The ink can be a significant piece of evidence in the evaluation of written documents. In this work we present a new way of detection and classification of inks from ballpoint pens of different producers, using the ion mobility spectrometry (IMS) technique and artificial intelligence methods. IMS enables fast, sensitive, on-site collection of spectrometric data, which can be immediately analyzed using machine learning methods. In this study, an Advanced IMS (AIMS) instrument was employed to gather ink data, and several machine learning classifiers (including Extra Trees, Gradient Boosting, Random Forest, and Support Vector Classifier) were applied, some of which were integrated into an ensemble voting classifier for enhanced performance. The method demonstrated a high accuracy of 95,23% in classifying inks and provided promising results in estimating the age of written text without compromising the integrity of the sample. These findings highlight the potential applications of the proposed method in forensic investigations, offering a portable, efficient, and non-destructive solution for ink analysis and ageing.

Background: The detection of explosives in crime scene investigations is critical for forensic science. This study explores the application of laser desorption (LD) ion mobility spectrometry (IMS) as a novel method for this purpose utilising a new IMS prototype developed by MaSaTECH. Methods: The LD sampling technique employs a laser diode module to vaporise explosive traces on surfaces, allowing immediate analysis by IMS without sample preparation. Chemometric approaches, including multivariate data analysis, were utilised for data processing and interpretation, including pre-processing of raw IMS plasmagrams and various pattern recognition techniques, such as linear discriminant analysis (LDA) and support vector machines (SVMs). Results: The IMS prototype was validated through experiments with pure explosives (TNT, RDX, PETN) and explosive products (SEMTEX 1A, C4) on different materials. The study found that the pre-processing method significantly impacts classification accuracy, with the PCA-LDA model demonstrating the best performance for real-world applications. Conclusions: The LD-IMS prototype, coupled with effective chemometric techniques, presents a promising methodology for the detection of explosives in forensic investigations, enhancing the reliability of field applications.

A highly efficient and sensitive ion mobility spectrometry (IMS) system with laser desorption sampling was applied for rapid explosive detection using different surface materials. This portable IMS detector, powered by a battery, offers mobility and is suitable for use in the field or combat zones. The laser desorption (LD) sampling of common explosives (Trinitrotoluene—TNT; Dinitrotoluenes—DNTs; Hexogene—RDX; pentaerythritol tetranitrate—PETN; plastic explosives—Compound 4 (C-4) and Semtex) on a wide range of common surface materials, such as metal, ceramic, plastic, glass, drywall, paper, wood, and textiles, was studied. Successful detection was achieved on nearly all surfaces except flammable materials (paper, wood, and textiles). The limit of detection (LOD) was determined for each explosive and specific surface, demonstrating an impressive LOD of 7 ng/mm² for TNT. RDX, C-4, PETN, and Semtex achieved LOD values of 15 ng/mm², while DNTs showed an LOD of approximately 50 ng/mm².

Hadamard multiplexing can increase ion throughput in ion mobility spectrometry (IMS) by encoding the ion gate with a pseudorandom modulation sequence and reconstructing the underlying signal by deconvolution. To our knowledge, no publicly available IMS dataset currently provides paired conventional and Hadamard-multiplexed drift tube measurements together with the full information required for reproducible reconstruction. Here we present an open drift tube IMS dataset for six ketones (2-butanone, 3-pentanone, 2-hexanone, 3-hexanone, 2-heptanone and 2-octanone) together with baseline measurements acquired without analyte. For each compound and acquisition mode, five independent 90 s measurements were recorded, yielding time-ordered sequences of individual spectra. The dataset includes conventional single-pulse IMS alongside Hadamard-multiplexed IMS acquired using a 12-bit modulation sequence under two averaging configurations. Spectra are provided on a common grid of 4095 points with drift times in milliseconds. Multiplexed records include raw spectra and deconvoluted/normalized Hadamard spectra (HT, HTi) and reconstructed inverse Hadamard spectra (IHT). The modulation sequence and deconvolution arrays are provided to enable reproducible reconstruction and downstream method development.