An algorithm for LET-analysis.

An algorithm for the derivation of LET-distributions from pulse-height spectra obtained with proportional counters is described. The method is based on Fourier transformation; it is applicable to spherical as well as non-spherical proportional counters. The relation between the energy mean, E,, of LET and the energy mean, g , , of the lineal energy density is given. 1. Introduction Tissue equivalent proportional counters are frequently used for the analysis of LET-distributions in neutron fields (Rossi and Rosenzweig 1955; Wilson and Field 1970). The method is limited by the fact that the experimentally observed pulse height distributions are the result of various random factors which cannot in general be easily separated. These limitations of the LET-analysis with proportional counters have been one of the starting points of microdosimetry when R'ossi and his co-workers realized that the observed distributions are more closely related to the biological effectiveness of a given type of radiation than the distribution of linear energy transfer. Linear energy transfer is only a mean value of the rate of energy loss of a charged particle along it's track, while the observed pulse height distributions reflect the actual variations of energy deposit'ion in the irradiated tissue. There are, however, cases where one needs an approximate overall evaluation of radiation quality for the purpose of comparing different types of radiation. I n these cases, and this refers particularly to radiotherapy and to radiation protection applications, LET-distributions offer a convenient frame of reference. It is for this reason that the present paper re-evaluates the algorithm used in the LET-analysis. The established method of LET analysis (Rossi and Rosenzweig 1955) is based on the assumption that energy loss straggling of the heavy charged particle, radial extension of its 8-ray halo and finite length of its track can be neglected. It is assumed that a charged particle traverses the sensitive region of the counter without change of its LET, and that the energy deposited is the product of the LET and of the chord length in the region. These assumptions are tenable only under the t'wo conditions: (i) that the neutron energies are high enough so that most of the recoil tracks are long compared with the dimensions of t'he region and (ii) that the stopping power of the particles and the diameter of the region are large enough that energy loss straggling and the efflux and influx of &rays can be neglected. Strictly …