The Trace Vapor Generator for Explosives and Narcotics (TV-Gen) is a portable and compact instrument designed to deliver a continuous source of trace-level vapors and vapor mixtures. It provides a tool to assist in the independent validation and verification of new materials and sensors under development for the vapor detection of explosives and narcotics. The design was conceived for use with a broad range of analytes, detection systems, materials, and sensors and to switch easily between the clean and analyte vapor streams. The TV-Gen system utilizes nebulization of aqueous analyte solutions, an oven to promote efficient transport, and a control box that provides dedicated computer control with logging capabilities. Resultant vapor streams are stable over several hours, with the vapor concentration controlled by a combination of aqueous analyte solution concentration, liquid flow rate through the nebulizer, and volume flow rate of air through the TV-Gen manifold.

Trace vapor detection of explosives and narcotics allows for non-contact sampling that has the potential to be less invasive, relieves logistic and safety concerns associated with direct contact sampling, and has the potential for improved detection of wrapped or obscured materials.1 These inherent benefits have been the impetus for the development of trace vapor detection capabilities, and for this purpose, it is important that reliable methods for the evaluation of new and emerging detectors and sensing materials/technologies be available.

The greatest challenge faced in the vapor generation of explosives and narcotics is their exceedingly low vapor pressures, often requiring vapor concentrations as low as the parts-per-quadrillion level in air (ppq by volume; femtomol of analyte/mol of air) for realistic detection scenarios.2 Additionally, many explosive analyte molecules exhibit strong interactions with many surfaces, making them appear to be “sticky.” For this reason, all surfaces that have contact with analyte vapor must be maintained at an elevated temperature in order to prevent any loss of analyte in cold spots encountered along the vapor path.3,4 Generally, an ideal vapor generation system should allow for the reproducible and continuous delivery of analytes (i.e., no pulsing effect) of varying vapor pressures in a vapor stream free of organic solvents. The concentration of the analyte stream should be adjustable across a large range of vapor concentrations, and the pathway from the analyte source to the detector should be passivated and free of cold spots to minimize attenuation of the analyte vapor.

The most common method of vapor generation is direct sampling from a bulk energetic/static source.5 In this approach, reproducibility and stability of the vapor stream concentration can be challenging and is not easily amenable to varying concentration levels. Of course, researchers must consider safety/regulatory limitations for handling bulk explosives as well as when implementing this mode of vapor generation. Alternatively, an analyte material may be deposited onto a solid substrate, such as glass beads.6 A flow of heated air is then swept across the substrate, carrying analyte vapor to the detector. In this method, the hazard of working with a bulk material is reduced, and there is a greater surface area for vaporization. Both of these generation techniques suffer from changing concentration due to a depletion of analyte vapor over time. Alternatively, in instances where the presence of solvent is not an issue, vaporization from analytical standards can be utilized.7 It is worth noting that some of these vapor generation technologies have been commercialized by companies including, but not limited to, Kintek and Microfab.

In order to address some of the operational limitations and concerns outlined above, the Naval Research Laboratory (NRL) developed two technologies: the Pneumatically Modulated Liquid Delivery System (PMLDS) coupled to a perfluoroalkoxy (PFA) total-consumption microflow nebulizer8 and the Trace Explosives Sensor Testbed (TESTbed)2 for the purposes of generating and delivering trace vapors of explosives to detectors/sensors/materials. The PMLDS-nebulizer introduces a pulse free aerosol of trace explosives in aqueous solution into a heated tube, causing the water to evaporate resulting in the generation of a trace vapor of explosives in humid air. With the PMLDS-nebulizer as one of several vapor sources, the TESTbed incorporates a dual-line manifold system allowing for rapid switching between the clean and analyte vapor streams at relative humidity levels between 20% and 85% and the capability to deliver trace vapor streams (<100 ppqv)9 in the presence of interferents for improved testing under more realistic situations.

