of these two experiments demonstrate that the luminescence mechanism in this dosimetric material is rather complex, and that additional recombination pathways affect the luminescence process during a TL measurement.

**139.Modelling the dependence of equivalent dose determined from a dose recovery test on preheating temperature: The intervention of shallow electron traps**

Jun Peng, Xulong Wang, Grzegorz Adamiec, Vasilis Pagonis, Jeong-Heon Choi

Radiation Measurements 142 (2021) 106566

**ABSTRACT**

In optically stimulated luminescence (OSL) dating, the performance of the single-aliquot regenerative-dose (SAR) protocol is commonly tested using a dose recovery test. The dependence of equivalent dose (De) determined in a dose recovery test on preheating temperature using quartz OSL was reported in previous studies which demonstrated that using a lower preheating temperature prior to the measurement of the test dose OSL response may result in serious underestimation of the recovered De. In this study, we simulate the variation of normalised sensitivity-corrected OSL signals as a function of cycle number and the dependence of the recovered De on preheating conditions, using a kinetic model consisting of four electron traps and two recombination centres. Experimentally observed phenomena are successfully reproduced using the model with a set of optimised kinetic parameters. We suggest that the intervention of shallow electron traps may play a significant role in affecting the accuracy of De determination. Relevant mechanisms are explained, and implications on dose measurements using the SAR protocol are discussed.

**138. Simulating feldspar luminescence phenomena using R**

Vasilis Pagonis, Christoph Schmidt, Sebastian Kreutzer

Journal of Luminescence 235 (2021) 117999

**ABSTRACT
**Kinetic models have been used extensively for modeling and numerical simulation of luminescence phenomena and dating techniques for various dosimetric materials. Several comprehensive models have been implemented for quartz, which allow simulation of complex sequences of irradiation and thermal/optical events in nature and in the laboratory. In this paper, we present a simple and accurate way of simulating similarly complex sequences in feldspars. We introduce the open-access R scripts Feldspar Simulation Functions (FSF) for kinetic model simulation of luminescence phenomena in feldspars. These R functions offer useful numerical tools to perform luminescence simulations in a user-friendly manner. The mathematical framework of four different types of previously published models is presented in a uniform way, and the models are simulated with FSF. While previously published versions of these four models require numerical integration of the differential equations, the FSF circumvent the need for numerical integration by using accurate summations over the finite range of the model parameters. The simulation process can be understood easily by creating transparent sequences of events consisting of these compact R functions. The key physical concept of the FSF is that irradiation and thermal/optical treatments of feldspars change the distribution of nearest neighbor (NN) distances in donor-acceptor pairs. These changes are described using analytical equations within the four models examined in this paper. The NN distribution at the end of one simulation stage becomes the initial distribution for the next stage in the sequences of events being simulated. Several practical examples and possible applications and extensions of the FSF are discussed.

**137. Thermoluminescence due to simultaneous recombination of two electrons into two-hole centers**

R. Chen, J.L. Lawless, V. Pagonis

Radiation Measurements 141 (2021) 106521

**ABSTRACT
**A model describing the possibility of simultaneous recombination of two electrons into two-hole recombination centers yielding thermoluminescence (TL) is presented. This follows previous reports on such transitions in some other solid-state phenomena. The simulated results may explain some unusual TL effects. These include the occurrence of very narrow peaks which results in the evaluation of very high effective activation energies and exceedingly high effective frequency factors, sometimes reported in the literature. The model can also explain the more-than-quadratic dose dependence of TL. Other effects are peaks with symmetry factors smaller than that of first-order peaks or intermediate between that of first and of second order. Another result of the simulations is the possibility of first-order-like peaks that shift with the excitation dose. Finally, the possibility of a dose-rate effect is also seen in the results of the simulations. In addition to the numerical simulations, limited to a small number of sets of parameters, an analytical treatment with some approximations has been developed. For a certain set of parameters and rather low excitation dose, the agreement between the results of the simulation and theory has been very good.

**136. Superlinearity revisited: A new analytical equation for the dose response of ****defects in solids, using the Lambert W function**

Vasilis Pagonis, George Kitis, Reuven Chen

Journal of Luminescence 227 (2020) 117553

**ABSTRACT**

In a recent paper we developed a new analytical equation for the dose response of dosimetric materials based on the Lambert W function, within the framework of the simplest model, the one trap one recombination center model (OTOR). In the present paper we extend this recent work to develop a new analytical equation describing the nonlinear trap filling process during irradiation of insulators. The new equations are developed within the framework of a more complex two trap and one recombination center model (2T1R). The model has been proposed previously to explain the nonlinear dose response function of luminescence signals in solids, and is based on competition between two traps during irradiation. The new analytical equation describes the superlinear behavior commonly exhibited by dosimetric materials and contains meaningful physical parameters. It is tested by fitting previously published dose response curves for dosimetric materials.

**135. Influence of scatter data and temperature lag on the analysis of
thermoluminescence glow peak: A Monte Carlo simulation study
**

A.M. Sadek, Vasilis Pagonis, G. Kitis

Applied Radiation and Isotopes 167 (2021) 109405

**ABSTRACT**

FThe effects of the scattering data that might appear at the low radiation doses and the temperature lag that might occur between the heater pan and the detector on the thermoluminescence (TL) glow-curve have been addressed. The scattering data were mathematically induced in the TL glow curve using the Monte Carlo (MC) algorithm. While the phenomenon of the temperature lag has been simulated, assuming an exponential function. The deconvolution analysis has been carried out by TL expressions based on a presumed kinetics order value in comparison with the general-order kinetics equation. It has been found that even though the glow peak is known to obey the first- or second-order kinetics model, using the general-order kinetics equation in the deconvolution analysis is better than using the corresponding model equation.

**134. Inherent statistics of glow curves from small samples and single grains **

J. L. Lawless, R. Chen, V. Pagonis

Journal of Luminescence 226 (2020) 117389

**ABSTRACT**

For measurement techniques, it is usually important to know the accuracy and detection limits. This issue is particularly important in thermoluminescence when studying either low doses or small samples, especially single grains. Our focus herein is not on instrumental measurement errors but instead on the inherent noise or uncertainty in the thermoluminescence process. To this end, we study a simple one–center, one–active-trap model that produces a first–order glow curve. A master equation is developed for the statistical noise and probability distributions for trap population during both irradiation and heating. In addition to quantifying the probability distributions, governing equations are developed for both the mean and the standard deviation of trap population during irradiation and heating.

**133. Sequential two-step optical stimulation in K-feldspars: Correlation among the luminescence signals and implications for modeling parameters
**

V. Angeli , G. Kitis, V.Pagonis, G.S. Polymeris

Journal of Luminescence 226 (2020) 117425

Abstract

Luminescence signals from feldspars are of importance in luminescence dosimetry and dating. The present study attempts a correlation between two specific experimental parameters, such as the lifetime of luminescence components and the dimensionless acceptor density parameter ρ0 , as well as the influence of the crystal structure on the luminescence signals. The analysis of the above parameters was completed in three different K-feldspar samples, each one belonging to the structural group of microcline, sanidine and orthoclase. The stimulations in this manuscript were carried out using blue optically stimulated luminescence (BLOSL) and infrared (IRSL) optically stimulated luminescence, in two different two step experimental protocols. A comparison of these two step protocols is applied for the first time in the literature. A de-convolution analysis was completed with the use of two components, a slow component which can be described by general order kinetics (GOK), and a tunneling component which is necessary to describe the blue light stimulation. The GOK component describes recombination processes taking place only through the conduction band. In order to describe the results of the IR stimulation protocol, two tunneling components were used. The results from this study are encouraging and are in full correlation with published studies in the luminescence literature.

**132. Simulation of thermoluminescence dose response in cluster systems with deep trap**

A.S. Merezhnikov, S.V. Nikiforov, V. Pagonis

Radiation Measurements 134 (2020) 106307

**ABSTRACT**

The TL dose response was simulated within the framework of a kinetic model with cluster defects which contain electron TL-active traps and deep traps. It was shown that superlinearity of the TL dose dependence arises due to the temperature-dependent competition between the intracluster capture of electrons by deep traps and their thermal release into the conduction band. The regularities of the influence of the main parameters of the model (heating rate, activation energies of the ground and excited levels of TL-active traps) on the TL yield and nonlinearity of the dose dependence were found.

**131.Competition between long time excitation and fading of
thermoluminescence (TL) and optically stimulated luminescence (OSL)
**

R. Chen, J.L. Lawless, V. Pagonis

Radiation Measurements 136 (2020) 106422

**Abstract**

In the present work, we consider the filling and thermal emptying of traps and centers in a simple one-trap-one recombination-center model in a small system such as a quartz grain in nature, when both the excitation and thermal release of electrons are very slow. Due to the nature of the very slow process, Monte Carlo simulations appear to be a very appropriate method. However, in parallel, we have applied an approximate analytical method and found practically the same results although the Monte Carlo results showed some small fluctuations due to the statistical nature of the procedure. This is in line with the experimental results which are also expected to have statistical fluctuations. The main result found is that after a long enough time, measured in hundreds or thousands years or more, the filling of the trap reaches a plateau which, depending on the parameters, may be very significantly smaller than the concentration of the trap in question. This equilibrium value is the same if we start from very low, e.g. zero concentration or very high, above the equilibrium value. This plateau level depends strongly on the relevant parameters. However, comparing simulations with activation energies of 1.2 eV and 1.3 eV shows strong dependence of the plateau level on the energy. Similarly, we can expect strong dependence on the temperature at which the sample is held. The results reached here and shown in Figs. 1–4 are based on the simplest OTOR model, but similar results of approaching a plateau level which are not due to the saturation of traps may occur in more complex systems as is demonstrated by simulations based on the Bailey model for quartz

which includes several traps and centers.

**130. Quantum tunneling processes in feldspars: Using thermoluminescence
signals in thermochronometry**

Vasilis Pagonis, George Kitis, George S. Polymeris

Radiation Measurements 134 (2020) 106325

**Abstract**

During the past decade, our understanding of the nature of quantum tunneling processes in luminescence materials has been advanced significantly, by both experimental work and modeling insights into the underlying luminescence mechanism. This paper provides examples of analysis of thermoluminescence (TL) signals in feldspars, based on recent modeling work which showed that the shape of the TL glow curves remains practically unchanged after thermal and/or optical treatment of the samples. Analysis of the experimental TL glow curves in thermally or optically pretreated feldspars shows that these signals can be described as the sum of several Gaussian curves. While the positions and height of these Gaussians depend on the thermal and optical history of the sample, the width of the Gaussians apparently remains the same across the TL glow curve. TL glow curves are analyzed for 10 different samples, and the width of the Gaussian curves is found to be a common property characterizing all samples, at least within experimental error. Specifically it is found that the Gaussian width stays practically unchanged when the irradiated samples undergo thermal treatments for different times and temperatures, or alternatively when they undergo optical bleaching treatments with blue LEDs. The common overall shape of the TL glow curves has important implications for the use of these signals in thermochronometry.

**129. A new analytical equation for the dose response of dosimetric materials,
based on the Lambert W function
**

Vasilis Pagonis,George Kitis , Reuven Chen

Journal of Luminescence 225 (2020) 117333

**Abstract**

The dose response of dosimetric materials is of fundamental importance in luminescence dosimetry and luminescence dating applications. In this paper we present a new analytical equation describing the trap filling process during irradiation of insulators, starting from the one trap and one recombination center model (OTOR). Even though this model has been studied extensively during the past 50 years, there are no published analytical solutions for the dose response n(D) in this model, where n is the concentration of filled traps and D is the irradiation dose. The new analytical equation contains the well-known Lambert function W, which has been used extensively during the past 20 years in diverse research areas. Under certain conditions, the new n(D) equation leads to the empirical saturating exponential function (SE). The new equation contains a smaller number of fitting parameters than two other commonly used fitting functions, the saturating exponential plus a linear function (SEL), and the double saturating exponential (DSE). In addition, the new equation contains physically meaningful parameters. Examples are shown of using the new equation to fit a variety of experimental signals, namely thermoluminescence (TL), optically simulated luminescence (OSL), isothermal TL (ITL) and electron spin resonance (ESR).

**128. Investigation of thermoluminescence processes during linear and isothermal heating of dosimetric materials**

George S. Polymeris, Vasilis Pagonis, George Kitis

Journal of Luminescence 222 (2020) 117142

**Abstract
**During experiments involving heating of dosimetric materials, trapped electrons are thermally excited and subsequently recombine with holes, producing a thermally stimulated luminescence signal. Thermal stimulation can take place either at a constant elevated temperature giving rise to a isothermal decay signal (PID), or with a constant heating rate which gives rise to a thermoluminescence (TL) signal. The recombination pathways during thermal stimulation stage (also called the readout stage), can be either of a delocalized nature involving the conduction band, or of a localized nature involving an excited state of the trapped electrons. The present work investigates the experimental conditions which can distinguish

between delocalized and localized transitions during the readout stage. The dosimetric materials used in this study are LiF:Mg,Ti, BeO, a natural apatite and artificial porcelain. The results show that during the readout stage with a constant heating rate, the prevalent recombination mechanism in all these materials involves delocalized transitions. However, the results show that during an isothermal decay experiment, the recombination mechanism in LiF:Mg,Ti and BeO involves delocalized transitions, whereas in the case of apatite and artificial porcelain the recombination takes place through localized transitions.

**127. A Monte-Carlo study of the fading of TL and OSL signals in the presence of deep-level competitors**

R. Chen, V. Pagonis

Radiation Measurements 132 (2020) 106257

**Abstract**

In a previous paper, the issue of the evaluated lifetimes of thermoluminescence (TL) and optically stimulated luminescence (OSL) has been studied for the one-trap-one-recombination-center (OTOR) case, using the Monte-Carlo simulation. It was shown that under these circumstances, the decay curve of the electron occupancy along many thousands of years may not be exponential. Therefore, a lifetime determined from the results at short periods of time may not apply at longer periods of time. The decay at longer times was found to be slower than exponential and thus, one may observe longer lifetimes than predicted by the evaluated trapping parameters. In the present work we demonstrate that with a more complex model, namely, when an additional deeper trap is involved, the probability of getting an exponential decay of the signal is much larger. We study the fading of OSL and TL signals with different times elapsing between excitation and read-out under these circumstances, using a Monte-Carlo procedure, and show that with a significant deep-trap competitor, the extrapolation leading to the evaluation of the long-term stability of the signal is more viable. The results are compatible with previously existing evidence that the chances of having a TL peak with first-order characteristics are significantly larger in cases where a large deep trap, acting as a competitor, is present.

**126. On the stochastic uncertainties of thermally and optically stimulated**

**luminescence signals: A Monte Carlo approach**

Vasilis Pagonis, Sebastian Kreutzer, Alex Roy Duncan, Ena Rajovic, Christian Laag, Christoph Schmidt

Journal of Luminescence 219 (2020) 116945

**Abstract**

Phenomenological models are frequently used to analyze experimental signals in thermally and optically stimulated luminescence experiments. Typically, these models consist of systems of differential equations describing various electronic transitions. An alternative to the differential equation approach is the use of Monte Carlo (MC) methods, which also allow an estimation of the theoretical stochastic uncertainty of the intensity of the luminescence signal. By running and averaging several MC variants, these stochastic uncertainties are estimated in this paper for various luminescence models. In the case of first-order kinetics processes, the MC results compare well with previously published analytical results for the coefficient of variation (CV) in stochastic linear pure death processes. By contrast, no analytical results are available for the more general one trap one recombination center model (OTOR), and MC is the only method available for estimating the stochastic uncertainties. In this paper the CV coefficients are simulated for three commonly used experimental stimulation modes, namely thermally stimulated luminescence (TL), continuous-wave optically stimulated luminescence (CW-OSL) and linearly modulated OSL (LM-OSL). The results of the simulations show that CW-OSL signals have the smallest CV values among the three stimulation modes, and therefore these signals are least likely to exhibit stochastic variations. The stochastic uncertainties in these phenomenological models are discussed in the context of single grain luminescence experiments and nanodosimetric materials, in which one deals with small numbers of charge carriers.

**125. Stimulated luminescence emission: From phenomenological models to ****master analytical equations
**

George Kitis, George S. Polymeris, Vasilis Pagonis

Applied Radiation and Isotopes 153 (2019) 108797

**Abstract
**This paper reviews developments in phenomenological models of stimulated luminescence phenomena. A set of five master equations is presented, which describe a wide variety of stimulated luminescence signals: thermoluminescence, isothermal luminescence, optically stimulated luminescence and infrared stimulated luminescence. Both delocalized and localized models are reviewed, and analytical solutions are presented for these models. The master equations are tested against the solutions of the differential equations in the models, as well by fitting experimental data for a variety of luminescencent dosimetric materials. Three out of the five master equations involve the Lambert W(z) function, thus establishing this function as the theoretical cornerstone of the phenomenological luminescence models. The applicability of the superposition principle is discussed, in connection with computerized curve deconvolution analysis.

**124. Comprehensive analysis of thermoluminescence signals in MgB4O7:Dy,Na dosimeter**

Vasilis Pagonis, Nathan Brown, George S. Polymeris, George Kitis

Journal of Luminescence 213 (2019) 334–342

**ABSTRACT
**Magnesium borate (MgB4O7) is a material of interest for thermoluminescence (TL) dosimetry of ionizing ra- diation, due to its low photon energy dependence and the possibility of developing neutron dosimeters, and its potential as a temperature sensor in passive temperature sensing applications. Recent experimental and mod- eling work has shown that TL glow curves and isothermal luminescence signals in MgB4O7:Dy,Na can be ana- lyzed using a radiative tunneling model, based on localized recombination processes. This paper presents a comprehensive analysis of TL signals in MgB4O7:Dy,Na dosimeters, based on a recently published model which incorporates simultaneous irradiation and tunneling processes. Four diﬀerent types of TL data are presented and analyzed using the proposed model: (a) TL glow curves at diﬀerent irradiation doses, (b) TL signals after sample irradiation at diﬀerent temperatures, (c) TL after isothermal heating for 500 s at various temperatures, and (d) TL after room temperature irradiation, followed by heating to a temperature Tstop. We also present an alternative fast and eﬃcient Monte Carlo method of analyzing the data, which is very useful for quick analysis of experi- mental data. The experimental data and modeling results show that the shape of the TL glow peaks remains essentially the same in all four experimental protocols, while the height and temperature of maximum TL in- tensity depends on the irradiation and preheating conditions.

**123. Simulation of TL kinetics in complex trap cluster systems: Some new
approaches**

A.S. Merezhnikov, S.V. Nikiforov, V. Pagonis

Radiation Measurements 125 (2019) 78–84

**ABSTRACT**

This paper presents an improved Monte Carlo algorithm for calculating thermoluminescence (TL) curves for complex cluster systems, which contain a large number of localized levels. The new algorithm is based on generation of two random numbers, instead of the many random number generations used in previous approaches. The first random number generation is associated with the choice of which type of energy transition takes place in the model, while the second random number generation is connected with the choice of a cluster group for which this transition occurred. The proposed new algorithm was tested on a new TL model consisting of clusters containing both electron traps and deep competitor traps. The luminescence centers in the model are treated as uniformly distributed defects. Applicability of the Monte Carlo method is demonstrated by simulating for the first time in this type of model the complete TL process, including the excitation, relaxation and heating stages. The new model explains the anomalous heating rate effect of TL, and describes the effect of localized trapping processes within the cluster on the intensity and temperature position of the TL peaks.

**122. Correlation between isothermal Tl and Irsl in K-Feldspars of various types
**

Ioanna K. Sfampa, George S. Polymeris, Vasilis Pagonis, George Kitis

Radiation Physics and Chemistry 165 (2019) 108386

**ABSTRACT
**Feldspars are used as natural dosimeters to date geological and archaeological materials. These minerals are widely used in dosimetric methods of dating using luminescence signals, such as Thermoluminescence (TL), Optically Stimulated Luminescence (OSL) and Infrared Stimulated Luminescence (IRSL). The present work is an effort to compare the Isothermal decay Thermoluminescence signals (ITL) to the IRSL signals from four Kfeldspar samples which belong to two different species, namely microcline, and sanidine. The experimental data were fitted by analytical expressions coming from a tunneling recombination model. The results prove that all ITL and IRSL signals can be fitted very accurately with the analytical equations and that only two tunneling

components are needed to obtain a high-quality fit. These tunneling components consist of a very fast and intense component, which is followed by an extended slowly decaying component. Significant differences were found between the best-fit parameters of the ITL and IRSL signals, indicating that thermal excitation processes may reach different traps from optical excitation processes in these samples.

**121. A model explaining the anomalous heating-rate effect in
thermoluminescence as an inverse thermal quenching based on
simultaneous thermal release of electrons and holes
**

R. Chen , V. Pagonis

Radiation Measurements 106 (2017) 20-25

**ABSTRACT
**A model is presented which explains the anomalous heating-rate effect of thermoluminescence (TL) in which the peak area increases with increasing heating rate. In a similar way to the Sch€on-Klasens model, the present model is based on delocalized transitions only. In addition to the occurrence of an electron trapping state and a hole recombination center, we assume the participation of a hole reservoir which competes with the other levels and participates in the process during both the excitation and the readout stages. Moreover, we assume that the reservoir is close enough to the valence band so that holes may be thermally released in the same temperature range in which electrons are thermally raised into the conduction band. Simulations with this model show that, with certain sets of trapping parameters, an increase of the heating rates results in an increase in the area under the normalized TL curve. Inverting the roles of the recombination center and the reservoir so that the recombination of a free electron with a hole in the reservoir is assumed to be radiative and the other recombination is radiationless yields opposite results. Increasing the heating rates causes a significant decrease in the area under the TL curve which is a demonstration of the well-known thermal quenching heating-rate effect of TL. An intuitive qualitative explanation of these two effects within the proposed model is given. A recently discovered case in which two consecutive TL peaks respond to the heating rate change in opposing directions, one

decreases and the other increases with increasing heating rate can also be explained by this model.

