Uncertainty budget for the dose range between 2 Gy and 25 Gy established

The limit is, on the one hand, determined by the individual background signal of the alanine probes and, on the other hand, by the so-called base line drift, a slow variation of the spectrometer sensitivity and background which may differ from one spectrum to the other. Various data analysis tools, such as singular value decomposition and Fourier- and wavelet transforms, were used to eliminate the unwanted background signal. Finally, the background treatment was indeed able to remove occasionally occurring, strong background for probes which could otherwise not have been used. However, no significant decrease of the uncertainty was achieved. As a consequence, the development of the measurement- and analysis method for alanine was regarded as being completed. A graphical user interface for the analysis software was built. The uncertainty budget was subjected to a further revision and the result published.

The uncertainty is estimated in two steps. In the first step, the uncertainty of the so-called dose-normalised amplitude A_D is determined. The latter is the result of an analysis method which was developed at the PTB on the basis of a method invented by the NPL. By means of an appropriate normalisation, A_D obtains the unit Gy. The repeatability of A_D turned out to have a constant value of 20 mGy over the whole range between 2 Gy and 25 Gy, which corresponds to a relative standard uncertainty of 1% at 2 Gy and 0.1% at 20 Gy, respectively. Further contributions to the uncertainty budget are due to the uncertainties of the mass of the probes, the irradiation temperature, the repeatability of the irradiations and the uncertainty of the primary standard.

The dose is determined with the aid of a calibration line, the parameters of which are calculated from five pairs of (D_W,A_D)-values. The uncertainty of the calculated dose is deduced from the uncertainties of the calibration line parameters (which again are obtained from the uncertainties of the A_D-values for the known dose values) and from the uncertainties of the A_D-values for the unknown values of the applied dose. The achievable uncertainties of the dose determined via alanine/ESR are summarized in figure.

In addition, investigations of the response of the alanine dosimeter with respect to high-energy electrons and of the possible influence of different materials surrounding the alanine probe during irradiation were carried out. Also, home-made alanine probes were used with success as well. Further on, possible alternatives to alanine as a detector material were investigated. An enhanced sensitivity, and thus a reduction of the uncertainty for low doses, will probably be obtained by using lithium formate monohydrate instead of alanine. The measurement- and analysis procedures would have to be adapted only slightly for this purpose.

Figure:	Relative standard measurement uncertainty of the dose determined via alanine/ESR (excluding the uncertainty of the primary standard of 0.2%) as a function of the applied dose. Red: calibration- and test probes irradiated on the same day; blue: time delay between irradiation of calibration- and test probes up to 8 months.