Decreased photoluminescence of the phosphor BaMgAL 10 O 17 :Eu due to oxidation of the europium dopant at high temperatures has been a subject of study for many years in relation to its use in lighting applications. However, understanding of the underlying effects that cause this reduction in photoluminescence remains incomplete and some of the mechanisms proposed in the literature are contradictory. Recent use of this phosphor as a thermal history sensor has extended the range of exposure

more »

... tions normally investigated in lighting applications to higher temperatures and multiple exposure times. The kinetics of the process were investigated by means of spectroscopy and material characterisation techniques. It was found that changes in the luminescence are the result of two simultaneous processes: the oxidation of Eu 2+ ions (through a process of diffusion) and a phase transition. The level of degradation of the phosphor is suggested to follow the Kolmogorov-Johnson-Mehl-Avrami (KJMA) model above 900 °C and thus can be predicted with knowledge of the exposure time and temperature. This is useful in applications of the phosphor as a temperature sensor. A C C E P T E D M A N U S C R I P T ACCEPTED MANUSCRIPT Rietveld refinement, only changed significantly above this temperature. Boolchand et al. [8] used Mössbauer spectroscopy to find Eu 3+ ions even in fresh samples, with the ions moving to new previously unoccupied sites after thermal degradation in air. However, the nature of this new site and the possibility that it is created as a consequence of the appearance of a new phase is not discussed by these authors. Bizarri and Moine [9] proposed a mechanism for the degradation of the phosphor based on luminescence, thermoluminescence and electron spin resonance (ESR) measurements. The mechanism involves three consecutive steps: the adsorption of gaseous oxygen into oxygen vacancies at the surface of the phosphor, the diffusion of europium ions through bulk material triggered by high temperatures, and electronic transfer from Eu 2+ ions to the adsorbed oxygen when the two elements are in close proximity. They simulated these three steps to account for the loss of luminescence and numerical results agreed well with the values of luminescence intensity obtained experimentally. The A is a pre-exponential constant, E a is the activation energy, k b is the Boltzmann constant and T the absolute temperature. The experimental data from the authors' previous work [11] are shown in Figure 5 fitted with only two parameters A and E a , and an Avrami exponent of 1. This value was chosen since it corresponds to a process in which no initiation phase exists i.e. it describes growth and saturation in a pre-nucleated material. This was done because it was earlier hypothesised by the authors that the rate of formation of the magnetoplumbite phase is controlled by diffusion of Eu ions rather than a probabilistic initiation process. Nevertheless, in order to test this hypothesis a fit was calculated with the exponent as one of the fitting variables and a value close to 1 (0.98) was determined. The fit agrees quite well with the experimental data, thereby supporting the hypothesis, at all except the lowest temperature of Highlights  Oxidation of BAM:Eu is investigated for a new temperature sensing application.  The oxidation follows the KJMA model above 900 °C.  Optical spectroscopy and materials characterisation techniques are employed.  Temperature can be sensed with knowledge of the exposure time.