Trace Level Detection of Explosives in Solution Using Leidenfrost Phenomenon Assisted Thermal Desorption Ambient Mass Spectrometry

Abstract

The present paper demonstrates the detection of explosives in solution using thermal desorption technique at a temperature higher than Leidenfrost temperature of the solvent in combination with low temperature plasma (LTP) ionization. Leidenfrost temperature of a solvent is the temperature above which the solvent droplet starts levitation instead of splashing when placed on a hot metallic surface. During this desorption process, slow and gentle solvent evaporation takes place, which leads to the pre-concentration of less-volatile explosive molecules in the droplet and the explosive molecules are released at the last moment of droplet evaporation. The limits of detection for explosives studied by using this thermal desorption LTP ionization method varied in a range of 1 to 10 parts per billion (ppb) using a droplet volume of 20 μL (absolute sample amount 90–630 fmol). As LTP ionization method was applied and ion–molecule reactions took place in ambient atmosphere, various ion–molecule adduct species like [M+NO₂]⁻, [M+NO₃]⁻, [M+HCO₃]⁻, [M+HCO₄]⁻ were generated together with [M−H]⁻ peak. Each peak was unambiguously identified using ‘Exactive Orbitrap’ mass spectrometer in negative ionization mode within 3 ppm deviation compared to its exact mass. This newly developed technique was successfully applied to detect four explosives contained in the pond water and soil sample with minor sample pre-treatment and the explosives were detected with ppb levels. The present method is simple, rapid and can detect trace levels of explosives with high specificity from solutions.

INTRODUCTION

Explosives are regarded as persistent environmental pollutants because most of them are associated with mutagenic or cytotoxic effects at various levels.^1–3) Different types of explosives are widely applied for various military and industrial purposes. Due to the long term operation and inappropriate handling/disposal procedure, explosive residues remain in soil, ground-water and other water resources at production sites as well as active or retired military operation sites. The exact area contaminated by explosives is difficult to identify as the residues are widely scattered in places of artillery practice and production and also they are disposed off without any documentation.⁴⁾ There is also possibility that these explosive-residue related pollutants can be enriched and enter in food chain.⁵⁾ Due to these facts, rapid, sensitive and unambiguous detection of various explosives from different environmental sources with minimal sample handling are highly demanding. In addition to this, improvised explosives are manufactured and used for global terrorist activity. It is also highly essential to identify trace levels of explosives from various surfaces and complex samples to prevent such nuisance activities.

Among the various analytical techniques available for explosive analysis, high performance liquid chromatography (HPLC) and gas chromatography (GC) have been used to detect the explosives. United States Environmental Protection Agency (US-EPA) adopted HPLC technique with ultra-violet (UV) detection for various explosives.⁶⁾ But the technique is unable to detect non-UV absorbing explosives. GC based method has also limitations to detect thermally labile explosives.⁷⁾ In addition to this, both HPLC and GC based methods need extensive sample preparation techniques which are laborious and time-consuming. Ion mobility spectrometry (IMS) is another rapid and sensitive technique that could be operated at atmospheric pressure for various explosive analyses.^8–10) But it has also a few drawbacks regarding its quantitative capability and selectivity.¹¹⁾

Mass spectrometry is a highly specific and sensitive technique which can overcome most of the disadvantages associated with previously mentioned techniques. Ambient mass spectrometry is a technique where samples are ionized in their native environment with minimal handling.¹²⁾ For the last few years a number of ambient mass spectrometric techniques have been developed to detect trace levels of explosives from various surfaces and complex matrices without or with minor sample pre-treatment. In 2005, Takats et al. reported the detection of several explosives from ambient surfaces by using desorption electrospray ionization (DESI) mass spectrometry.¹³⁾ Since then several other ambient mass spectrometric techniques like thermal desorption ambient ion/molecule reactions,¹⁴⁾ direct analysis in real time (DART),¹⁵⁾ extractive electrospray ionization (EESI),¹⁶⁾ dielectric barrier discharge ionization (DBDI),¹⁷⁾ low-temperature plasma (LTP),^18,19) etc. have been developed for direct detection of explosives from various surfaces and mixtures. A recent technique applied thermal sampling direct current atmospheric pressure glow discharge source for detection of explosives from soil sample.²⁰⁾ Rapid response times of these techniques made them high-throughput and applicable to various airport check-points and public places. Overall, the ambient mass spectrometric techniques are quite promising for explosive detection and presently its application is growing rapidly.

