Development of LiF:Mg,Cu,Si TL material (new KLT-300) with a low-residual signal and high-thermal stability

LiF-based thermoluminescence (TL) materials have been widely used for radiation dosimetry due to their attractive features. LiF:Mg,Cu,P is one of the most sensitive tissue-equivalent TL materials, approximately 40 times more sensitive than LiF:Mg,Ti (TLD-100), but it has two main drawbacks: a thermal loss of the TL sensitivity when annealed at temperatures >240 degrees C, and a relatively high-residual signal. Recently, LiF:Mg,Cu,Na,Si TL material was developed to overcome these drawbacks at the Korea Atomic Energy Research Institute, but it provided only marginal improvements in reducing the residual signal. The newly developed LiF:Mg,Cu,Si TL material has a significantly lower residual signal and a better stability to thermal treatments. In this article, the preparation method and some dosimetric properties (sensitivity and residual signal) of the new LiF:Mg,Cu,Si TL material are presented. At the end of the preparation procedures, a dual-step annealing method is introduced and this has proved as a very efficient method to reduce the high-temperature peak and is the cause of residual signal. Therefore, the high-temperature peak in the glow curve was significantly reduced. The sensitivity is approximately 20 times higher than that of TLD-100 and the residual signal was estimated to be approximately 0.04%.     

UV-induced bleaching of deep traps in Harshaw TLD LiF:Mg,Cu,P and LiF:Mg,Ti

The effects of UV-induced bleaching of deep traps on Harshaw thermoluminescent (TL) LiF:Mg,Cu,P and LiF:Mg,Ti materials were investigated. During a normal heating cycle, LiF:Mg,Cu,P is limited to a maximum temperature of 240 °C. LiF:Mg,Ti can be read to higher temperatures; however, encapsulation in polytetrafluoroethylene limits the maximum readout temperature to 300 °C. Generally, for both materials, these respective temperatures are sufficient for emptying traps corresponding to the main dosemetric peaks. However, when the dosemeters are subjected to a high dose level, such as 1 Gy (much higher than individual monitoring dose levels), higher temperature traps are filled that cannot be emptied without exceeding the above-mentioned maximum temperatures. These high temperature traps tend to be unstable during normal readout and can significantly increase the residual TL signal. The purpose of this study was to investigate the applicability of a UV-induced bleaching technique for emptying higher temperature traps following high-dose applications. In addition, in the case of LiF:Mg,Cu,P, where the maximum readout temperature is significantly lower, we investigated the possibility of reducing the residual signal using the application of repeated readout cycles. The optical bleaching approach was found to be effective in the case of LiF:Mg,Ti; however, for LiF:Mg,Cu,P, no reduction in the residual signal was observed. For this latter material, the application of repeatable readout cycles is very effective and residual signals equivalent to dose levels as low as 0.01 mGy were observed following an initial dose of 5 Gy. To the best of our knowledge, this work is the first attempt to apply an 'optical annealing' technique to the Harshaw thermoluminescent dosemeter (TLD) materials.     

The application of LiF:Mg,Cu,P to large scale personnel dosimetry: current status and future directions

LiF:Mg,Cu,P is starting to replace LiF:Mg,Ti in a variety of personnel dosimetry applications. LiF:Mg,Cu,P has superior characteristics as compared to LiF:Mg,Ti including, higher sensitivity, improved energy response for photons, lack of supralinearity and insignificant fading. The use of LiF:Mg,Cu,P in large scale dosimetry programs is of particular interest due to the extreme sensitivity of this material to the maximum readout temperature, and the variety of different dosimetry aspects and details that must be considered for a successful implementation in routine dosimetry. Here we discuss and explain the various aspects of large scale LiF:Mg,Cu,P based dosimetry programs including the properties of the TL material, new generation of TLD readers, calibration methodologies, a new generation of dose calculation algorithms based on the use of artificial neural networks and the overall uncertainty of the dose measurement. The United States Navy (USN) will be the first US dosimetry processor who will use this new material for routine applications. Until June 2002, the Navy used two types of thermoluminescent materials for personnel dosimetry, CaF2:Mn and LiF:Mg,Ti. A program to upgrade the system and to implement LiF:Mg,Cu,P, started in the mid 1990s and was recently concluded. In 2002, the new system replaced the LiF:Mg,Ti and is scheduled to start replacing the CaF2:Mn system in 2006. A pilot study to determine the dosimetric performance of the new LiF:Mg,Cu,P based dosimetry system was recently completed, and the results show the new system to be as good or better than the current system in all areas tested. As a result, LiF:Mg,Cu,P is scheduled to become the primary personnel dosimeter for the entire US Navy in 2006.     

