Elsevier

Journal of Chromatography A

Volume 1182, Issue 2, 29 February 2008, Pages 205-214
Journal of Chromatography A

Identification of homemade inorganic explosives by ion chromatographic analysis of post-blast residues

https://doi.org/10.1016/j.chroma.2008.01.014Get rights and content

Abstract

Anions and cations of interest for the post-blast identification of homemade inorganic explosives were separated and detected by ion chromatographic (IC) methods. The ionic analytes used for identification of explosives in this study comprised 18 anions (acetate, benzoate, bromate, carbonate, chlorate, chloride, chlorite, chromate, cyanate, fluoride, formate, nitrate, nitrite, perchlorate, phosphate, sulfate, thiocyanate and thiosulfate) and 12 cations (ammonium, barium(II), calcium(II), chromium(III), ethylammonium, magnesium(II), manganese(II), methylammonium, potassium(I), sodium(I), strontium(II), and zinc(II)). Two IC separations are presented, using suppressed IC on a Dionex AS20 column with potassium hydroxide as eluent for anions, and non-suppressed IC for cations using a Dionex SCS 1 column with oxalic acid/acetonitrile as eluent. Conductivity detection was used in both cases. Detection limits for anions were in the range 2–27.4 ppb, and for cations were in the range 13–115 ppb. These methods allowed the explosive residue ions to be identified and separated from background ions likely to be present in the environment. Linearity (over a calibration range of 0.05–50 ppm) was evaluated for both methods, with r2 values ranging from 0.9889 to 1.000. Reproducibility over 10 consecutive injections of a 5 ppm standard ranged from 0.01 to 0.22% relative standard deviation (RSD) for retention time and 0.29 to 2.16%RSD for peak area. The anion and cation separations were performed simultaneously by using two Dionex ICS-2000 chromatographs served by a single autoinjector. The efficacy of the developed methods was demonstrated by analysis of residue samples taken from witness plates and soils collected following the controlled detonation of a series of different inorganic homemade explosives. The results obtained were also confirmed by parallel analysis of the same samples by capillary electrophoresis (CE) with excellent agreement being obtained.

Introduction

Homemade explosive devices (sometimes also referred to as improvised explosives or improvised explosive devices) are employed frequently in terrorist attacks and the identification of these explosives is of major importance to counter-terrorism authorities. Rapid and reliable identification of such explosives is a vital step in the investigation of a terrorist attack, and the analytical challenges involved in this process have been discussed using the 2002 Bali bombings as a case study [1]. The broad area of explosives analysis can be divided into two sub areas: detection and identification of explosives or their major ingredients prior to detonation (pre-blast analysis), or identification of explosives by analysis of debris and residues following detonation (post-blast analysis). The present study focuses only on post-blast analysis.

Homemade explosive devices used by terrorists are normally based on inorganic salts and/or peroxides since these materials are readily available, are of low cost and, importantly, can often be purchased legally. The major ingredients in inorganic explosives are fuels and oxidisers. Commonly used fuels include fuel oil, charcoal, sulfur, sugar, and metals such as aluminium(III), magnesium(II) and zinc(II). Common oxidisers include nitrate, chlorate and perchlorate. Additional components may also be added to increase the heat of combustion or improve the handling properties, such as stabilisers, thickeners, etc. Four typical examples of homemade inorganic explosives are shown in Table 1: ammonium nitrate-fuel oil (ANFO), black powders (consisting of nitrate/sulfur/charcoal mixtures), chlorate/sugar, and perchlorate/sugar devices. After detonation, inorganic explosives leave residues which contain a variety of components produced from the explosive reaction itself, as well as unconsumed components of the original explosive mixture. Most of the chemical species present in these post-blast residues are inorganic anions and cations and considerable effort has been expended (see below) in order to develop analytical techniques capable of identifying the original explosive through determination of the levels of inorganic anions and cations present. Moreover, there is a strong preference for those analytical methods which can provide rapid results and which can also be employed in the field.

Some of the analytical methods used for the analysis of inorganic explosives include ion chromatography (IC) [2], [3], [4], [5], [6], [7], [8], capillary electrophoresis (CE) [4], [9], mass spectrometry [7], X-ray powder diffraction [10] and various spot tests [11], [12], [13]. Of these, CE and IC show most promise because of their sensitivity and separation selectivity, as well as their potential to be used in portable or field-deployable formats. CE methods using various techniques such as capillary electrochromatography [14], [15], [16], electrokinetic chromatography [17], micellar electrokinetic chromatography [18], [19] and mixed-mode methods [20] have been applied in capillary and chip-based formats. Detection has typically been performed using ultraviolet absorbance [19], indirect and direct laser-induced fluorescence [21] and electrochemical methods [16], [22], [23], [24]. Recently, we have reported a capillary zone electrophoresis method utilising light-emitting diodes as light sources for indirect photometric detection. This work has demonstrated highly sensitive and selective detection of the widest reported range of anions and cations pertinent to post-blast residues from inorganic IEDs [25].