Although several ambient mass spectrometric techniques have been developed for explosive detection, still new techniques are desired to increase its area of application and use to some specific challenging field for high sensitive detection. The present technique is applicable to the detection of explosive molecules from various complex environmental matrices with high sensitivity under ambient experimental conditions. Explosives are generally less-volatile, thermally unstable and possess very low vapor pressure, which creates much difficulty for desorption in ambient condition.^14,17) Here, we report the application of a thermal desorption technique at a temperature above the Leidenfrost temperature²¹⁾ of the solvent containing the explosives. Under this particular condition, smooth and gentle evaporation of solvents can be realized and explosive molecules are desorbed at the last moment of solvent evaporation. This technique solves the problem associated with desorption of less-volatile explosives and due to spontaneous sample enrichment during the Leidenfrost phenomenon, desorption of less-volatile explosive molecules takes place at a short time domain. This characteristic makes the technique high sensitive. Low temperature plasma (LTP),¹⁸⁾ a modified version of dielectric barrier discharge (DBD)²²⁾ was used as an ionization source for the desorbed explosives. As the temperature of the liquid droplet remains much lower than the solvent boiling point and ion–molecule reactions take place in ambient atmosphere, various cluster ions are generated with little thermal fragmentation. Soft and high sensitive detection of explosive molecules contaminated in soil and pond water indicates diverse applicability of the technique.

EXPERIMENTAL SECTION

Reagents and chemicals

The structures and exact molecular masses of all four explosives used in this study are provided in Fig. 1. All four explosives—trinitrotoluene (TNT), 1,3,5-trinitroperhydro-1,3,5-triazine (RDX), octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX), and pentaerythritol tetranitrate (PETN) were obtained from Chugoku Kataku Co., Ltd. (Hiroshima, Japan). HPLC-grade acetonitrile was purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan).

Fig. 1. Structures and exact molecular masses of the explosives studied.

Instrumentation

A home-made low-temperature plasma (LTP) ion-source was used for all the experiments reported here. Figure 2 shows the picture of the experimental set-up. The details of experimental set-up have been reported elsewhere.²³⁾ In brief, a copper tape placed near the open end of the glass tube (with outer and inner diameter of 3.0 and 1.5 mm, respectively) served as the outer electrode and an axially centered grounded stainless steel wire in the glass tube was used as an inner electrode. An alternating high voltage, 15 kHz, V_p–p: 3.0 kV was applied to the outer electrode. Helium gas (450 mL min⁻¹) was used to generate helium plasma to initiate the ion–molecule reactions. The horizontal distance between the inlet of mass spectrometer and the exit of glass tube was 7.0 mm. The sample holder (Arios Corporation, Tokyo, Japan) was placed below the inlet of mass spectrometer at 90° angle and the vertical distance between the sample holder and the mass spectrometer inlet was 7.0 mm.

Fig. 2. Picture of the experimental set-up.

High-resolution ‘Exactive Orbitrap’ mass spectrometer (Thermo Fisher Scientific, Bremen, Germany) was used to conduct the experiments and the acquired data were analyzed with the Xcalibur software (version 2.1) (Thermo Fisher Scientific, Bremen, Germany). The instrument was used in high resolution mode (resolution: 50,000 at 2 Hz). Capillary temperature and voltage were 200°C and 30 V, respectively. A fixed sample injection time of 100 ms was used and tube lens and skimmer voltages were maintained at 30 V and 20 V, respectively. All of the four explosives showed less than 3 ppm deviation compared to their exact mass.

Standard solution preparation

1 mg mL⁻¹ standard stock solutions in acetonitrile were obtained for all four explosives from manufacturer. Further dilutions were made using acetonitrile and the stock solutions were preserved at 4°C. A 20 μL solution was introduced to the heated sample holder manually using a pipet tip (Eppendorf, Hamburg, Germany).

Sample preparation

Pond water: Pond water was collected from a pond situated at the University of Yamanashi. One milliliter of the raw pond water was spiked with 100 μL of the diluted explosives standard and served for the analyses.

Soil: The soil sample was collected from the University of Yamanashi. One gram of soil was spiked with 10 μL of diluted stock solutions of each explosive and kept for 1 h for drying in air. Then, 1 mL acetonitrile was added and was centrifuged for 2 min and the supernatant was decanted for analysis.