Lithium fluoride: from LiF:Mg,Ti to LiF:Mg,Cu,P

Differences and similarities between LiF-based LiF:Mg,Ti and LiF:Mg,Cu,P are discussed, with respect to their dosimetric properties--sensitivity, non-linearity of dose response and heavy charged particle efficiency, as related to the concentration and the individual role of the Mg, Ti, Cu and P dopants. To study further the role of these dopants, the properties of some new, 'hybrid' phosphors: LiF:Mg,Cu,Ti and LiF:Mg,P, specially developed for this purpose, are also discussed. In the glow curve of LiF:Mg,Cu,P with a low concentration of Mg a new peak was found, which appears to be an analogue of peak 4 in LiF:Mg,Ti, Magnesium apparently controls most of the dosimetric properties of LiF-based phosphors. For instance, charged-particle efficiency appears to be anti-correlated with the concentration of Mg, being much less dependent on the content of other dopants. On the other hand, some properties of LiF-based systems seem to be correlated with changes in the emission spectra. It is suggested that Ti hampers the acceptance of any increased amount of Mg into more traps in LiF:MgTi. The absence of Ti, not the presence of P or Cu, is therefore a key to the high sensitivity of LiF:MgCuP.     

Thermoluminescence properties of LiF:Mg,Cu,Na,Si pellets in radiation dosimetry

Sintered LiF:Mg,Cu,Na,Si thermoluminescence (TL) pellets have been developed for application in radiation dosimetry. LiF:M,Cu,Na,Si TL pellets were made from TL powders using a sintering process, that is, pressing and heat treatment. These pellets have a diameter of 4.5 mm, and a thickness of 0.8 mm are blue in colour and have a mass of 28 mg each. After 400 pellets had been produced they were irradiated with 137Cs gamma radiation and samples having a sensitivity within a +/-5% standard deviation were selected for experimental use. In the present study, the physical and dosimetric properties of LiF:Mg,Cu,Na,Si TL pellets were investigated for their emission spectrum, dose response, energy response and fading characteristics. Photon irradiation for the experiments was carried out using X ray beams and a 137Cs gamma source at the Korea Atomic Energy Research Institute (KAERI). The average energies and the dose were in the range of 20-662 keV and 10(-6) - 10(2) Gy respectively. The glow curves were measured with a manual type thermoluminescence dosimetry reader (system 310, Teledyne) at a constant nitrogen flux and a linear heating rate. For a constant heating rate of 5 degrees C.s(-1). the main dosimetric peak of the glow curve appeared at 234 degrees C, its activation energy was 2.34 eV and the frequency factor was 1.00 x 10(23). The TL emission spectrum appeared at the blue region centred at 410 nm. A linearity of photon dose response was maintained up to 100 Gy. The photon energy responses relative to the 137Cs response were within +/-20% in the overall photon energy region. No fading of the TL sensitivity of the pellets stored at room temperature was found over the course of a year. Therefore LiF:Mg,Cu,Na,Si TL pellets can be used for personal dosimetry, but more research is needed to improve the characteristics for repeated use.     

Pre- and post-irradiation fading effect for LiF:Mg,Ti and LiF:Mg,Cu,P materials used in routine monitoring

LiF is a well-known thermoluminescent (TL) material used in individual monitoring, and its fading characteristics have been studied for years. In the present study, the fading characteristics (for a period of 150 d) of various commercial LiF materials with different dopants have been evaluated. The materials used in the study are those used in routine procedures by the Personal Dosimetry Department of Greek Atomic Energy Commission and in particular, LiF:Mg,Ti (MTS-N, TL Poland), LiF:Mg,Cu,P (MCP-N, TL Poland), LiF:Mg,Cu,P (MCP-Ns, thin active layer detector, TL Poland) and LiF:Mg,Cu,P (TLD100H, Harshaw). The study showed that there is a sensitivity loss in signal of up to 20 % for the MTS-N material for a 150-d period in the pre-irradiation fading phase. The MCP-N has a stable behaviour in the pre-irradiation fading phase, but this also depends on the readout system. As far as the post-irradiation fading effect is concerned, a decrease of up to 20 % for the MTS-N material is observed for the same time period. On the other hand, the LiF:Mg,Cu,P material presents a stable behaviour within ± 5 %. These results show that the fading effect is different for each material and should be taken into account when estimating doses from dosemeters that are in use for >2 months.     