**120. Influence of the infrared stimulation on the optically stimulated
luminescence in four K-feldspar samples
**

G. Kitis , G.S.Polymeris , E. Şahiner , N.Meriç , V.Pagonis

Journal of Luminescence 176 (2016) 32–39

**ABSTRACT**

Many dating protocols involve combined measurements of thermoluminescence (TL), optically stimu- lated luminescence (OSL) and infrared stimulated luminescence (IRSL). Although the physics behind each stimulation mode is entirely different, it is possible that all of these stimulation modes act on the same set of electron trapping levels existing in dosimetric materials. Therefore, one might expect that the experimental results of the combined TL, OSL and IRSL measurements may be correlated. The present work deals with the correlation between IRSL and OSL signals. The experimental protocol involves an initial exposure of the irradiated sample to infrared radiation (IR) for various stimulation times, and then recording the continuous-wave OSL signal (CW-OSL) of the same sample. In this way one can follow how the exposure to IR inﬂuences the trapping levels responsible for the CW-OSL signal. The results of the experiments showed that the IR stimulation inﬂuences these trapping levels which are responsible for the fast components of the CW-OSL decay curve, and not the trapping levels responsible for the slow CW-OSL components. The IRSL decay curves are described by analytical equations derived from a loca- lized tunneling recombination model. On the other hand, the CW-OSL decay curves were ﬁtted suc- cessfully with either one of two following methods; Firstly they were ﬁtted by using an analytical expression derived from the solution of the one trap one recombination center model (OTOR), which describes delocalized recombination processes. Secondly, the CW-OSL decay curves were also ﬁtted successfully using the same analytical expressions used for ﬁtting the IRSL decay curves, based on localized transition processes. A more detailed analysis of the CW-OSL signals for stimulation durations up to 100 s, showed that the description of the CW-OSL using the localized tunneling recombination expression is superior to the description which uses the delocalized OTOR model.

**119. Evaluated thermoluminescence trapping parameterseWhat do they
really mean?**

R. Chen , V. Pagonis , J.L. Lawless

Radiation Measurements91 (2016) 21-27

**ABSTRACT**

The main two trapping parameters in thermoluminescence (TL), the activation energy and the frequency factor, are often calculated and used for the evaluation of the stability f the TL signal at a given temperature. In several cases, “anomalous” values of these parameters, either very high or very low have been reported in the literature. In practically all of these cases, the values reported have been recognized to be effective values which resulted from some special circumstances related to the specific materials in hand. Obviously, these effective values are not associated directly with the real rate of thermal release of carriers from traps at the ambient temperature, prior to heating, and therefore, they do not indicate the real decay time of the TL signal or, in other words, the stability of the signal which may be used in TL dosimetry or dating of archaeological or geological samples. In the present paper, we discuss briefly some of these cases and add, in more detail, a rather elementary situation of very low effective activation energy and frequency factor. A model with two trapping states and one kind of recombination center is used and the simulation includes the numerical solution of the relevant sets of coupled differential equations in the three stages of the measurement, namely, excitation, relaxation and heating for a given set of the trapping parameters. The parameters are chosen in such a way that two overlapping TL peaks occur, which look together like a single first-order peak, but with anomalously low evaluated effective activation energy and frequency factor. Implications regarding the possible results in glow curve

deconvolution are discussed.

**118. On the unchanging shape of thermoluminescence peaks in preheated feldspars: Implications for temperature sensing and thermochronometry**

Vasilis Pagonis, Nathan Brown

Radiation Measurements 124 (2019) 19–28

**ABSTRACT**

Significant advances have been made during the past 10 years in our understanding of the nature of thermoluminescence (TL) signals from feldspars. Modeling insights into the luminescence mechanism have been obtained based on quantum tunneling, and this has led to the development of new analytical equations for quantifying the luminescence signals. In addition, Monte Carlo studies have helped researchers understand the effect of random distributions of charges and of energy distributions on the TL signals. From an experimental point of view, there has been renewed interest in using these signals for particle temperature sensing in the microseconds-to-seconds time scale, and in thermochronometry applications involving a much larger geological time scale (103–105 yr). One of the fundamental issues in these field applications is the difficulty of calibrating the TL signals from feldspars as a function of the preheating temperature and of the preheating time. This paper presents new analytical expressions describing the changes in the TL glow curves in feldspar samples, when they are heated at different temperatures and for different preheat times. The behavior of the TL glow curves is described and some universal characteristics of TL glow curves are pointed out, which can be useful in analyzing regenerative TL signals. Good agreement is found between the analytical equations and

experimental data for several geological feldspar samples. The insights obtained from the new equations and the model can help in understanding how feldspar TL signals record recent or geologic thermal signatures in temperature sensing and thermochronometry applications.

**117. Thermoluminescence governed by the Auger-recombination process
**J.L. Lawless, R. Chen, V. Pagonis

Radiation Measurements 124 (2019) 40–47

**ABSTRACT**

In the present work, we study the possibility that a thermoluminescence (TL) peak is governed by the effect of Auger recombination, an effect which has been considered for other luminescence phenomena. In Auger recombination in the form of interest here, two conduction-band electrons are involved in the recombination of one of them with a hole in a center. The two electrons collide in the presence of the center, one loses energy and recombines, yielding a TL photon, and the other gains energy and speeds away. As mentioned with regard to other luminescence phenomena, in this case, in the set of differential equations governing this process, a term proportional to the square of the free-electron concentration should be included in analogy to the law of mass action. The relevant set of simultaneous differential equations has been solved numerically for feasible sets of parameters. The results yield a relatively narrow TL peak which is somewhat asymmetric, with the fall-off half being larger than the low-temperature half. Under appropriate conditions, the set of equations is shown to reduce to an approximate third-order kinetic equation, the solution of which has a very similar symmetry. The third-order approximate curve has an effective activation energy which is twice as large as the original. Such asymmetric peaks have been described in the literature. Also, when using standard peak-shape methods for evaluating the effective activation energy and frequency factor very high values of these magnitudes have been found due to the narrowness of the simulated peak. This model may explain the occurrence of such TL peaks previously reported in the literature. Also is discussed the possible concurrent regular Randall-Wilkins recombination and Auger recombination within the one-trap-one-recombination center (OTOR) model. In another version of the model, an additional thermally disconnected trap is considered. With certain sets of parameters, the simulations yield a cubic dependence of TL

intensity on the excitation dose, an effect previously reported in some materials.

**116. Excited state luminescence signals from a random distribution of defects: A new Monte Carlo simulation approach for feldspar**

Vasilis Pagonis, Johannes Friedrich, Michael Discher, Anna Müller-Kirschbaum, Veronika Schlosser,Sebastian Kreutzer, Reuven Chen, Christoph Schmidt

Journal of Luminescence, Volume 207, March 2019, Pages 266-272

**ABSTRACT**

This paper presents Monte Carlo simulations of tunneling recombination in random distributions of defects. Simulations are carried out for four common luminescence phenomena in solids exhibiting tunneling recombination, namely continuous wave infrared stimulated luminescence (CW-IRSL), thermoluminescence (TL), isothermal thermoluminescence (iso-TL) and linearly modulated infrared stimulated luminescence (LM-IRSL). Previous modeling work has shown that these phenomena can be described by the same partial differential equation, which must be integrated numerically over two variables, the elapsed time and the donor-acceptor distance. We here present a simple and fast Monte Carlo approach which can be applied to these four phenomena, and which reproduces the solution of the partial differential equation, without the need for numerical integrations. We show that the method is also applicable to cases of truncated distributions of nearest neighbor distances, which characterize samples that underwent multiple optical or thermal pretreatments. The accuracy and precision of the Monte Carlo method are tested by comparing with experimental data from several feldspar samples.

**115. Thermoluminescence glow curves in preheated feldspar: A Monte Carlo study**

Vasilis Pagonis, Francisco Marques dos Santos Vieira, Antony Chambers, Lucas Anthony

Nuclear Inst. and Methods in Physics Research B 436 (2018) 249–256

**ABSTRACT**

Thermoluminescence (TL) glow curves from feldspars have been the subject of numerous experimental and modeling studies, because of their importance in dosimetry and luminescence dating. Recently there also has been increased interest in using these signals for temperature sensing and for thermochronometry studies. It is now generally accepted that these materials exhibit anomalous fading phenomena due to quantum mechanical tunneling, and that several of their luminescence signals can be described by localized energy transitions taking place in a randomly distributed system of trapped electrons and recombination centers. Our recent modeling work showed that the TL signals of freshly irradiated feldspar samples can also be described from a completely microscopic point of view, by using Monte Carlo methods. This paper extends this recent work, and shows how the Monte Carlo method can also describe TL signals from thermally pretreated feldspars. Specifically, the simulations show that the Monte Carlo method can describe several types of TL experiments for irradiated samples that underwent partial thermal cleaning, and for samples that underwent more complex multistage isothermal procedures. The results from the Monte Carlo simulations are compared quantitatively with experimental data from several types of feldspars, which were preheated at temperatures above 200–300 °C. Common experimental characteristics are pointed out for these preheated feldspars, and the experimental data suggest the possibility of a universal description of the thermal behavior of TL glow curves in feldspars. Specifically, it is found that the shape and width of the experimental TL glow curves do not change significantly for different preheat temperatures, and also do not change when different preheat times are used at a fixed preheat temperature. The relevance of these results for dosimetric and thermochronometry studies is discussed.

**114. On the resolution of overlapping peaks in complex thermoluminescence glow curves
**

George Kitis , Vasilis Pagonis

Nuclear Inst. and Methods in Physics Research, A 913 (2019) 78–84

**ABSTRACT**

Experimental Thermoluminescence (TL) glow curves for most dosimetric materials are composites, consisting of weakly or strongly overlapping peaks. The ability to discriminate between two overlapping peaks defines the resolution of a TL glow curve. Although the concept of resolution has been defined and used widely in many areas of science, there have been no previous attempts to define and study the resolution of overlapping TL peaks. In this work the resolution parameter R is defined in terms of the geometrical characteristics of TL peaks. A large number of TL glow curves consisting of two peaks were numerically evaluated for various degrees of overlapping between them. These numerically generated glow curves were analyzed by a computerized glow curve deconvolution analysis, in s, in order to determine the reliability of the deconvolution results as a function of the resolution parameter R.

**113. Thermoluminescence associated with two-hole recombination centers**

R. Chen, J.L. Lawless, V. Pagonis

Radiation Measurements 115 (2018) 1–6

**ABSTRACT**

A model of thermoluminescence (TL) is presented based on a double-occupancy hole-recombination center and a single-electron trap. The concepts of double-electron traps and double-hole centers have been established before with regard to different solid state phenomena and also mentioned as a possible occurrence in connection with TL. In a recent paper, we dealt with the TL associated with a two-electron trap and a one-hole center. Here, we consider the case of one-electron traps and two-hole centers. A new set of simultaneous differential equations governing the three stages of excitation, relaxation and readout of TL in this new framework is developed. This situation is dealt with by solving these sets of equations sequentially for reasonable sets of chosen trapping parameters. Also, an analytical treatment using plausible simplifying assumptions is given in parallel. The outcome of these procedures yields a two-peak TL curve and, in a sense, to some extent, the two-hole center behaves as two centers with different recombination probabilities. The results of the simulations and the approximate analytical approach show that the lower-temperature peak has usually features of first-order peak and its intensity is superlinear with the dose of excitation. With the appropriate choice of parameters, the dose dependence of the first peak has been found to be initially quadratic with the excitation dose. This may explain experimental results of quadratic dose dependence previously reported in the literature. The second peak has

second-order features and both peaks shift to lower temperatures with increasing dose. This effect is explained by a change in the effective frequency factor with the excitation dose.

**112. Localized transition models in luminescence: A reappraisal**

George Kitis and Vasilis Pagonis

Nuclear Instruments ad Methods in Physics Research B 432 (2018) 13-19

**ABSTRACT**

Localized energy levels within the forbidden energy band are the source of various stimulated luminescence phenomena. The present study deals with the case of electrons stimulated from localized levels and recombining with a hole at a luminescence center, without the mediation of the conduction band. Previous research was based on three different assumptions as follows. Firstly, it was assumed that the recombination rate is independent of the concentration of recombination centers, and is proportional only to the concentration of electrons in the excited state of the trap. Secondly, it is assumed that the principle of detailed balance holds for these localized transitions. A third common assumption is that the system is in quasi-equilibrium condition. When these three conditions are applied to the system of differential equations describing the localized transitions, it was shown that the resultant thermoluminescence (TL) signals follow first order kinetics. This paper examines the assumptions used in these previous studies, and extensive simulations are carried out for a wide range of parameters in the localized transitions model. The results of the simulations show that the TL peaks in the localized model have very similar characteristics with TL peaks derived from delocalized models, including non-first order kinetic characteristics. The differential equations describing the localized transition model are solved analytically using the Lambert W function, and the resulting analytical master equation can describe a

variety of optically and thermally stimulated phenomena.

**111. On the half-life of luminescence signals in dosimetric applications: A unified presentation**

Vasilis Pagonis , George Kitis, George Polymeris

Physica B: Condensed Matter 539 (2018) 35–43

**ABSTRACT**

Luminescence signals from natural and man-made materials are widely used in dosimetric and dating applications. In general, there are two types of half-lives of luminescence signals which are of importance to experimental and modeling work in this research area. The first type of half-life is the time required for the population of the trapped charge in a single trap to decay to half its initial value. The second type of half-life is the time required for the luminescence intensity to drop to half of its initial value. While there a handful of analytical expressions available in the literature for the first type of half-life, there are no corresponding analytical expressions for the second type.

In this work new analytical expressions are derived for the half-life of luminescence signals during continuous wave optical stimulation luminescence (CW-OSL) or isothermal luminescence (ITL) experiments. The analytical expressions are derived for several commonly used luminescence models which are based on delocalized transitions involving the conduction band: first and second order kinetics, empirical general order kinetics (GOK), mixed order kinetics (MOK) and the one-trap one-recombination center (OTOR) model. In addition, half-life expressions are derived for a different type of luminescence model, which is based on localized transitions in a random distribution of charges. The new half-life expressions contain two parts. The first part is inversely proportional to the thermal or optical excitation rate, and depends on the experimental conditions and on the cross section of the relevant luminescence process. The second part is characteristic of the optical and/or thermal properties of the material, as expressed by the parameters in the model. A new simple and quick method for analyzing luminescence signals is developed, and examples are given of applying the new method to a variety of dosimetric materials. The new test allows quick determination of whether a set of experimentally measured luminescence signals originate in a single trap, or in multiple traps.

**110. Thermoluminescence due to tunneling in nanodosimetric ****materials: A Monte Carlo study**

Vasilis Pagonis , Phuc Truong

Physica B: Condensed Matter 531 (2018) 171–179

**ABSTRACT**

Thermoluminescence (TL) signals from nanodosimetric materials have been studied extensively during the past twenty years, especially in the area of nanomaterials doped with rare earths. One of the primary effects being studied experimentally have been possible correlations between the nanocrystal size and the shape and magnitude of TL signals. While there is an abundance of experimental studies attempting to establish such correlations, the underlying mechanism is not well understood. This paper is a Monte Carlo simulation study of the effect of nanocrystal size on the TL signals, for materials in which quantum tunneling is the dominant recombination mechanism. TL signals are simulated for a random distribution of electrons and positive ions, by varying the following parameters in the model: the radius of the crystal *R*, tunneling length *a,* and the relative concentrations of electrons and ions. The simulations demonstrate that as the radius of the nanocrystals becomes larger, the peaks of the TL glow curves shift towards lower temperatures and changes occur in both peak intensity and peak width. For large crystals with a constant density of positive ions, the TL glow curves reach the analytical limit expected for bulk materials. The commonly used assumption of nearest neighbor interactions is examined within the model, and simulated examples are given in which this assumption breaks down. It is demonstrated that the Monte Carlo method presented in this paper can also be used for linearly modulated infrared stimulated luminescence (LM-IRSL) signals, which are of importance in luminescence dosimetry and luminescence dating applications. New experimental data are presented for Durango apatite, a material which is known to exhibit strong anomalous fading due to tunneling; the experimental data is compared with the model. The relevance of the simulated results for luminescence dosimetry is discussed.

**109. Anomalous fading in TL, OSL and TA – OSL signals of Durango apatite for
various grain size fractions; from micro to nano scale
**

G.S. Polymeris, I.K. Sfampa, M. Niora, E.C. Stefanaki, L. Malletzidou, V. Giannoulatou,V. Pagonis, G. Kitis

Journal of Luminescence 195 (2018) 216–224

**ABSTRACT**

Anomalous fading (AF) of luminescence signals has been studied extensively both experimentally and by simulations. This paper reports a new type of study of anomalous fading in grains of Durango apatite, a naturally occurring luminescent material that yields very intense anomalous fading. Grains of Durango apatite were ball milled (BM) for various durations, up to 48 hours. Different ball milling durations resulted in different average grain size fractions as low as 200 nm, as it was indicated by Scanning Electron Microscopy (SEM) measurements. The anomalous fading effect was studied for optically stimulated luminescence (OSL), thermoluminescence (TL), as well as thermally-assisted OSL signals (TA-OSL). Anomalous fading was found to be ubiquitous for all luminescence signals, and for all apatite grain size fractions. The anomalous fading rate is weakly affected by the grain size for the cases of OSL and TL, while the TA-OSL signals were found to fade in a much slower rate than either the TL or the conventional OSL signals. An important experimental result is that the fading rate of TA-OSL decreases as the grain size fraction is decreased. For average grain size fractions between 200 and 350 nm, the TA-OSL signal is unaffected by the AF effect. A differential analysis on the TL glow curves showed that the AF rate decreases with increasing temperature along the glow curve, and also with increasing BM time. Finally, a component resolved de-convolution analysis was performed for both OSL and TA-OSL decay curves and recombination lifetimes are reported for both localized and delocalized components. FTIR analysis indicates that the ball milling procedure does not induce a new phase in this material.

**108. The effect of crystal size on tunneling phenomena in luminescent
nanodosimetric materials
**

**Vasilis Pagonis, Shannon Bernier, Francisco Marques dos Santos Vieira, Shane Steele**

Nuclear Inst. and Methods in Physics Research, B 412 (2017) 198–206

**ABSTRACT**

The study of luminescence signals from nanodosimetric materials is an active research area, due to the many possible practical applications of such materials. In several of these materials it has been shown that quantum tunneling is a dominant mechanism for recombination processes associated with luminescence phenomena. This paper examines the effect of crystal size on quantum tunneling phenomena in nanocrystals, based on the assumption of a random distribution of electrons and positive ions. The behavior of such random distributions is determined by three characteristic lengths: the radius of the crystal R, the tunneling length a, and the initial average distance 〈d〉 between electrons and positive ions (which is directly related to the density of charges in the material). Two different cases are examined, depending on the relative concentrations of electrons and ions. In the first case the concentration of electrons is assumed to be much smaller than the concentration of positive ions. Examination of a previously derived analytical equation demonstrates two different types of crystal size effects. When the tunneling length a is much smaller than both R and 〈d〉, the analytical equations show that smaller crystals exhibit a faster tunneling recombination rate. However, when the tunneling length a is of the same order of magnitude as both R and 〈d〉, the opposite effect is observed, with smaller crystals exhibiting a slower tunneling recombination rate. As the crystal size increases, the rate of tunneling in both cases reaches the limit expected for bulk materials. In the second case we examine the situation where the concentrations of electrons and positive ions are equal at all times. In this situation there is no analytical equation available to describe the process, and the crystal size effects are simulated by using Monte Carlo (MC) techniques. The two opposite behaviors as a function of the crystal size are also observed in these MC simulations. The effect of sample temperature is also studied by extending the MC simulations to include thermal characteristics of the defects. The relevance of the simulated results for luminescence dosimetry is discussed.

**107. Can thermoluminescence be used to determine soil heating from a
wildfire?
**

**Francis K. Rengers, Vasilis Pagonis, Shannon A. Mahan**

Radiation Measurements (2017), http://dx.doi.org/10.1016/j.radmeas.2017.09.002

**ABSTRACT**

The Silverado wildfire occurred from September 12 to 20, 2014, burning 960 acres in Orange County, California. Soil samples from within the burn area were obtained and the thermoluminescence (TL) properties of those samples were compared against a control sample to understand wildfire heating. We performed a series of experiments investigating the degree to which the control differed from the wildfire soil samples. This work showed that soil heated by a wildfire had a distinctly different glow curve shape than the unburned soil sample. Moreover, it was possible to see changes in the TL signal as a function of soil depth in wildfire-heated samples. Our experiments suggest that minimal soil heating occurred below approximately 10 cm. Estimates of wildfire temperatures, however, were nuanced.

**106. Sublinear dose dependence of thermoluminescence as a result of
competition between electron and hole trapping centers
**

**S.V. Nikiforov , V. Pagonis , A.S. Merezhnikov**

Radiation Measurements 105 (2017) 54-61

**ABSTRACT**

This paper presents a model of sublinearity of thermoluminescence (TL) dose response, based on the competitive interaction between active electron trapping centers and deep hole centers during excitation and heating of a phosphor. It is shown by analytical and numerical calculations that the sublinear response is observed at low doses, when the trap occupancy is far from saturation. It is found that the linearity index of TL dose response depends on the concentration of the deep hole traps, and on the coefficient of non-radiative electron recombination. The TL dose response is derived in parametric form. A new analytical expression is derived for the threshold doses, at which the TL response changes from a sublinear growth to an almost linear behavior. This phenomenon of a sublinear TL response changing into a linear behavior, is explained by the deep traps filling up to saturation. The dependence of the linearity index on the model parameters is found by numerical methods.

**105. Quantum tunneling recombination in a system of randomly distributed trapped electrons and positive ions
**

**Vasilis Pagonis, Christopher Kulp, Charity-Grace Chaney and M Tachiya**

J. Phys.: Condens. Matter 29 (2017) 365701

**ABSTRACT**

During the past 10 years, quantum tunneling has been established as one of the dominant mechanisms for recombination in random distributions of electrons and positive ions, and in many dosimetric materials. Specifically quantum tunneling has been shown to be closely associated with two important effects in luminescence materials, namely long term afterglow luminescence and anomalous fading. Two of the common assumptions of quantum tunneling models based on random distributions of electrons and positive ions are: (a) An electron tunnels from a donor to the nearest acceptor, and (b) the concentration of electrons is much lower than that of positive ions at all times during the tunneling process. This paper presents theoretical studies for arbitrary relative concentrations of electrons and positive ions in the solid. Two new differential equations are derived which describe the loss of charge in the solid by tunneling, and they are solved analytically. The analytical solution compares well with the results of Monte Carlo simulations carried out in a random distribution of electrons and positive ions. Possible experimental implications of the model are discussed for tunneling phenomena in long term afterglow signals, and also for anomalous fading studies in feldspars and apatite samples.

