Screening of volatiles from explosive initiators and plastic‐bonded explosives (PBX) using headspace solid‐phase microextraction coupled with gas chromatography – mass spectrometry (SPME/GC–MS)

The detection of explosives and explosive devices based on the volatile compounds they emit is a long‐standing tool for law enforcement and physical security. Toward that end, solid‐phase microextraction (SPME) combined with gas chromatography–mass spectrometry (GC–MS) has become a crucial analytical tool for the identification of volatiles emitted by explosives. Previous SPME studies have identified many volatile compounds emitted by common explosive formulations that serve as the main charge in explosive devices. However, limited research has been conducted on initiators like fuses, detonating cords, and boosters. In this study, a variety of SPME fiber coatings (i.e., polydimethylsiloxane (PDMS), polydimethylsiloxane/divinylbenzene (PDMS/DVB), divinylbenzene/carboxin/polydimethylsiloxane (DVB/CAR/PDMS), and carboxin/polydimethylsiloxane (CAR/PDMS)) were employed for the extraction and analysis of volatiles from Composition C‐4 (cyclohexanone, 2‐ethyl‐1‐hexanol, and 2,3‐dimethyl‐2,3‐dinitrobutane (DMNB)) and Red Dot double‐base smokeless powder (nitroglycerine, phenylamine). The results revealed that a PDMS/DVB fiber was optimal. Then, an assortment of explosive items (i.e., detonation cord, safety fuse, slip‐on booster, and shape charge) were analyzed with a PDMS/DVB fiber. A variety of volatile compounds were identified, including plasticizers (tributyl acetyl citrate, N‐butylbenzenesulfonamide), taggants (DMNB), and degradation products (2‐ethyl‐1‐hexanol).

Identifying the various volatile compounds emitted from explosives through headspace solid-phase microextraction (SPME) and gas chromatography -mass spectrometry (GC-MS) has revealed hundreds of compounds that could be viable targets for either instrumental analysis or canine detection.Previous research in this area has examined common explosive formulations such as 2,4,6-Trinitrotolunene (TNT), RDX (i.e., Composition C-4), and Pentaerythritol Tetranitrate (PETN) [1][2][3].Many of these formulations emit volatile compounds that are degradation products, taggants, or plasticizers, such as 2-ethyl-1-hexanol, dibutyl phthalate, 2,3-dimethyl-2,3-dinitrobutane (DMNB), dioctyl adipate, and tributyl acetyl citrate [1][2][3].In contrast to explosives used as main charges, explosives used as initiators (such as denotating cords and safety fuses) are often encased in materials such as manufactured textile tubing and/or layers of asphalt.These materials are the area of focus for this paper as they have not been extensively studied.
A summary of the volatile compounds identified in previous SPME studies are summarized in Table S1.This table provides a comprehensive list of common explosives, the detected volatile compounds via SPME/GC-MS associated with each explosive, and the pertinent references documenting the identification of individual volatiles.Every volatile compound detected by SPME/GC-MS within the references has been incorporated in the table.However, it is noteworthy that certain volatiles, while detected, manifest as sporadic occurrences, or deviate from anticipated observations.For example, 2-ethyl-1-hexanol (a degradation product of plasticizers) is not expected to be found in Ammonium Nitrate plus calcium prills based on SPME/GC-MS analysis.Although it was reported by DeGreeff, Peranich, and Simon in their studies on the detection of ammonium nitrate variants, it is important to note that the presence of 2-ethyl-1-hexanol could be attributed to plasticizers or contamination [1].
Detection of initiator and booster materials is crucial as they are not a common focus of analysis, as highlighted by Table S1 where only five of the 48 references analyze these materials.Performing distinct measurements on initiator materials, separate from characterizing the main charge, is a critical endeavor to comprehend explosive devices.Initiator materials often have significantly less explosive material compared to the main charge, making them challenging to detect due to their lower concentration of characteristic explosive volatile compounds.This lower concentration can result in a subtler olfactory signature, necessitating more sensitive instrumentation methodologies.Moreover, initiator materials can encompass a diverse array of components beyond the explosive core.For instance, some initiators are composed of textiles, asphalt, or plastics making them challenging to detect due to the potentially overpowering scent of the additional components.Analyzing the specific volatiles within initiators is instrumental in streamlining explosive detection techniques, enabling accurate identification and differentiation from the primary charge.Additionally, this information is indispensable in forensic investigations, contributing to robust investigative procedures.