On the roles of the dopants in LiF: Mg,Cu,Na,Si thermoluminescent material

In this paper, some results of the study on the roles of the dopants in the LiF:Mg,Cu,Na,Si thermoluminescent (TL) material that was developed at the Korea Atomic Energy Research Institute for radiation protection are presented. Although there have been many studies to investigate the roles of the dopants in LiF:Mg,Cu,P TL material in the TL process, there are some discrepancies in the understanding of the roles of Cu and P between various researchers. In case of LiF:Mg,Cu,Na,Si TL material, there are a few studies on the roles of the dopants. Three kinds of samples in each of which one dopant is excluded, and the optimised sample, were prepared for this study. The measurements and analysis of the three-dimensional TL spectra, based on the temperature, wavelength and intensity, and the glow curves for those samples are used in this study. The results show that Mg plays a role in the trapping of the charge carriers and Cu plays a role in the luminescence recombination process; however, the effect of Na and Si on the glow curve structure and the TL emission spectra is much less than that of Mg and Cu. It is considered that Na and Si each plays a role in the improvement of the luminescence efficiency.     

A comparative study on the susceptibility of LiF:Mg,Ti (TLD-100) and LiF:Mg,Cu,P (TLD-100H) to spurious signals in thermoluminescence dosimetry

It is well known that spurious signals can occur in thermoluminescence dosimetry (TLD) whenever contaminants (i.e. dirt, oil, dust) are present on the surface of the TLD card or crystal during the read-out process. For TLD cards, the Teflon material can also contribute to the background noise and this contribution has been found to depend on the material's light absorption. These non-radiation-induced signals contribute to the total light output during TLD read-out and can lead to incorrect dosimetry especially for low-dose measurements such as personal dosimetry. However, these spurious signals are generally in the low-temperature channels and are mostly accompanied by abnormal glow curves. Most of the published reports dealing with this type of spurious TL signal are on the LiF:Mg,Ti (TLD-100) material. The relatively new TLD material, LiF:Mg,Cu,P, is more sensitive and has higher signal-to-noise ratio than the traditional LiF:Mg,Ti. In this study, the effects of disturbing signals to the LiF:Mg,Cu,P (TLD-100H) cards used in personal dosimetry are investigated and compared with those of LiF:Mg,Ti (TLD-100).     

Study of real time temperature profiles in routine TLD read out--influences of detector thickness and heating rate on glow curve shape

This paper reports the results of a study using a commercial routine read out system with non-contact hot nitrogen heating and linear heating gas profiles. Glow curves of LiF:Mg,Ti as well as LiF:Mg,Cu,P were analysed for different linear heating rates beta from 1 to 30 degrees K s(-1). Different thermoluminescent detectors (TLDs) of different thicknesses (0.38-0.90 mm) were studied and compared. By means of the application of CGCD program considering kinetic parameters of the used TL-material the analysis of the peak temperature of the individual TL peaks lead to the approximation of the real heating profile T(chip)(t) in the TL chip. The real heating profile deviates strongly from linearity and can be characterised by the solution of a differential equation T(chip)(t) = F [T(gas)(t)]. The model of this equation is discussed in the paper. The difference between gas and chip temperatures are heating rate and chip thickness dependent and reach values of up to 100 degrees C (for thick detectors and fast heating rates). Especially for LiF:Cu,P, knowledge of the real chip temperature is essential, since read out shall be performed at the highest possible temperature, without destroying the dosimetric properties of the material. On the basis of this work, an optimisation of the readout parameters for LiF:Cu,P is possible.     

Dosimetry at the Portuguese research reactor using thermoluminescence measurements and Monte Carlo calculations

This work presents an extensive study on Monte Carlo radiation transport simulation and thermoluminescent (TL) dosimetry for characterising mixed radiation fields (neutrons and photons) occurring in nuclear reactors. The feasibility of these methods is investigated for radiation fields at various locations of the Portuguese Research Reactor (RPI). The performance of the approaches developed in this work is compared with dosimetric techniques already existing at RPI. The Monte Carlo MCNP-4C code was used for a detailed modelling of the reactor core, the fast neutron beam and the thermal column of RPI. Simulations using these models allow to reproduce the energy and spatial distributions of the neutron field very well (agreement better than 80%). In the case of the photon field, the agreement improves with decreasing intensity of the component related to fission and activation products. (7)LiF:Mg,Ti, (7)LiF:Mg,Cu,P and Al(2)O(3):Mg,Y TL detectors (TLDs) with low neutron sensitivity are able to determine photon dose and dose profiles with high spatial resolution. On the other hand, (nat)LiF:Mg,Ti TLDs with increased neutron sensitivity show a remarkable loss of sensitivity and a high supralinearity in high-intensity fields hampering their application at nuclear reactors.     