**104. Tunnelling recombination in conventional, post-infrared and post-infrared multi-elevated temperature IRSL signals in microcline K-feldspar
**

**Eren Şahiner, George Kitis, Vasilis Pagonis, Niyazi Meriç, George S. Polymeris**

Journal of Luminescence 188 (2017) 514–523

**ABSTRACT**

In this manuscript, we study three different types of infrared stimulated luminescence signals (IRSL) from a microcline K-feldspar. These signals consist of conventional infrared (IR) stimulation measured at elevated temperatures, multi-elevated temperature post-infrared IRSL signals (MET-pIRIR), and post-infrared IRSL (pIRIR) signals. All three types of signals were analysed using analytical expressions previously derived according to a model based on quantum tunnelling. For the purpose of identifying various tunnelling components, de-convolution analysis was used based on continuous wave infrared stimulation (CW-IRSL), as well as de-convolution based on pseudo linearly modulated transformation of the signals (pLM-IRSL). The variation of the parameters with stimulation temperature was determined using both of these de-convolution methods of analysis. The luminescence curves and normalized signals were investigated at various temperatures using several related protocols. The results of this analysis have implications for the understanding of the emission of luminescence in this material during the above-mentioned protocols, and they highlight the strengths and shortfalls of the tunnelling recombination models. The results from this study contribute to a better understanding of the origin of pIRIR and MET-pIRIR luminescence signals.

**103. Quartz radiofluorescence: a modelling approach
**

Johannes Friedrich, Vasilis Pagonis, Reuven Chen, Sebastian Kreutzer, Christoph Schmidt

Journal of Luminescence, 186 (2017) 318-325

**ABSTRACT**

Modelling (natural) quartz luminescence (TL/OSL) phenomena appears to be quite common nowadays. The corresponding simulations are capable of giving valuable insights into the charge transport system. By contrast, simulating radiofluorescence (RF) of quartz has rather been neglected in the past. Here we present and discuss (1) the RF signals of natural quartz measured in the UV band and (2) simulations of these experiments executed using a three-energy-level model to explain the experimentally obtained results. Two natural quartz samples were investigated at room temperature (RT) following different preheat procedures: (a) consecutively increasing preheat temperatures from 50 °C to 700 °C and (b) repeating a 500 °C preheat with subsequent UV-RF measurement at RT for eleven times. Based on the measurement and modelling results, we finally confirm theoretically the dependency of the UV-RF signal of quartz on the burial dose.

**LINK TO PDF FILE- PART 1**

**LINK TO PDF FILE- PART 2
LINK TO PDF FILE-PART 3**

**102. The influence of competition effects on the initial rise method during thermal stimulation of luminescence: A simulation study**

George Kitis, Vasilis Pagonis, S. E. Tzamarias

Radiation Measurements 100 (2017) 27-36

**Abstract**

The initial rise (IR) method is a widely used method to evaluate the activation energy of complex thermoluminescence (TL) glow-curves. It is generally accepted that competition effects take place among the various electrons traps responsible for the TL peaks of a glow curve. In the present study the potential influence of such competition effects on the IR method is studied. The simulation is divided in two parts. In the first part it is assumed that retrapping probabilities are lower than recombination probability. This simulation was carried out assuming either strong or weak competition from a thermally disconnected deep trap (TDDT). In the second part of the simulation the retrapping probabilities are assumed to be greater than the recombination probability. The simulation results showed that competition has serious effects on the resulting glow-curve shapes. Furthermore, the simulation showed that in most cases the IR method is able to reproduce accurately the input values of the activation energies of the three TL peaks. However, it was found that the IR method underestimated the activation energy of the last peak corresponding to the highest temperature by 8-20%, when the retrapping probabilities were higher than the recombination probability.

**101. Thermoluminescence associated with two-electron traps**

R. Chen , J.L. Lawless, V. Pagonis

Radiation Measurements 99 (2017) 10-17

**ABSTRACT**

A model of thermoluminescence (TL) is presented based on a double-occupancy electron trap and a

single-hole recombination center. The concept of double-electron traps has been established before with regard to different solid state phenomena and briefly mentioned as a possible occurrence in connection with TL. A new set of simultaneous differential equations governing the three stages of excitation, relaxation and readout of TL in this new framework is developed. This situation is dealt with by solving these sets of equations sequentially for reasonable sets of chosen trapping parameters. Also, an analytical treatment using plausible simplifying assumptions is given in parallel. The outcome of these procedures yields a two-peak TL curve and, in a sense, the two-electron trap behaves as two traps with different activation energies, frequency factors and retrapping probabilities. The results of the simulations and the approximate analytical approach show that the lower-temperature peak has features of first-order peak and is strongly superlinear with the dose of excitation. With the appropriate choice of parameters, the dose dependence of the first peak has been found to be cubic and slightly more than cubic with the excitation dose. This may explain experimental results of cubic and somewhat stronger superlinearity previously reported in the literature. The second peak has second-order features and it shifts dramatically to lower temperatures with increasing dose.

**100. An overview of recent developments in luminescence models with a
focus on localized transitions**

Vasilis Pagonis, Reuven Chen, Christopher Kulp, George Kitis

Radiation Measurements xxx (2017) 1-10:

http://dx.doi.org/10.1016/j.radmeas.2017.01.001

**ABSTRACT
**Advances in modeling during the past 20 years have contributed to better understanding of the luminescence properties of dosimetric materials. Three types of models have been used extensively in the literature: delocalized models based on transitions involving the conduction and valence bands, localized models usually involving different energy levels of the same trap, and semilocalized models which are based on a combination of localized and delocalized energy levels. The purpose of this paper is to provide an overview of recent developments in luminescence models, with an emphasis on the importance of delocalized transitions. Two recent theoretical developments are discussed, namely analytical equations based on the Lambert W-function which are applicable for delocalized models, and analytical equations based on tunneling in a random distribution of defects which are applicable for localized models. A new model for luminescence in quartz is proposed, which is applicable for time scales ranging from microseconds to seconds. Recent Monte Carlo simulations of ground state tunneling in a random distribution of traps and centers are discussed, which are based on a modified version of a previously published model. Some of the current challenges associated with luminescence signals measured at elevated temperatures are pointed out, and suggestions are made for future work in this research area.

**99. New expressions for half life, peak maximum temperature, activation energy and kinetic order of a thermoluminescence glow peak based on the Lambert W function**

George Kitis , Vasilis Pagonis

Radiation Measurements 97 (2017) 28-34

**ABSTRACT
**In this work approximate analytical expressions are derived for four different aspects of Thermoluminescence (TL) kinetics, by using the Lambert Wfunction. Firstly, analytical expressions are derived for the half life of trapped charge in a single trap within the one trap one center model (OTOR), and these are compared with analytical equations based on the empirical general order (GO) kinetics. Secondly, approximate analytical expressions are obtained for the temperature Tm of maximum TL intensity in a glow peak and for the activation energy E, by solving the condition for the maximum. Finally, an analytical expression was derived for evaluating the kinetic order b as a function of the integral symmetry factor of an experimental TL peak. Where necessary, the analytical expressions were successfully verified by numerical simulations of the OTOR model by using a wide range of possible values of the parameters in this model.

**98. Thermoluminescence glow curves in preheated feldspar samples:
An interpretation based on random defect distributions**

George S. Polymeris, Vasilis Pagonis, George Kitis

Radiation Measurements 97 (2017) 20-27

**ABSTRACT**

Thermoluminescence (TL) glow curves from feldspar samples have been studied extensively in luminescence dating and dosimetry applications. However, the mechanism responsible for their unusual shape and behavior in preheated samples is not well understood. This paper presents new experimental TL glow curves for four types of preheated feldspar samples; an orthoclase, a sanidine and two microclines. Both the preheat temperature and the duration of the preheat are varied, before measurement of the remnant TL glow curve. Kinetic analysis of the experimental results is carried out using a recently proposed physical kinetic model which describes localized electronic recombination in a random distribution of donor-acceptor pairs, with nearest-neighbor interaction. The experimental data are analyzed using two different methods within the model. In the first method one follows the development of the distribution of nearest neighbor distances for each of the four stages of the experiment, namely irradiation followed by heating to a temperature and holding the sample at this temperature for certain time, and finally measurement of the TL glow curve. In the second method the TL glow curves are analyzed by using a single adjustable parameter in the model, which characterizes the thermal history of the sample.Good agreement is found between the two methods. However, the second method is much simpler to use in practical situations, and in principle it can be applied for any thermally or optically pretreated feldspar sample.

**97. Quantitative analysis of time-resolved infrared stimulated luminescence in feldspars**

Vasilis Pagonis, Christina Ankjærgaard, Mayank Jain, Makaiko L. Chithambo

Physica B 497 (2016) 78–85

**ABSTRACT**

Time-resolved infrared-stimulated luminescence (TR-IRSL) from feldspar samples is of importance in the ﬁeld of luminescence dating, since it provides information on the luminescence mechanism in these materials. In this paper we present new analytical equations which can be used to analyze TR-IRSL signals, both during and after short infrared stimulation pulses. The equations are developed using a recently proposed kinetic model, which describes localized electronic recombination via tunneling be-tween trapped electrons and recombination centers in luminescent materials. Recombination is assumed to take place from the excited state of the trapped electron to the nearest-neighbor center within a random distribution of luminescence recombination centers. Different possibilities are examined within the model, depending on the relative importance of electron de-excitation and recombination. The equations are applied to experimental TR-IRSL data of natural feldspars, and good agreement is found between experimental and modeling results.

**96. On the intrinsic accuracy and precision of the standardised growth curve (SGC) and global-SGC (gSGC) methods for equivalent dose determination: A simulation study**

Jun Peng, Vasilis Pagonis , Bo Li

Radiation Measurements 94 (2016) 53-64

**ABSTRACT
**

In optically stimulated luminescence (OSL) dating, the single aliquot regenerative-dose (SAR) method has been used extensively for determining equivalent doses (De) in quartz. A variation of the SAR method is the “standardised growth curve” (SGC) method, which has been used as an efficient procedure to save measurement time during dating studies. During the application of the SGC method one establishes the SGC and calculation of the De of an aliquot requires only measurement of the standardised natural dose signal. Recently, a “global standardised growth curve” (gSGC) method was developed as an improved version of the SGC procedure. During the application of the gSGC method, the growth curves are renormalized using sensitivity-corrected signal corresponding to one of the regenerative doses. Subsequently the De of an aliquot is estimated using the sensitivity-corrected natural dose signal and an additional sensitivity-corrected regenerative dose signal as well as the established gSGC. In the present study, simulations are performed to assess the intrinsic accuracy and precision of the SGC and gSGC De estimates. The results of our simulations validate that the gSGC method is intrinsically more precise than the SGC method and is also more accurate for doses greater than 210 Gy. Several factors which affect the reliability of the two methods are investigated.

**95. Monte Carlo simulations of tunneling phenomena and nearest neighbor hopping mechanism in feldspars**

Vasilis Pagonis, Christopher Kulp

Journal of Luminescence 181 (2017) 114–120

**ABSTRACT**

The study of luminescence signals from geological materials is of importance in dosimetric and dating applications. Monte Carlo simulations are often used to describe the charge creation, charge trapping and donor/acceptor recombination processes in luminescent and dosimetric materials. While such stochastic methods are in wide use in many applied areas of physics, there have been few studies of such phe- nomena on the origin and production of luminescence signals for feldspars, apatites and similar mate- rials exhibiting loss of charge due to quantum tunneling phenomena. Previous Monte Carlo work on feldspars has been based on the assumption that the number density of donors and acceptors are equal at any time. These previous studies were able to get good agreement with experiment only when they assumed that the crystal consisted of small sub-volumes, and that charge carriers were only allowed to recombine within these nanocrystals. This paper provides a different version of this previously suggested model, in which the number density of acceptors far exceeds that of donors. The new version of the model describes the loss of charge due to ground state tunneling, as well as the charge creation by natural irradiation of samples. The results from the model compare well with previously derived approximate analytical equations for feldspars. In addition, the Monte Carlo simulations provide valuable insight into the various factors which affect the luminescence mechanism in these materials. The simulations can describe the loss of charge on a wide variety of time scales, from microseconds to thousands of years. The effect of crystal size, charge carrier density, natural irradiation dose rate and total number of charge carriers is studied in a quantitative manner. Finally we examine the possibility of extending the present version of the model to describe luminescence signals originating in the nearest neighbor hopping mechanism in feldspars. The results from the model are compared with experimental data from time-resolved infrared stimulated luminescence (TR-IRSL) in these materials.

**94. Reliability of single aliquot regenerative protocol (SAR) for dose estimation in quartz at different burial temperatures: A simulation study**

D.K. Koul , V. Pagonis, P. Patil

Radiation Measurements 91 (2016) 28-35

**ABSTRACT**

The single aliquot regenerative protocol (SAR) is a well-established technique for estimating naturally acquired radiation doses in quartz. This simulation work examines the reliability of SAR protocol for samples which experienced different ambient temperatures in nature in the range of 10 to 40 C. The contribution of various experimental variables used in SAR protocols to the accuracy and precision of the method is simulated for different ambient temperatures. Specifically the effects of paleo-dose, test dose, pre-heating temperature and cut-heat temperature on the accuracy of equivalent dose (ED) estimation are simulated by using random combinations of the concentrations of traps and centers using a previously published comprehensive quartz model. The findings suggest that the ambient temperature has a significant bearing on the reliability of natural dose estimation using SAR protocol, especially for ambient temperatures above 0 C. The main source of these inaccuracies seems to be thermal sensitization of the quartz samples caused by the well-known thermal transfer of holes between luminescence centers in

quartz. The simulations suggest that most of this inaccuracy in the dose estimation can be removed by delivering the laboratory doses in pulses (pulsed irradiation procedures).

**93. Spectral and kinetic analysis of thermoluminescence from
manganiferous carbonatite**

M.L. Chithambo, V. Pagonis, F.O. Ogundare

Journal of Luminescence 145 (2014) 180–187

**ABSTRACT**

Thermoluminescence spectra of manganiferous carbonatite has been studied from 30 1C to 400 1C over the wavelength range from 200 nm to 800 nm. The natural thermoluminescence appears above 200 1C and shows continuous spectral distribution from 240 nm to 800 nm. Above 500 nm the emission appears in the form of line structure emission. These are reproduced by laboratory irradiation which also produces broad bands near 100 1C. The emission features are attributed to presence of Mn2+ impurities in the carbonatite matrix. The spectral study was supplemented by kinetic analysis of the most prominent peaks and their kinetic features are reported.

**92. Prompt isothermal decay of thermoluminescence in MgB4O7:Dy, Na
and LiB4O7:Cu, In dosimeters
**

G. Kitis , G.S. Polymeris , I.K. Sfampa, M. Prokic, N. Meriç , V. Pagonis

G. Kitis , G.S. Polymeris , I.K. Sfampa, M. Prokic, N. Meriç , V. Pagonis

Radiation Measurements 84 (2016) 15-25

**ABSTRACT**

According to standard delocalized kinetic models of thermoluminescence (TL), when an irradiated

sample is held at a high temperature T, the isothermal TL signal will decay with a characteristic thermal decay constant l which depends strongly on the temperature T. This prediction of standard delocalized kinetic theory is investigated in this paper by studying two TL dosimeters, MgB4O7:Dy, Na and LiB4O7:Cu, In (hereafter MBO and LBO correspondingly). In the case of LBO it was found that the thermal decay constant l of the main dosimetric TL peak follows exactly the predictions of standard delocalized kinetic theory. Furthermore, the thermal activation energy of the main peak evaluated by the isothermal decay method is in full agreement with values obtained from initial rise and glow curve fitting methods. However, in the case of MBO it was found that the thermal decay constant l varies little with the isothermal decay temperature T. In order to explain these unusual results for MBO, the TL glow curves and isothermal decay curves were analyzed using analytical expressions derived recently from a radiative tunneling recombination model. Based on the different behavior of the two TL dosimeters, it is suggested that the isothermal decay of TL at high temperatures can be used to discriminate between radiative delocalized recombination and radiative localized recombination processes.

**91. Simulating comprehensive kinetic models for quartz luminescence using the R program KMS**

Jun Peng, Vasilis Pagonis

Radiation Measurements, Vol. 86, (2016), 63–67

**ABSTRACT**

Kinetic models have been used extensively for modeling and numerical simulation of luminescence

phenomena and dating techniques in quartz. In this paper we introduce the open access R program KMS for kinetic model simulation of luminescence phenomena in quartz. Several handy and compact functions are designed to simulate events in the geological history of quartz: crystallization, irradiation, optical illumination, and heating processes. These R functions provided by program KMS offer useful numerical tools to perform quartz luminescence simulations in a flexible manner. The simulation process can be understood easily by creating transparent programs by creating sequences of these compact R functions. With this contribution, we provide several practical examples of using the R functions in KMS to simulate the geological history of either sedimentary quartz samples, or of quartz samples which were fired in antiquity. In addition, examples are provided for simulations of commonly used dating protocols, like the multiple aliquot additive-dose technique (MAA) and the single aliquot regenerative-dose protocol (SAR).

**90. Mathematical aspects of ground state tunneling models in luminescence materials
**

**Vasilis Pagonis , George Kitis**

Journal of Luminescence 168 (2015) 137–144

**ABSTRACT**

Luminescence signals from a variety of natural materials have been known to decrease with storage time at room temperature due to quantum tunneling, a phenomenon known as anomalous fading. This paper is a study of several mathematical aspects of two previously published luminescence models which describe tunneling phenomena from the ground state of a donor–acceptor system. It is shown that both models are described by the same type of integral equation, and two new analytical equations are presented. The first new analytical equation describes the effect of anomalous fading on the dose response curves (DRCs) of naturally irradiated samples. The DRCs in the model were previously expressed in the form of integral equations requiring numerical integration, while the new analytical equation can be used immediately as a tool for analyzing experimental data. The second analytical equation presented in this paper describes the anomalous fading rate (g-Value per decade) as a function of the charge density in the model. This new analytical expression for the g-Value is tested using experimental anomalous fading data for several apatite crystals which exhibit high rate of anomalous fading. The two new analytical results can be useful tools for analyzing anomalous fading data from luminescence materials. In addition to the two new analytical equations, an explanation is provided for the numerical value of a constant previously introduced in the models.

**89. Time-resolved luminescence from quartz: An overview of contemporary developments and applications**

M.L. Chithambo, C.Ankjærgaard, V.Pagonis

Physica B 481 (2016) 8–18

**ABSTRACT**

Time-resolved optical stimulation of luminescence has become established as a key method for measurement of optically stimulated luminescence from quartz, feldspar and a-Al_{2}O_{3}:C, all materials of interest in dosimetry. The aim of time-resolved optical stimulation is to separate in time the stimulation and emission of luminescence. The luminescence is stimulated from a sample using a brief light pulse and the emission monitored during stimulation in the presence of scattered stimulating light or after pulsing, over photomultiplier noise only. Although the use of the method in retrospective dosimetry has been somewhat limited, the technique has been successfully applied to study mechanisms in the processes leading up to luminescence emission. The main means for this has been the temperature dependence of both the luminescence intensity as well as the luminescence lifetimes determined from time-resolved luminescence spectra. In this paper we review some key developments in theory and applications to quartz including methods of evaluating lifetimes, techniques of evaluating kinetic parameters using both the dependence of luminescence intensity and lifetime on measurement temperature, and of lifetimes on annealing temperature. The paper concludes with an overview of some notable applications such as separation of quartz signals from a quartz-feldspar admixture and the utility of the dynamic throughput, a measure of luminescence measured as a function of the pulse-width.

**88. Time and dose-rate dependence of TL and OSL due to competition
between excitation and fading**

R. Chen , V. Pagonis , J.L. Lawless

Radiation Measurements 82 (2015) 115-121

**ABSTRACT**

During the excitation period of thermoluminescence (TL) and optically stimulated luminescence (OSL), competing effects of fading may take place, in particular in situations of long irradiation with a relatively low dose rate. In this work, we study, by the use of numerical simulations, this possible occurrence, using a model with two trapping states and one recombination center. The dependence on the time of excitation of the filling of the active trap, to which the TL or OSL signal is proportional, has been followed. Using plausible sets of trapping parameters, the solution of the relevant set of coupled differential equations revealed a time dependence of an increase up to a maximum value and then a decrease toward an equilibrium value where the rates of production and decay are equal. The equilibrium value reached by the simulations has been found to be consistent with a direct comparison of the excitation and the thermal decay rates. The results are somewhat similar to previous reports on non-monotonic dose dependence with two main differences. The previous model included two trapping states and two centers whereas the present one has only one center. Also, previously, the dependence was on the dose with no distinction whether its variation is by changing the dose rate of excitation or the length of irradiation. With the present work, the non-monotonic and final equilibrium behavior have been specific to the time dependence whereas when the dose was varied by changing the dose rate, the curve increased monotonically until it reached the saturation value. Similar results have been reached by a quasi-analytical method, using some plausible simplifying assumptions.

**87. Dating quartz near saturation – Simulations and application at archaeological sites in South Africa and South Carolina
**

James K. Feathers , Vasilis Pagonis

Quaternary Geochronology xxx (2015) 1-6

**ABSTRACT**

Single-grain dating of quartz near saturation is shown to have the potential of under-estimating

equivalent dose. Experimental work shows that dose recoveries can be under-estimated when the

administered dose approaches saturation, an observation also seen by Duller (2012). Duller (2012) found that by calculating the ratio between the fast and medium bleaching components, the “fast ratio”, for each grain, the under-estimation can be corrected by removing those grains with low fast ratios. Similar results are shown for samples from archaeological sites in South Africa and South Carolina. To understand why grains with low fast ratios might lead to equivalent dose under-estimation, simulations using a comprehensive quartz model was employed. It was found that large grain-to-grain variation in the decay constants for the fast and medium components can cause this effect.

**86. On the effect of optical and isothermal treatments on luminescence signals from feldspars
**

Vasilis Pagonis, George Polymeris, George Kitis

Radiation Measurements 82 (2015) 93-101

ABSTRACT

During luminescence dosimetry and luminescence dating applications it is often necessary to precondition the geological samples by applying a thermal or optical treatment before measuring the luminescence signal. In luminescence applications using apatites or feldspars, measurement of continuous wave infrared or optically stimulated signals (CW-IRSL and CW-OSL) are customarily preceded by either an isothermal heating of the samples at a fixed temperature for a short time interval, or alternatively by optically bleaching the samples using light from LEDs with the appropriate wavelength. This paper presents new analytical equations which can be used to describe these commonly employed double experimental procedures. The equations are based on a recently published model which assumes that tunneling processes are taking place in random distributions of donoreacceptor pairs. The concentration of charge carriers during the CW-IRSL or CW-OSL experiment is expressed in terms of the parameters of the preceding thermal or optical bleaching procedure, and depends also on the distribution of distances between electron and hole pairs. The analytical equations in this paper are compared with experimental data from a feldspar sample which undergoes an isothermal procedure followed by measurement of the CW-IRSL signal. Additional comparisons with experiment are provided using a feldspar sample which undergoes an infrared bleaching process, followed by measurement of the CWOSL signal. These results and conditions under which the equations can be used are discussed within the framework of the model.

