Classification of Gunshot Residue Using Laser Electrospray Mass Spectrometry and Offline Multivariate Statistical Analysis

Abstract

Nonresonant laser vaporization combined with high-resolution electrospray time-of-flight mass spectrometry enables analysis of a casing after discharge of a firearm revealing organic signature molecules including methyl centralite (MC), diphenylamine (DPA), N-nitrosodiphenylamine (N-NO-DPA), 4-nitrodiphenylamine (4-NDPA), a DPA adduct, and multiple unidentified features not observed in previous mass spectral measurements. Collision-induced dissociation measurements of unknown GSR signature ions reveals inorganic barium and derivatives BaOH, BaOHCH, BaCHCOO remaining from the primer. Both hydrophilic and hydrophobic signatures are detected using water-methanol electrospray solution. Offline principal component analysis and discrimination of the laser electrospray mass spectral (LEMS) measurements resulted in perfect classification of the gun shot residue with respect to the manufacturer. Principal component analysis of recycled and reloaded casings resulted in classification of the penultimate manufacturer with an accuracy of 89%.

Applications of Direct Injection Soft Chemical Ionisation-Mass Spectrometry for the Detection of Pre-blast Smokeless Powder Organic Additives

“…As stated in the introduction, owing to the complex composition of smokeless powders, the presence of unidentified peaks was expected, the majority coming from the plasticizers used in the manufacturing process. This was confirmed by additional unidentified peaks for all the powders at m/z 149.02 (reported by Scherperel et al as a dibutyl phthalate fragment) [29], 205.09 and 279.16 (the latter is assigned to be protonated dibutyl phthalate [DBP.H] + by Reese et al [8] and Perez et al [15]). It is evident that a more detailed study dealing with these is needed if a complete chemical analysis is required.…”

“…[2,[9][10][11][12][13] The aim of these additives is to increase the shelf-life and modify the burning characteristics of the powder [5,14]. Different concentrations and/ or different additives are characteristic of a given manufacturer, producing therefore a chemical fingerprint for each powder [15]. It is thus also important to determine their content throughout the manufacturing quality control process.…”

“…Several analytical techniques have been used for the qualitative and/or quantitative detection of smokeless powders, either in their pre-and/or post-blast forms [9,10], including highperformance liquid chromatography (HPLC) [17][18][19], liquid chromatography-mass spectrometry (LC-MS) [8,[20][21][22], Fourier transform infrared spectroscopy [23], gas chromatography (GC) [12,14,24], capillary electrophoresis (CE) [25,26], ion m o b i l i t y s p e c t r o m e t r y ( I M S ) [ 2 7 ] , s o l i d -p h a s e microextraction-ion mobility spectrometry (SPME)-IMS [12,28], (nano)electrospray ionization (nESI)-tandem mass spectrometry [29][30][31], laser electrospray-mass spectrometry (LEMS) [15,32], desorption electrospray ionization-mass spectrometry (DESI) [33,34], direct analysis in real timemass spectrometry (DART-MS) [35], time-of-flight secondary ion-mass spectrometry (ToF-MS) [36] and Raman spectroscopy [23,37]. Most of the abovementioned techniques require time-consuming sample preparation step(s)-exception of DESI and DART-or if not, they require complicated setups, such as the use of lasers as the means for sample vaporization (LEMS) or heated purified gases (DART).…”

Toward Locard’s Exchange Principle: Recent Developments in Forensic Trace Evidence Analysis

“…Meanwhile, Perez et al. used laser electrospray MS to analyze the residues found inside spent cartridge casings . They were able to achieve high classification rates for the ammunition manufacturers investigated by means of PCA.…”

Consensus Classification Using Non-Optimized Classifiers

“…Classifying a sample to particular category is common in analytical chemistry. − The meaning to classification varies between fields. Classification in this paper is the problem of determining which class (category) a sample belongs to regardless of the type of classifier (traditional generative and discriminative groupings or other approaches not fitting this binary grouping paradigm).…”