The Trace Vapor Generator for Explosives and Narcotics (TV-Gen) utilizes two separate but identical PLMDS-nebulizers for the analyte and clean manifold sides, thus eliminating the need for multiple clean air generation systems and no longer requiring humidity matching. Additionally, the manifold into which the nebulizer sprays the micro-droplets had its diameter increased (relative to the TESTbed) to allow for the nebulizer to spray directly into the center of the tube. Additionally, carrier gas is introduced coaxially around the tip of the nebulizer. These modifications eliminate the need for the vortex mixer2 and allow the micro-droplets’ additional time to interact with the carrier gas. Detailed TV-Gen design elements and validation of the TV-Gen unit are described herein.

The TV-Gen was designed to deliver trace vapor concentrations continuously through a single sample port. It consists of three main components: the manifold, the oven, and the instrument control box. Briefly, aqueous solutions of explosives and narcotics are introduced from the control box, whereupon it is nebulized and sent through the manifold to a vapor output port protruding from the oven, as illustrated in Fig. 1. The manifold is contained within the oven and has two isolated flow paths for the clean and analyte vapor streams. The control box contains all the hardware necessary for the operation of two nebulizers: one for the clean side of the manifold, and one for the analyte side. The design of each component will be detailed in Secs. II B and II C, as will the analytical evaluations and quantitation of resultant vapor streams.

FIG. 1.

Diagram illustrating the air and liquid flow paths allowing for the generation of trace vapors using the TV-Gen. The control box shows the computer control system, four mass flow controllers (MFCs) for air flow control, and two liquid flow meters. The oven section shows the nebulizer interface, dual manifold system, and exit port.

FIG. 1.

Diagram illustrating the air and liquid flow paths allowing for the generation of trace vapors using the TV-Gen. The control box shows the computer control system, four mass flow controllers (MFCs) for air flow control, and two liquid flow meters. The oven section shows the nebulizer interface, dual manifold system, and exit port.

Close modal

An illustration of the dual TV-Gen manifold is presented in Fig. 2. The manifold can be classified into three distinct sections from left to right in Fig. 2: the back wall of the manifold, which also serves as a removable oven wall, the pair of vapor generation cylinders, and the four-way crossover valve that connects the cylinders and allows for the vapor output at the front of the TV-Gen.

FIG. 2.

Illustration of the dual manifold housed within the TV-Gen oven.

FIG. 2.

Illustration of the dual manifold housed within the TV-Gen oven.

Close modal

The pair of vapor generation cylinders is 10 in. long, with a 1.37 in. diameter. The cylinder and all surfaces that will come into contact with the analyte are SilcoNert 2000-coated (Silcotek). The two vapor generation cylinders make up the manifold, meeting at a Swagelok four-port two-way crossover valve. This valve directs vapors from each cylinder to either the exhaust line (the connection between the bypass line and the exhaust port is not illustrated in Fig. 2 for clarity) or the vapor output port at the face of the TV-Gen. The vapor output port has a secondary exhaust option that allows for finer control of the vapor output flow rate. The valve is actuated by a drive shaft, which protrudes through the back wall of the manifold (seen in Fig. 2).

The back wall of the manifold incorporates an air balancing system and consists of two halves, each containing channels for the carrier or sheath gas (Fig. 3). The channels allow the carrier gas to be slightly warmed, prior to introduction into the vapor generation cylinders, while keeping the back of the manifold cooler to the touch. The carrier gas air flow enters the vapor generation cylinders from the back wall of the manifold system through eight evenly spaced holes surrounding the inlet port of the nebulizer (Fig. 4).

FIG. 3.

Vapor generation cylinders attached to the back wall of the manifold, showing (left) both halves of the wall and (right) only the inner most portion of the back wall, exposing the sheath air flow path.

FIG. 3.

Vapor generation cylinders attached to the back wall of the manifold, showing (left) both halves of the wall and (right) only the inner most portion of the back wall, exposing the sheath air flow path.

Close modal
FIG. 4.

Diagram of eight carrier flow gas ports organized concentrically around the nebulizer at head of vapor generation cylinders.

FIG. 4.

Diagram of eight carrier flow gas ports organized concentrically around the nebulizer at head of vapor generation cylinders.