**85. Correlation of basic TL, OSL and IRSL properties of ten K-feldspar samples
of various origins**

I.K. Sfampa, G.S. Polymeris, V. Pagonis, E. Theodosoglou, N.C. Tsirliganis, G. Kitis

Nuclear Instruments and Methods in Physics Research B 359 (2015) 89–98

ABSTRACT

Feldspars stand among the most widely used minerals in dosimetric methods of dating using thermoluminescence (TL), optically stimulated luminescence (OSL) and infrared stimulated luminescence (IRSL). Having very good dosimetric properties, they can in principle contribute to the dating of every site of archaeological and geological interest. The present work studies basic properties of ten naturally occurring K-feldspar samples belonging to three feldspar species, namely sanidine, orthoclase and microcline. The basic properties studied are (a) the influence of blue light and infrared stimulation on the thermoluminescence glow-curves, (b) the growth of OSL, IRSL, residual TL and TL-loss as a function of OSL and IRSL bleaching time and (c) the correlation between the OSL and IRSL signals and the energy levels responsible for the TL glow-curve. All experimental data were fitted using analytical expressions derived from a recently developed tunneling recombination model. The results show that the analytical expressions provide excellent fits to all experimental results, thus verifying the tunneling recombination mechanism in these materials and providing valuable information about the concentrations of luminescence centers.

TO PDF FILE-Part 1

TO PDF FILE-Part 2

**84. Monte Carlo simulations of TL and OSL in nanodosimetric materials and feldspars**

Vasilis Pagonis, Reuven Chen

Radiation Measurements (2015),

http://dx.doi.org/10.1016/j.radmeas.2014.12.009

ABSTRACT

The study of luminescent materials consisting of nanoclusters is an increasingly active research area. It

has been shown that the physical properties of such nanodosimetric materials can be very different from

those of similar conventional microcrystalline phosphors. In addition, it has been suggested that traditional

energy band models may not be applicable for some of these nanodosimetric materials, because of

the existence of strong spatial correlations between traps and recombination centers. The properties of

such spatially correlated materials have been previously simulated by using Monte Carlo techniques and

by considering the allowed transitions of charge carriers between the conduction band, electron traps

and recombination centers. This previous research demonstrated successfully the effect of trap clustering

on the kinetics of charge carriers in a solid, and showed that trap clustering can significantly change the

observed luminescence properties. This paper presents a simplified method of carrying out Monte Carlo

simulations for thermoluminescence (TL) and optically stimulated luminescence (OSL) phenomena,

based on a recently published model for feldspar. This model is based on tunneling recombination

processes involving localized near-neighbor transitions. The simulations show that the presence of small

clusters consisting of a few traps can lead to multiple peaks in both the TL and linearly modulated OSL

signals. The effects of donor charge density, initial trap filling and cluster size are simulated for such

multi-peak luminescence signals, and insight is obtained into the mechanism producing these peaks.

**83. Radiation-induced growth and isothermal decay of infrared-stimulated luminescence**

**from feldspar**

Benny Guralnik, Bo Li, Mayank Jain, Reuven Chen, Richard B. Paris, Andrew S.

Murray, Sheng-Hua Li, Vasilis Pagonis, Pierre G. Valla, Frédéric Herman

Radiation Measurements (2015),

DOI: 10.1016/j.radmeas.2015.02.011

ABSTRACT

Optically stimulated luminescence (OSL) ages can determine a wide range of geological events or processes, such as the timing of sediment deposition, the exposure of a rock surface, or the cooling of bedrock. The accuracy of OSL dating critically depends on our capability to describe the growth and decay of laboratory regenerated luminescence signals. Here we review a selection of common models describing the response of infrared stimulated luminescence (IRSL) of feldspar to constant radiation and temperature as administered in the laboratory. We use this

opportunity to introduce a general-order kinetic model that successfully captures the behaviour of different materials and experimental conditions with a minimum of model parameters, and thus appears suitable for future application and validation in natural environments. Finally, we evaluate all the presented models by their ability to

accurately describe a recently published feldspar multi-elevated temperature post-IR IRSL (MET-pIRIR) dataset, and highlight each model’s strengths and shortfalls.

**82. Study of the stability of the TL and OSL signals**

R. Chen, V. Pagonis

Radiation Measurements (2015),

http://dx.doi.org/10.1016/j.radmeas.2015.01.006

ABSTRACT

In the study of thermoluminescence (TL) and optically stimulated luminescence (OSL), and in particular

in the applications of archaeological and geological dating as well as dosimetry, the issue of stability of

the signal at ambient temperature following excitation is of paramount importance. In many cases, one

determines the activation energy (E) and frequency factor (s) of a TL peak, and tries to evaluate the

lifetime of the excited signal. This is meaningful if the process is of pure first order, and may not be so in

non-first-order situations. In the present work, we study this matter for both first-order and the more

general one-trap-one-recombination-center (OTOR) cases using numerical simulations. The conventional

numerical solution of the relevant set of coupled differential equations may not work when the traps are

deep and the length of time is, say, thousands of years or more, and we therefore resort to a Monte-Carlo

approach. It is obvious that in instances of dominating recombination, the long-time decay is exponential,

and the decay constant is as expected from the first-order behavior and the E and s values.

However, in cases of substantial retrapping, the fading is slower, sometimes very significantly, and is not

exponential. Thus, one may deduce from the evaluated E and s shorter decay times than occur in fact.

This may lead to an apparent effect of unexpected stability, namely, that a signal is stable much longer

than expected from the evaluated trapping parameters. Possible implications concerning applications in

archaeological and geological dating are obvious.

**81. Intrinsic superlinear dose dependence of thermoluminescence and
optically stimulated luminescence at high excitation dose rates**

R. Chen, J.L. Lawless, V. Pagonis

Radiation Measurements 71 (2014) 220-225

ABSTRACT

Superlinear dose dependence of thermoluminescence (TL) and optically stimulated luminescence (OSL)

has been reported for many materials. The theoretical explanation has been ascribed to competition of

either traps or recombination centers, during the excitation stage or during the read-out phase. There has

been an account in the literature on superlinearity of OSL associated with merely one trapping state and

one kind of recombination center. This had to do with the process taking place during the read-out stage,

namely the optical stimulation. In the present work, we report on a model of one trapping state and one

kind of recombination center which results in a superlinear filling of the center. Thus, one can expect a

superlinear dose dependence of the area under the resulting TL glow peak as well as the OSL signal. We

follow this situation by writing the simultaneous nonlinear rate equations for the one-trap-onerecombination-

center (OTOR) model and study the expected results by numerical simulation consisting

of solving the equations with sets of the trapping parameters. We also present analytical results

based on simplifying assumptions, and compare the analytical and numerical results. The effect is significant

at relatively high dose rates. The main implication is that when one tries to evaluate by TL

dosimetry a dose applied at a high rate, calibration of the TL dosimeter using much smaller dose rates

may result in inaccurate results.

**80. Mathematical characterization of continuous wave infrared stimulated luminescence signals(CW-IRSL) from feldspars**

V. Pagonis, HuyPhan, RebeccaGoodnow, SaraRosenfeld, P.Morthekai

Journal of Luminescence 154 (2014) 362–368

ABSTRACT

Continuous-wave infrared stimulated luminescence signals (CW-IRSL) from feldspars have been the subject of many experimental studies, due to their importance in luminescence dating and dosimetry. Accurate mathematical characterization of the shape of these CW-IRSL signals in feldspars is of practical and theoretical importance, especially in connection with “anomalous fading” of luminescence signals in dating studies. These signals are known to decay in a non-exponential manner and their exact mathematical shape as a function of stimulation time is an open research question. At long stimulation times the IRSL decay has been shown experimentally to follow a power law of decay, and previous researchers have attempted to fit the overall shape of these signals empirically using the well known Becquerel function (or compressed hyperbola decay law). This paper investigates the possibility of fitting CW-IRSL curves using either the Becquerel decay law, or a recently developed analytical equation based on localized electronic recombination of donor-acceptor pairs in luminescent materials. It is shown that both mathematical approaches can give excellent fits to experimental CW-IRSL curves, and the precision of the fitting process is studied by analyzing a series of curves measured using a single aliquot of a feldspar sample. Both fitting equations are solutions of differential equations involving numerically similar time dependent recombination probabilities k(t). It is concluded that both fitting equations provide approximately equivalent mathematical descriptions of the CW-IRSL curves in feldspars, and can be used as mathematical representations of the shape of CW-IRSL signals.

**79. Monte Carlo simulations of luminescence processes under quasiequilibrium (QE) conditions**

Vasilis Pagonis, Ethan Gochnour, Michael Hennessey, Charles Knower

Radiation Measurements 67 (2014) 67-76

ABSTRACT

Previous researchers have carried out Monte Carlo simulations of thermoluminescence (TL) phenomena by considering the allowed transitions of charge carriers between the conduction band, electron traps and recombination centers. Such simulations have demonstrated successfully the effect of trap clustering on the kinetics of charge carriers in a solid, and showed that trap clustering can significantly change the observed luminescence properties. While such Monte Carlo simulations have been carried out for TL, there has been no such trap clustering studies for optically stimulated luminescence phenomena (OSL). This paper presents a simplified method of carrying out Monte Carlo simulations for TL and linearly modulated optically stimulated luminescence (LM-OSL) phenomena, based on the General One Trap (GOT) model, which is a special case of the one trap one recombination center model (OTOR) when quasi-equilibrium conditions (QE) hold. The simulated results show that the presence of small clusters consisting of a few traps in a solid can lead to multiple peaks in both the TL and LM-OSL signals. The

effects of retrapping and degree of trap filling are simulated for such multi-peak luminescence signals, and insight is obtained into the mechanism producing these peaks. The method presented in this paper can be easily generalized for other types of luminescence solids in which the recombination probability varies with time.

**78. Kinetic analysis of thermoluminescence glow curves in feldspar: evidence of a continuous distribution of energies **

Vasilis Pagonis, P. Morthekai and George Kitis

GEOCHRONOMETRIA 41(2) 2014: 168–177

ABSTRACT

The thermoluminescence (TL) glow curves from feldspars have been the subject of numerous studies, because of their importance in luminescence dating and dosimetry. This paper presents new experimental TL glow curves in a plagioclase feldspar, measured using the T_{max}-T_{stop} technique of glow curve analysis. Kinetic analysis of the experimental results is carried out for a freshly irradiated sample, as well as for a sample which has undergone optical treatment using infrared light for 100 s at 50^{o}C. Application of the initial rise method of analysis indicates that the TL signals from both samples can be characterized by a continuous distribution of energy levels . By subtracting the TL glow curves measured at successive T_{stop} values, a series of TL glow curves is obtained which are analyzed using the empirical general order kinetics. It is found that all TL glow curves obtained by this subtractive procedure can be described accurately by the same general order parameter b~1.7. In a second attempt to analyze the same TL glow curves and possibly extract information about the underlying luminescence process, the shape of TL glow curves is analyzed using a recently proposed physical kinetic model which describes localized electronic recombination in donor-acceptor pairs. Within this model, recombination is assumed to take place via the excited state of the donor, and nearest-neighbor recombinations take place within a random distribution of centers. This recent model has been used to describe successfully several types of luminescence signals. This paper shows that it is possible to obtain good fits to the experimental data using either one of these two approaches

**77. On the shape of continuous wave infrared stimulated luminescence signals from feldspars: A case study**

Vasilis Pagonis, M.Jain, K.J.Thomsen , A.S.Murray

Journal of Luminescence 153 (2014) 96–103

ABSTRACT

The continuous-wave IRSL signals from feldspars (CW-IRSL) are known to decay in a non-exponential manner, and their exact mathematical description is of great importance in dosimetric and dating studies. This paper investigates the possibility of fitting experimental CW-IRSL curves from a variety of feldspar samples, by using an analytical equation derived within the framework of a new model based on localized electronic recombinations of donor-acceptor pairs. 24 different types of feldspars were studied and their CW-IRSL signals are analyzed in order to establish the range and precision of numerical values for the fitting parameters in the analytical equation. The study searches for systematic trends in the fitting parameters, and for possible systematic differences between K and Na rich extracts from the same feldspar samples. Furthermore, results are compared from natural samples, freshly irradiated samples, and samples which had undergone anomalous fading. The results of this analysis establish broad numerical ranges for the fitting parameters in the model. Specifically the possible range of the dimensionless density rho was found to be rho~0.002-0.01. These experimentally established ranges of will help guide future modeling work on luminescence processes in feldspars. Small statistical differences were found between K-rich and Na-rich fractions of the same sample. However, the experimental data shows that the parameters depend on the irradiation dose, but do not depend on the time elapsed after the end of the irradiation process. All samples exhibited the power law of decay, with the range of the power law coefficient beta=0.6-1.1

**76. Properties of thermoluminescence glow curves from tunneling recombination processes in random distributions of defects**

George Kitis and Vasilis Pagonis

Journal of Luminescence 153 (2014) 118–124

ABSTRACT

Localized electronic recombination processes in donor-acceptor pairs of luminescent materials have been recently modeled using a new kinetic model based on tunneling. Within this model, recombination is assumed to take place via the excited state of the donor, and nearest-neighbor recombinations take place within a random distribution of centers. An approximate semi-analytical version of the model has been shown to simulate successfully thermally and optically stimulated luminescence (TL, OSL), linearly modulated OSL (LM-OSL) and isothermal TL processes. This paper presents a detailed analysis of the geometrical properties of the TL glow curves obtained within three different published versions of the model. The dependence of the shape of the TL glow curves on the kinetic parameters of the model is examined by allowing simultaneous random variations of the parameters, within wide ranges of physically reasonable values covering several orders of magnitude. It is found that the TL glow curves can be characterized according to their shape factors m_{g}, as is commonly done in TL theory of delocalized transitions. The values of the shape factor are found to depend rather weakly on the activation energy E and the frequency factor s, but they have a strong dependence on the parameter which characterizes the concentration of acceptors in the model. It is also shown by simulation that both the variable heating rate and initial rise methods are applicable in this type of model and can yield the correct value of the activation energy E. However, the initial rise method of analysis for the semianalytical version of the model fails to yield the correct E value, since it underestimates the low temperature part of the TL glow curves. Two analytical expressions are given for the TL intensity, which can be used on an empirical basis for computerized glow curve deconvolution analysis (CGCD).

**75. The role of simulations in the study of thermoluminescence (TL)**

Reuven Chen, Vasilis Pagonis

Radiation Measurements (2014), http://dx.doi.org/10.1016/j.radmeas.2013.12.011

ABSTRACT

The traffic of charge carriers in a luminescent material during its excitation by irradiation and during readout either in the measurement of thermoluminescence (TL) or optically stimulated luminescence (OSL) is governed by sets of coupled nonlinear differential equations. The analytical solution of these sets is usually not possible, and one can resort to one of two options. Some researchers preferred to make simplifying assumptions and thus got approximate solutions whereas others performed simulations by solving the simultaneous equations numerically. Each of these routes has its pitfalls. The simplifying assumptions, mainly the quasi-equilibrium assertion or the assumption that certain relations between the relevant parameters and functions hold, may be valid in certain ranges of the TL or OSL curve, and may cease to be valid, say at the high-temperature range in TL. Performing simulations using the numerical solution of the relevant set of equations may yield results which are accurate, but cannot be considered as being general because they depend on the specific choice of the parameters. Repeating the simulations with several sets of the physically plausible parameters would add credibility to the conclusions drawn. The combination of the two approaches is highly recommended, i.e, if similar results are found by approximations and simulations, the validity of the conclusions is strengthened. Evidently, the comparison of these theoretical results to experimental effects is essential. In the present work we consider the occurrence of unusually high and unusually low values of the activation energy and the effective frequency factor. In particular, we can simulate a recently discovered behavior of TL in LiF:Mg, Cu, P at the ultra-high dose range and get qualitatively the main elements of the experimentally found results.

**74. Prompt isothermal decay of thermoluminescence in an apatite exhibiting strong anomalous fading**

I.K. Sfampa, G.S. Polymeris, N.C. Tsirliganis, V. Pagonis, G. Kitis

Nuclear Instruments and Methods in Physics Research B 320 (2014) 57–63

ABSTRACT

Anomalous fading (AF) is one of the most serious drawbacks in thermoluminescence (TL) and optically stimulated luminescence (OSL) dating. In the present work the isothermal decay of TL signals from Durango apatite is studied for temperatures located on the rising part of the main TL peak. This material is known to exhibit strong AF phenomena, and its isothermal TL decay properties have not been studied previously. The experimental results show that the characteristic decay time of the isothermal signal does not depend of the temperature, and that this signal does not exhibit the strong temperature dependence expected from conventional TL kinetic theories. This is further direct experimental evidence for the possible presence of tunneling phenomena in this material. The isothermal decay curves are analyzed and discussed within the framework of conventional theories of TL, as well as within the context of a recently developed tunneling kinetic model for random distributions of electron-hole pairs in luminescent materials.

**73. Further investigations of tunneling recombination processes in random distributions of defects**

Vasilis Pagonis, Huy Phan, David Ruth, George Kitis

Radiation Measurements (2013), http://dx.doi.org/10.1016/j.radmeas.2013.08.006

ABSTRACT

The shape of infrared stimulated luminescence signals (IRSL) from feldspars has been the subject of numerous studies in the field of luminescence dating. Specifically linearly modulated IRSL signals (LM-IRSL) are commonly assumed to consist of several first order components corresponding to distinct optical stimulation cross sections. This paper models the shape of LM-IRSL signals using a recently proposed kinetic model, which describes localized electronic recombination in donoreacceptor pairs of luminescent materials. Within this model, recombination is assumed to take place via the excited state of the donor, and nearest-neighbor recombinations take place within a random distribution of centers. The model has been used previously successfully to describe both thermally and optically stimulated luminescence (TL, OSL). This paper shows that it is possible to obtain approximate solutions for the distribution of donors in the ground state as a function of two variables, time and the distance between donors and acceptors. Approximate expressions are derived for several possible modes of optical and thermal stimulation, namely TL, OSL, linearly modulated OSL (LM-OSL), LM-IRSL and isothermal TL (ITL). Numerical integration of these expressions over the distance variable yields the distribution of remaining donors at any time t during these experimental situations. Examples are given for the derived distributions of donors in each experimental case, and similarities and differences are pointed out. The paper also demonstrates how LM-IRSL signals in feldspars can be analyzed using the model, and what physical information can be extracted from such experimental data. The equations developed in this paper are tested by fitting successfully a series of experimental LM-IRSL data for Na- and K-feldspar samples available in the literature. Finally, it is shown that the equations derived in this paper are a direct generalization of an equation previously derived for the case of ground state tunneling.

**72. Modeling TL-like thermally assisted optically stimulated luminescence (TA-OSL)**

R. Chen, V. Pagonis

Radiation Measurements 56 (2013) 6-12

ABSTRACT

In this work we present a model for thermally assisted optically stimulated luminescence (TA-OSL), an effect previously observed in experimental work. The model consists of one trapping state with an excited state beneath the conduction band and one kind of recombination center. In one version, we assume that an electron is thermally elevated to the excited state. From there it can either be raised optically into the conduction band or it can be de-excited back into the ground state of the trap. Once in the conduction band, the electron may retrap in the excited state or recombine with a hole in a center, yielding a luminescence photon. The other version is similar except that once the electron is in the excited state, it may be elevated to the conduction band either optically or thermally, with activation energy E2 and frequency factor s2. Using approximations along with analytical considerations as well as numerical simulations consisting of the solution of the simultaneous differential equations with linear heating function, the resulting TL-like curves are studied. First-order like and second-order like cases as well as intermediate cases are identified. The conditions for reaching these situations are considered; it is shown that they are not simply the same as in regular TL, and in addition to the trapping parameters, the intensity of the stimulating light also plays a role.

**71. On the expected order of kinetics in a series of thermoluminescence (TL) peaks**

R. Chen and V. Pagonis

Nuclear Instruments and Methods in Physics Research B 312 (2013) 60–69

ABSTRACT

As reported in the literature, both in experimental results and in simulated glow curves, in a series of TL peaks associated with a series of trapping states and a single recombination center, the peaks tend to be of first order. In the present work we show theoretically and demonstrate by examples of numerical simulations that the last peak may be of second order whereas the lower-temperature peaks are usually of first order. This is the case even when retrapping is significantly faster than recombination. In some cases, the last peak has a long tail, longer than that of second-order peaks, which has to do with a different mechanism that has been discussed in a recent paper. Similar simulations of a more complex and more realistic situation of a model with multiple trapping states and multiple recombination centers have been performed. The prevalence of first-order appearance of both the curves of free electrons, associated with thermally stimulated conductivity (TSC) and of TL, evaluated for randomly chosen sets of trapping parameters, is shown by histograms. The occurrence of a small number of very high values of the symmetry factor is also discussed

**70. THERMAL DEPENDENCE OF LUMINESCENCE LIFETIMES AND RADIOLUMINESCENCE IN QUARTZ**

V. Pagonis, M.L. Chithambo, R. Chen, A. Chruścińska, M. Fasoli, S.H. Li, M. Martini, K. Ramseyer

Journal of Luminescence 145(2014)38–48

ABSTRACT

During time-resolved optical stimulation experiments (TR-OSL), one uses short light pulses to separate the stimulation and emission of luminescence in time. Experimental TR-OSL results show that the luminescence lifetime in quartz of sedimentary origin is independent of annealing temperature below 500^{o}C, but decreases monotonically thereafter. These results have been interpreted previously empirically on the basis of the existence of two separate luminescence centers L_{H} and L_{L} in quartz, each with its own distinct luminescence lifetime. Additional experimental evidence also supports the presence of a non-luminescent hole reservoir R, which plays a critical role in the predose effect in this material. This paper extends a recently published analytical model for thermal quenching in quartz, to include the two luminescence centers L_{H} and L_{L}, as well as the hole reservoir R. The new extended model involves localized electronic transitions between energy states *within* the two luminescence centers, and is described by a system of differential equations based on the Mott-Seitz mechanism of thermal quenching. It is shown that by using simplifying physical assumptions, one can obtain analytical solutions for the intensity of the light during a TR-OSL experiment carried out with previously annealed samples. These analytical expressions are found to be in good agreement with the numerical solutions of the equations. The results from the model are shown to be in quantitative agreement with published experimental data for commercially available quartz samples. Specifically the model describes the variation of the luminescence lifetimes with (a) annealing temperatures between room temperature and 900°C, and (b) with stimulation temperatures between 20 and 200°C. This paper also reports new radioluminescence (RL) measurements carried out using the same commercially available quartz samples. Gaussian deconvolution of the RL emission spectra was carried out using a total of seven emission bands between 1.5 and 4.5 eV, and the behavior of these bands was examined as a function of the annealing temperature. An emission band at ~3.44 eV (360 nm) was found to be strongly enhanced when the annealing temperature was increased to 500°C, and this band underwent a significant reduction in intensity with further increase in temperature. Furthermore, a new emission band at ~3.73 eV (330 nm) became apparent for annealing temperatures in the range 600-700^{o}C. These new experimental results are discussed within the context of the model presented in this paper.