| Materials
There are many potential explosive boosters, initiators, and detonating cords that could be included in this study.Therefore, we set out to assemble a collection that is not exhaustive but representative.The samples selected and reasonings for selection are summarized in Table 1.These samples were either purchased from Omni Explosives or received as exemplars from local law enforcement.Examples of the boosters, initiators, and detonating cords we used were bubble gum boosters, cast booster, PETN detonating cord, hexahy-dro-1,3,5-trinitro-1,3,5-triazine (RDX) detonating cord, shape charge, and Safety Fuse.The explosive boosters were kept in one Type II Explosive Magazine with other explosive materials including SEMTEX 10 and SEMTEX 1H.The bubble gum boosters were kept in separate glass vials with a screw on lid that was sealed with parafilm after each time being opened.The cast booster, SEMTEX 10, and SEMTEX 1H were stored in separate Ziplock plastic bags that were then stored together in a larger Ziplock plastic bag.
• Several volatile compounds not previously observed in explosive formulations were identified.

TA B L E 1
Explosive materials selected to be analyzed.Table 1 presents the chosen explosive materials for analysis, categorized based on the rationale behind their selection.

| SPME fiber study
Composition C4 and Red Dot double-base smokeless powder were run in triplicate with each SPME fiber.The same GC-MS parameters described in Section 2.2 were followed with an additional fiber bakeout of 5 min at 220°C with optimal operating temperatures between 200°C and 280°C for all fibers.Three air blanks were run between each sample.

| SPME fiber study
Overall, the PDMS SPME fiber demonstrated the worst extraction efficiency as seen in Figure 1.PDMS is recommended to be used with volatile compounds [4].However, explosive volatiles are generally polar compounds, whereas PDMS is a nonpolar phase appropriate for the analysis of nonpolar and semipolar compounds [5].Four of the five selected volatiles analyzed are polar compounds and would be expected to have weaker desorption.2-ethyl-1-hexanol is a semipolar compound that would be expected to have a higher affinity to the PDMS fiber.This could be due to the long extraction time as it has been demonstrated that PDMS has less extraction efficiency during longer extractions when compared to other coatings [6,7].
PDMS is also a film fiber which has been shown to have less absorption compared to particle-based fibers [8].One study has shown that the PDMS fiber was better for the detection of 2-ethyl-1-hexanol compared to PDMS/DVB and DVB/CAR/PDMS [9].The observed phenomenon can be attributed to the incubation temperature of 240 degrees Celsius, enhancing the concentration in the headspace and facilitating improved extraction.
Polyacrylate was less successful in its extraction of the semi-polar compounds, cyclohexanone and 2-ethyl-1-hexanol, as it is a polar polymer allowing it to interact more readily with polar compounds.Which can be seen in the extractions of DMNB, nitroglycerin, and diphenylamine in Figure 1.However, analytes featuring high branching or aromaticity, along with substitution groups, such as nitroglycerin and diphenylamine, can exhibit diminished interactions with adsorbents [4].Notably, high electronegativity within substitution groups can further diminish this interaction, as illustrated in Figure 1.
Coatings containing DVB proved effective in extracting the polar compounds, DMNB, nitroglycerin, and diphenylamine through π-π interactions due to the presence of an aromatic ring and vinyl groups.This is reinforced as the use of PDMS/DVB is recommended for detecting volatile nitro-aromatics and amines [4].A study has shown that PDMS/DVB demonstrated the most favorable recovery for the 14 studied explosives, with outcomes varying depending on the specific explosive being extracted [5].
Carboxin is an adsorbent material that has a high affinity for certain volatile organic compounds.It contains carboxylic acid functional groups (-COOH) that are highly polar and can interact with other polar compounds through hydrogen bonding and dipole-dipole interactions.
Both cyclohexanone and 2-ethyl-1-hexanol had similar extraction efficiencies to all the fibers, except PDMS.Both are highly volatile and semi-polar, creating an abundance in the headspace to be extracted by most fibers [6].Although nitroglycerin and diphenylamine are polar compounds, highly branched or aromatic analytes with substitution groups may have reduced interaction with adsorbents, and their effective size can be smaller compared to an all-hydrocarbon structure of the same molecular weight [4].If the substitution group has high electronegativity, it further reduces the interaction with adsorbents.This can be seen in Figure 1 where nitroglycerin and diphenylamine are not extracted as efficiently with the polyacrylate fiber.bonded explosives [10].Previous studies have demonstrated its presence in the headspace of some polymer bonded explosives such as SEMTEX and Detasheet [11].It has also been seen in nail polishes and cosmetics [12].