Close modal

The nebulizers are crucial to the continuous and reproducible supply of analyte vapor and have been shown to be capable of producing stable vapor streams for a variety of trace explosives (measured vapor concentrations as low as ppqv).2,9 Explosives are nebulized as aqueous solutions at oven set point temperatures from 30 °C to 130 °C, preventing thermal decomposition of thermally liable analytes, such as pentaerythritol tetranitrate (PETN). The nebulizers used in the TV-Gen are the same as those used in the TESTbed and are further described elsewhere.2,9 This method of introducing the nebulized vapor with the surrounding carrier gas improves the vaporization efficiency due to more complete evaporation of aerosolized water droplets from the nebulizer in the wider bore vapor generation cylinder (1.37 in. diameter) when compared to the much smaller 0.375 in. diameter tubing used in the TESTbed nebulizer system. In addition, it should be noted that trace contamination of analyte in the manifold can be mitigated by an oven bakeout at 130 °C with analytical validation of manifold cleanliness.

The back wall of the manifold also serves as the back wall of the TV-Gen oven (Fig. 5), which is housed within custom inner and outer anodized aluminum walls separated by a 2 in. gap filled with ceramic insulation. The inner oven walls are connected to the outer shell by four polyether ether ketone (PEEK) standoffs and an Ultem ring, which seals the inner core up against the back of the manifold.

FIG. 5.

TV-Gen oven with the manifold: (left) external view of the oven; (right) oven with the shell and core wall removed exposing the in-place manifold.

FIG. 5.

TV-Gen oven with the manifold: (left) external view of the oven; (right) oven with the shell and core wall removed exposing the in-place manifold.

Close modal

The control box (Fig. 6) was designed to house the air flow controllers, a proportional–integral–derivative (PID) controller to regulate oven temperature, and a pneumatic switch to actuate the manifold valve, power supplies, liquid flow sensors, electronic pressure controllers, and conical tube reservoirs for the analyte solution and clean solvent (water). Clean compressed air (60 psi–80 psi) for the sheath gas and nebulizer airflows is supplied to the control box via a 1/4 in. Swagelok compression fitting shown in Fig. 6 (left). The control box holds two 50 ml conical tube vials containing the clean and analyte solutions, as can be seen in Fig. 6 (center). Liquid from the vials is carried to the nebulizer via 1/16 in. PEEK tubing.

FIG. 6.

TV-Gen control box and sample introduction: (left) front view of the control box with the user interface, (center) side view of the control box with sample introduction, and (right) back view of control box.

FIG. 6.

TV-Gen control box and sample introduction: (left) front view of the control box with the user interface, (center) side view of the control box with sample introduction, and (right) back view of control box.

Close modal

The TV-Gen is controlled using a touch screen computer running a custom graphical user interface (GUI) designed in-house. The software allows the user to control all vapor generation parameters, including the sheath gas and liquid nebulizer flow rates, in addition to controlling the manifold valve. The program displays the last hour of data from the nebulizer flow rate and sheath flow, while continuously saving all data in daily log files. Upon turning on the TV-Gen control box, the user is presented with a screen that displays all the flow rates for the TV-Gen (Fig. 7). The top left of the control screen displays the liquid flow rate (μl min−1), the nebulizer carrier gas flow rate (ml min−1), and the sheath flow rate (l min−1). Dropdown boxes allow for the entry of the liquid flow rate (30 µl min−1–100 µl min−1), the sheath flow rate (0 l min−1–20 l min−1), and the oven temperature set point (0 °C–130 °C). Both nebulizers, clean and analyte, utilize the same set point values for the liquid and sheath flow rates. Toggle buttons are present to control the status of both the air flow and the nebulizer liquid flow, as well as to switch the output between the two manifold pathways. Both the air flow and nebulizer flow switches are displayed in green if the flows are on and white if turned off. The pathway switch is green if the clean pathway is connected to the output and red if the analyte pathway is connected.

FIG. 7.

TV-Gen control box GUI (screenshot).

FIG. 7.

TV-Gen control box GUI (screenshot).