Role of impurities in the thermoluminescence of LiF:Mg,Cu,P

The thermoluminescent properties of lithium fluoride doped with magnesium, copper and phosphorous are reviewed. This material shows specific advantages over LiF (TLD-100) in that is has higher sensitivity, less energy dependence and no supralinear behaviour with dose. However, it suffers from an irreversible loss in sensitivity when heated above 270 degrees C. Possible causes for this loss in sensitivity are discussed and experimental results are shown that indicate potential roles for the Cu, P and O impurities in this process. The present work focuses on the understanding of the impurity contents, inclusions, cluster formation and valence changes in this material by using differential scanning calorimetry (DSC) and glow discharge mass spectroscopy (GDMS). This research concludes that phosphorus inclusions are formed and a valence change of the Cu ions occurs after annealing above 270 degrees C, which degrades the TL sensitivity of LiF:Mg,Cu,P phosphors. It is also shown that a high content of oxygen in the material and control of the oxygen content during processing are critical to maintaining the sensitivity of these phosphors.     

Influence of background exposure on detection and determination limits for a TL dosimetry system based on LiF:Mg,Cu,P(GR-200A)

Thermoluminescence dosemeters are widely used to monitor personal doses. For these low dose range applications, it is important to determine the detection limit L(D) and the determination limit L(Q) of the dosimetric system. The influence of background exposure on these limits for LiF:Mg,Cu,P(GR-200A) based TL dosimetry was investigated. Both the conventional analysis and the glow curve analysis methods were used to determinate these limits. The detection limit L(D) was compared with the recording level and the investigation level. A systematic error can occur in the occupational dose evaluation when the detection limit L(D) is more than the recording level. It was found that the L(D) of the dosimetric system-based LiF:Mg,Cu,P(GR-200A) was less than the recording level for exposure time tau > or = 10 days considering an annual dose limit of 1 mSv for the public recommended in ICRP Publication 60.     

Thermoluminescence properties of LiMgF3 doped with Ce,Er and Dy

The main dosimetric properties are reported of a new perovskite, LiMgF3, doped with Ce, Er and Dy impurities. An annealing temperature of 400 degrees C for one hour is necessary to erase any previous signal and to stabilise its sensitivity. A readout up to 450 degrees C gives the same result. The glow curve structure consists of two intense and isolated glow peaks at 170 degrees C and 315 degrees C. The sensitivity of both glow peaks to beta irradiation increases as the dopant concentration increases. The sensitivity is higher than that of LiF:Mg,Ti using the glow peaks 4+5 integral. Other properties like TL dose response, fading, sensitivity versus successive readout and ammealing cycles, sensitivity versus the heating rate, sensitivity to light and trapping parameter evaluation were also carried out.     

Developments in the synthesis of LiF:Mg,Cu,Na,Si TL material

A procedure for synthesis of the highly sensitive pellet-type LiF:Mg,Cu,Na,Si thermoluminescent (TL) detector has been newly developed. It was found that the optimum concentrations of dopants for a pellet-type LiF:Mg,Cu,Na,Si TL detector were found to be Mg: 0.2 mol %, Cu: 0.05 mol %, Na: 0.9 mol%, and Si: 0.9 mol%. The TL sensitivity of this new detector was about 30 times higher than that of the TLD-100 by light integration measurements. Reusability study of the detector was carried out for 10 cycles. The results show that the coefficients of variation for each detector separately did not exceed 0.016, and that for all 10 detectors collectively was 0.0054.     

Sensitivity loss of Lif:Mg,Cu,P thermoluminescence dosemeters caused by oven annealing

LiF:Mg,Cu,P is a very sensitive thermoluminescence material that can be used for personal dosimetry in radiology. But if this material is heated too much during annealing or reading it quickly loses its sensitivity. This study shows that an annealing oven in wide use can cause thermal damage to the dosemeters owing to inhomogeneous temperature distribution in the annealing tray. At annealing temperatures>240 degrees C, differences of only 1 degrees C lead to significant losses of sensitivity. Therefore, it is necessary to measure the distribution of temperature in the annealing tray for correct placement of the dosemeters in the tray.     

