Abstract
The main methods for dating ballpoint pen inks in questioned documents used up to day are based mainly on the analysis of ink volatile components. These methods have some limitations still unresolved, such as the impossibility for dating documents older than two to five years and the destruction in part or totally of the questioned document after the analysis. This study aims to overcome these drawbacks by exploring the feasibility of dating inks based on their spectroscopic UV–vis-NIR diffuse reflectance spectra monitoring combined with a Partial Least-Squares (PLS) multivariate modelling. Inoxcrom® ink samples were exposed to artificial aging, their reflectance spectra were measured and a multivariate calibration was applied. Mathematical pretreatments of the spectroscopic data were carried out to enhance the prediction ability of the models and the qualitative interpretation of the spectra. The best PLS model was obtained after SNV spectral filter. Accurate predictions (RSD of 25%) were obtained for two of the five pen inks analysed and a correlation between the natural aging and the accelerated artificial aging could be established. Moreover, the spectra region more closely related to the ink aging process could be delimited, in which younger inks were characterized by modifications in the visible and NIR spectra region, while after aging the most influential region was the NIR spectrum. The mismatch with the other three pen inks tested could be attributed to consistent differences in ink formulations. Inks sharing a common chromatographic profile proved to fit correctly in the predictive model indicating that the methodology shows a great potential for future applications in the field of questioned documents dating.
| Original language | English |
|---|---|
| Pages (from-to) | 158-166 |
| Number of pages | 9 |
| Journal | Microchemical Journal |
| Volume | 140 |
| DOIs | |
| Publication status | Published - Jul 2018 |
| Externally published | Yes |
Keywords
- Ballpoint pen ink
- Diffuse reflectance
- Ink dating
- Multivariate regression