Close modal

The humidity and flow rate capabilities of the TV-Gen were validated by running a matrix of trials with the nebulizer flow rate set to 20 µl min−1, 30 µl min−1, 40 µl min−1, 50 µl min−1, and 60 µl min−1, and the total airflow rates controlled between 2 l min−1 and 20 l min−1. The oven was heated to 130 °C, and the humidity was measured by attaching a length of PTFE tubing to the Swagelok tee at a distance far enough removed from the sample port that the air is at room temperature. The back exhaust port was capped off, allowing the full flow of the TV-Gen to exit out the front port. The majority of airflow was exhausted to the room, while the sample chamber holding the humidity sensor (Sensirion SHT1, custom electronics package) was set to sample at a flow rate of 500 ml min−1 from the tee. It should be noted that the only source of humidity in the TV-Gen is the water being nebulized; therefore, humidity levels are controlled by the combination of the nebulizer liquid flow rate and the diluent air flow rate.

Absolute humidity levels from 1 g m−3 to 30 g m−3 were achievable depending on the liquid flow rate and total airflow (Fig. 8). Figure 8 includes data from both the left and right sides of the manifold, demonstrating an excellent overlap between the two. The dashed lines represent ideal humidity levels calculated based on the liquid flow rate (g) divided by the total airflow (m3) measured from both sides of the manifold (the time terms in the liquid and air flow rates cancel out). At low total airflow rates, less than 4 l min−1, humidity levels greater than 17 g m−3 were generated at liquid flow rates of 40 ml min−1 or higher, producing condensation at room temperature. Should the TV-Gen be used in the testing of water-sensitive materials, care must be taken when approaching such high humidity levels.

FIG. 8.

Plot of TV-Gen humidity (g/m3) vs nebulizer flow and total airflow rates. Dashed lines give the ideal humidity based on the airflow and liquid spray from the nebulizer, while the measured humidity is plotted for the left (dashed) and right (diamond) sides of the manifold.

FIG. 8.

Plot of TV-Gen humidity (g/m3) vs nebulizer flow and total airflow rates. Dashed lines give the ideal humidity based on the airflow and liquid spray from the nebulizer, while the measured humidity is plotted for the left (dashed) and right (diamond) sides of the manifold.

Close modal

The performance of the TV-Gen to generate continuous and adjustable vapor streams was analytically evaluated for the generation of a number of vapors of explosives, including 2,4,6-trinitrtolune (TNT), hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), PETN, nitromethane (NM), nitroglycerine (NG), ethylene glycol dinitrate (EGDN), and triacetone triperoxide (TATP). Evaluation parameters included the efficiency (experimentally determined vs theoretical vapor concentration as a function of solution concentration) and linear dynamic range (demonstrated using air flow dilution). Finally, the scale and duration of any carryover from the analyte solution to clean water was assessed. All explosive materials, with the exception of the TATP, were purchased in solution as analytical standards from AccuStandard. Aqueous solutions were prepared by starting with the analytical standards, evaporating off the solvent, and reconstituting in water to the appropriate concentration.

All vapor samples, with the exception of TATP, were generated using the microflow nebulizer. Aqueous explosive solutions were delivered to the nebulizer at 40 µl min−1. The resulting vapor streams were then carried to the TV-Gen at 1 l min−1. Diluent airflow was incorporated into the vapor stream and was adjusted as a means for manipulating the final vapor concentration delivered. Alternatively, the TATP bulk sample was placed in a stainless steel dynamic headspace vapor generator that was coupled to the airline associated with the nebulizer system. In this configuration, TATP vapor was presented through the nebulizer, with humidity arising from the nebulized water. The TV-Gen oven was heated between 70 °C and 100 °C depending on whether or not the analyte was thermally labile.

For online detection and quantification of explosive vapors generated through the TV-Gen, the vapor output port was connected directly to the gas chromatograph/mass spectrometer (GC/MS) [7890A GC/5975C Mass Selective Detector (MSD); Agilent Technologies] by a heated transfer line (225 °C; Clayborn Lab). The GC contains a 15 m RTX-5MS column (0.25 mm × 250 nm; Restek Co.). Typically, vapor was concentrated before the head of the GC column in a cooled programmable temperature vaporizing (PTV) inlet from Gerstel, Inc. The Gerstel online cold injection system (CIS; CIS G4) housed a liner of either deactivated baffled glass or packed with Tenax-TA (Gerstel, Inc.), which was cooled for analyte trapping.

