Studying ancient glass to bring to light new insights into the mechanism of glass corrosion

This work reports new insight into the mechanism of natural and long-term glass corrosion acquired by laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) analysis, an innovative technique for collecting high-resolution elemental maps of corroded archaeological glass. The theory that explains the mechanisms ruling glass corrosion has been revisited multiple times across the years thanks to the increasing availability of more advanced analytical techniques. However, that of glass corrosion remains a not completely disclosed question because of the multiple factors involved in the phenomenon: the glass composition, temperature, relative humidity, and the pH of the environment. In addition, glass alteration is a process that lasts for extremely long periods in natural conditions making its artificial simulation strikingly complicated [1]. For this reason, concrete evidence of the transformation of vitreous structures can be obtained only by the characterisation of altered ancient glass: such data can unlock new records on the nature of glass alteration layers and make the possible comparison with the results of laboratory tests. The 2D elemental maps of the cracked sample show a sharp boundaries between the composition of the glass and that of the cracks that seem to be filled with mineralized material that acts as cement among the fragments of the cracked glass. The presence of Si, Ca, and high levels of Al and K into the cracks suggests that this filling is probably part of the soil where the sample was found and aged for centuries, that has penetrated the cracks during the process of their formation. Assessment of the LA-ICP-MS 3D elemental maps obtained in this work by the analysis of corroded glass fragments has established that de-alkalinization is the main phenomenon occurring to glass surface and shown the changes in glass composition from the samples’ surface to the bulk, highlighting differences in lateral and in-depth distribution of glass network formers and modifiers. The images reveal enrichment of silicon in the first nm of the glass surface and a lower concentration of modifier ions. In particular, the concentration of modifier ions increases at different depths depending on the ion nature: examining the surface from the outside layers toward the bulk, it is possible to observe the increase of Ca content first, then that of Fe, Co, and Mn, and only after a micron and a half that of Na. These results indicate the different mobility of ions into the silica network and their different exchange capability during the alteration process providing additional information about the kinetics of the leaching process, a situation rather different from the currently accepted glass dissolution mechanism theories [2]. Moreover, the silica gel layer formed on the surface of the glass seems to block the ions migration outside the silica network since the 3D images show the presence of Na in the surface where the iridescent patina is not present. These results are the starting point to change the analytical approach for investigating the glass corrosion process by focusing attention on the study of archaeological glass as unique evidence of natural and long-term glass alteration and as fundamental method to gain new insight into the not yet clarified glass corrosion mechanism.