IC is ideally suited to the analysis of inorganic explosives, and the FBI began developing IC methods for the analysis of slurry explosives and bombing debris as early as 1980 [26]. The simple sample preparation and additional information provided by IC methods compared to previously used techniques were important benefits. A hydrochloric acid/methanol eluent was employed to separate the monovalent cations sodium(I), ammonium, monomethylammonium and potassium(I). In response to concerns from US law enforcement agencies, an IC method for perchlorate and chlorate in perchlorate/chlorate/sugar devices was developed in 1985 [27]. Pipe bomb and soil samples were analysed on a Wescan 269-001 Anion Column (low capacity, silica gel based) using a potassium hydrogen phthalate eluent. Chloride, chlorate, nitrate, sulfate and perchlorate were separated in under 20 min. Around the same time, the Explosion and Flame Laboratory of the Health and Safety Executive (UK) [28] developed IC methods for the analysis of ammonium, methylammonium, potassium(I), sodium(I), chloride, chlorate, perchlorate, nitrate and thiosulfate. A range of columns and eluents was presented to achieve the separation of the target analytes. Henderson and Saari-Nordhaus [3] presented simultaneous determination of nitrate, ammonium, hexamethylenetetramine, calcium(II) and magnesium(II) in under 35 min, and compared the chromatographic results of ammonium and nitrate content of commercial explosive slurries with traditional wet methods. McCord et al. [29] examined post-blast analysis of explosives using IC and they separated chloride, nitrite, chlorate, nitrate, sulfate, thiocyanate, perchlorate and hydrogen carbonate on a Vydac 302IC 4.6 column with an isocratic isophthalic acid eluent in under 20 min. Indirect photometric detection was employed at 280 nm. Sodium(I), ammonium, potassium(I), magnesium(II), calcium(II), lithium(I), ethanolamine and monomethylamine were separated on a Waters M/D column with nitric acid–EDTA eluent in 17 min using conductivity detection. The separation of sodium(I), ammonium, potassium(I), magnesium(II) and calcium(II) was also demonstrated on an Interaction Ion 210 column with a cerium(III) sulfate eluent with indirect photometric detection. The two cation separation methods were used in a complementary manner. Pipe bomb residues were analysed using the IC methods for anions and cations. Burns et al. [7] described methods for the analysis of salts in slurry explosives via the identification of sodium(I), calcium(II), ammonium, methylammonium and nitrate. Nitrate was analysed on a Dionex AS9 column with sodium carbonate eluent, although perchlorate was not adequately resolved from nitrate and was not determined. The target cations were separated on a Dionex CS12 column set with suppressed conductivity detection and a methanesulfonic acid eluent. Four commercial slurry explosives were analysed and identified using the IC methods, in conjunction with chemical spot tests, gravimetric analysis, electrospray MS and liquid chromatography. Bender [10] used two mobile phases (5% acetonitrile/water and 5% acetonitrile/water/50 mM NaOH) to create a hydroxide gradient of 10–50 mM hydroxide in 10 min for the anion analysis of black powder and black powder substitutes. Chloride, nitrite, cyanate, nitrate, sulfate, sulfide, thiocyanate, chlorate, perchlorate and hydrogen carbonate were separated in under 20 min. The FBI developed an IC method targeting anions typically associated with low explosives [8]. Chloride, nitrite, chlorate, nitrate, sulfate, thiocyanate and perchlorate were separated in less than 16 min on a Waters Ion Pak Anion HR column using an isocratic eluent containing boric acid, d-gluconic acid, lithium hydroxide, glycerol, octanesulfonic acid (ion-exclusion reagent to adjust perchlorate selectivity), acetonitrile (to improve resolution) and tetrapropylammonium hydroxide (ion-interaction reagent). The addition of the ion-exclusion and ion-interaction reagents was designed to facilitate the elution of more strongly retained analytes without resorting to gradient elution. Post-blast residues from pipe bombs were analysed using this method. Finally, a thorough review of IC methods used at Sandia National Laboratories was released in 2002 [30]. Methods for the analysis of propellants, organic explosives, pyrotechnics and smoke compositions were presented. A separation (in 18 min) of fluoride, chloride, nitrate, sulfate, thiocyanate, phosphate and perchlorate was presented using a Dionex AS16 column set with hydroxide gradient elution and suppressed conductivity detection. A method for the separation of chlorate and nitrate on a Dionex AS9HC column with an isocratic sodium carbonate eluent was also presented. The important characteristics of the IC methods reported for post-blast analysis of explosives residues are summarised in Table 2.