For sake of brevity, we have chosen to forgo an exhaustive demonstration of TV-Gen capabilities related to the generation of trace vapors of explosives and narcotics. Sections IV AIV E and V serve to highlight the capabilities of the TV-Gen to generate trace vapors across a wide range of vapor pressures (ppmv–pptv) and analyte functionality (e.g., nitroaromatic, nitroamine, peroxide, and nitrate ester). They illustrate the ability to control vapor concentration through the nebulization solution concentration and total airflow rate—capabilities that are not analyte specific.

Liquid solutions of RDX and TNT in water from 0.1 µg ml−1 to 0.001 µg ml−1 were sprayed at a constant flow rate of 40 µl min−1 with a total airflow of 10 l min−1. The resulting vapor concentrations were stable for more than five hours [Relative Standard Deviation (RSD) <5%] after reaching equilibrium (Fig. 9). The approximate time to reach equilibrium for TNT was 15 min and 45 min for RDX, which is consistent with vapor equilibrium times associated with those observed in the original TESTbed vapor generation system. The performance was linear within the range of concentrations tested. The vapor stream achieved from the 0.1 µg ml−1 liquid solution was further diluted proportionally to the total volume of air using total flow rates ranging from 5 l min−1 to 20 l min−1. The vapor concentration also scaled linearly with a 1/total airflow rate. Again, an observation consistent with the behavior was observed in the TESTbed system and attributed to changes in analyte surface equilibrium at variable flow rates. The measured vapor concentration was compared to the theoretical vapor concentration to determine efficiency. Total vaporization efficiency was >82% for all TNT concentrations and >85% for RDX, with efficiency decreasing with an increased total airflow rate.

FIG. 9.

TNT and RDX vapor concentration produced by using the TV-Gen from 0.1 µg ml−1 aqueous solutions and a total flow rate of 10 l min−1 over time.

FIG. 9.

TNT and RDX vapor concentration produced by using the TV-Gen from 0.1 µg ml−1 aqueous solutions and a total flow rate of 10 l min−1 over time.

Close modal

The PETN tests were performed using the same parameters as the TNT and RDX tests, with a 0.1 µg ml−1 aqueous solution of PETN and 40 µl min−1 liquid flow. Sampling was performed by collecting vapors onto Tenax sorbent tubes for 30 min at 100 ml min−1 air flow and analyzed using a GC-Electron Capture Detector (ECD) instrument setup with a Gerstel Thermal Desorption System (TDS) sorbent tube sampler.10 Like TNT and RDX, resulting concentration of PETN vapor was linear with the 1/total airflow rate. It took less than 1 h to reach the equilibrium concentration and was steady, provided that there was no disruption in the nebulized solution flow rate. The standard deviations of PETN vapor concentrations were higher than TNT or RDX (relative standard deviations between 5% and 15% across all generated vapor concentrations), with the increase in deviation attributed to variability of Tenax sampling tube performance. The vaporization efficiency ranged from 48% at 20 l min−1 to 78% at 5 l min−1.

Nitromethane (NM) has a much higher volatility than any of the explosives previously evaluated using the TESTbed system. As such, a much higher, 304 µg ml−1, aqueous concentration of nitromethane was required for vapor generation in order to approach expected real-world vapor concentrations. Like the compounds previously tested, the vapor concentration was linear with respect to the inverse air volume flow rate. A few points to note with respect to NM vapor generation are the equilibrium vapor concentration was observed in ∼10 min and the RSD of vapor concentration was ∼1%, apparent in Fig. 10, where the concentration over 80 min is shown at various flow rates. This was attributed to the high vapor pressure of NM, which contributes significantly to the ease of vaporization. That being said, the observed vaporization efficiency was 50%; however, it was believed that collection inefficiency on the small online-CIS Tenax tube was the underlying reason why the measured efficiency is low.

FIG. 10.

NM vapor concentration generated over time at 5 l min−1, 10 l min−1, 15 l min−1, and 20 l min−1.

