As described in McKeever (1985), the use of TL in the study of radiation accelerated in the beginning decades of the twentieth century. A key area of research was to examine the relationship between the point defects within the materials studied (e.g. color centers) and their role in localizing (trapping) the electrons and holes ionized from their host atoms during the absorption of radiation. A feature of TL is that the luminescence at first increases and then decreases, forming a series of characteristic TL peaks as the temperature increases. It was realized that the cause of the TL peaks was the thermal release of trapped charge from lattice defects – with the larger the trapping energy, the higher the temperature of the TL peak. In 1930, in Vienna, Austria, Urbach discussed the connection between the energy needed to release the trapped charge and the TL peak position in a series of papers on luminescence from the alkali halides (Urbach 1930). However, it was not until the work of the group at the University of Birmingham in the United Kingdom that the relationship was quantified through the development of mathematical descriptions of the process (Randall and Wilkins 1945a, 1945b; Garlick and Gibson 1948).
Not long afterwards, Farrington Daniels and the research group at the University of Wisconsin, United States of America, discussed several applications in which TL could be a useful research tool. Among them was radiation dosimetry. Daniels and colleagues wrote: “Since in many crystals the intensity of thermoluminescence is nearly proportional to the amount of γ-radiation received, a considerable effort has been devoted to developing a practical means of measuring the exposure to gamma radiation.” (Daniels et al. 1953) – and so was born the field of thermoluminescence dosimetry. These authors specifically highlighted lithium fluoride as being the best crystal for this purpose and their work also initiated the parallel search for other TL dosimetry materials.
The growth of optically stimulated luminescence (OSL) as a method of radiation dosimetry had a similar genesis to that of TL and emerged as a potential dosimetry tool at about the same time. As described by Yukihara and McKeever (2011), the birth of OSL stems from the early work of the Becquerels, father and son, Edmond and Henri (E. Becquerel 1843; H. Becquerel 1883). These and similar studies through the late nineteenth and early twentieth centuries observed that phosphorescence could either be enhanced or quenched by the application of light to an irradiated material, the precise effect being dependent on the wavelength of the stimulating light. Observations of photoconductivity on some of these materials lead to the realization that free electrons were being produced during photostimulation and quenching of the phosphorescence (as discussed by Harvey 1957). Leverenz (1949) noted that when luminescence is enhanced by stimulating with an external light source, the eventual decay of the luminescence is unrelated to the characteristic fluorescence lifetime of the emitting species. As Leverenz (1949) states, for the luminescence to be produced, an “additional activation energy must be supplied to release the trapped electrons … This activation energy may be supplied by heat … or it may be supplied by additional photons.” When the energy is supplied by heat, TL results; when it is supplied by photons, OSL results.
It seems that the name, optically stimulated luminescence (OSL), appeared in the literature only in 1963 with the work of Fowler (Fowler 1963). Earlier names for the phenomenon included photophosphorescence, radiophotostimulation, photostimulation phosphorescence, co-stimulation phosphorescence, and photostimulated emission (see Yukihara and McKeever 2011). Even today, one often sees the phrase photostimulated luminescence (PSL) instead of OSL, the two names being synonymous, referring to the same phenomenon. For clarity, the more popular and most frequently used name of OSL will be used throughout this book.
As with TL, the connection between these optically stimulated effects and the initial absorption of energy from radiation was also established in the mid-twentieth century. The first suggested use of OSL in radiation dosimetry appeared with the work of Antonov-Romanovsky and colleagues (Antonov-Romanovsky et al. 1955). These authors examined infra-red (IR) stimulated luminescence from irradiated sulfides and related the IR-induced luminescence to the initial dose of radiation absorbed. Other similar works followed, but OSL did not emerge as a popular radiation dosimetry tool at this time, primarily because the emphasis of these studies was on sulfide materials and infra-red stimulation. These materials contained defects from which the energy required to release the trapped electrons was quite small (and, hence, the electrons could be released through absorption of low-energy, infra-red light). As a result, room temperature thermal stimulation could also release the trapped charges, which were thus observed not to be stable. As a result, the OSL signal is said to have faded with time since irradiation.
With the advent of studies into wider-band-gap materials (e.g. oxides, alkali halides, and sulfates) it was found that OSL could be stimulated by shorter-wavelength, visible light from defects that required larger activation energies to release the trapped charge. Hence, the OSL signal was more stable and did not fade. Nevertheless, the breakthrough in OSL’s application in radiation dosimetry came not in the radiation dosimetry field itself, but in the related field of geological dating. Huntley and colleagues (Huntley et al. 1985) demonstrated that OSL from quartz deposits in geological sedimentary layers could be used to determine the dose of natural radiation absorbed by the quartz grains since they were deposited in the layer. Analysis of the natural environmental dose rate then leads to a calculation of the age of the sediment (age = dose/dose rate). This paper, more than any other, opened the flood gates for the development of OSL in dosimetry. “Optical dating” is now an established technique (Aitken 1998; Bøtter-Jensen et al. 2003) and the method demonstrated that the OSL signal stimulated from defects with large activation energies could be stable for thousands of years in the right environmental circumstances.
Use of OSL in conventional radiation dosimetry started with the development of oxygen-deficient Al2O3, doped with carbon. This material was first suggested as a sensitive TL dosimeter at the Urals Polytechnical Institute in Russia (Akselrod and Kortov 1990), but the TL signal from this material was found to be very sensitive to visible light, such that exposure to daylight after irradiation reduced the subsequent TL signal. The group at Oklahoma State University in the United States then turned this apparent disadvantage into an advantage and showed that the material was a very sensitive OSL material (McKeever et al. 1996; Akselrod and McKeever 1999). The era of OSL dosimetry and the search for new OSL dosimetry materials had begun.
The term radiophotoluminescence (RPL) appears in the literature in the early 1920s with the work of Przibram and colleagues in Vienna (Przibram and Kara-Michailowa 1922; Przibram 1923). These researchers showed that photoluminescence (PL) can be induced in some materials only after exposure to ionizing radiation. Without irradiation, no PL is observed and, therefore, these authors gave their observation the name radiophotoluminescence. The distinction between RPL and OSL lies in the stability of the radiation-induced luminescence centers during stimulation with visible or infra-red light. In OSL, the luminescence signal decays under continued light stimulation, whereas in RPL, the signal remains constant. OSL is a destructive readout process (involving ionization) whereas RPL is a non-destructive process involving electron excitation, but not ionization. (These concepts will be discussed in more detail in later sections and chapters.)
The association between the RPL signal induced by the radiation