**69. ****Anomalous heating rate effect in thermoluminescence intensity using a simplified**

**semi-localized transition (SLT) model**

Vasilis Pagonis, Leigh Blohm, Mark Brengle, Gina Mayonado, Patrick Woglam

Radiation Measurements 51-52 (2013) 40-47

ABSTRACT

During thermoluminescence (TL) measurements carried out with different heating rates, one expects the

total number of emitted photons to be constant. However, for many luminescence materials one observes

a decreased intensity of luminescence at elevated temperatures, due to the presence of the well-known

phenomenon of thermal quenching. Recent experimental work on the dosimetric material YPO4 double

doped with lanthanides demonstrated the exact opposite behavior, in which the total luminescence

intensity increases with the heating rate during the TL experiments. This anomalous TL behavior was

recently explained by using the Mandowski model of semi-localized transitions (SLTs). In this paper it is

shown that this anomalous heating rate or “anti-quenching” phenomenon can also be described by using

a simplified SLT model of TL with approximated kinetic equations. The simulated glow curves show that

as the probability of the non-radiative processes increases, the anomalous heating rate effect becomes

dominant. The dependence of the anomalous heating rate effect on the values of the kinetic parameters

in the model is examined by allowing random variations of the parameters in the model, within wide

ranges of physically reasonable values covering several orders of magnitude. It is shown by simulation

that the variable heating rate method can systematically underestimate the value of the activation energy

E, while by contrast the initial rise method of analysis almost always yields the correct E value. These

simulated results are discussed in relation to recent experimental work on the double doped dosimetric

material YPO4.

**68. Analytical solutions for stimulated luminescence emission from tunneling
recombination in random distributions of defects
**

George Kitis and Vasilis Pagonis

Journal of Luminescence 137 (2013) 109–115

ABSTRACT

Recently a new kinetic model was presented in the literature, which describes localized electronic

recombination in donor–acceptor pairs of luminescent materials. Within this model, recombination is

assumed to take place via the excited state of the donor, and nearest-neighbor recombinations take

place within a random distribution of centers. Two versions of the model were presented which were

found to be in good agreement with each other, namely an exact model that evolves both in space and

in time, and an approximate semi-analytical model evolving only in time. The model simulated

successfully both thermally stimulated luminescence (TL) and optically stimulated luminescence (OSL),

and also demonstrated the power law behavior for simulated OSL signals. This paper shows that the

system of simultaneous differential equations in the semi-analytical model can be approximated to an

excellent precision by a single differential equation. Furthermore, analytical solutions are obtained for

this single differential equation, and for four different experimental modes of stimulation: TL, OSL,

linearly modulated OSL (LM-OSL) and isothermal TL processes. The exact form of the power law for the

model is found in analytical form for both OSL and isothermal TL processes. The analytical equations are

tested by successfully fitting typical infrared stimulated luminescence (IRSL) signals, as well as

experimental TL glow curves from feldspar samples. The dimensionless number density of acceptors

in the model is estimated from fitting the experimental IRSL and TL data. The analytical expressions

derived in this paper apply also to stimulated emission via the excited state of the donor–acceptor

system. However, the same analytical expression, with different numerical values for its constants, can

also be applied in the case of ground state tunneling, with important implications for luminescence

dating.

**67. Anomalous fading of OSL signals originating from very deep traps in Durango
apatite**

G. Kitis , G.S. Polymeris, V. Pagonis , N.C. Tsirliganis

Radiation Measurements 49 (2013) 73-81

ABSTRACT

Athermal or anomalous fading (AF) is one of the most serious problems in thermoluminescence (TL) and optically stimulated luminescence (OSL) dating. Several possible ways have been suggested for choosing a non-fading signal in feldspars, including high temperature thermal preconditioning of the samples. The aim of the present work is to search for stable OSL signals in Durango apatite, a material which exhibits very strong AF effects, by focusing on the existence of optically sensitive very deep traps (VDT). These are

traps which are responsible for TL peaks above 500 C. Two experimental protocols are used to establish the optimal stimulation temperature for these OSL signals from VDT, in order to estimate the activation energy for these signals, and to study their AF properties. It is found that the OSL signals from VDT are more stable than conventional signals from traps responsible for TL peaks below 500 C, and that they exhibit smaller anomalous fading effects. A new luminescence signal is reported in this material, consisting

of a long-lived luminescence decay, observed after the end of short 1 s optical pulses. This new relaxation signal is observed after both blue light stimulation and infrared stimulation of the sample.

**66. Thermal dependence of time-resolved blue light stimulated luminescence in a-Al2O3:C**

Vasilis Pagonis, Christina Ankjærgaard, Mayank Jain, Reuven Chen

Journal of Luminescence 136 (2013) 270–277

ABSTRACT

This paper presents time-resolved optically stimulated luminescence (TR-OSL) experiments in the important dosimetric material Al_{2}O_{3}:C. During these experiments short pulses (0.5 s) of light from blue LEDs (470 nm) are followed by relaxation periods (2.5 s) of the charge carriers at different stimulation temperatures. During the pulse excitation period the integrated TR-OSL signal increases with the stimulation temperature between 50 and 150^{o}C, while between 160 and 240^{o}C the signal intensity decreases. This behavior is interpreted to arise from competing effects of thermal assistance (activation energy, E_{th}= 0.067±0.002 eV) and thermal quenching (activation energy *W=*(1.032+0.005) eV). Changes in the shape of the TR-OSL curves were analyzed at different stimulation temperatures using analytical expressions available in the literature. The TR-OSL signals contain a slower temperature-dependent phosphorescence signal, the “delayed-OSL” described previously for this material. The temperature dependent luminescence lifetimes obtained from analysis of the optical stimulation period are identical to those obtained from the corresponding relaxation period. However, the values of these luminescence lifetimes are systematically higher than previously reported values from time-resolved photoluminescence (TR-PL) studies carried out in this important dosimetric material. These results are discussed within the context of a recently published kinetic model.

LINK TO PDF FILE____________________________________________________________________________________________________

**65. On the quasi-equilibrium assumptions in the theory of thermoluminescence(TL)**

R. Chen and V.Pagonis**
**Journal of Luminescence 143 (2013) 734–740

ABSTRACT

The phenomenon of thermoluminescence (TL) is known to be governed by a set of simultaneous differential equations. When one studies the properties of a single peak, resulting from the thermal release of electrons from a trap into the conduction band, followed by radiative recombination with holes in centers, the set consists of three non-linear equations. Even in this relatively simple one trap-one center (OTOR) case, the equations cannot be solved analytically. In order to be able to get approximate solutions, the conventional way has been to make the “quasi-equilibrium” assumptions, namely that |*dn _{c}*/

*dt*| is significantly smaller than |

*dn/dt*|; |

*dm/dt*| where

*n*and

*m*are the occupancies of traps and centers, respectively and

*n*is the concentration of electrons in the conduction band, and

_{c}*n*<<

_{c}*n*and

*n*<<

_{c}*m*. We show here, using simulations as well as analytical arguments that the former condition often does not take place however, its consequences are usually valid. The reason is that the conventional quasi-equilibrium must be replaced by a somewhat different condition. As for the second condition related to the smallness of the concentration of free electrons, we show that it may not be fulfilled at the high-temperature end of a single glow peak or in the highest-temperature peak in a series. In some cases, this condition results in a very broad high-temperature tail of the TL peak, as previously observed experimentally in several materials.

**64. Time-resolved infrared stimulated luminescence signals in feldspars: Analysis based on exponential and stretched exponential functions**

V. Pagonis, P.Morthekai, A.K.Singhvi, J.Thomas, V.Balaram, G.Kitis, R.Chen

Journal of Luminescence 132 (2012) 2330–2340

ABSTRACT

Time-resolved infrared-stimulated luminescence (TR-IRSL) signals from feldspar samples have been the subject of several recent experimental studies. These signals are of importance in the field of luminescence dating, since they exhibit smaller fading effects than the commonly employed continuous-wave infrared signals (CW-IRSL). This paper presents a semi-empirical analysis of TR-IRSL data from feldspar samples, by using a linear combination of exponential and stretched exponential (SE) functions. The best possible estimates of the five parameters in this semi-empirical approach are obtained using five popular commercially available software packages, and by employing a variety of global optimization techniques. The results from all types of software and from the different fitting algorithms were found to be in close agreement with each other, indicating that a global optimum solution has likely been reached during the fitting process. Four complete sets of TR-IRSL data on well- characterized natural feldspars were fitted by using such a linear combination of exponential and SE functions. The dependence of the extracted fitting parameters on the stimulation temperature is discussed within the context of a recently proposed model of luminescence processes in feldspar. Three of the four feldspar samples studied in this paper are K-rich, and these exhibited different behavior at higher stimulation temperatures, than the fourth sample which was a Na-rich feldspar. The new method of analysis proposed in this paper can help isolate mathematically the more thermally stable components, and hence could lead to better dating applications in these materials.

**63. Reconstruction of thermally quenched glow curves in quartz**

Bhagawan Subedi, George S. Polymeris, Nestor C. Tsirliganis, Vasilis Pagonis and George Kitis

Radiation Measurements 47 (2012) 250-257

ABSTRACT

The experimentally measured thermoluminescence (TL) glow curves of quartz samples are influenced by the presence of the thermal quenching effect, which involves a variation of the luminescence efficiency as a function of temperature. The real shape of the thermally unquenched TL glow curves is completely unknown. In the present work an attempt is made to reconstruct these unquenched glow curves from the quenched experimental data, and for two different types of quartz samples. The reconstruction is based on the values of the thermal quenching parameter W (activation energy) and C (a dimensionless constant), which are known from recent experimental work on these two samples. A computerized glow-curve deconvolution (CGCD) analysis was performed twice for both the reconstructed and the

experimental TL glow curves. Special attention was paid to check for consistency between the results of these two independent CGCD analyses. The investigation showed that the reconstruction attempt was successful, and it is concluded that the analysis of reconstructed TL glow curves can provide improved values of the kinetic parameters E, s for the glow peaks of quartz. This also leads to a better evaluation of the half-lives of electron trapping levels used for dosimetry and luminescence dating.

**62. Modeling of the shape of infrared stimulated luminescence signals in feldspars**

Vasilis Pagonis, Mayank Jain, Andrew S. Murray, Christina Ankjærgaard and Reuven Chen

Radiation Measurements (2012), doi:10.1016/j.radmeas.2012.02.012

ABSTRACT

This paper presents a new empirical model describing infrared (IR) stimulation phenomena in feldspars. In the model electrons from the ground state of an electron trap are raised by infrared optical stimulation to the excited state, and subsequently recombine with a nearest-neighbor hole via a tunneling process, leading to the emission of light. The model explains the experimentally observed existence of two distinct time intervals in the luminescence intensity; a rapid initial decay of the signal followed by a much slower gradual decay of the signal with time. The initial fast decay region corresponds to a fast rate of recombination processes taking place along the infrared stimulated luminescence (IRSL) curves. The subsequent decay of the simulated IRSL signal is characterized by a much slower recombination rate, which can be described by a power-law type of equation. Several simulations of IRSL experiments are carried out by varying the parameters in the model. It is found that the shape of the IRSL signal is remarkably stable when the kinetic parameters are changed within the model; this is in agreement with several previous studies of these signals on feldspars, which showed that the shape of the IRSL curves does not change significantly under different experimental

conditions. The relationship between the simulated IRSL signal and the well-known power-law dependence of relaxation processes in solids is also explored, by fitting the IRSL signal at long times with a power-law type of equation. The exponent in this power-law is found to depend very weakly on the various parameters in the model, in agreement with the results of experimental studies. The results from the model are compared with experimental IRSL curves obtained using different IR stimulating power, and good quantitative agreement is found between the simulation results and experimental data.

**61. Superlinear dose response of thermoluminescence (TL) and optically stimulated luminescence (OSL) signals in luminescence materials: An analytical approach**

Vasilis Pagonis, Reuven Chen and John L. Lawless

Journal of Luminescence 132 (2012) 1446–1455

ABSTRACT

The phenomenon of superlinear dose response of thermoluminescence (TL) and optically stimulated luminescence (OSL) signals has been reported for several important dosimetric materials. We develop new analytical equations for the filling of traps and centers during irradiation and for the read-out stage of annealed luminescence materials, within the context of a two-trap and two-center model. The equations are applicable for both TL and OSL signals in annealed dosimetric materials, and are derived under the assumptions of low irradiation doses and dominant strong retrapping (weak recombination) processes. For low doses all traps and centers display linear dose response, which leads to quadratic dose response of the integrated TL/OSL signals. A new analytical expression is presented for this well-known quadratic dose dependence, in terms of the kinetic parameters in the model. The effect of elevated irradiation temperature on the integrated TL/OSL signals is also considered, and analytical expressions are obtained for this situation as well. A new type of dose-rate effect is reported based on the modeling results, which is due to irradiation during elevated temperatures. The accuracy of the analytical expressions is verified by comparing with the results of numerical simulations.

**60. A model for explaining the concentration quenching of thermoluminescence**

R. Chen , J.L. Lawless, V. Pagonis

Radiation Measurements 46 (2011) 1380-1384

ABSTRACT

The effect of concentration quenching (CQ) of luminescence has been reported in the literature. The luminescence efficiency dependence on the concentration of a specific impurity was found to reach a maximum intensity for a certain concentration, and decline at higher concentrations. A formula has been developed for the dependence of the efficiency on the concentration, assuming that only activators not adjacent to other activators can emit luminescence. Curve fitting of the CQ experimental curves to the theoretical function resulted in very large values of the parameter z, the number of nearest neighbors, of up to 4000, which is not feasible. A similar effect was found in TL of some materials, and the same formula for explaining the effect was used. Medlin has described the TL properties of calcite and dolomite. For a 300 K peak in Pb++ doped calcite, he used the same function and found z = 700, and for a 410 K peak, he got z= 150; the maxima occurred at different concentrations. In the present work, we propose a possible, alternative model to explain the QC of thermoluminescence (TL). The model includes 3 trapping states and one recombination center (3T1C model). We assume that the 3 traps have a constant concentration, and the variable concentration is that of the recombination center M. An important assumption made is that the initial occupancy of M is not zero, and we assume that m(0)= 0.1M. The results yield the concentration dependence of the area under two simulated peaks reached by solving numerically the relevant set of six simultaneous rate equations. The maximum intensities of the two peaks occur at different concentrations, similarly to experimental results in Pb++ doped calcite and Mn++ doped dolomite. Approximate analytical derivations support these results.

**59. Two-stage thermal stimulation of thermoluminescence**

R. Chen, J.L. Lawless, V. Pagonis

Radiation Measurements 47 (2012) 809-813

ABSTRACT

A thermoluminescence (TL) model of two-stage stimulation of electrons into the conduction band is discussed. This release of the carriers is assumed to take place via an intermediate localized excited state. Electrons are thermally stimulated from the trap into an excited state and then thermally released into the conduction band from which they may either be retrapped or recombine with holes in centers. The model resembles the previous “semi localized” model, but we concentrate only on recombination of electrons that go through the conduction band. It also bears similarity to the effect of thermally-assisted optically stimulated luminescence (OSL) previously discussed in the literature. The model is studied by solving the set of the relevant four simultaneous differential equations which govern the process during heating or isothermal decay. Using different sets of parameters, we can get pseudo-first-order, pseudo-second-order as well as intermediate cases, which are identified by their symmetry coefficient. Once the effective order is established, different analytical methods are used to determine the effective activation energy and frequency factor. We used the peak-shape methods, the various heating rate (VHR) method and the method based on the change of phosphorescence decay with temperature. The results are compared to the parameters used in the simulation. In many cases, the effective activation energy is equal to E1 þ E2 where E1 and E2 are, respectively, the activation energies for the first and second stage of thermal stimulation. The numerical simulation results are accompanied by an analytical treatment using the usual quasi-steady assumption. Unusual cases, in which the effective frequency factor and the effective retrapping probability coefficient are temperature dependent, are identified. Some cases in which the effective activation energy is close to E1 rather than E1 þ E2 are identified and discussed. The relevance of this possible situation to the evaluation of the stability of TL signals is also considered, and a possible effect of anomalous stability is predicted.

**58. Prevalence of first order kinetics in thermoluminescence materials: An explanation based on multiple competition processes**

Vasilis Pagonis and George Kitis Phys. Status Solidi B, 1–12 (2012) / DOI 10.1002/pssb.201248082

ABSTRACT

Typical materials used in thermoluminescence (TL) dosimetry exhibit the following common characteristics: (i) the temperature of glow peak maximum of individual glow peaks remains practically constant over a wide dose range, (ii) there are no systematic changes in the glow curve shapes with the irradiation dose, and (iii) higher order kinetics is rarely seen in dosimetric materials, while first-order kinetics is a common occurrence in experimental TL work. Theoretical explanation of these experimental characteristics is an open topic of TL research. In the present work these three characteristics are studied by using several models of increasing complexity. The simplest model studied is based on the empirical analytical general order (GO) expressions, followed by two commonly used models, the wellknown

one trap one recombination center models (OTOR) and the interactive multiple trap system (IMTS). Previous researchers have studied the behavior of these models using arbitrary values of the kinetic parameters in the models, and by varying these parameters within limited physically reasonable ranges. In this paper, a new method of analyzing the results from such models is presented, in which the average behavior of real dosimetric materials is simulated by allowing simultaneous random variations of the kinetic parameters, within several orders of magnitude. The simulation results lead to the conclusion that the presence of many competitive processes during the heating stage of TL, may be correlated to the remarkable stability of the glow curve shapes exhibited by most materials, and to the prevalence of first-order kinetics. This correlation is demonstrated further by a series of simulations in which the number of competitor traps is increased systematically, by adding up to 12 competitor traps in the IMTS model. As the number of competitor traps increases, the average behavior of theTL glow curves tends progressively toward first order kinetics, and this in turn results in very small average variations in the shape of the TL glow peak. The simulation results in this paper provide a convincing demonstration and explanation of the stability of the shape of TL glow curves in dosimetric materials, and for the prevalence of first-order kinetics in TL.

**57. Preliminary results towards the equivalence of transformed continuous wave OSL (CW-OSL) and LM-OSL signals in quartz**

G. Kitis, G. Polymeris, N. Kiyak and V.Pagonis

GEOCHRONOMETRIA (2011), DOI 10.2478/s13386-011-0031-8

ABSTRACT

The present paper presents a comparative experimental study of two commonly measured Optically Stimulated Luminescence (OSL) signals in quartz. The experimental study measures both the continuous wave OSL (CW-OSL) and the linearly modulated (LM-OSL) signals from the same quartz sample for a range of stimulation temperatures between 180 and 280°C, while the former is transformed to pseudo LM-OSL (ps LM-OSL). A computerized deconvolution curve analysis of the LM-OSL and ps LM-OSL signals was carried out, and the contributions of several OSL components to the initial OSL signal (0.1 s) were shown to be independent of the stimulation temperature used during the measurement. It was also found that the composite OSL (0.1 s) signal consists mainly of the first two OSL components present in the OSL curves. The equivalence of the ps LM-OSL (trans-formed CW-OSL) and of LM-OSL measurements was also examined by an appropriate choice of the experimental stimulation times, and of the stimulation power of the blue LEDs used during the meas-urement.

**56. Analytical expressions for time-resolved optically stimulated luminescence experiments in quartz**

V. Pagonis, J. Lawles, R. Chen, M.L. Chithambo

Journal of Luminescence 131 (2011) 1827–1835

ABSTRACT

Optically stimulated luminescence (OSL) signals can be obtained using a time-resolved optical stimulation (TR-OSL) method, also known as pulsed OSL. During TR-OSL measurements, the stimulation and emission of luminescence are experimentally separated in time using short light pulses. This paper presents analytical expressions for the TR-OSL intensity observed during and after such a pulse in quartz experiments. The analytical expressions are derived using a recently published kinetic model which describes thermal quenching phenomena in quartz samples. In addition, analytical expressions are derived for the concentration of electrons in the conduction band during and after the TR-OSL pulse, and for the maximum signals attained during optical stimulation of the samples. The relevance of the model for dosimetric applications is examined, by studying the dependence of the maximum TR-OSL signals on the degree of initial trap filling, and also on the probability of electron retrapping into the dosimetric trap. Analytical expressions are derived for two characteristic times of the TR-OSL mechanism; these times are the relaxation time for electrons in the conduction band, and the corresponding relaxation time for the radiative transition within the luminescence center. The former relaxation time is found to depend on several experimental parameters, while the latter relaxation time depends only on internal parameters characteristic of the recombination center. These differences between the two relaxation times can be explained by the presence of localized and delocalized transitions in the quartz sample. The analytical expressions in this paper are shown to be equivalent to previous analytical expressions derived using a different mathematical approach. A description of thermal quenching processes in quartz based on AlO4-/AlO4 defects is presented, which illustrates the connection between the different descriptions of the luminescence process found in the literature.