FIG. 10.

NM vapor concentration generated over time at 5 l min−1, 10 l min−1, 15 l min−1, and 20 l min−1.

Close modal

Like NM, EGDN and NG have high volatilities when compared to TNT, RDX, and PETN. In addition, as was the case with PETN, these explosives are thermally labile; hence, much care in temperature optimization of the TV-Gen for generation and the analytical system for validation was required. Vaporization efficiencies of 86% for EGDN and 60% for NG were observed when nebulizing 1 µg ml−1 solutions at air flow rates between 10 l min−1 and 20 l min−1. However, at the lowest air flow rate of 5 l min−1, decomposition was observed. It takes ∼30 min for EGDN to reach equilibrium concentration at the detector, whereas NG takes ∼1 h; once at equilibrium, the vapor concentrations remained steady within 1% for EGDN and 5% for NG.

It is worth noting that the time to achieve an equilibrium vapor concentration generally tracts with analyte vapor pressure, with more volatile analytes reaching equilibrium more quickly. With that said, the time necessary to reach an equilibrium vapor concentration depends on the equilibrium being established in the liquid handling system (control box and nebulizer), the manifold, and the on-line PTV-GC/MS vapor handling infrastructure. While a given analyte may have a high vapor pressure, its affinity for the surfaces it comes into contact with can have an influence on equilibrium time and overall vaporization efficiency (in the case of the liquid handling system and manifold). Detailed studies of those interactions are beyond the scope of this work.

TATP, a peroxide-based explosive, is unique among the vapors of explosives generated on the TV-Gen as it is only fleetingly soluble in water. Rather than dissolving TATP in water, bulk TATP (∼800 mg) was placed in a dynamic headspace chamber as described previously.11,12 In this instance, the TATP vapor replaced the air source of the nebulizer at a flow rate of 1 l min−1, and the humidity was provided by nebulizing water in the TATP vapor. TATP was successfully detected in this manner using the previously developed analytical method, at concentrations consistent with past results of ∼20 ppbv at a volume flow rate of 20 l min−1.12 

In a similar fashion to the work described in Sec. IV, trace vapors of narcotics and drugs of abuse were generated using the TV-Gen; specifically, vapors of 3,4-methylenedioxy methamphetamine (MDMA), methamphetamine, cocaine, heroin, and THC were successfully nebulized, resulting in stable vapors at ppbv concentrations. Work generating trace vapors from this analyte subset is limited by the fact that very few analytical methods exist demonstrating the quantitation of trace vapors of these analytes. As such, a series of analytical methods using TDS-PTV-GC and online PTV-GC were developed in house to detect these vapors. As was the case with explosives, vapor concentration could be attenuated by the nebulized solution concentration, liquid flow rate, and volume flow rate. Similar trends in the vaporization efficiency were observed with the narcotics and drugs of abuse as a function of vapor pressure; specifically, the lower the vapor pressure, the lower the measured vaporization efficiency, with ∼90% efficiency for MDMA and methamphetamine and ∼50% efficiency for cocaine, THC, and heroin.

Table I summarizes the lowest concentrations quantified for each of the target analytes tested on the TV-Gen. Bold text indicates that a peer-reviewed manuscript describing the analytical method for validation is in press.9,10,12 Italicized text indicates that an unpublished in-house method was used in validation.13 In those instances, the nominal vapor concentration is reported based on the nebulized solution concentration and the volume flow rate of air. To be clear, these concentrations do not represent a limitation of the TV-Gen to produce a vapor at a lower concentration, but a limitation associated with the analytical methods used for analysis. Increases in sample time/volume and improved/optimized detector choices would result in lowering the lowest concentration quantified (LCQ) values.

TABLE I.

Summary of lowest vapor concentrations quantified. Bold indicates that the vapor validation method has been published in a peer-reviewed manuscript, and italics indicate that an in-house method has been utilized for validation.

