A critical evaluation of the degradation state of dry archaeological wood from Egypt by SEM, ATR-FTIR, wet chemical analysis and Py(HMDS)-GC-MS

An in-depth investigation was undertaken of the chemical changes to wood induced by degrading agents in dry burial environments. The degradation state of eleven wood samples from dry archaeological sites in Egypt was evaluated by combining the information obtained by scanning electron microscopy (SEM), wet chemical analysis (WCA), attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) and analytical pyrolysis gas chromatography mass spectrometry with in situ silylation (Py(HMDS)-GC-MS). The samples belonged to five different wood species (Faidherbia albida, Ficus sycomorus, Taxus baccata, Pinus sylvestris and Tamarix sp.) and came from three different archaeological sites corresponding to three different time periods (from ca. 1700 BC to ca. 1700 AD). The results were compared with sound reference wood of the same species. SEM enabled a fungal attack to be identified in some of the samples. WCA highlighted the preferential loss of lignin or carbohydrates in these samples, but also showed unusually high values of water-soluble substances in most of them. FTIR spectra acquired before and after the extraction of the samples revealed that this soluble fraction generally contained depolymerised carbohydrates and/or lignin units. Py(HMDS)-GC-MS was applied without any sample pre-treatment and was used to assess the alteration/depolymerisation of the individual wood components, thus complementing the picture regarding the chemical changes. The results showed that, unlike waterlogged archaeological wood, most of the degraded wood components - depolymerised carbohydrates and/or lignin - had not leached away from the dry wood matrix, and thus complicated the interpretation of data. The commonly used parameters, such as the H/L (holocellulose/lignin) ratio, failed to give a correct evaluation of the wood degradation when both carbohydrates and lignin were degraded. The preservation conditions of the samples were very variable, from very good to a high depletion of carbohydrates, to the preferential depletion of lignin, or comparable levels of carbohydrate and lignin degradation. This was sometimes observed within the same wood species and the same archaeological site. This highlighted how difficult the interpretation can be in order to correctly evaluate the multiple causes of degradation affecting dry archaeological wood. An analytical approach using complementary techniques appears to be essential.