**55. Simulations of thermally transferred OSL signals in quartz: Accuracy and precision of the protocols for equivalent dose evaluation**

Vasilis Pagonis, Grzegorz Adamiec, C. Athanassas, Reuven Chen, Atlee Baker, Meredith Larsen, Zachary Thompson

Nuclear Instruments and Methods in Physics Research B 269 (2011) 1431–1443

ABSTRACT

Thermally-transferred optically stimulated luminescence (TT-OSL) signals in sedimentary quartz have been the subject of several recent studies, due to the potential shown by these signals to increase the range of luminescence dating by an order of magnitude. Based on these signals, a single aliquot protocol termed the ReSAR protocol has been developed and tested experimentally. This paper presents extensive numerical simulations of this ReSAR protocol. The purpose of the simulations is to investigate several aspects of the ReSAR protocol which are believed to cause difficulties during application of the protocol. Furthermore, several modified versions of the ReSAR protocol are simulated, and their relative accuracy

and precision are compared. The simulations are carried out using a recently published kinetic model for quartz, consisting of 11 energy levels. One hundred random variants of the natural samples were generated by keeping the transition probabilities between energy levels fixed, while allowing simultaneous random variations of the concentrations of the 11 energy levels. The relative intrinsic accuracy and precision of the protocols are simulated by calculating the equivalent dose (ED) within the model, for a given natural burial dose of the sample. The complete sequence of steps undertaken in several versions of the dating protocols is simulated. The relative intrinsic precision of these techniques is estimated by fitting Gaussian probability functions to the resulting simulated distribution of ED values. New simulations are presented for commonly used OSL sensitivity tests, consisting of successive cycles of sample irradiation with the same dose, followed by measurements of the sensitivity corrected L/T signals. We investigate

several experimental factors which may be affecting both the intrinsic precision and intrinsic accuracy of the ReSAR protocol. The results of the simulation show that the four different published versions of the ReSAR protocol can reproduce accurately the natural doses in the range 0–400 Gy with approximately the same intrinsic precision and accuracy of 1–5%. However, these protocols underestimate doses above 400 Gy; possible sources of this underestimation are investigated. Two possible explanations are suggested for the modeled underestimations, possible thermal instability of the TT-OSL traps, and the presence of thermally unstable medium OSL components in the model.

**54. On the intrinsic accuracy and precision of luminescence dating techniques for fired ceramics**

Vasilis Pagonis, Reuven Chen, George Kitis

Journal of Archaeological Science 38 (2011) 1591-1602

ABSTRACT

Luminescence dating techniques have been used extensively for archaeological and geological samples. Such techniques are based on thermally or optically stimulated signals. This paper presents simulations of several luminescence techniques for equivalent dose (ED) estimation for ceramic materials containing quartz. The simulations are carried out using a recently published comprehensive kinetic model for quartz, consisting of 11 electron and hole traps and centers. The complete sequence of the experimental protocols for several thermoluminescence (TL) and optically stimulated luminescence (OSL) techniques are simulated using the same set of kinetic parameters. The specific simulated protocols are: additive dose TL protocol, predose technique (both additive and multiple activation versions), phototransfer protocol, single aliquot regenerative optically stimulated luminescence (SAR-OSL) protocol, and SAR thermoluminescence protocol (SAR-TL). One hundred random variants of the natural samples were generated by keeping the transition probabilities between energy levels fixed, while allowing simultaneous random variations of the concentrations of the 11 energy levels. The relative intrinsic accuracy and precision of the protocols are simulated by calculating the equivalent dose (ED) within the model, for

a given natural burial dose of the sample. The intrinsic accuracy of these techniques is estimated by simulating natural irradiation of the samples with a known burial dose, followed by simulation of the luminescence method used to recover the estimated dose ED. The percent difference between the burial dose and the ED value represents the simulated accuracy of the luminescence technique. The relative intrinsic precision of these techniques is estimated by fitting Gaussian probability functions to the ED values obtained with the 100 model variants. It is found that the various techniques can reproduce natural paleodoses in the range 10 mGye10 Gy with a typical intrinsic accuracy of þ1 to 10%. Techniques based on single aliquot protocols were found in general to be more precise than techniques requiring the use of multiple aliquots. In addition, techniques based on interpolation of experimental data were found to be consistently both more precise and accurate than those based on extrapolation of experimental data.

**53. Dissolution and subsequent re-crystallization as zeroing mechanism, thermal properties and dose response of salt (NaCl) for retrospective dosimetry**

G. Polymeris, G. Kitis, N. Kiyak, I. Sfamba,B.Subedi, V.Pagonis

Applied Radiation and Isotopes 69 (2011) 1255–1262

ABSTRACT

In the present study we report dosimetric properties of iodized salt aiming at using it as an accidental luminescent dosimeter. It was found that the very good sensitivity of its main dosimetric peak is strongly affected by thermal treatments. This is also the case for OSL emission. The sensitivity loss due

to heating implies that caution should be exercised while applying single aliquot protocols for dose evaluation. The sequence of dissolution and subsequent re-crystallization was established to be an extremely effective zeroing mechanism for the TL signal. The linearity in the dose response was also monitored in the case of dissolved and subsequently re-crystallized salt. In the case of naturally occurring salt, zeroing of the TL signal due to dissolution as well as the linearity of dose response up to doses as large as 100 Gy were found to be very promising features for dating applications.

**52. Precision and accuracy of two luminescence dating techniques for retrospective dosimetry: SAR-OSL and SAR-ITL**

Vasilis Pagonis, Atlee Baker, Meredith Larsen, Zachary Thompson

Nuclear Instruments and Methods in Physics Research B 269 (2011) 653–663

ABSTRACT

Luminescence techniques based on thermally or optically stimulated signals are used extensively for estimating the equivalent dose (ED) of quartz samples for dating and retrospective dosimetry. This paper presents simulations of two luminescence dating protocols which use single aliquots of the quartz samples. The first protocol is the well-known single-aliquot regenerative optically stimulated luminescence (SAR-OSL) protocol for quartz. The second protocol was developed more recently and is based on a thermoluminescence (TL) signal measured under isothermal conditions (termed the SAR-ITL technique). The simulations are carried out using a recently published comprehensive kinetic model for quartz, consisting of 11 electron and hole traps and centers. The complete sequence of the two experimental protocols is simulated using the same set of kinetic parameters. The simulated dose response curves for the two protocols are found to be very similar to published experimental data. The relative intrinsic accuracy and precision of the two techniques is estimated by simulating one hundred random variants of the natural samples, and by calculating the equivalent doses using each technique. The 100 simulated natural variants are generated by keeping the transition probabilities between energy levels fixed, while allowing simultaneous random variations of the concentrations of the 11 energy levels. The SAR-OSL protocol was found to be intrinsically both more accurate and more precise than the SAR-ITL protocol. We investigate several experimental factors which affect the precision and accuracy of the two protocols. New simulations are presented for commonly used sensitivity tests consisting of successive cycles of sample irradiation with the same dose, followed by measurements of the sensitivity corrected L/T signals. These new simulations provide valuable insight into the previously reported sensitivity changes taking place during application of the SAR-ITL protocol.

**51. Simulations of time-resolved photoluminescence experiments in a-Al2O3:C**

Vasilis Pagonis, Reuven Chen, John W. Maddrey, Benjamin Sapp

Journal of Luminescence 131 (2011) 1086–1094

ABSTRACT

This paper presents simulations of time-resolved photoluminescence (TR-PL) experiments in a-Al2O3:C, which is one of the main dosimetric materials. During TR-PL experiments, short pulses of UV-light are followed by relaxation periods of the charge carriers. The model used in these simulations was

previously used to explain radioluminescence (RL), thermoluminescence (TL) and photoluminescence (PL) experiments for this material, and is based on optical and thermal ionizations of excited F-centers. There are no published simulations of TR-PL experiments in this important dosimetric material in the literature. In this paper, we present new simulations using two slightly different versions of the model. In the first original version of the model, thermal quenching is explained via thermal ionization of the recombination centers. In the new proposed modified version of the model, thermal quenching is described by a Mott–Seitz type of mechanism, based on competitions between radiative and radiationless electronic transitions occurring within the recombination center. We simulate a typical TR-PL experiment in Al2O3:C at different stimulation temperatures, and compare the simulation results with available experimental data. It is found that both versions of the model provide a reasonable quantitative description of the luminescence lifetime and luminescence intensity as a function of the stimulation temperature. However, very significant differences between the two models are found for the behavior of the simulated integrated thermoluminescence (TL) and thermally stimulated conductivity (TSC) as a function of the heating rate used during such experiments. Only the results from the modified version of the model, (which is based on Mott–Seitz mechanism), are in good agreement with previously reported experimental results. The two models also predict very different behaviors for the dependence of the optically integrated stimulated luminescence (OSL) and TSC signals, as a function of the stimulation temperature. The results from the two models suggest that it may be possible to decide between the two mechanisms of thermal quenching in this material, by carrying out accurate measurements of the TL, TSC and OSL signals under different experimental conditions. The effect of shallow traps on the luminescence lifetimes is also studied and compared with available

experimental data.

**50. Simulation of the Nonlinear Dose Dependence of Stabilized Point Defects**

R Chen, V Pagonis and J L Lawless

IOP Conf. Series: Materials Science and Engineering 15 (2010) 012071

doi:10.1088/1757-899X/15/1/012071

ABSTRACT

The dose dependence of the concentration of point defects in alkali-halides as well as other crystals, as exhibited by the dependence of the thermoluminescence (TL), optical absorption and ESR on the dose of non-ionizing UV excitation is studied using numerical simulation. The relevant set of coupled rate equations are first written and plausible sets of trapping parameters are chosen. Instead of using simplifying assumptions previously used for reaching conclusions concerning this dose behavior, exact numerical solutions have now been reached. Depending on the parameters chosen, different dose dependencies are seen. In some cases, linear dose dependence is reached in a broad range. Sublinear dose dependence, close to

a D1/2 dependence when D is the dose of excitation can be reached when retrapping is stronger than trapping in other traps stabilizing the defects. When strong competition between stabilizing traps takes place, an initial linear range is observed followed by strong superlinearity and an approach to saturation. All these behaviors have been observed experimentally in TL measurements as well as ESR and optical absorption in different

materials. Similarities and dissimilarities to linear and non-linear dose dependencies obtained experimentally and by simulations when ionizing irradiation is used for excitation are discussed.

**49. Simulation of the influence of thermal quenching on thermoluminescence glow-peaks**

B. Subedi, G. Kitis and V. Pagonis

Phys. Status Solidi A 207, No. 5, 1216–1226 (2010)

ABSTRACT

The thermal quenching of luminescence efficiency is an effect which is present in many thermoluminescent (TL) materials. It

causes a significant decrease of the luminescence signal and disturbs the shape of the glow-peaks. Therefore, in principle,

the thermoluminescence kinetics theory cannot describe TL glow-peaks influenced by thermal quenching. In the present

work a detailed simulation of the influence of the thermal quenching effect on thermoluminescence glow-peaks is

presented. Specifically we study the shift of the quenched glow-peak with heating rate and the effect on the various

heating rate methods, the influence on the symmetry factor and the kinetic order of the glow-peak, and the effect of thermal

quenching on the initial rise and peak shape methods for evaluating kinetic parameters. Furthermore, the evaluation of

the thermal quenching parameters using the quenched glowpeak and the possibility of using the conventional expression

describing a single glow-peak to fit the quenched glow peaks are also investigated.

**48. Simulations of isothermal processes in the semilocalized transition (SLT) model of thermoluminescence (TL)**

V. Pagonis and C. Kulp

J. Phys. D: Appl. Phys. 43 (2010) 175403 (8pp)

ABSTRACT

Semilocalized transition (SLT) kinetic models for thermoluminescence (TL) contain characteristics of both a localized transition (LT) and of a single trap model. TL glow curves within SLT models typically contain contain two TL peaks; the first peak corresponds to the intra-pair luminescence due to LTs and the second TL peak corresponds to delocalized transitions involving the conduction band (CB). The latter delocalized TL peak has also been found to exhibit non-typical double-peak structure, in which the main TL peak is accompanied by a smaller peak called the displacement peak. This paper describes the simulation of isothermal luminescence signals using a previously published SLT model. It is found that these simulated isothermal signals exhibit several unusual time characteristics. Isothermal signals associated with the LTs follow first order kinetics and are therefore described by single decaying exponentials. However, isothermal signals associated with delocalized transitions show a non-typical complex structure characterized by several time regions with different decay characteristics. For certain values of the parameters in the SLT model the isothermal signals can also exhibit non-monotonic behaviour as a function of time. Another notable result from the simulations is that isothermal currents (which are proportional to the concentration of electrons in the CB) can persist for very long periods of time, even after the apparent termination of the isothermal luminescence signals. It is concluded that isothermal processes described by the SLT model depend strongly on the presence of SLTs, in contrast to previous studies using Monte Carlo simulations, which showed a weak interdependence of these phenomena. The simulations in this paper suggest that isothermal experiments offer a sensitive method for detecting the presence of SLTs in a dosimetric material.

**47. On the initial-occupancy dependence of some luminescence phenomena under the one-trap-one-recombination-center (OTOR) model**

R.Chen, V. Pagonis, J. Lawless, 2010

Radiation Measurements 45 (2010) 147–150

ABSTRACT

We discuss the expected dependence of the maximum signal of a number of luminescence phenomena,

used in dosimetry and archaeological and geological dating, on the initial occupancy of the relevant

traps, n0, within the OTOR (also called General One Trap (GOT)) model. This, in turn, has important

bearing on the dose dependence of these phenomena. We discuss the dependence of the linearlymodulated

optically-stimulated-luminescence (LM-OSL) as well as the non-linear optical stimulation

(NL-OSL) on the initial concentration n0 of trapped carriers. We also consider the behavior of CW-OSL

and phosphorescence in a transformed form, as well as TL measured under hyperbolic and linear

heating rates. Using an appropriate presentation, the maximum of the signal is seen to depend nearly

linearly on n0, which in many cases means nearly linear dependence on the dose, a property important

for dosimetry and dating.

**46. Nonlinear dose dependence of TL and LM-OSL within the one trap-one center model**

R.Chen, V. Pagonis, J. Lawless

Radiation Measurements 45 (2010) 277–280

ABSTRACT

In a recent paper, it has been shown that strong sub-linearity of the occupancy n0 of trapping states far

from saturation can be explained by the simplest model of one trap-one recombination center (OTOR). In

the present work we report on results of numerical simulation of dose dependence of the TL maximum

under similar conditions. In some cases, the TL maximum is found to be strictly proportional to the filling

of the traps, but this is not always the case. Different sublinear dose-dependence functions of the trap

occupancy and the maximum TL are demonstrated. With the same sets of parameters, curves of LM-OSL

have also been simulated; superlinear as well as sublinear dependencies on the excitation dose have

been found

**45. Investigation of OSL signals from very deep traps in unfired and fired quartz samples**

G. Kitis, N.G. Kiyak, G.S. Polymeris, V. Pagonis

Nuclear Instruments and Methods in Physics Research B 268 (2010) 592–598

This paper presents an attempt to isolate experimentally optically stimulated luminescence (OSL) signals which may originate from very deep traps (VDT) in quartz samples. As VDT we consider those traps which are responsible for TL glow peaks with a peak maximum temperature above a TL readout temperature of 500 degC. The basic experimental procedure used to isolate OSL signals from VDT is heating the quartz samples to 500 degC immediately before measuring the OSL signal. The study was carried out on eight quartz samples of very different origins; it is found that all eight samples exhibit OSL signals from VDT, and for a wide region of OSL stimulation temperatures. The OSL signal from VDT depends strongly on the type of quartz sample studied and on whether the sample was fired at high temperatures or not. The behavior of the OSL signal from VDT as a function of the stimulation temperature is found to be very different in fired and unfired samples. The thermal activation energy E for the OSL signals from VDT is obtained in both fired and unfired samples. The OSL signal from VDT in quartz samples fired at 800 deg 1 h is very high, and the OSL curves consist of three well-defined components and a fourth slow component hich is rather poorly resolved. The dose response of these components is obtained using a computerized econvolution procedure for the dose region 0.5–300 Gy. The results are of importance for ating of ancient fired ceramics, since OSL signals from VDT could potentially extend appreciably the quivalent dose region toward both lower and higher values.

**44. Mixed-order kinetics model for optically stimulated luminescence**

G. Kitis, C. Furetta and V.Pagonis

Modern Physics Letters B, Vol. 23, No. 27 (2009) 3191–3207

ABSTRACT

The theory of mixed-order kinetics is well-established for the description of single

thermoluminescence (TL) glow-peaks. The main advantage of mixed-order kinetics relative

to the more widely used general-order kinetic theory is that the former is physically

meaningful whereas the latter is entirely empirical. In the case of optically stimulated

luminescence (OSL) either non-first-order or second-order kinetics are studied using the

empirical general-order kinetics theory. In the present work, expressions for mixed-order

kinetics are derived for OSL curves. A peak shape parameter for linear modulation OSL

is developed and special mixed-order expressions are derived for use in the computerized

OSL curve deconvolution analysis.

**43. Modelling the thermal quenching mechanism in quartz based on time-resolved optically stimulated luminescence**

V. Pagonis, C. Ankjærgaard, A.S. Murray, M. Jain, R. Chen, J. Lawless, S. Greilich

Journal of Luminescence 130 (2010) 902–909

Abstract

This paper presents a new numerical model for thermal quenching in quartz, based on the previously suggested Mott–Seitz mechanism. In the model electrons from a dosimetric trap are raised by optical or thermal stimulation into the conduction band, followed by an electronic transition from the conduction band into an excited state of the recombination center. Subsequently electrons in this excited state undergo either a direct radiative transition into a recombination center, or a competing thermally assisted non-radiative process into the ground state of the recombination center. As the temperature of the sample is increased, more electrons are removed from the excited state via the non-radiative pathway. This reduction in the number of available electrons leads to both a decrease of the intensity of the luminescence signal and to a simultaneous decrease of the luminescence lifetime. Several simulations are carried out of time-resolved optically stimulated luminescence (TR-OSL) experiments, in which the temperature dependence of luminescence lifetimes in quartz is studied as a function of the

stimulation temperature. Good quantitative agreement is found between the simulation results and new experimental data obtained using a single-aliquot procedure on a sedimentary quartz sample.

**42. Sublinear dose dependence of thermoluminescence and optically stimulated luminescence prior to the approach to saturation level**

J.L. Lawless, R. Chen , V. Pagonis

Radiation Measurements (2009), doi:10.1016/j.radmeas.2009.03.003

Abstract

In the use of thermoluminescence (TL) and optically stimulated luminescence (OSL) for dosimetry and for geological and archaeological dating, the nature of the dose dependence of the luminescence signal is of great importance. Non-linear dependence has been shown to result either from non-linear filling of the relevant traps or recombination centers during excitation, or by a combined effect of the linear filling of traps and centers due to processes taking place during the read-out stage. Sublinearity, which had been found in several materials, was usually attributed to saturation effects during excitation of either the relevant traps or centers. Sometimes, the competition effects during the excitation between traps result in superlinearity of some TL peaks and sublinearity of others. In the present work, we show that sublinear dose dependence may take place even in the simplest possible case of one trap–one recombination center (OTOR), even when the traps and centers are far from saturation. Analytical derivations as well as simulations consisting of the numerical solution of the relevant sets of coupled differential equations show the occurrence of the sublinear dose dependence under these circumstances. The filling of the traps is shown to behave like D^1/2 where D is the excitation dose, for an appropriate choice of the trapping parameters. This, in turn, may result in a similar dose dependence of the TL and OSL signals.

**41. Simulations of thermally transferred OSL experiments and of the ReSAR dating protocol for quartz**

Vasilis Pagonis , Ann G. Wintle, Reuven Chen, X.L. Wang

Radiation Measurements (2009), doi:10.1016/j.radmeas.2009.02.009

Abstract

In several recent studies, a thermally transferred OSL (TT-OSL) signal has been proposed as the basis of a new OSL dating procedure. In this paper we present simulations of several published TT-OSL experiments. The first simulation is of repeated cycles of preheating (260degC for 10 s) and TT-OSL measurements in order to separate the recuperated OSL (ReOSL) and basic-transferred OSL (BT-OSL) signals. The second simulation is of the temperature dependence of the TT-OSL signal after the aliquots were annealed for 1 s at temperatures from 200 to 400 degC. In addition, we simulate the construction of a dose response curve using the ReOSL signal in a single-aliquot regenerative-dose (SAR) protocol used for dating. The results of the three simulations are in general qualitative agreement with the experiments and confirm that a single charge transfer mechanism is responsible for the ReOSL signal.

**40. Radioluminescence in Al2O3 : C – analytical and numerical simulation results**

V Pagonis, J Lawless, R Chen and C Andersen

J. Phys. D: Appl. Phys. 42 (2009) 175107 (9pp)

Abstract

The phenomenon of radioluminescence (RL) has been reported in a number of materials including Al2O3 :C, which is one of the main dosimetric materials. In this work, we study RL using a kinetic model involving two trapping states and two kinds of recombination centres. The model has been previously used to provide a quantitative description of the thermoluminescence and optically stimulated luminescence processes in Al2O3 :C. Using appropriate sets of trapping parameters for the kinetic model, the RL signal along with the occupancies of the relevant traps and centres are simulated numerically. The set of differential equations is also solved analytically by assuming dynamic balance during sample irradiation. Analytical expressions are obtained for the concentrations of traps and centres in the material during irradiation with short irradiation pulses, by assuming that quasi-steady conditions hold during irradiation. Several experimentally observed characteristics of the RL signals are explained by using the model. Good quantitative agreement is found between the analytical expressions and the numerical solutions of the model for short irradiation pulses.

**39. On the theoretical basis for the duplicitous thermoluminescence peak**

J L Lawless, R Chen and V Pagonis

J. Phys. D: Appl. Phys. 42 (2009) 155409 (8pp)

Abstract

The simultaneous release of electrons and holes by what seems to be a single trap has been observed experimentally. We previously performed numerical simulations on a phenomenological model which showed similar behaviour. Here, we provide an analytical solution to this model. This model explains trends in radioluminescence, thermoluminescence and thermally stimulated conductivity of a material with one electron trap, one hole trap and one radiative recombination centre, in which thermal excitation of the electron trap occurs before that of the hole trap. It is shown that TL emission due to electron recombination at centres can be controlled by a hole trap and the electron recombination will have a peak shape associated with the hole trap’s parameters. When this happens, the peaks in free electron concentration, free hole concentration and TL all occur nearly simultaneously. The analytical model allows this to be explained along with scaling laws and initial rise behaviour. Under the conditions illustrated by this model, the usual methods used to distinguish between electron traps and hole traps will give incorrect results.

