"Datazione al radiocarbonio di malte storiche: messa a punto di un protocollo di purificazione del legante = Radiocarbon dating of historical mortars: optimization of the binder purification protocol"

The radiocarbon dating of inorganic materials such as lime mortars proved to be an innovative approach for the dating of the architectures, that allows to confirm data from the evaluation of the typological criteria, based on the archaeological stratigraphy. The tested method provides a an essential support where the dating obtained by the traditional method presents considerable limits. The protocol developed in this study is based on the CryoSoniC method, a physical-mechanical separation of the binder, focused on the separation of the binder from the aggregates and its purification from the potential contaminants that could affect the dating. The binder separation procedure, preceding the mortar dating, is followed to numerous analysis for the characterization and the control of the samples in order to test the quality of the isolation of the binder and the presence of contaminants. The procedure has been tested on 16 samples from archaeological site of Torba (province of Varese, Lombardy) which presents many construction phases between the Roman age and the early Middle age, therefore are represented various production techniques. The characterization of the samples prove the success of the binder purification protocol by sedimentation for the separation of the binder fraction from the sand fraction (aggregate), that permits to obtain reliable radiocarbon dates when the process of carbonation is prevalent on the pozzolanic reaction (that produces hydrate phases (C-S-H) and anionic clays). The samples dated through the accelerator mass spectrometry have been choosen based on three parameters: the ratio between the carbonate fraction and the silicate fractions of the sample, the stable carbon isotopes ratio δ13C, the intensity of the luminescence observed under microscope. In the cases where at least two parameters of three are satisfied, the results of the dating are in agree with the archaeological datas (Brogiolo et al. 2013), otherwise, the results from the binder taken directly from the mortar, the lime lumps and the samples made up of a high amount of clay are postdated if compared the expected date. The results show, on one hand, the successful of the separation and purification of the binder fraction from the sand fraction (both carbonate and silicate) of the mortars, on the other hand the real possibility to predict the reliability of the results through the mineralogical and isotopic characterization of the sedimented fraction