A few general conclusions can be drawn from this overview of the literature relating to IC analysis of inorganic explosive residues. Thus far, the methods that have been reported have been used for the separation and detection of only a fairly limited range of analytes of interest. Often, complicated eluents have been required to provide the required separation selectivity and to avoid excessively long run times. Generally, the analytical performance characteristics of the developed methods (limits of detection, calibration linearity, reproducibility, etc.) have not been reported. The use of gradient elution in association with suppressed conductivity detection (and the associated benefits of enhanced limits of detection and ease of online eluent generation) has not been fully realised.

In this paper we present optimised IC methods for the separation and detection of a broad range of target analytes (18 anions and 12 cations). The optimised methods were then applied to the analysis of real samples collected following the controlled detonation of a series of homemade inorganic explosive devices. The use of micro-bore columns with simultaneous anion and cation separations allowed shorter analysis times and reduced the volumes of eluents required. The developed methods address the widest range of anions and cations reported to date for IC of explosive residues, and are shown to permit reliable identification of the type of the explosive used.

Section snippets

Instrumentation

Dionex ICS-2000 (Sunnyvale, CA, USA) ion chromatographs were used throughout this work. Instrument control and data acquisition were performed using Chromeleon® software. Two ICS-2000 instruments were used in conjunction with a single Dionex AS autosampler to enable simultaneous separation of anions and cations. A 5 μL sample loop was employed for all the work.

The first ICS-2000 system was used for the separation and suppressed conductivity detection of anions. Separation was performed on a 2 mm

Selection of target analytes

Fuels and oxidisers are the critical components of inorganic homemade explosive devices. Fuel oil, sugar, charcoal, carbon, zinc and sulfur are examples of commonly used fuel sources. Inorganic oxidisers, such as nitrate, chlorate or perchlorate salts are also typically present. The heat of combustion of the explosion can be increased by the addition of lighter elements, such as beryllium, aluminium, and magnesium. Cross-linking agents, gums, thickeners and water can be added to vary the

Conclusions

Anions and cations produced from the detonation of inorganic homemade explosive devices can be separated and identified with suitable selectivity and sensitivity to allow determination of the type and composition of the explosive used. The range of analytes chosen and separated covers the most common types of inorganic homemade explosives and the IC separations shown here offer separation of more components and provide lower limits of detection than methods which have been presented previously.

Acknowledgements

The authors would like to thank the National Security Science & Technology Unit, Department of Prime Minister and Cabinet, Australia, for funding this project. Field testing of explosives and exposed witness plates were provided courtesy of the Australian Bomb Data Centre in conjunction with the Australian Federal Police.

References (32)

  • D. Royds et al.

    Talanta

    (2005)
  • P. Kolla

    Forensic Sci. Int.

    (1991)
  • I.K. Henderson et al.

    J. Chromatogr.

    (1992)
  • K.A. Hargadon et al.

    J. Chromatogr.

    (1992)
  • P. Kolla

    J. Chromatogr. A

    (1994)
  • K.G. Hopper et al.

    Talanta

    (2005)
  • S.A.H. Amas et al.

    J. Forensic Sci. Soc.

    (1966)
  • C.A. Groom et al.

    J. Chromatogr. A

    (2003)
  • S.A. Oehrle

    J. Chromatogr. A

    (1996)
  • J.H.T. Luong et al.

    J. Chromatogr. A

    (1998)
  • B.R. McCord et al.

    Anal. Chim. Acta

    (1994)
  • S.E. Klassen et al.

    Thermochim. Acta

    (2002)
  • P.R. Haddad

    J. Chromatogr. A

    (1997)
  • K.E. Hall et al.

    J. Forensic Sci.

    (1993)
  • D.T. Burns et al.

    Anal. Chim. Acta

    (1998)
  • J.M. Doyle et al.

    Anal. Chem.

    (2000)
  • Cited by (89)

    • Application of stabilized hypobromite for controlling membrane fouling and N-nitrosodimethylamine formation

      2020, Chemosphere
      Citation Excerpt :

      Concentrations of trihalomethanes, haloacetic acids and bromo chloroacetonitrile were determined using gas chromatography mass spectrometry (MS) (Xie, 2001; Nikolaou et al., 2002). Concentrations of chloric acid and bromate were determined using an ion chromatography (Johns et al., 2008) and liquid chromatography-MS/MS (Snyder et al., 2005), respectively. Concentrations of NDMA and NMOR were analyzed using a high-performance liquid chromatography-inline anion exchange reaction-photochemical reaction-chemiluminescence (HPLC-AEM-PR-CL) (Kodamatani et al., 2009, 2018).

    • Identification of inorganic oxidizing salts in homemade explosives using Fourier transform infrared spectroscopy

      2019, Spectrochimica Acta - Part A: Molecular and Biomolecular Spectroscopy
    View all citing articles on Scopus
    View full text