| Explosive boosters
In addition to examining the boosters, the study also analyzed Semtex.The study incorporated Semtex that was stored in the same magazine as the boosters, to investigate the potential occurrence of contamination in the booster samples.The headspace SPME results for SEMTEX 1H and SEMTEX 10, shown in Figure 3, were shown not to contain tributyl acetyl citrate, but dibutyl phthalate.The taggant DMNB was also identified in both samples.

| Initiators
The SPME headspace results for a shape charge, shown in Figure 4, exhibited the common taggant DMNB and a degradation product of a plasticizer, 2-ethylhexyl ester acetic acid.2-ethylhexyl ester acetic acid has been detected in Detasheet and double-base smokeless powders [13].
Safety fuse was also analyzed and confirmed in a previous study, with the most prominent volatile being sulfur, which is a component of the black powder filling in the fuse [14] (Figure 5).

| Detonating cords
In the case of detonating cords, SPME can be used to detect and quantitate the various components of the cord, such as the explosive material, the binder, and any additional additives or impurities.
Due to the plastic casings of detonating cords, it is expected that plasticizers will be seen in the headspace.Explosive plasticizers are chemical compounds that are added to explosive materials to make them more flexible and moldable.This allows the explosive to be shaped into various forms, such as slabs, pellets, or even cord, which can be useful for specific applications.Plasticizers work by decreasing the viscosity of the explosive material, making it easier to work with.They can also help to improve the sensitivity and stability of the explosive and can be used to increase the density or reduce the sensitivity to shock and impact.Common explosive plasticizers include dioctyl sebacate, dioctyl adipate, di-n-octyl phthalate, and tri-n-butyl citrate [6].These compounds are often used in combination with other ingredients, such as binders and stabilizers, to create the desired properties in the explosive.
Different companies create unique mixtures of these additives which affect the volatiles in the headspace.In Figure 6, the headspace of sample B has a peak labeled as DMNB which is not an odor found in PETN detonating cords.It is also at a low abundance,   and N-butylbenzenesulfonamide. TBP is an organophosphorus compound that is primarily used as a solvent and plasticizer.TBP is a highly effective solvent for a wide range of organic compounds, particularly for polar and ionic compounds, and it is also used as a solvent for the extraction of nuclear fuel from spent fuel rods, in the enrichment of uranium.In addition to its use as a solvent, TBP is also used as a plasticizer for a variety of materials, including polyvinyl chloride (PVC), nitrocellulose, and cellulose acetate.
TBP could be used as a plasticizer in the manufacturing of certain types of explosives.Like other plasticizers added to the explosive mixture it would increase its flexibility and reduce its viscosity, making it easier to shape and mold.It would also help to improve the sensitivity and stability of the explosive and reduce the sensitivity to shock and impact.TBP has been detected on the surface of plastic bonded explosives, such as RDX, by direct analysis in real time mass spectrometry (DART-MS) [16].These types of explosives are made by bonding a high explosive material, such as RDX or HMX, to a plastic binder, such as TBP or a mixture of TBP and other plasticizers.
There is no published literature regarding TBP being detected in the headspace of plastic bonded explosives via SPME.
The other compound detected in the headspace of Figure 6; sample A was N-butylbenzenesulfonamide. N-butylbenzenesulfonamide is an organic liquid compound that is used as an intermediate in the synthesis of other chemicals, particularly dyes and pigments.
It is also used as an intermediate to produce pesticides and drugs.
Additionally, it can be used as an additive in lubricants, plastic, and rubber.The most common use of N-butylbenzenesulfonamide as a plasticizer is in polyamides such as nylon [17].Some detonating cords use nylon fibers in their core [18].This has never been detected in headspace analysis of detonating cords.

