Answer to Question #12587 Submitted to "Ask the Experts"

When calibrating thermoluminescent dosimeters (TLDs) to high-energy fields such as from a ¹³⁷Cs source, they will have accurate readings for high energies but will overestimate at low energies. Is there a correction factor that may be applied such as using differences in linear energy deposition that can counteract this error at low energies?

Your statement is generally correct as relates to dosimeters used to assess doses to individuals. Most phosphors used in TLDs have effective atomic numbers somewhat greater than that of typical soft tissue, the medium to which we would normally like to estimate dose for personal dosimetry purposes. The increased response at lower photon energies in such cases is associated with the enhanced photoelectric effect in the dosimeter compared to soft tissue because of the strong dependence on atomic number; the photoelectric cross-section increases as the atomic number to the fourth or fifth power.

For a phosphor such as LiF:Mg,Ti, the effective atomic number is slightly greater than that for soft tissue, resulting in a modest dosimetric over-response, typically by a maximum  of roughly 20% to 40% at about 30 keV, depending on physical characteristics of the dosimeter LiF element. Other less tissue-equivalent phosphors, such as CaSO₄:Dy exhibit considerably greater overresponses; at 25 keV the cited phosphor overresponds by a maximum of typically between a factor of about 10 to 15 (900% to 1,400% overresponse). Such phosphors as the latter are typically used in dosimeters with the phosphor shielded to reduce low energy response.

The added response is associated with electrons produced by the excess photoelectric events and depositing energy in the TLD phosphor. This excess energy imparted can be interpreted through the respective values of the mass energy absorption coefficient ratio for the phosphor compared to soft tissue at any particular energy of interest. If you know the energies of the photons involved, you can then make a reasonable estimate of the expected overresponse of the innate TL phosphor. For example, using values from the National Institute for Standards and Technology (NIST) for LiF and International Commission on Radiation Units and Measurements (ICRU) values for soft tissue (available in Table 4 at this link) at 20 keV, the respective mass energy absorption coefficients for LiF and soft tissue are 0.6494 cm²g^-1 and 0.5663 cm²g^-1, leading to a ratio of 1.15. This implies that the phosphor would yield a dose response 15% greater than that of soft tissue. The actual value may differ from this calculation, depending on the specific physical characteristics of the LiF element, especially its thickness; the thicker the element, the greater will be the attenuation of the low-energy photons in the element and the greater will be the likely difference between the LiF and soft tissue dose response. The directionality of the photon field on the dosimeter is also a factor that affects the dosimeter response and can contribute to larger differences between phosphor and tissue doses at lower photon energies compared to responses at ¹³⁷Cs energy.

Given the above considerations, it may not be practical to attempt to make energy corrections to dosimeter responses except under very well-defined conditions. Even then, uncertainties may remain. In most practical personal dosimetry situations, the histories of exposure, in terms of photon energies, intensities, and directions, are not tracked very well during the time between changeouts of dosimeters and attempting corrections to account for energy variations is impractical.

I hope this is helpful to you.

George Chabot, PhD, CHP