RESULTS AND DISCUSSION

Desorption and ionization are two significant terms in ambient mass spectrometry. In various ambient mass spectrometric techniques developed so far, desorption and ionization take place simultaneously or separately. Heat is a very common source for desorption and can increase the detection sensitivity of explosives with several orders of magnitude.¹⁹⁾ When heat is supplied to a solution containing a non-volatile compound, the solution evaporates fast, leaving the sample residue on sample holder. If the sample is thermally labile in nature, then heat-associated dissociation could occur.²⁴⁾ But when the same solution droplet is placed on a hot plate having temperature above the Leidenfrost temperature of the solvent, the droplet starts levitating on the hot surface and a slow and gentle solvent evaporation takes place. Despite the relatively high temperature of the hot plate, the droplet temperature remains lower than the solution’s boiling point due to the presence of heat-insulating vapor layer between the heater and the droplet. As the less-volatile explosive molecules remain in the droplet, they experience the same temperature as the droplet solution. Thus, the whole evaporation process minimizes the possibility of heat associated dissociation of the analytes. At the last stage of solvent evaporation, desorption of concentrated explosive molecules takes place. Therefore, the use of present desorption technique has two distinct benefits—high sensitivity and little thermal dissociation for the less-volatile analytes. In the present experiments, acetonitrile was used as a solvent for explosives whose Leidenfrost temperature was 250°C. Thus, all of the experiments were performed with heater temperature at 250°C, except pond water where 400°C (Leidenfrost temperature of water) was used. Figure 3 represents the temperature dependent response curve for [RDX+NO₂]⁻ peak using acetonitrile as solvent. After 250°C a sharp decrease in response was observed, although the droplet showed Leidenfrost phenomenon above 250°C. This may partly be due to the dissociation of cluster ion [RDX+NO₂]⁻ into RDX and NO₂⁻ at higher temperature.

Fig. 3. Temperature dependent response curve of RDX in acetonitrile (1 ppm) where [RDX+NO₂]⁻ peak area was plotted against sample holder temperature. Each data point is the average peak area of three independent experiments and error bars represent the corresponding standard errors. Normalized responses at each data point are generated by dividing the average area at each data point by lowest average area obtained at 350°C (i.e. normalized response is 1 at 350°C). The straight lines are guide lines.

The high electron affinities of the nitrite and nitro compounds favor the formation of negatively charged ions.²⁵⁾ In this work negative mode ionization has been exploited for all four explosives—TNT, RDX, HMX, and PETN. The aromatic nitro compound TNT was easily detected as deprotonated molecular ion peak [M−H]⁻ (m/z 226) (Fig. 4a) and weaker fragment peak at m/z 197, [M−NO]⁻. The LOD was 1 ppb with S/N>10 [Fig. 4a (inset)]. RDX and HMX, both cyclic aliphatic nitro compounds were detected as [M−H]⁻ peaks and various cluster ions [Fig. 4(b, c)]. For both cases the cluster ions were identified as [M+NO₂]⁻, [M+HCO₃]⁻, and [M+HCO₄]⁻. The LODs for RDX and HMX were 1 and 5 ppb, respectively, with S/N>3 [Inset (Fig. 4b, 4c)]. The present technique was also able to detect 10 ppb PETN [Fig. 4d (inset)]. The various peaks present in the PETN mass spectrum were identified as [M−H]⁻, [M+NO₃]⁻, and [M+HCO₄]⁻ (Fig. 4d). Peak identification was based on their exact masses obtained by using high-resolution mass spectrometer. The precision of the present technique was studied using 100 ppb RDX solution in acetonitrile. Three separate measurements were performed and a relative standard deviation (RSD) of 13.7% was obtained based on the area of [RDX+NO₂]⁻ peak (m/z: 268).

Fig. 4. Negative ion mode mass spectra of pure chemical solutions (1 ppm) in acetonitrile for (a) TNT, (b) RDX, (c) HMX, and (d) PETN. Inset of each of the spectrum shows the LOD level chronograms for selected ion monitoring (SIM) peak.