An
Agilent 6890 GC coupled to an Agilent 5975 Mass Selective Detector with an attached Gerstel multipurpose sampler was used for all experiments.The GC column was an Agilent Technologies HP-5MS Ultra Inert column with a length of 30 m, a 0.250 mm inner diameter, and a 0.25μm-film thickness.Tapered inlet liners of 2.0 mm inner diameter from Restek and 65 um polydimethylsiloxane/divinylbenzene StableFlex/ss from Supelco were used for SPME analyses.The incubation temperature was 60°C and the incubation time was 2 min.It is noted that this incubation temperature exceeds the temperatures at which canines would typically encounter these explosives.The sample extraction and desorption times were 20 min and 60 s, respectively.The inlet temperature was set to 220°C and was operated with a 20:1 split ratio.The initial oven temperature was 60°C and was held for 1 min, then the temperature was ramped at 20°C/min to 300°C where it was held for 1 min.The transfer line was set to 250°C.The quadrupoles were kept at 150°C.A scan range of m/z 40 -m/z 400 was used, with no solvent delay.Approximately, 1 gram of each sample was analyzed in 20 mL headspace vials.The sample mass ranged from 0.9391 to 1.10479 grams.
Bar graph representing the peak area of cyclohexanone, 2-ethyl-1-hexanol, and 2,3-dimethyl-2,3-dinitrobutane in C4 samples for each SPME fiber used with error bars based on the standard deviation of the peak areas.(B) Bar graph representing the peak area of nitroglycerin, and diphenylamine in double-base smokeless powder for each SPME fiber used with error bars based on the standard deviation of the peak areas.
Our collection of explosive materials was divided into three groups: boosters, detonating cords, and initiators.The bubble gum boosters and cast booster were analyzed by SPME-GC-MS; their headspace SPME results are shown in Figure 2A,B.Our analysis of a cast booster indicates the presence of TNT combined with other nitroaromatics (Figure 2C) [7].Sample C contained 2,4,6-trinitrotoluene (TNT) and 2,4-dinitrotoluene (DNT), indicators of the parent explosive TNT.2,3-dimethyl-2,3-dinitrobutane (DMNB) was determined to be a contaminant.No explosive components were detected in two of the samples but tributyl acetyl citrate was detected.Tributyl acetyl citrate (also known by the trade name Citraflex) is commonly utilized to substitute for phthalate plasticizers.Alkyl esters, including tributyl citrate, dioctylsebacate, bis(2-ethylhexyl)adipate, and bis(2ethylhexyl)sebacate are commonly used as plasticizers in plastic

F I G U R E 2
Chromatograms of SPME GC-MS analysis of the following explosives: (A) Bubble gum booster; (B) Bubble gum booster red; (C) Cast booster.Peaks labeled with an asterix (*) are cyclic siloxane peaks seen in the blanks.F I G U R E 3 Chromatograms of SPME-GC-MS analysis of the following explosives: (A) SEMTEX 1H; (B) SEMTEX 10.F I G U R E 4 Chromatogram of SPME-GC-MS analysis of shape charge.Peaks labeled with an asterix (*) are cyclic siloxane peaks seen in the blanks.

FIGURE 5
FIGURE 5 Chromatogram of SPME GC-MS analysis of safety fuse.Peaks labeled with an asterix (*) are cyclic siloxane peaks seen in the blanks.

Figure 6
Figure 6 indicates two non-explosive compounds not previously seen in the headspace of RDX; Tributyl phosphate (TBP) SPME/GC-MS is an important analytical tool for identifying volatile compounds emitted from explosive formulations.To date, previous SPME studies have focused on explosives found in the main charge of an explosive device.Our research has aimed to identify volatile compounds in explosives that have not been extensively studied and investigate the impact of different SPME fiber coatings on the detection of these compounds.In general, expanding our understanding of the expansive array of explosive volatiles and their diverse chemical profiles can significantly advance explosives detection.Future research should strive to incorporate real-life packaging materials and conditions to simulate practical explosive detection scenarios.By creating an extensive guide cataloging these compounds, and their potential variations across different explosive types and packaging materials, we can develop a comprehensive resource for the explosives community.For example, this guide could serve as a reference for tailoring instrumental or canine training programs.