Abstract
This study presents the application of real-time mid-infrared spectroscopy for the non-destructive characterisation of heritage materials, supported by multivariate analysis. A novel, ultra-fast portable mid-IR spectrometer was used to analyse a range of heritage material samples, including minerals, pigments, paints and polymers commonly found in artworks and historical objects. Part of the spectral datasets were inspected using principal component analysis and hierarchical cluster analysis, with selected cases further examined using partial least squares discriminant analysis and variable importance in projection scoring. The results enabled reliable identification of calcium sulfate hydration states, differentiation between blue and green pigment types, subtle separation of binder-based groupings between aqueous and acrylic paints, as well as classification of synthetic foams into broad polymer families. Additionally, key discriminative wavenumbers were identified at 1900–1600 cm⁻¹, 4500, 3700, 2900–2800, and 2100 cm⁻¹, corresponding to diagnostic functional groups crucial for heritage material identification. Overall, the technique showed significant potential for monitoring and classification of heritage materials in situ, with particular relevance for museums, archaeological sites and conservation studios, especially as future developments aim to expand infrared coverage, and integrate scanning systems and complementary analytical techniques.
| Original language | English |
|---|---|
| Journal | Heritage Science |
| ISSN | 2050-7445 |
| Publication status | - - 30 Apr 2026 |
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver
- MLA