Regarding the ionization process, we expect similar types of ionization take place as reported for low temperature plasma probe.^18,19) In this case the excited helium (He*) reacted with atmospheric gases and generated secondary ions that reacted with explosive molecules. The strong appearance of [M+NO₂]⁻, [M+NO₃]⁻, [M+HCO₃]⁻, [M+HCO₄]⁻ indicates that ion–molecule clustering reactions take place in ambient atmosphere. The possible roots behind the formation of product ions in negative mode are the interaction of analyte ions via proton abstraction, electron transfer or adduct formation.²⁶⁾ The characteristics of analytes and reactant ions will determine the nature of product formation.

Compared with the other techniques, one of the most interesting facts of the present technique is that all the explosives generated their [M−H]⁻ peaks despite of their different chemical properties. In the present method, all explosives showed similar desorption process and desorbed molecules interacted with reactant ions generated by the DBD plasma. The whole process must be much softer compared to direct interaction of LTP with explosive molecules desorbed from the surface. Although a high temperature was applied to the sample holder, the dominant appearance of [M−H]⁻ ions for labile explosives confirm that the droplet temperature was much lower than that of the sample holder.

The application of present technique was extended to soil and pond water that are highly desirable for environmental monitoring and post explosion residue analysis. Figures 5 and 6 show the mass spectra obtained from soil and pond water spiked with four explosives. All of the explosives were detected unambiguously and simultaneously based on their characteristic peaks. Soil sample has higher interferences compared to pond water sample due to the higher abundances of organic matters and salts. As all of the four explosives were detected in ppb levels from soil and pond water with a rapid sample handling procedure, this technique could be considered as a viable alternative technique for rapid detection of explosives from various complex environmental matrices.

Fig. 5. Mass spectrum for extracted acetonitrile solution from soil spiked with 100 ppb TNT, 200 ppb RDX, 500 ppb HMX and PETN.

Fig. 6. Mass spectrum for pond water spiked with 50 ppb TNT, 100 ppb RDX and HMX, 200 ppb PETN.

CONCLUSION

The introduction of Leidenfrost phenomenon for desorption of explosive molecules in ambient atmosphere has been made. This method was found to be sensitive and specific for rapid detection of explosive molecules from complex environmental matrices. Basically the present technique is a desorption technique in combination with LTP; analytes could be detected with high sensitivities. A potential advantage of this technique is the soft and instant desorption characteristics, which facilitates the production of molecular ions as major ions with little thermal decomposition. Due to its high sensitivity, this technique could also be used for swab sample analysis in security check-points. The technique may open a new era of thermal desorption technique for less- or even non-volatile explosives.

Acknowledgment

The authors acknowledge the financial support for this work by the “Strategic Funds for the Promotion of Science and Technology” from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan.