**38. A new look at the linear-modulated optically stimulated luminescence (LM-OSL) as a tool for dating and dosimetry**

Reuven Chen, Vasilis Pagonis, John L. Lawless

Radiation Measurements 44 (2009) 344–350

Abstract

Optically stimulated luminescence (OSL) has been in use for dosimetry and dating in the last two decades. Since the OSL dependence on time is a featureless decaying function, a linear-modulation of the stimulating-light intensity has been suggested [Bulur, E., 1996. An alternative technique for optically stimulated luminescence. Radiat. Meas. 26, 701–709.], which resulted in a peak-shaped curve. The properties of this curve have been studied, assuming first-, second- and general-order kinetics. In a recent paper we have shown [Chen, R., Pagonis, V., 2008. A unified presentation of thermoluminescence (TL), phosphorescence and linear-modulated OSL (LM-OSL). J. Phys. D: Appl. Phys. 41, 035102 (1–6).] that for general-order curves, the peak maximum cannot be expected to depend linearly on the dose of excitation. A new presentation of the LM-OSL has been suggested, in which the peak maximum is linear with the filling of trapping states, which, in turn, may be expected to be linear with the dose under appropriate conditions. In the present work, we report on results of numerical simulation of the LM-OSL using the one trap-one recombination center (OTOR) model, dealing with the traffic of carriers between one trapping state, one kind of recombination center and the conduction and valence bands during excitation and read-out, and without making any simplifying assumptions. The process during optical read-out has been followed in the simulation that consisted of the numerical solution of the relevant sets of coupled differential equations, and also by analytical treatment. Sets of parameters leading to approximately first- and second-order kinetics, and to intermediate cases, have been used and the results presented in the original and the new ways are shown. The consequences concerning dating and dosimetry are discussed.

**37. Optically stimulated exoelectron emission processes in quartz: comparison of experiment and theory**

V. Pagonis, C. Ankjærgaard, A.S. Murray, R. Chen

Journal of Luminescence 129 (2009) 1003–1009

Abstract Recent experiments have demonstrated that it is possible to measure optically stimulated exoelectron emission (OSE) signals simultaneously with optically stimulated luminescence (OSL) from quartz samples. These experiments provide valuable information on the charge movement in quartz grains. Two specific experiments measured the temperature dependence of the OSL and OSE signals on preheat and stimulation temperature. This paper provides a quantitative description of these experiments by using a previously published theoretical model for photostimulated exoelectron emission (PSEE). The experimental data yield a value of w~1.2 eV for the work function of quartz. The experimental temperature dependence of the OSE signals is interpreted on the basis of a photo-thermostimulated (PTSEE) process involving the main OSL trap at 320oC; this process takes place with a thermal assistance energy estimated at W=0.29+0.02 eV. Good quantitative agreement is obtained between theory and experiment by assuming a thermal broadening of the thermal depletion factor for the OSL traps, described by a Gaussian distribution of energies. ‘

**36. Simulations of the predose technique for retrospective dosimetry and authenticity testing**

V. Pagonis, E. Balsamo, C. Barnold, K. Duling, S. McCole

Radiation Measurements 43 (2008) 1343–1353

Abstract

The predose technique of thermoluminescence (TL) for quartz has been used extensively for retrospective dosimetry and archaeological authenticity testing. In this paper, we use a previously published comprehensive kinetic model for quartz, to simulate the complete sequence of experimental steps taken during the additive dose and the multiple activation versions of the predose technique. The simulation results show how both versions of the predose technique can reproduce the paleodose received by the sample with an accuracy of 1–5% in the low dose region of 0–2Gy. For doses greater than ~ 2Gy the non-linear dose dependence of the sensitivity of the “110 0C” TL peak causes significant inaccuracies in the technique. The solution of the kinetic differential equations elucidates several electron and hole processes taking place during the experimental predose procedure; these processes include the thermal transfer of holes from the Zimmerman hole reservoirs to the luminescence center, the radiation quenching of the TL sensitivity and the radiation-enhanced sensitivity of quartz samples. Specific numerical examples are given for samples exhibiting the thermal activation characteristics of “low-S0” and “high-S0” values. Quantitative results are presented for the effect of the test dose and of the calibration beta dose on the accuracy of both versions of the predose technique. Results are also presented for the sensitivity of the predose technique to the natural variations of the hole concentrations in the luminescence center. Finally, the results of the predose technique simulations are compared with those from simulating the popular single aliquot SAR/OSL technique based on optically stimulated luminescence signals.

**35. Thermoluminescence kinetic study of binary lead-silicate glasses**

V. Pagonis, S. Mian, R. Mellinger, K. Chapman

Journal of Luminescence 129 (2009) 570–577

Abstract

This paper describes a detailed experimental study of the thermoluminescence (TL) properties of four binary lead-silicate glasses, with PbO concentrations ranging from 32% to 62% in mole percent. The TL glow peaks between room temperature and 300 oC were analyzed using a systematic thermal cleaning technique. The Tmax–Tstop and E–Tstop methods of analysis were used to identify the number of peaks under the glow curves, and to obtain the activation energy E for each TL trap. A computerized glow curve fitting analysis is used to fit the experimental data to four first-order peaks with maxima at temperatures of 54, 80, 110 and 210 oC, as measured with a heating rate of 2 oC/s. The kinetic parameters of the glow-peak at 210oC were confirmed by using phosphorescence decay methods of analysis. The TL traps associated with the low-temperature TL peak at 54oC are found to depend strongly on the PbO concentration of the samples, while the higher-temperature TL peaks show a behavior independent of the PbO concentration. The activation energy E and frequency factor s of the low-temperature TL trap associated with the peak at 54oC are consistent with a trap involving a delocalized transition through the conduction band. However, the activation energies and frequency factors for the higher-temperature TL traps are consistent with traps involving localized transitions via an excited state below the conduction band. The data suggest that these higher-temperature TL traps are associated with the common silicate matrix in these binary silicate glasses.

**34. Experimental and modelling study of pulsed optically stimulated luminescence in quartz, marble and beta irradiated salt**

V Pagonis, S M Mian, M L Chithambo, E Christensen and C Barnold

J. Phys. D: Appl. Phys. 42 (2009) 055407 (12pp)

Abstract

Optical stimulation luminescence (OSL) signals can be obtained using continuous-wave optical stimulation (CW-OSL), the linear modulation optical stimulation method (LM-OSL) and the time-resolved optical stimulation (TR-OSL) method. During TR-OSL measurements, the stimulation and emission of luminescence are experimentally separated in time by using short light pulses. This paper presents new TR-OSL data for annealed high purity synthetic quartz, for marble and for commercially available iodized salt. A new type of behaviour for TR-OSL signals for quartz and iodized salt is presented, in which the OSL signal exhibits a nonmonotonic behaviour during optical stimulation; this type of behaviour has not been reported previously in the literature for quartz. Furthermore, a luminescence component with very long luminescence lifetime is reported for some quartz aliquots, which may be due to the presence of a delayed-OSL (DOSL) mechanism in quartz. A new kinetic model for TR-OSL in quartz is presented, which is based on a main electron trap and on several luminescence

centres. The model is used to quantitatively fit several sets of experimental data of pulsed optically stimulated luminescence from quartz.

**33. A simulation of OSL pulse annealing at different heating rates: Conclusions concerning the evaluated trapping parameters and lifetimes**

V. Pagonis, R. Chen

GEOCHRONOMETRIA, 30 (2008), pp xx-xx: DOI 10.2478/v10003-008-0009-6

Abstract

Pulse annealing has been the subject of several studies in recent years. In its basic form, it consists of relatively short-time optically stimulated luminescence (OSL) measurements of a given sample after annealing at successively higher temperatures in, say, 10°C increments. The result is a decreasing function with a maximum OSL at low temperatures and gradually decreasing to zero at high temperature. Another presentation is that of the percentage OSL signal lost per annealing phase, associated with minus the derivative of the former curve, which yields a thermoluminescence (TL)- like peak. When the heating is performed at different heating rates, the TL various heating rates (VHR) method can be utilized to evaluate the trapping parameters. Further research yielded more complex pulse-annealing results in quartz, explained to be associated with the hole reservoir. In the present work, we simulate numerically the effect, following the experimental steps, in the simpler form when no reservoir is involved, and in the more complex case where the reservoir plays an important role. The shapes of the reduction-rate curves resemble the experimental ones. The activation energies found by the VHR method are very close to the inserted ones when the retrapping probability is small, and deviate from them when retrapping is strong. The theoretical reasons for this deviation are discussed.

**32. A theoretical model for a new dating protocol for quartz based on thermally transferred OSL (TT-OSL)**

V. Pagonis, A.G.Wintle, R. Chen, X.L.Wang

Radiat. Meas. (2008), doi: 10.1016/j.radmeas.2008.01.025

Abstract

Recently a new dating procedure has been suggested, which is based on the thermally transferred optically stimulated luminescence (TT-OSL) signal that is measured after irradiated quartz is optically bleached and then preheated. Experimentally the TT-OSL signal was measured after a high temperature preheat (260 0C for 10 s) following an optical bleach at 125 0C for 270 s to deplete the fast and medium OSL components. The TT-OSL signal was measured for 90 s at 125 0C in order to avoid the effect of re-trapping of electrons in the 110 0C trap of quartz. The luminescence sensitivity changes were monitored by the OSL response to a test dose. In this paper, we use a modi0ed version of a comprehensive model to simulate the complete experimental sequence of the new protocol, and to 0t the experimental dose–response graphs of the OSL, TTOSL and basic-TT-OSL signals for doses up to 4000 Gy. Two possible mechanisms for the production of the TT-OSL signals in this quartz are discussed, namely the double transfer mechanism suggested for the recuperation effect, and a single transfer mechanism in which the TT-OSL signal is due to the thermal transfer of charge from a “source trap” into the fast OSL trap of quartz. The results of the simulation indicate that the latter mechanism is more likely to be responsible for the observed TT-OSL dose growth in 0ne-grained quartz extracted from Chinese loess.

**31. Computerized curve deconvolution analysis for LM-OSL**

G. Kitis, V. Pagonis

Radiat. Meas. (2008), doi: 10.1016/j.radmeas.2007.12.055

The computerized curve deconvolution analysis (CCDA) technique is well known in the case of thermoluminescence (TL). In the present work we investigate the application of CCDA to the linear modulated optically stimulated luminescence curves (LM-OSL). We derive single LM-OSL peak equations which are based on variables which can be extracted directly from the experimental OSL curve, for both 0rst order and general order LM-OSL peaks. The similarities and differences between TL and OSL CCDA analysis are discussed. The resolution of the technique is also examined in the cases of synthetic curves consisting of two or four constituent components. Finally a new experimental procedure is suggested which can be used to separate composite LM-OSL curves into their constituent components.

**30. Duplicitous thermoluminescence peak associated with a thermal release of electrons and holes from trapping states**

R. Chen, V. Pagonis, J.L. Lawless

Radiat. Meas. (2007), doi: 10.1016/j.radmeas.2007.09.021

A thermoluminescence (TL) glow peak may result from a transition of electrons from traps into the conduction band, followed by a recombination with holes trapped in a luminescence center. Another possibility is that holes trapped in a hole trap are thermally released into the valence band and recombine with electrons in an electron recombination center. A series of glow peaks emitted from a given sample may include peaks of both kinds. In some cases, peaks may be identi0ed as being of one kind or the other, say, by using thermally stimulated electron emission (TSEE), which can take place when the free carriers are electrons. In the present work, we demonstrate by the use of simulation that two peaks may result from one electron and one hole trapping states and a single hole recombination center. The first TL peak is observed when thermally stimulated electrons recombine with holes in the center. The TL peak is terminated when the holes in the center are exhausted. At higher temperatures, holes from a hole trap are released into the valence band and then captured by the hole center, thus this center is replenished. More electrons from the electron trap are thermally released now and recombine with the newly arrived holes in centers. A second TL peak may be observed which carries some information concerning the hole trap. It is thus demonstrated that some of the usual methods for distinguishing between electron and hole traps can lead to incorrect conclusions. It is possible for a hole trap, for example, to induce an increase in electron recombination in such a way that it produces a peak that looks nearly identical to TL from an electron trap. This simulation may bring about a new look at TL peaks occurring in materials used in TL dosimetry and dating. A new interpretation may also be given to “Auger” TSEE associated with the thermal release of electrons from the surface of a material, which indirectly results from the thermal release of holes from traps. The performance of some methods for evaluating the activation energies and the signi0cance of the results in the present situation are discussed.

**29. Dependence of the anomalous fading of the TL and blue-OSL of 0uorapatite on the occupancy of the tunnelling recombination sites**

N.C. Tsirliganis, G.S. Polymeris, G. Kitis, V. Pagonis

Journal of Luminescence 126 (2007) 303–308

ABSTRACT

The anomalous fading (AF) of thermoluminescence (TL) and optically stimulated luminescence (OSL) signals in Durango apatite is attributed to tunnelling effects. Electrons from the TL and OSL traps in this material are transferred, via a tunnelling effect, to the recombination sites. The availability of recombination sites for tunnelled electrons is of major importance for the degree of AF rate observed in this material. It is expected that a variation of the number of the electron recombination sites will be re0ected in the experimentally measured AF rate. In the present work an investigation of the recombination sites for the tunnelled electrons is attempted by studying the AF effect using a special technique, in which the anomalously faded TL (OSL) is replaced by an equal amount of TL (OSL) induced by a beta dose.

**28. A quantitative kinetic model for Al2O3:C: TL response to UV-illumination**

Pagonis, V., Chen, R. and Lawless, J.L.

Radiation Measurements (2007), doi:10.1016/j.radmeas.2007.10.046

ABSTRACT

This paper presents a quantitative kinetic model applicable to the important dosimetric material Al2O3:C. The model describes successfully the thermoluminescence (TL) response of the material to UV illumination (206 nm). The energy levels in this model consist of the main dosimetric trap, two competing deep hole and deep electron traps, and the luminescence center. The model also describes successfully the experimental variation of the optical absorption coefficient K with UV fluence. The values of the kinetic parameters are not arbitrary, but are obtained either from published experimental data, or by using reasonable physical assumptions. A correction factor is applied to the calculated UV-fluence to account for the fact that the samples used in the experiments were several optical lengths thick. By using this correction factor, the experimental data can be shown on the same graph as the calculated curves of TL vs. UV-fluence and K vs. UV-fluence, with the UV fluences given in photons/cm2 and not in some arbitrary units.

**27. Simulations of the effect of pulse annealing on optically-stimulated luminescence of quartz**

V. Pagonis, A.G.Wintle, R. Chen

Radiation Measurements 42 (2007) 1587 – 1599

Pulse annealing techniques are commonly used in OSL studies of quartz to obtain information on the kinetic parameters of OSL traps and hole reservoirs. In this paper, simulations of pulse annealing experiments are carried out using the comprehensive model for quartz developed by Bailey [2001. Towards a general kinetic model for optically and thermally stimulated luminescence of quartz. Radiat. Meas. 33, 17–45] for both natural and laboratory irradiated aliquots. The results of the simulations are in qualitative agreement with, and reproduce, several unusual features of the experimental data of Wintle and Murray [1998. Towards the development of a preheat procedure for OSL dating of quartz. Radiat. Meas. 29, 81–94]. The simulations are also carried out using different heating rates, and show that pulse annealing experiments can be used to recover appropriate kinetic parameters for both the OSL traps and the hole reservoirs known to exist in quartz. The results of the

simulations show the importance of these hole reservoirs in determining how the OSL signal depends upon the preheat temperature.

**26. Thermoluminescence glow-peak shape methods based on mixed order kinetics**

George Kitis, Reuven Chen, and Vasilis Pagonis

phys. stat. sol. (a), 1– 9 (2008) / DOI 10.1002/pssa.200723470

The peak shape methods used in thermoluminescence (TL) theory to evaluate the activation energy E, are based on first and second order kinetics equations. For the intermediate kinetic orders the peak shape methods are based on the empirical theory of general order kinetics. In the present work we derive peak shape methods based on the physically meaningful mixed order kinetics model. The derived equations are tested for their accuracy and are compared with other peak shape methods existing in the TL literature.

**25. A unified presentation of thermoluminescence (TL), phosphorescence and linear-modulated optically stimulated luminescence (LM-OSL)**

Reuven Chen and Vasilis Pagonis

J. Phys. D: Appl. Phys. **41** (2008) 035102

It has been shown in the past that a thermoluminescence (TL) peak has a relatively simple shape if the heating function is hyperbolic. Also, two different ways have been given for the transformation of the featureless decaying phosphorescence and optically stimulated luminescence (OSL) functions into peak-shaped curves. In the former case, one plots *L* · *t *versus ln*(t)*, where L is the phosphorescence intensity and t the time. In the latter, linear-modulated OSL (LM-OSL) is reached by changing the stimulated light intensity linearly with time, and pseudo-LM-OSL by manipulating the regular CW-OSL curve. In this work, a modified presentation of LM-OSL is suggested and a unified presentation of these phenomena is offered, which may help in analysing the results and elucidating the details of the relevant kinetics process. The implications on the use of LM-OSL for dosimetry are also discussed.

**24. Peak shape methods for general order TL glow-peaks: A reappraisal**

**
**G. Kitis and V Pagonis

Nuclear Instruments and Methods in Physics Research B 262 (2007) 313–322s

This paper presents a reappraisal of the well known peak shape expressions for calculating the activation energy E in a thermoluminescence (TL) glow-peak. This study leads to new insights as to the meaning of the coefficients used in the original equations. The reappraisal leads to new equations for the coefficients of the peak shape expressions which contain the general order parameter b, instead of the experimentally determined geometrical shape factor which is used in the original equations. Previously only the coefficients for first and second order kinetics were determined on the basis of existing theory and the coefficients for intermediate kinetics order were determined empirically using a linear interpolation–extrapolation method. In the present work the improved peak shape coefficients are evaluated in analytical form as a function of the kinetic order b, by using the general order kinetics expression for the TL intensity. The intrinsic errors in the newly derived expressions for E are evaluated and their relevance to experimental work is discussed in detail. A method for a further improvement of the accuracy of the peak shape methods is suggested.

**23. Modelling thermal transfer in optically stimulated luminescence of quartz**

V Pagonis, R Chen and AG Wintle

J. Phys. D: Appl. Phys. 40 (2007) 998–1006

A previously published kinetic model for the production of luminescence signals in quartz is used to investigate the production of thermally transferred optically stimulated luminescence (TT-OSL) signals. This paper provides a mathematical description of the thermal transfer mechanism for two different phenomena that have been observed in previously published experiments (Aitken and Smith 1988 Quat. Sci. Rev. 7 387–93). The starting point is the model proposed by Bailey (2001 Radiat. Meas. 33,17–45). The numerical values of some of the parameters are varied so that they match the experimental data. The effect caused by varying these values is investigated.

The first of these phenomena takes place after storing optically bleached samples at room temperature; this involves the traps responsible for the 110 0C thermoluminescence (TL) peak of quartz acting as a refuge trap. The second phenomenon concerns OSL signals that are induced by heating the samples after the bleaching of the OSL signal and involves a putative TL peak at ~230 0C associated with the refuge trap; specifically, the paper presents a simulation of the temperature dependence of the OSL signal measured by successively heating the quartz samples to higher temperatures up to ~400 0C.

**22. Thermoluminescence response and apparent anomalous fading factor of Durango fluorapatite as a function of the heating rate**

G. Kitis, G. Polymeris, V. Pagonis, and N. Tsirliganis

phys. stat. sol. (a) 203,No. 15, 3816-3823 (2006)

The anomalous fading (AF) of the Thermoluminescence (TL) signal has been observed in Durango apatite, and is believed to be caused by electrons transferred to the recombination sites via a tunnelling effect. The AF factor is a property unique for each glow-peak of a glow-curve and shows a remarkable stability, even when extreme experimental conditions are applied to this material. In the present work the AF is studied as a function of the rate at which the material is heated in order to obtain its TL. In this experimental study it is found that the TL response of Durango apatite is drastically increased as the rate of heating increases, contrary to what is expected from standard kinetic models of TL. An important consequence of this result is that, since the AF rate is monitored through the TL signal, the variations of TL as a function of the heating rate will be passed on to the evaluation of the AF rate. The result is an apparent decrease of the AF with an increasing heating rate, although the actual AF effect has no relation with the TL readout process. A possible explanation is proposed for the increase of the TL response and for the apparent decrease of the AF rate with the rate of heating.

**21. The effects of annealing and irradiation on the sensitivity and superlinearity properties of the 110 0C thermoluminescence peak of quartz**

G. Polymeris, G. Kitis, V. Pagonis

Radiation Measurements 41 (2006) 554 – 564

Quartz samples which undergo heating and irradiation treatments exhibit changes in their sensitivity to irradiation, as well as in their TL dose response. These changes of thermoluminescence (TL) sensitivity and superlinearity of the 110 0C TL peak of quartz have been the subject of several experimental and theoretical studies, because they form the basis of the predose technique for dating ceramics and for accident dosimetry. In an effort to separate experimentally the effects of irradiation and annealing on the predose effect, three quartz samples of different origin were prepared under three different conditions: unannealed samples, samples annealed at 500 0C, and samples annealed at 900 0C.

Complete TL versus dose and sensitivity S versus predose curves were obtained for the dose range of 0.1<D<400 Gy. Additional complete sets of data were obtained for samples that underwent a combined predose irradiation and a heat treatment to 500 0C. Although the TL versus dose curves and the sensitivity versus predose data showed very different behaviors, preannealing the samples at 900 0C removed the observed differences that are due to the thermal or irradiation history of the quartz samples. The experimental data is consistent with the assumption that high-temperature anneals and/or high dose irradiation of the samples reduces the concentration of available competitor sites. The concentration of these competitor sites, as described by the Zimmerman model of quartz, is identified as the most important factor in causing the observed differences in predose behavior between quartz samples of different origin. Strong evidence in support of this competitor theory is provided by the estimated equivalent doses (EDs) for the three quartz samples.

**20. Thermoluminescence under exponential heating function I: Theory**

G. Kitis, R. Chen, V. Pagonis, E. Carinou and V. Kamenopoulou

J. Phys. D: Appl. Phys. 39, 1500-1507 (2006).

Constant temperature hot gas readers are widely employed in thermoluminescence dosimetry. In such readers the sample is heated according to an exponential heating function. The single glow-peak shape derived under this heating condition is not described by the TL kinetics equation corresponding to a linear heating rate. In the present work TL kinetics expressions, for first and general order kinetics, describing single glow-peak shapes under an exponential heating function are derived. All expressions were modified from their original form of I (n0,E, s, b, T ) into I (Im,E, Tm, b, T ) in order to become more efficient for glow-curve deconvolution analysis. The efficiency of all algorithms was extensively tested using synthetic glow-peaks.