TDS-CIS-GCOnline CIS-GC
AnalyteSat. vapor conc.aSample time (Vol.)LCQSample time (Vol.)LCQ
TNT ∼9 ppbv 60 min (6 l)b 3.4 pptv 4 min (0.66 l)c 100 ppqv 
RDX ∼5 pptv 60 min (6 l)b 4.3 pptv 4 min (0.66 l)c 500 ppqv 
PETN ∼11 pptv 30 min (3 l)b 12 pptv 10 min (3 l)c 260 ppqv 
TATP ∼63 ppmv   1 min (0.025 l)c 5.5 ppbv 
NM ∼46 800 ppmv   0.5 min (0.0125 l)c 3000 ppbv 
EGDN ∼102 ppmv   1 min (0.1 l)c 17 ppbv 
NG ∼645 ppmv   1 min (0.1 l)c 2.6 ppbv 
Meth ∼210 ppmv   5 min (0.125 l)c 8 ppbv 
MDMA ∼2 ppmv   5 min (0.125 l)c 6 ppbv 
Cocaine ∼160 pptv   10 min (4 l)c 3.2 ppbv 
Heroin ∼80 pptv (50 °C)   4 min (1.1 l)c 6.2 ppbv 
THC ∼60 pptv 60 min (7.4 l)d 3.1 ppbv   
TDS-CIS-GCOnline CIS-GC
AnalyteSat. vapor conc.aSample time (Vol.)LCQSample time (Vol.)LCQ
TNT ∼9 ppbv 60 min (6 l)b 3.4 pptv 4 min (0.66 l)c 100 ppqv 
RDX ∼5 pptv 60 min (6 l)b 4.3 pptv 4 min (0.66 l)c 500 ppqv 
PETN ∼11 pptv 30 min (3 l)b 12 pptv 10 min (3 l)c 260 ppqv 
TATP ∼63 ppmv   1 min (0.025 l)c 5.5 ppbv 
NM ∼46 800 ppmv   0.5 min (0.0125 l)c 3000 ppbv 
EGDN ∼102 ppmv   1 min (0.1 l)c 17 ppbv 
NG ∼645 ppmv   1 min (0.1 l)c 2.6 ppbv 
Meth ∼210 ppmv   5 min (0.125 l)c 8 ppbv 
MDMA ∼2 ppmv   5 min (0.125 l)c 6 ppbv 
Cocaine ∼160 pptv   10 min (4 l)c 3.2 ppbv 
Heroin ∼80 pptv (50 °C)   4 min (1.1 l)c 6.2 ppbv 
THC ∼60 pptv 60 min (7.4 l)d 3.1 ppbv   
a

Saturated vapor concentrations from Ref. 1.

b

Sample was collected on a Tenax-TA sorbent at 25 °C.

c

Sample was collected on a SilcoNert coated glass tube at 10 °C.

d

Sample was collected on a Tenax-TA sorbent and solvent extracted; analysis by liquid injection GC–MS.

The TV-Gen was designed as a portable, compact system capable of reproducibly and accurately generating trace vapors (ppqv to ppmv) of troublesome, low vapor pressure compounds such as explosives, narcotics, and drugs of abuse. Key features of the TV-Gen include the ability to generate vapors from thermally labile analytes, rapid switching between the clean and analyte manifolds, a linear dynamic range of more than three orders of magnitude, and nebulization of analytes in water preventing interference from organic solvents. Additionally, the placement of the dual manifold in a custom oven in addition to the passivation of the entirety of the vapor flow path ensures efficient transport of low vapor pressure explosives.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

This work was funded by the Department of Homeland Security Science and Technology Directorate through Agreement No. HSHQPM-14-X-00176. This work was supported, in part, by the National Research Council Postdoctoral Fellowship Program.