REFERENCES

• 1)  J. Yinon. Toxicity and metabolism of explosives, CRC Press, Boca Raton, FL, 1990.
• 2)  P. Y. Robidoux, C. Svendsen, J. Caumartin, J. Hawari, G. Ampleman, S. Thiboutot, J. M. Weeks, G. I. Sunahara. Chronic toxicity of energetic compounds in soil determined using the earthworm (Eisenia andrei) reproduction test. Environ. Toxicol. Chem. 19: 1764–1773, 2000.
• 3)  E. L. Rylott, A. Lorenz, N. C. Bruce. Biodegradation and biotransformation of explosives. Curr. Opin. Biotechnol. 22: 434–440, 2011.
• 4)  R. Mu, H. Shi, Y. Yuan, A. Karnjanapiboonwong, J. G. Burken, Y. Ma. Fast separation and quantification method for nitroguanidine and 2,4-dinitroanisole by ultrafast liquid chromatography-tandem mass spectrometry, Anal. Chem. 84: 3427–3432, 2012.
• 5)  W. J. Buttner, M. Findlay, W. Vickers, W. M. Davis, E. R. Cespedes, S. Cooper, J. W. Adams. In situ detection of trinitrotoluene and other nitrated explosives in soils. Anal. Chim. Acta 341: 63–71, 1997.
• 6)  EPA. Method 8330—Nitroaromatics and nitroamines by high performance liquid chromatography (HPLC), 1994.
• 7)  M. E. Walsh. Determination of nitroaromatic, nitramine, and nitrate ester explosives in soil by gas chromatography and an electron capture detector. Talanta 54: 427–438, 2001.
• 8)  G. R. Asbury, J. Klasmeier, H. H. Hill, Jr. Analysis of explosives using electrospray ionization/ion mobility spectrometry (ESI/IMS). Talanta 50: 1291–1298, 2000.
• 9)  R. G. Ewing, D. A. Atkinson, G. A. Eiceman, G. J. Ewing. A critical review of ion mobility spectrometry for the detection of explosives and explosive related compounds. Talanta 54: 515–529, 2001.
• 10)  G. A. Eiceman, J. A. Stone. Ion mobility spectrometers in national defence. Anal. Chem. 76: 390A–397A, 2004.
• 11)  R. Waddell, D. E. Dale, M. Monagle, S. A. Smith. Determination of nitroaromatic and nitramine explosives from a PTFE wipe using thermal desorption-gas chromatography with electron-capture detection. J. Chromatogr. A 1062: 125–131, 2005.
• 12)  G. A. Harris, A. S. Galhena, F. M. Fernández. Ambient sampling/ionization mass spectrometry: Applications and current trends. Anal. Chem. 83: 4508–4538, 2011.
• 13)  Z. Takats, I. Cotte-Rodriguez, N. Talaty, H. Chen, R. G. Cooks. Direct, trace level detection of explosives on ambient surfaces by desorption electrospray ionization mass spectrometry. Chem. Commun. 1950–1952, 2005.
• 14)  I. A. Popov, H. Chen, O. N. Kharybin, E. N. Nikolaev, R. G. Cooks. Detection of explosives on solid surfaces by thermal desorption and ambient ion/molecule reactions. Chem. Commun. 1953–1955, 2005.
• 15)  R. B. Cody, J. A. Laramée, H. D. Durst. Versatile new ion source for the analysis of materials in open air under ambient conditions. Anal. Chem. 77: 2297–2302, 2005.
• 16)  H. Chen, A. Venter, R. G. Cooks. Extractive electrospray ionization for direct analysis of undiluted urine, milk and other complex mixtures without sample preparation. Chem. Commun. 2042–2044, 2006.
• 17)  N. Na, C. Zhang, M. Zhao, S. Zhang, C. Yang, X. Fang, X. Zhang. Direct detection of explosives on solid surfaces by mass spectrometry with an ambient ion source based on dielectric barrier discharge. J. Mass Spectrom. 42: 1079–1085, 2007.
• 18)  J. D. Harper, N. A. Charipar, C. C. Mulligan, X. Zhang, R. G. Cooks, Z. Ouyang. Low-temperature plasma probe for ambient desorption ionization. Anal. Chem. 80: 9097–9104, 2008.
• 19)  J. F. Garcia-Reyes, J. D. Harper, G. A. Salazar, N. A. Charipar, Z. Ouyang, R. G. Cooks. Detection of explosives and related compounds by low-temperature plasma ambient ionization mass spectrometry. Anal. Chem. 83: 1084–1092, 2011.
• 20)  L. Ma, B. Xin, Y. Chen. Direct mass spectrometric detection of trace explosives in soil samples. Analyst (Lond.) 137: 1730–1736, 2012.
• 21)  B. S. Gottfried, C. J. Lee, K. J. Bell. The Leidenfrost phenomenon: Film boiling of liquid droplets on a flat plate. Int. J. Heat Mass Transf. 9: 1167–1188, 1966.
• 22)  N. Na, M. Zhao, S. Zhang, C. Yang, X. Zhang. Development of a dielectric barrier discharge ion source for ambient mass spectrometry. J. Am. Soc. Mass Spectrom. 18: 1859–1862, 2007.
• 23)  S. Saha, L. C. Chen, M. K. Mandal, K. Hiraoka. Leidenfrost phenomenon-assisted thermal desorption (LPTD) and its application to open ion sources at atmospheric pressure mass spectrometry. J. Am. Soc. Mass Spectrom., 2012, in press.
• 24)  T. R. Covey, B. A. Thomson, B. B. Schneider. Atmospheric pressure ion sources. Mass Spectrom. Rev. 28: 870–897, 2009.
• 25)  I. Cotte-Rodríguez, Z. Takáts, N. Talaty, H. Chen, R. G. Cooks. Desorption electrospray ionization of explosives on surfaces: Sensitivity and selectivity enhancement by reactive desorption electrospray ionization. Anal. Chem. 77: 6755–6764, 2005.
• 26)  M. J. Waltman, P. Dwivedi, H. H. Hill, Jr., W. C. Blanchard, R. G. Ewing. Characterization of a distributed plasma ionization source (DPIS) for ion mobility spectrometry and mass spectrometry. Talanta 77: 249–255, 2008.