Measured cross sections for the production of the PET isotopes , and from carbon and oxygen targets induced by protons (40–220 ) and carbon ions (65–430 ) are presented. These data were obtained via activation measurements of irradiated graphite and beryllium oxide targets using a set of three scintillators coupled by a coincidence logic. The measured cross sections are relevant for the PET particle range verification method where accurate predictions of the emitter distribution produced by therapeutic beams in the patient tissue are required. The presented dataset is useful for validation and optimization of the nuclear reaction models within Monte Carlo transport codes. For protons the agreement of a radiation transport calculation using the measured cross sections with a thick target PET measurement is demonstrated.

The $$^\text {nat}$$ nat C(p,x)$$^{11}$$ 11 C reaction has been discussed in detail in the past [EXFOR database, Otuka et al. (Nuclear Data Sheets 120:272–276, 2014)]. However, measured activation cross sections by independent experiments are up to 15% apart. The aim of this study is to investigate underlying reasons for these observed discrepancies between different experiments and to determine a new consensus reference cross section at 100 MeV. Therefore, the experimental methods described in the two recent publications [Horst et al. (Phys Med Biol https://doi.org/10.1088/1361-6560/ab4511, 2019) and Bäcker et al. (Nuclear Instrum Methods Phys Res B 454:50–55, 2019)] are compared in detail and all experimental parameters are investigated for their impact on the results. For this purpose, a series of new experiments is performed. With the results of the experiments a new reference cross section of (68±3) mb is derived at (97±3) MeV proton energy. This value combined with the reliably measured excitation function could provide accurate cross section values for the energy region of proton therapy. Because of the well-known gamma-ray spectrometer used and the well-defined beam characteristics of the treatment machine at the proton therapy center, the experimental uncertainties on the absolute cross section could be reduced to 3%. Additionally, this setup is compared to the in-beam measurement setup from the second study presented in the literature (Horst et al. 2019). Another independent validation of the measurements is performed with a PET scanner.