**——————————————————————————–**

**19. Thermoluminescence under exponential heating function II: Deconvolution of experimental glow-curves**

G. Kitis, R. Chen, V. Pagonis, E. Carinou, P. Ascounis and V. Kamenopoulou

J. Phys. D: Appl. Phys. 39, 1508-1514 (2006).

Thermoluminescence (TL) glow-curves measured using an exponential heating function (EHF) in constant temperature hot gas readers, cannot be analysed using the existing single TL glow-peak equations derived assuming a linear heating rate. In the present work single TL glow-peak equations, which were recently derived assuming an EHF, are used to perform a computerized glow-curve deconvolution analysis of experimental glow-curves measured using a stable temperature hot gas reader. Glow-curves of the most commonly used dosimetric material LiF:Mg,Ti were analysed using the first order kinetics glow-peak equations. The glow-curves were analysed for samples that were pre-irradiation annealed at 400 0C for 1 h and 100 0C for 2 h, with and without a post-irradiation annealing at 80 0C for 1 h. TL glow-peak equations of the general order kinetics were used to analyse experimental glow-curves of the dosimetric material Li_{2}B_{4}O_{7} : Mn,Si. The results showed that the recently derived TL equations are very efficient for analysing glow-curves measured using stable temperature hot gas readers.

——————————————————————————–

**18. A comprehensive study of the predose effect for three quartz crystals of different origin.**

G. Kitis, V. Pagonis, R. Chen and G. Polymeris

Radiation Protection Dosimetry (2006), Vol. 119, No. 1–4, pp. 438–441

The study of the thermoluminiscence (TL) sensitivity of quartz due to heat and irradiation treatments is of importance in dating and retrospective dosimetry. A comprehensive comparative study of the predose effect was carried out for three types of quartz of different origin. Complete TL vs. dose and sensitivity S vs. predose curves were obtained for the dose range of 0.1<D<400 Gy. Additional complete sensitivity vs. predose curves were obtained for samples which underwent a combined predose irradiation and a subsequent heat treatment to 500C. Although the TL vs. dose curves showed very different behaviours, the sensitivity vs. predose curves showed several common characteristics. The sensitivity vs. predose curves showed abrupt changes at ~10 Gy. The sensitivity after a combined predose irradiation and heat treatment to 500C showed a very gradual change in the whole dose range studied. These results are explained qualitatively by using the modified Zimmerman model for quartz.

——————————————————————————-

**17. Theoretical modeling of experimental diagnostic procedures employed during predose dosimetry of quartz**

V. Pagonis, R. Chen and G. Kitis

Radiation Protection Dosimetry (2006), Vol. 119, No. 1–4, pp. 111–114

The pre-dose technique in thermoluminescence (TL) is used for dating archaeological ceramics and for accident dosimetry. During routine applications of this technique, the sensitisation of the quartz samples is measured as a function of the annealing temperature, yielding the so-called thermal activation characteristic (TAC). The measurement of multiple TACs and the study of the effect of UV-radiation on the TL sensitivity of quartz are important analytical and diagnostic tools. In this paper, it is shown that a modified Zimmerman model for quartz can successfully model the experimental steps undertaken during a measurement of multiple TACs.

——————————————————————————–

**16. Non-monotonic dose dependence of OSL intensity due to competition during irradiation and read-out**

V. Pagonis, R. Chen and J.L. Lawless

Radiation Measurements 41 (2006) 903– 909

The nonmonotonic dose dependence of thermoluminescence has been observed in several materials; a recent publication (Lawless, J.L., Chen, R., Lo, D., Pagonis, V., 2005. A model for non-monotonic dose dependence of thermoluminescence (TL). J. Phys. Condens. Matter 17,737–753.) gave a theoretical account based on competition between trapping states or recombination centers during the excitation and/or readout stages. A similar effect has been observed in the optically stimulated luminescence (OSL) of some materials such as quartz and Al2O3:C(e.g., Yukihara, E.G., Whitley, V.H., McKeever, S.W.S., Akselrod, A.E., Akselrod, M.S., 2004a. Effect of high-dose irradiation on the optically stimulated luminescence of Al2O3:C. Radiat. Meas. 38, 317–330; Yukihara, E.G., Gaza, R., McKeever, S.W.S., Soares, C.G., 2004b. Optically stimulated luminescence and thermoluminescence efficiencies for high-energy heavy charged particle irradiation in Al2O3:C. Radiat. Meas. 38, 59–70.). The model of competition has now been developed to explain the nonmonotonic dose dependence of OSL. A distinction is made between two cases. In one, the competition during excitation causes the filling of the relevant radiative center to be nonmonotonic with the dose, and as a result, the OSL intensity behaves in a similar way. This can take place with a “minimal” model including one trapping state and two kinds of recombination centers, one radiative and the other nonradiative. In the other case, all the relevant concentrations increase monotonically with the dose, and the nonmonotonic dose dependence of the OSL signal is mainly due to competition in the readout stage. In this case, it appears that the requirement for nonmonotonic dose dependence is a system with two trapping states and two kinds of recombination centers.

——————————————————————————–

**15. Comparison of experimental and modelled quartz thrmal-activation curves obatined using multiple-and single-aliquot procedures**

G. Kitis, V. Pagonis and R. Chen

Radiation Measurements 41 (2006) 910 – 916

This paper presents the results of a broad study of the thermal-activation characteristics of three quartz samples of different origin. The thermal-activation characteristics of synthetic, sedimentary and Arkansas quartz are measured as a function of several experimental parameters and over the temperature range 20.500 0C using both multiple-aliquot and single-aliquot techniques. The studies are carried out also as a function of the thermal preparation of the samples, for “as-is” samples, for samples fired at temperatures of 500 and 900 0C. Additional experimental parameters varied in this study include the test dose used in measuring the thermal activation curves and the temperature interval between successive heating of the samples. The results of this experimental study for all three quartz samples are interpreted by using a simple modified Zimmerman model for quartz consisting of 2 electron traps, a luminescence center and three hole reservoirs. The results are also discussed within the recently published complex theoretical quartz model by Adamiec (2005. Investigation of a numerical model of the pre-dose mechanism in quartz. Radiat. Meas. 39, 175–189), which is based on a luminescent center, three electron traps and three reservoir hole traps.

——————————————————————————–

**14. A quantitative kinetic model for Al2O3:C: TL response to ionization radiation**

V. Pagonis, R. Chen and J.L. Lawless

Radiation Measurements (In press)

This paper presents a quantitative kinetic model for the important dosimetric material Al2O3:C. The model consists of two traps and two centers, and reproduces the experimental thermoluminescence (TL) vs. dose behavior, as well as the experimental variation of the optical absorption coefficient K with beta dose. Initial estimates of the kinetic parameters in the model are obtained either from published experimental data, or by using reasonable physical assumptions. Good agreement between the experimental data and calculations from the model are obtained for three different types of samples of alumina. This is achieved by keeping the trapping and recombination probabilities constant for all three samples, while the concentrations of the carriers are varied. The kinetic model provides also a quantitative description of the experimentally observed nonmonotonic behavior of the TL dose–response curves for all three samples.

——————————————————————————–

**13. ABSORBED DOSE MEASUREMENTS OF A HANDHELD 50 kVP X-RAY SOURCE IN WATER WITH THERMOLUMINESCENCE DOSEMETERS**

Christopher Soares, Chris Drupieski, Brian Wingert, Garey Pritchett, Vasilis Pagonis, Michelle O’Brien, Alan Sliski, Pawel Bilski and Pawel Olko

RADIATION PROTECTION DOSIMETRY (2006), Vol. 120, No. 1–4, pp. 78–82

Absorbed dose rate measurements of a 50 kV p handheld X-ray probe source in a water phantom are described. The X-ray generator is capable of currents of up to 40 lA, and is designed for cranial brachytherapy and intraoperative applications with applicators. The measurements were performed in a computer-controlled water phantom in which both the source and the detectors are mounted. Two different LiF thermoluminescence dosemeter (TLD) phosphors were employed for the measurements, MTS-N LiF:Mg,Ti) and MCP-N (LiF:Mg,Cu,P). Two small ionisation chambers (0.02 and 0.0053 cm^{3}) were also employed. The TLDs and chambers were positioned in watertight mounts made of water-equivalent plastic. The chambers were calibrated in terms of air-kerma rate, and conventional protocols were used to convert the measurements to absorbed dose rate. The TLDs were calibrated at National Institute of Standards and Technology (NIST) in terms of absorbed dose rate using a ^{60}Co teletherapy beam and narrow-spectrum X-ray beams. For the latter, absorbed dose was inferred from air-kerma rate using calculated air-kerma-to-dose conversion factors. The reference points of the various detectors were taken as the center of the TLD volumes and the entrance windows of the ionisation chambers. Measurements were made at distances of 3–45 mm from the detector reference point to the source center. In addition, energy dependence of response measurements of the TLDs used was made using NIST reference narrow spectrum X-ray beams. Measurement results showed reasonable agreement in absorbed dose rate determined from the energy dependence corrected TLD readings and from the ionisation chambers. Volume averaging effects of the TLDs at very close distances to the source were also evident.

——————————————————————————–

**12. The nonmonotonic dose dependence of optically stimulated luminescence in Al2O3 :C: Analytical and numerical simulation results**

R. Chen, V. Pagonis and J. L. Lawless

JOURNAL OF APPLIED PHYSICS 99, 033511,2006

Nonmonotonic dose dependence of optically stimulated luminescence (OSL) has been reported in a number of materials including Al_{2}O_{3} :C which is one of the main dosimetric materials. In a recent work, the nonmonotonic effect has been shown to result, under certain circumstances, from the competition either during excitation or during readout between trapping states or recombination centers. In the present work, we report on a study of the effect in a more concrete framework of two trapping states and two kinds of recombination centers involved in the luminescence processes in Al_{2}O_{3} :C. Using sets of trapping parameters, based on available experimental data, previously utilized to explain the nonmonotonic dose dependence of thermoluminescence including nonzero initial occupancies of recombination centers _F+ centers_, the OSL along with the occupancies of the relevant traps and centers are simulated numerically. The connection between these different resulting quantities is discussed, giving a better insight as to the ranges of the increase and decrease of the integral OSL as a function of dose, as well as the constant equilibrium value occurring at high doses.

LINK TO PDF FILE

——————————————————————————–

**11. Evaluation of activation energies in the semi-localized transition model of thermoluminescence**

V Pagonis

J. Phys. D: Appl. Phys. 38 (2005) 2179-2183

Recently a semi-localized transition (SLT) kinetic model was developed for thermoluminescence (TL), which is believed to be applicable to important dosimetric materials like LiF : Mg,Ti. This model contains characteristics of both a localized transition model and a single trap model and is characterized by two distinct activation energy levels. This paper describes the simulation of several standard methods of analysis for the TL peaks calculated using the SLT model in an effort to extract the two activation energy parameters of the model. The methods of analysis are applied to both possible types of transitions within the model, namely the direct recombination of the hole?electron pairs as well as the delocalized transitions involving the conduction band. In the former case of direct recombination, the methods of analysis give consistent results for the activation energy E. In the latter case of transitions involving the conduction band, it was found that extra caution must be exercised when applying standard methods of analysis to the SLT model because of the possibility of strongly overlapping TL peaks. Specifically the peak shape methods consistently fail to yield the correct value of E, while careful application of the fractional glow, thermal cleaning and variable heating rate methods can yield the correct energy values when no retrapping is present within the localized energy levels. A possible explanation is given for the previously reported failure of the peak shape methods to yield the correct activation energies within the SLT model. The heating rate methods of analysis consistently yield the correct activation energies E with an accuracy of a few per cent.

——————————————————————————–

**10. A model for non-monotonic dose dependence of thermoluminescence**

J L Lawless, R Chen, D Lo and V Pagonis

J. Phys.: Condens. Matter 17 (2005) 737-753

In the applications of thermoluminescence (TL) in dosimetry and archaeological and geological dating, a desirable dose dependence of TL intensity is a monotonically increasing function, preferably linear. It is well known that in many dosimetric materials, nonlinear dependence is observed. This may include a superlinear dependence at low doses and/or sublinear dose dependence at higher doses, where the TL intensity approaches saturation. In quite a number of materials, non-monotonic dose dependence has been observed, namely, the TL intensity reached a maximum value at a certain dose and decreased at higher doses. This effect is sometimes ascribed to “radiation damage” in the literature. In the present work we show, both quasi-analytically and by using numerical simulation, that such dose dependence may result from a simple energy level scheme of at least one kind of trapping state and two kinds of recombination centers. One does not necessarily have to assume a destruction of trapping states or recombination centers at high doses. Instead, the main concept involved is that of competition which takes place both at the excitation stage and the readout stage during the heating of the sample. This may explain the fact that the phenomenon in question, although very often ignored, is rather common. Cases are identified in which competition during excitation dominates, and others in which competition during read-out dominate

——————————————————————————–

**9. Modelling thermal activation characteristics of the sensitization of thermoluminescence in quartz**

Reuven Chen and Vasilis Pagonis

J. Phys. D: Appl. Phys. 37 (2004) 159-164

The sensitization of the ?110?C thermoluminescence peak in quartz, also termed the ?pre-dose? effect, was previously explained using an energy level model including two electron trapping states and two hole centers. The experimental procedure includes a stage of high temperature activation following a relatively large irradiation of the sample. The response to a small test-dose was found to depend on this activation temperature. With different quartz samples, different behaviors of the thermal activation characteristics (TACs) were found. In typical TACs, the sensitivity reached a maximum at 500^{o}C, followed by a rather sharp decline in some samples; in others a maximum was reached at 350^{o}C followed by a slight decline towards a plateau level. In this work, we show that these behaviors can rather easily be explained within the framework of the two traps?two centers model. This is done by numerical solution of the relevant sets of differential equations governing the different stages of the experimental procedure. The different kinds of dependence were simulated with different sets of trapping parameters. A better insight into the processes taking place is reached, which may have implications in the application of pre-dose dating of archaeological quartz samples and in retrospective dosimetry.

——————————————————————————–

**8. SIMULATION OF THE EXPERIMENTAL PRE-DOSE TECHNIQUE FOR RETROSPECTIVE DOSIMETRY IN QUARTZ**

Vasilis Pagonis and Hezekiah Carty

Radiation Protection Dosimetry (2004), Vol. 109, No. 3, pp. 225-234

The pre-dose technique of thermoluminescence for quartz has been used extensively for retrospective dosimetry of quartz and other natural materials. A recently published model that is a modification of the well-known Zimmerman theory is used here to simulate the complete sequence of experimental steps taken during the additive dose version of the pre-dose technique. The results of simulation show how the method can reproduce accurately the accumulated dose or paleodose received by the sample. The solution of the kinetic differential equations elucidates the various electron and hole processes taking place during the experimental pre-dose procedure and shows clearly the mechanism of hole transfer from the reservoir to the luminescence centre caused by heating to the activation temperature. The numerical results show that the pre-dose technique can reproduce the paleodose with an accuracy of ~1-5%, even when the paleodose is varied over more than an order of magnitude. New quantitative results are presented for the effect of the test dose and of the calibration beta dose, b, on the accuracy of the pre-dose technique. The conclusions drawn from the simple model for quartz can be used to make improvements to more general quartz models.

——————————————————————————–

**7. Applicability of the Zimmerman predose model in the thermoluminescence of predosed and annealed synthetic quartz samples**

Vasilis Pagonis, George Kitis, Reuven Chen

Radiation Measurements, 37, (2003), 267-274.

The “110^{o}C” thermoluminescence (TL) peak of unfired synthetic quartz is known to exhibit a highly superlinear growth with absorbed dose. In this paper, it is shown that the well-known Zimmerman predose model can explain recent experimental results on the superlinearity of annealed synthetic quartz, as well as experimental results on the superlinearity of heavily predosed samples at room temperature. In the case of the predosed samples, the simulation solves the kinetic rate equations for the various stages in the experimental TL predose process. The results of the simulation explain the behavior of the TL versus dose curves at different predoses, as well as the detailed behavior of the superlinearity coefficient k as a function of the predose amount. In the case of the annealed samples, the simulation solves the kinetic equations for different values of the initial concentration of holes in the recombination center. The results of the simulation explain the behavior of the TL versus dose curves at different annealing temperatures, as well as the detailed behavior of the superlinearity coefficient k in each of the two distinct superlinearity regions. The simulation also produces the correct order of magnitude for the large sensitivity changes of the TL intensity observed in both sets of experiments.

——————————————————————————–

**6. Cooling rate effects on the thermoluminescence glow curves of Arkansas quartz**

G. Kitis, V. Pagonis, and C. Drupieski

Phys. stat. sol. (a) 198, No. 2, 312? 321 (2003)

Samples of quartz annealed at high temperatures are known to exhibit thermoluminescence (TL) properties which depend on the rate of cooling of the samples to room temperature. Powder samples of Arkansas quartz were annealed in air at temperatures between 500 and 900^{o}C and were cooled to room temperature at different cooling rates. The TL of both slowly and fast cooled samples was measured at various doses of beta radiation; a fast cooling rate leads to significant enhancements of the TL intensity for the “110 ^{o}C” TL peak, as well as a change in the ratio of the relative intensities of the main TL peaks. The well-known Tm – Tstop method of analysis resulted in several well-defined plateaus at different temperatures for the fast cool and slow cool samples, while the thermal quenching parameters C and W and kinetic parameters of the TL peaks were found to be independent of the cooling rate. This new result provides further evidence for the cooling rate effects being due to slow ionic processes, rather than the much faster electronic processes involved in thermal quenching.

——————————————————————————–

**5. SEARCH FOR COMMON CHARACTERISTICS IN THE GLOW CURVES OF QUARTZ OF VARIOUS ORIGINS**

V. Pagonis, E. Tatsis, G. Kitis and C. Drupieski

Radiation Protection Dosimetry (2002), Vol. 100, Nos. 1-4, pp. 373-376

The thermoluminescence glow curves of quartz of various origins were measurd under two different conditions, (1) unannealed samples and (2) samples annealed at 500^{o}C and 900^{o}C. The different glow curves were analyzed using first order kinetics and glow curve deconvolution analysis (GCD). The comparison of the glow curves obtained was mainly concentrated in studying the sensitivities of the glow peaks as a function of the annealing temperature, and in obtaining the kinetic parameters of the glow peak at 110 oC. Furthermore, in four samples the detailed comparison was extended to the trapping parameters of all existing glow peaks. It was found that despite their different origin and the different shapes of the glow curves, there are several basic characteristics which are common to all samples studied.

——————————————————————————–

**4. DETAILED KINETIC STUDY OF THE THERMOLUMINESCENCE GLOW CURVE OF SYNTHETIC QUARTZ
**

G. Kitis, V. Pagonis, H. Carty and E. Tatsis

Radiation Protection Dosimetry (2002), Vol. 100, Nos. 1-4, pp. 225-228

A detailed kinetic analysis has been performed of the thermoluminescence glow curve of high purity synthetic quartz. The kinetic parameters of the glow peak at 110^{o}C were evaluated for doses ranging from 0.1 Gy to 100 Gy using glow curve deconvolution (GCD) methods, initial rise, variable heating rate and phosphorescence decay methods. All the methods gave results that agree within the experimental errors.

——————————————————————————–

**3. An improved experimental procedure of separating a composite thermoluminescence glow curve into its components**

V. Pagonis and C. Shannon

Radiation Measurements, 32, 805-812 (2000).

We present an improved experimental procedure of separating a composite thermoluminescence glow curve into its components. Careful monitoring of the isothermal cleaning process using the initial rise method ensures the complete thermal removal of TL peaks. Digital subtraction of two experimental TL glow curves yields individual experimental TL glow peaks. Several standard methods (initial rise and whole glow curve) are used to obtain the energy values and frequency factors of the traps. The method has been used successfully to analyze the well-known composite TL glow curve of the dosimetric material LiF (TLD-100). The limitations of the method are illustrated by analyzing the highly complex TL glow curve of a UV irradiated synthetic calcite consisting of at least 6 TL peaks. Although the method works best for TL glow curves described by first order kinetics, it should also be applicable to more general kinetics.

——————————————————————————–

**2. FIT OF SECOND ORDER THERMOLUMINESCENCE GLOW PEAKS USING THE LOGISTIC DISTRIBUTION FUNCTION**

V. Pagonis and G. Kitis

Radiation Protection Dosimetry, 93, No.3, 225-229 (2001).

A new Thermoluminescence glow-curve deconvolution (GCD) function is introduced that accurately describes second order thermoluminescence (TL) curves. The logistic asymmetric (LA) statistical probability function is used with the function variables being the maximum peak intensity, the temperature at the maximum peak intensity and the LA width parameter a_{2}. An analytical expression is derived from which the activation energy E can be calculated a sa function of Tm and the LA width parameter a_{2} with an accuracy of 2% or better. The accuracy of the fit was tested for values of E between 0.7 and 2.5 eV, for s values between 10^{5} and 10^{25} s-1, and for trap occupation numbers no/N between 1 and 10^{6}. The goodness of fit of the LA function is described by the Figure of Merit (FOM) which is found to be of the order of 10^{-2}.

——————————————————————————–

**1. FIT OF FIRST ORDER THERMOLUMINESCENCE GLOW PEAKS USING THE WEIBULL DISTRIBUTION FUNCTION**

V. Pagonis, S.M. Mian and G. Kitis

Radiation Protection Dosimetry, 93, 11-17 (2001).

A new Thermoluminescence glow-curve deconvolution (GCD) function is introduced that accurately describes first order thermoluminescence (TL) curves. The GCD functions are found to be accurate for first order TL peaks with a wide variety of the values of the TL kinetic parameters E and s. The 3-parameter Weibull probability function is used with the function variables being the maximum peak intensity (Im), the temperature of the maximum peak intensity (Tm) and the Weibull width parameter b. An analytical expression is derived from which the activation energy E can be calculated as a function of Tm and the Weibull width parameter b. The accuracy of the Weibull fit was tested using the ten reference glow-curves of the GLOCANIN intercomparison program and the Weibull distribution was found to be highly effective in describing both single and complex TL glow curves. The goodness of fit of the Weibull function is described by the Figure of Merit (FOM) which is found to be of comparable accuracy to the best FOM values of the GLOCANIN program. The FOM values are also comparable to the FOM values obtained using the recently published GCD functions of Kitis et al [3]. It is found that the TL kinetic analysis of complex first-order TL glow curves can be performed with high accuracy and speed by using commercially available statistical packages.