Abstract
Barkcloths, non-woven textiles originating from the Pacific Islands, form part of many museum collections and date back to the 18th and 19th centuries. The ability to determine different plant species which have been used for producing barkcloth is required by art historians to help understand the origin and use of the cloths and by conservators for whom the species type may have an impact on textile durability, deterioration and hence conservation. However, to date the development of a non-destructive, robust analytical technique has been elusive. This article describes the use of Fourier transform infrared spectroscopy with attenuated total reflection (ATR -FTIR) and principal component analysis (PCA) to differentiation between historic barkcloths. Three distinct groups of historic cloths were identified using PCA of the FTIR region between 1200 and 1600 cm-1 where molecular vibrations associated with tannins and lignins are dominant. Analysis of contemporary cloths only identified Pipturus albidus cloth as different and highlighted the difficulties around producing a representative textile sample to mimic the historic cloths. While the methodology does not itself identify species, the use of historically well-provenanced samples allows cloths showing similarities to group together and is a significant aid to identification.
Original language | English |
---|---|
Pages (from-to) | 32-41 |
Number of pages | 10 |
Journal | Journal of Cultural Heritage |
Volume | 39 |
Early online date | 23 Apr 2019 |
DOIs | |
Publication status | Published - Oct 2019 |
Keywords
- Pacific barkcloth
- ATR-FTIR
- multivariate analysis
- principal component analysis
- species differentiation
- microscopy
- Microscopy
- Species differentiation
- Multivariate analysis
- Principal component analysis
ASJC Scopus subject areas
- General Chemistry
- General Arts and Humanities
- Economics, Econometrics and Finance(all)
- Chemistry (miscellaneous)
- Archaeology
- Spectroscopy
- Conservation
- Computer Science Applications
- Materials Science (miscellaneous)