ARMAX Forecast Model for Estimating the Annual radon Activity Concentration in Confined Environment by Short Measurements Performed by Active Detectors.

This work aims to implement a forecast model that, combined with the use of active instrumentation for a rather limited time, and with the knowledge of a set of data referring to the environmental parameters of the place to be monitored, can estimate the concentration of indoor radon activity for longer time periods. This model has been built through the MATLAB program, exploiting the theories of time series and, in particular, ARMAX models, to reproduce the variation in the concentration of radon activity. The model validation has been carried out by comparing real vs. simulated values. In addition, analytic treatment of input data, such as temperature, pressure, and relative humidity, can reduce the influence of sudden transients allowing for better stability of the model. The final goal is to estimate the annual radon activity concentration on the basis of spot measurements carried out by active instrumentation, such to avoid the need to measure for an entire calendar year by the use of passive detectors. The first experimental results obtained in conjunction with active radon measurement demonstrates the applicability of the method not only for forecasting future average concentrations, but also for optimizing remedial actions.

Monte Carlo simulation and radiometric characterization of proton irradiated [18O]H2O for the treatment of the waste streams originated from [18F]FDG synthesis process.

The aim of this work is quantifying the radionuclidic impurities of the irradiated [(18)O]water originated by the [(18)F]FDG synthesis process, and characterizing, from a radioprotection point of view, the waste streams produced. Two samples of 2.4ml [(18)O]H(2)O, contained in two different target cells, have been irradiated with a proton current of 37µA in a PETtrace cyclotron for about one hour each; after irradiation, without performing any chemical purification process but waiting only for the (18)F decay, they have been transferred in two vials and measured by HPGe gamma spectrometry and, subsequently, by Liquid Scintillation Counting. Previously, Monte Carlo calculations had been carried out in order to estimate the radionuclides generated within the target components ([(18)O]H(2)O, silver body and Havar® foil), with the aim to identify the nuclides expected to be found in the irradiated water. Experimental results for the two samples, normalized to the same irradiation time, show practically the same value of tritium concentration (about 36kBq/ml) while gamma emitters activity concentrations exhibit a greater spread. Considering that tritium derives from water activation while other pollutants are caused by activated cell materials released into water through erosion/corrosion mechanisms, such a spread is likely to be attributable to differences in the proton beam shape and position (production of different natural circulation patterns inside the target and different erosion mechanisms of the target cell walls). Both tritium and the other radioactive pollutants exhibit absolute values of activity and activity concentrations below the exemption limits set down in EURATOM Council Directive 96/29.