Prediction of neutron induced radioactivity in the concrete walls of a PET cyclotron vault room with MCNPX.

PURPOSE

The authors want to assess the relevance of the neutron activation of the concrete vault of the PET cyclotron at CIMES (Universidad de Malaga) by predicting specific activities of the main activation products in the vault and their variation profiles as a function of penetration depth into concrete at present and after 10 yr of cyclotron operation.

METHODS

The dual proton cyclotron is used for PET isotopes production, mainly 18F. During the years 2006 and 2008, the using rate has been 1 h/day at single beam (40 microA). From January 2008, using rate is 4 h/day at dual beam (80 microA). The energy of the cyclotron proton beam is 18 MeV. Four point locations were chosen on the walls of the cyclotron room to assess neutron induced activity concentrations. In each wall point location, neutron induced radionuclide specific activity was assessed from the wall surface to a depth of 120 cm within concrete. Simulations were carried out with the Monte Carlo based radiation transport code MCNPX (v2.6.0).

RESULTS

According to MCNPX calculations, activity depth profiles of activation products studied, except 54Mn, have a maximum at variable depths from the wall surface never beyond 12 cm. 54Mn activity decreases exponentially in all the studied depth ranges within wall concrete. The activity of 152Eu, 154Eu, 60CO, 134Cs, 46Sc, and 65Zn decreases exponentially beyond a 30 cm depth into concrete. 54Mn activity presents the faster decrease within a concrete vault with an attenuation length of 21 cm. According to MCNPX estimations, present activity in the cyclotron vault is mostly due to 46Sc and 60Co, with highest specific activity near the vault surface of 146 +/- 16 and 50 +/- 4.6 Bq/kg, respectively. 46Sc and 60Co activity measurements near the surface wall present an acceptable match with the estimation within the uncertainties, but measured activities of the other radionuclides are quite over the MCNPX estimations. The calculations after 10 yr of cyclotron operation predict a slight increase for short half-life radionuclides (46Sc, 54Mn, and 65Zn). However, long half-life neutron induced radionuclides importantly increase their activities, especially 60Co and 152Eu. These radionuclides and 46Sc give the main contribution to the wall activity in a 10 yr period. Estimated highest 46Sc and 60Co activities in 10 yr of cyclotron operation are in acceptable agreement with published measurements, but MCNPX calculated activities are lower than the measurements for the rest of the radionuclides.

CONCLUSIONS

MCNPX estimates acceptably present activity levels of 46Sc and 60Co as confirmed by activity measurements, but underestimates activity for the rest of the neutron induced radionuclides in the wall. Activity measurements have revealed the inhomogeneity of wall concrete atomic composition since large differences in activity values were found in two near wall surface locations with similar neutron flux. Such inhomogeneity cannot be modeled with the program that considers the material composition homogeneous. Highest specific activities estimated in a 10 yr operation are under exemption limits and therefore the cyclotron vault can be discarded as radioactive waste.