Review: The Coupling of Capillary Microextraction of Volatiles (CMV) Dynamic Air Sampling Device with DART-MS Analysis for the Detection of Gunshot Residues

Emily C. Lennert





smokeless powder, gunshot residue, headspace, volatile, capillary microextraction of volatiles, CMV, direct analysis in real time – mass spectrometry, DART-MS

Article Reviewed

Williamson, R.; Gura, S.; Tarifa, A.; Almirall, J. R. The coupling of capillary microextraction of volatiles (CMV) dynamic air sampling device with DART-MS analysis for the detection of gunshot residues. Forensic Chemistry. 2018, 8, 49-56.


The opinions expressed in this review are an interpretation of the research presented in the article. These opinions are those of the summation author and do not necessarily represent the position of the University of Central Florida or of the authors of the original article.


Smokeless powders consist of a nitrocellulose base charge, i.e. single base powder, with the addition of nitroglycerin in double base powders. In addition to the base charge, smokeless powders contain a number of additives including stabilizers, plasticizers, flash inhibitors, and others which serve to stabilize and lengthen the shelf life of smokeless powders. These components are classified as volatile organic components (VOC) and semi volatile organic components (SVOC). VOCs and SVOCs from smokeless powders and gunshot residue (GSR) are analyzed several ways, including by headspace extraction sampling such as solid phase microextraction (SPME) and capillary microextraction of volatiles (CMV) followed by analysis with gas chromatography – mass spectrometry (GC-MS) or ion mobility spectrometry (IMS). This study presents the coupling of CMV, traditionally paired with GC-MS, to direct analysis in real time – mass spectrometry (DART-MS) for a rapid analytical technique allowing for the trace detection of VOCs and SVOCs from smokeless powders.

CMV has previously been reported for the sampling of smokeless powder headspace as well as GSR from shooters’ hands. The CMV is inserted into a thermal separation probe (TSP), which is at the injection port of the GC-MS. A CMV sampler is composed of glass microfibers, coated with a thin sol-gel polydimethylsiloxane (PDMS) film, which are packed into a 2 mm by 2 cm glass capillary. This allows for high surface area for analyte adsorption as well as dynamic sampling, i.e. sampling in which the headspace is actively drawn through the capillary by an air pump. Thus far, DART-MS has been limited in its application to smokeless powder analysis, and to date, according to the authors, CMV has not been coupled to DART-MS. Therefore, the authors coupled CMV to DART-MS for the analysis of VOCs and SVOCs from smokeless powders. Additionally, the authors followed DART-MS analysis of CMVs with GC-MS analysis to provide complimentary analysis.

A CMV apparatus was developed for DART-MS sampling in this study. This apparatus for introducing the CMV into the DART ion stream was designed to allow for introduction in a reproducible and fixed manner. The setup of the CMV introduction apparatus is depicted in Figure 1 within the study. For GC-MS, CMV samplers were introduced in a TSP. For sample generation, a vapor source was designed as depicted in Figure 2 within the study. Samples were injected manually into a preheated injection port, where the sample would volatilize and travel through a short column into a disposable plastic pipette. A CMV device was connected at the bottom of the pipette, and was also connected to an air pump, which drew the sample through the CMV device to allow for adsorption.

Standard solutions were prepared at 500 ng/μL concentrations for initial analysis by DART-MS. Samples were spiked onto a glass capillary tube and inserted directly into the DART ionization stream for identification of the molecular weight, adducts, and fragments for each compound. Standard solutions prepared at 300 ng/μL were prepared for CMV-DART-MS method development, and 15 ng/μL for performance evaluation. Method development included optimized DART temperatures for both positive and negative mode, distance between the DART and CMV device, and the dimensions of the ceramic tube. Performance evaluation included the determination of limits of detection in positive and negative modes. Finally, to determine the amount of sample being adsorbed onto the CMV devices, correlated measurements were performed. CMV samples were prepared using the vapor source and analyzed directly by GC-MS. These samples were compared to CMV samples spiked directly with the corresponding sample, which were analyzed by GC-MS.