1.
R. G.
Ewing
,
M. J.
Waltman
,
D. A.
Atkinson
,
J. W.
Grate
, and
P. J.
Hotchkiss
, “
The vapor pressures of explosives
,”
TrAC, Trends Anal. Chem.
42
,
35
48
(
2013
).
2.
G. E.
Collins
,
M. P.
Malito
,
C. R.
Tamanaha
,
M. H.
Hammond
,
B. C.
Giordano
,
A. L.
Lubrano
,
C. R.
Field
,
D. A.
Rogers
,
R. A.
Jeffries
,
R. J.
Colton
, and
S. L.
Rose-Pehrsson
, “
Trace explosives sensor testbed (TESTbed)
,”
Rev. Sci. Instrum.
88
(
3
),
034104
(
2017
).
3.
G. E.
Collins
,
B. C.
Giordano
,
V.
Sivaprakasam
,
R.
Ananth
,
M.
Hammond
,
C. D.
Merritt
,
J. E.
Tucker
,
M.
Malito
,
J. D.
Eversole
, and
S.
Rose-Pehrsson
, “
Continuous flow, explosives vapor generator and sensor chamber
,”
Rev. Sci. Instrum.
85
(
5
),
054101
(
2014
).
4.
J. W.
Grate
,
R. G.
Ewing
, and
D. A.
Atkinson
, “
Vapor-generation methods for explosives-detection research
,”
TrAC, Trends Anal. Chem.
41
,
1
14
(
2012
).
5.
X. G.
Yang
,
X. X.
Du
,
J. X.
Shi
, and
B.
Swanson
, “
Molecular recognition and self-assembled polymer films for vapor phase detection of explosives
,”
Talanta
54
(
3
),
439
445
(
2001
).
6.
J. P.
Davies
,
L. G.
Blackwood
,
S. G.
Davis
,
L. D.
Goodrich
, and
R. A.
Larson
, “
Design and calibration of pulsed vapor generators for 2,4,6-trinitrotoluene, cyclo-1,3,5-trimethylene-2,4,6-trinitramine, and pentaerythritiol tetranitrate
,”
Anal. Chem.
65
(
21
),
3004
3009
(
1993
).
7.
B. V.
Antohe
,
D. J.
Hayes
,
D. W.
Taylor
,
D. B.
Wallace
,
M. E.
Grove
, and
M.
Christison
, “
Vapor generator for the calibration and test of explosive detectors
,” in
2008 IEEE Conference on Technologies for Homeland Security
(
IEEE
,
2008
), Vol. 1-2, pp.
384
389
.
8.
C. R.
Field
,
A. V.
Terray
,
A. L.
Lubrano
,
D. A.
Rogers
,
S. J.
Hart
, and
S. L.
Rose-Pehrsson
, “
Pneumatically modulated liquid delivery with feedback control
,”
Rev. Sci. Instrum.
83
(
7
),
076102
(
2012
).
9.
B. C.
Giordano
,
C. R.
Field
,
B.
Andrews
,
A.
Lubrano
,
M.
Woytowitz
,
D.
Rogers
, and
G. E.
Collins
, “
Trace explosives vapor generation and quantitation at parts per quadrillion concentrations
,”
Anal. Chem.
88
(
7
),
3747
3753
(
2016
).
10.
A. L.
Lubrano
,
C. R.
Field
,
G. A.
Newsome
,
D. A.
Rogers
,
B. C.
Giordano
, and
K. J.
Johnson
, “
Minimizing thermal degradation in gas chromatographic quantitation of pentaerythritol tetranitrate
,”
J. Chromatogr. A
1394
,
154
158
(
2015
).
11.
L.
DeGreeff
,
D. A.
Rogers
,
C.
Katilie
,
K.
Johnson
, and
S.
Rose-Pehrsson
, “
Technical note: Headspace analysis of explosive compounds using a novel sampling chamber
,”
Forensic Sci. Int.
248
,
55
60
(
2015
).
12.
B. C.
Giordano
,
A. L.
Lubrano
,
C. R.
Field
, and
G. E.
Collins
, “
Dynamic headspace generation and quantitation of triacetone triperoxide vapor
,”
J. Chromatogr. A
1331
,
38
43
(
2014
).
13.
S.
Rose-Pehrsson
,
G. E.
Collins
,
M.
Hammond
,
B. C.
Giordano
,
L.
DeGreeff
,
M.
Malito
, and
C.
Katilie
, “
Trace vapor generator for explosives and narcotics (TV-GEN)
,” Report No. NRL/MR/6181–18-9829,
Naval Research Laboratory
, https://apps.dtic.mil/dtic/tr/fulltext/u2/1065565.pdf, December 8, 2018.