Method of thermoluminescent measurement of radiation doses from micrograys up to a megagray with a single LiF:Mg,Cu,P detector

On the basis of the newly discovered behaviour of LiF:Mg,Cu,P detectors at high and ultra-high doses, a new method of thermoluminescence (TL) measurement of radiation doses ranging from micrograys up to a megagray, has been recently developed at the Institute of Nuclear Physics (IFJ). The method is based on the relationship between the TL signal, integrated in the given temperature range and dose. It is quantified by a parameter called the 'ultra-high temperature ratio'. It has been demonstrated that this new method can measure radiation doses in the range of about 1 μGy to 1 MGy, using a single LiF:Mg,Cu,P detector. This method was recently successfully blindly tested for 10 MeV electrons up to doses of 200 kGy. It can be used for dosimetry in high-energy accelerators, especially in the Large Hadron Collider at CERN, and has great potential for accident dosimetry in particular.     

Influence of readout parameters on TL response,re-usability and residual signal in LiF:Mg,Cu,P

It has recently been recommended that heating rates do not exceed 10 K.s(-1) and that the maximum temperature of readout should not exceed 265 degrees C for LiF:Mg,Cu,P. In some cases, a decrease of sensitivity in this material in the first of several re-use cycles had been reported. Influence of heating rates up to 30 K.s(-1), duration time up to 40s and maximum readout temperatures up to 270 degrees C on TL response, re-usability and residual signal was investigated. It was found that the maximum readout temperatures above 240 degrees C may lead to the thermoluminscent response decrease in the first several re-use cycles. The readout parameters can be optimised to minimise the residual signal (less than 0.4%) and to retain a constant sensitivity at the same time at high heating rates up to 30 K.s(-1) in a short time (less than 1 min per TL chip) without the necessity of heating above 240 degrees C. A concept of 'efficient residual signal' was put forward to quantify more accurately the real residual signal which affects the precision of the next measurement.     

TL emission spectra from differently doped LiF:Mg detectors

There are two widely applied types of thermoluminescent detectors based on LiF:Mg luminophor: Lif:Mg,Ti and highly sensitive LiF:Mg,Cu,P. The role of luminescence centres in these materials is usually attributed to defects connected with, respectively, titanium and phosphorus dopants. In order to check how composition of dopants introduced into the LiF lattice influences emission spectra, measurements on a series of variously doped LiF:Mg samples were performed. Apart from LiF:Mg,Cu,P and LiF:Mg,Ti detectors with different concentration of activators, an experimental sample being a kind of a 'hybrid' between both standard materials was also prepared. It was synthesised with concentrations of magnesium and copper identical to those used for LiF:Mg,Cu,P preparation. but instead of phosphorus it was doped with titanium (LiF:Mg,Cu,Ti). The measurements of the emission spectra were performed by using a liquid nitrogen cooled CCD 1024E detector with an SP150 spectrograph. During the measurements the samples were placed inside a cryostat in a vacuum. Resulting data were numerically deconvoluted for individual peaks with respect to the wavelength and the temperature. The glow curve shape of this material resembles that of LiF:Mg,Cu,P, while sensitivity is at the level of LiF:Mg,Ti. Preliminary results indicate that emission of the LiF:Mg,Cu,Ti sample is similar to that of LiF:Mg,Cu,P rather than to LiF:Mg,Ti, showing a maximum for wavelengths well below 400 nm.     

Properties of the 4.45 eV optical absorption band in LiF:Mg,Ti

The optical absorption (OA) and thermoluminescence (TL) of dosimetric LiF:Mg,Ti (TLD-100) as well as nominally pure LiF single crystal have been studied as a function of irradiation dose, thermal and optical bleaching in order to investigate the role of the 4.45 eV OA band in low temperature TL. Computerised deconvolution was used to resolve the absorption spectrum into individual gaussian bands and the TL glow curve into glow peaks. Although the 4.45 eV OA band shows thermal decay characteristics similar to the 4.0 eV band its dose filling constant and optical bleaching properties suggest that it cannot be associated with the TL of composite peaks 4 or 5. Its presence in optical grade single crystal LiF further suggests that it is an intrinsic defect or possibly associated with chance impurities other than Mg, Ti.