Development of the CMV-DART-MS apparatus is described in section 3.1.1 of the study. Several parameters were evaluated. A ceramic tube diameter of 3.18 mm was determined to provide for better detection of target compounds. The positioning of the CMV within the ceramic tube was also evaluated, i.e. the depth, and the highest sensitivity was reported when the CMV was completely within the tube. However, for practical purposes, i.e. removal, the CMV was positioned such that a few mm remained outside of the tube. Shortened CMV devices were evaluated, down to 1 cm long, and better desorption was observed with the shorter devices. However, to allow for complimentary analysis by GC-MS, the original 2 cm CMV device was selected. Distance between the DART ion source and ceramic interface, i.e. CMV device, was evaluated. A distance of 3 cm was selected to allow for the removal and insertion of CMV devices without the need to reposition the ion source. Temperatures were evaluated in positive and negative modes between 200 and 300 ˚C. In negative mode, 200 ˚C was considered optimal, whereas in positive mode 250 ˚C was optimal. This temperature difference allowed for a temperature gradient, during which the ionization mode was switched, to allow for positive and negative mode analysis in a single screening. Duration of desorption was evaluated at 10 seconds and 1 min, with 1 min providing higher sensitivity. Finally, CMV devices were compared to sampling directly from a capillary tube. A loss in sensitivity was reported for CMV devices, and was attributed to incomplete desorption of analytes from the CMV device.

Development of the vapor source for sample generation is described in section 3.1.2 within the study. Injector temperature was optimal at 210 ˚C. This temperature prevented loss of nitroglycerin due to degradation, while providing the highest evaporation efficiency. High gas flows were required to ensure delivery of the vapor to the CMV, and a flow of 150 mL/min was selected. 5 min sampling time was considered optimum.

Compounds were analyzed three ways: direct sampling from glass capillary tubes, direct spiking onto the CMV device, and vapor sampling onto the CMV device. Direct sampling onto glass capillaries allowed for characterization of each compound. Spiked CMV devices were then compared to the glass capillaries. Vapor sampled CMV devices were analyzed by DART-MS for method evaluation, as well as to estimate the sensitivity of the method.

To determine the amount of compounds adsorbed onto the CMV device using the vapor source, samples were prepared using the vapor source and compared to results obtained from spiked CMV devices, analyzed by CMV-GC-MS. Table 2 within the study displays the absolute amounts of compound sampled onto the CMV. Samples prepared from the low concentration standard showed the adsorbance of 1-6 ng of compound, with the exception of diethyl phthalate, which was later reported to have originated in the sampling system. Recoveries of approximately 20% up to 40% were reported for SVOCs. 4-nitrodiphenylamine, a less volatile compound, showed lower recoveries of 10% or less. 4-nitrosodiphenylamine was only detected by GC-MS at the higher concentration sampled, 300 ng/μL. However, even at the lower concentration of 15 ng/μL, this compound is detected by CMV-DART-MS. At 15 ng/μL sampling, recoveries of other low volatility compounds were low as well: dibutyl phthalate and 2-nitrodiphenylamine had a 13% recovery rate. Diethyl phthalate was detected in the blank samples, likely originating from the plastic pipette used in the sampling device, resulting in poor repeatability for its measurement in the samples. However, other compounds showed precisions of 15-39% at high concentrations and 9-32% at low concentrations.
All target compounds were successfully detected by CMV-DART-MS in positive and negative mode. After analysis, CMV samples were wrapped in foil and allowed to sit t room temperature for 18 hours prior to CMV-GC-MS analysis. Subsequent, reliable identification of compounds was reported by CMV-GC-MS following CMV-DART-MS analysis and aging for 18 hours.

Scientific Highlights

  • This study presented a novel pairing of CMV with DART-MS, and demonstrated the CMV-DART-MS technique for the analysis of smokeless powder compounds.
  • CMV-DART-MS allowed for rapid detection of compounds and allowed for subsequent analysis by CMV-GC-MS.
  • Results obtained by CMV-DART-MS were confirmed by CMV-GC-MS analysis, even after an 18 hour aging period.


Rapid and reliable techniques are desirable, which may help to prevent or reduce backlogs in laboratories.

Potential Conclusions

CMV-DART-MS may be a rapid and reliable option for the analysis of smokeless powder constituents.