Dynamite Analysis by Raman Spectroscopy As a Unique Analytical Tool

Technical Note pubs.acs.org/ac

Dynamite Analysis by Raman Spectroscopy As a Unique Analytical Tool María López-López,†,‡ Jose Luis Ferrando,‡,§ and Carmen García-Ruiz*,†,‡ †

University Institute of Research in Police Sciences and ‡Department of Chemistry I, Multipurpose Building of Chemistry, University of Alcalá, Ctra. Madrid-Barcelona Km. 33.600, 28871 Alcalá de Henares (Madrid) Spain § Criminalistic Service of Guardia Civil, C/Guzmán el Bueno 110, 28003 Madrid, Spain ABSTRACT: Apart from powerful explosives, dynamites are complex samples with an intricate analysis. These mixtures of compounds of diverse chemical nature present a challenge to the analyst, and as a result, several analytical techniques need to be applied currently for their analysis. Taking into account that presently there are almost no methods for dynamite analysis in the literature, it is crucial to develop analytical methods that could be applied for the analysis of these samples. This study introduces the use of Raman spectroscopy to analyze dynamites. Two different dynamites made up of ethylene glycol dinitrate and ammonium nitrate, among other minor components, were analyzed by Raman spectroscopy. First, confocal Raman spectroscopy allowed the identification of different components easily distinguished by eye (ammonium nitrate, ethylene glycol dinitrate, and sawdust). Then, Raman mapping was used to show the distribution of the main components throughout the dynamite mass. Finally, several minor components were identified after flocculation (nitrocellulose) or precipitation (sawdust, CaCO3, and flour). The results obtained demonstrate the huge potential of this technique for the analysis of such a complex and tricky sample.

D

literature describing the analysis of the main active components that could be present in dynamites. In the case of EGDN, NG, and AN, several separation techniques were employed for their analysis. High-performance liquid chromatography (HPLC) was used, after previous extraction, to recover and determine the EGDN content in one dynamite sample.3 HPLC coupled with electrospray ionization and atmospheric pressure chemical ionization in negative-ion mode was investigated for the analysis of NG and EGDN along with other seven additional explosive compounds. Then, three additional forensic samples were used to demonstrate the validity of the developed methods.4 Gas chromatography using cool on-column injection and negative chemical ionization was employed to determine EGDN and NG, both explosives extracted from soil samples.5 Portable capillary electrophoresis methods were developed for field screening of EGDN, NG, and AN to identify the type of explosive used.6,7 Other analytical techniques, such as ion mobility spectrometry (IMS) or isotope ratio mass spectrometry (IRMS), were also used to determine AN and EGDG. In the case of IMS, few works described the trace detection of AN8,9 and both NG and AN directly absorbed to hair10 or using planar solid-phase microextraction-ion mobility.11 In regards to IRMS, one study described the analysis of five different unreacted AN samples and then classified them based on nitrogen, oxygen, and hydrogen isotopes values to determine if

ynamites are explosives commonly used in blasting rock, construction, demolition, and mining. However, unfortunately, they are sometimes used for noncivilian purposes. Dynamites gave worldwide fame to Alfred Nobel in 1867 when he discovered that combining nitroglycerine (NG) with a porous siliceous earth (kieselgur) the resulting paste was much safer to handle than NG.1 This new product could be shaped into sticks suitable for insertion into drilling holes and allowed the detonation under controlled conditions by using blasting caps. However, this first composition has changed over the years to make way for dynamite with different formulations. In fact, some years later, in 1875, Nobel incorporated nitrocellulose (NC) into dynamites producing gelatinous formulations more water resistant. Another significant development was the substitution of NG by ethylene glycol dinitrate (EGDN) in order to lower the dynamite freezing point. Today, mixtures of EGDN/NG ranging from 100/0 to 60/40 are commonly used as the active ingredients of dynamites.2 These active ingredients could be absorbed in different materials, such as sawdust and flour. Ammonium nitrate (AN) is added to dynamites to obtain stronger, cheaper, and safer dynamites. In fact, AN formulations are among the most commonly used. Other compounds may be added to commercial dynamites depending on their use and manufacturer, making them complex samples difficult to analyze. It is for that very reason that the analysis of dynamite samples must be carried out using different analytical techniques, usually with the previous extraction of the components from the sample. Although there are very few articles reporting the analysis of dynamites, there are several articles referred to in the scientific © 2013 American Chemical Society

Received: September 24, 2012 Accepted: January 29, 2013 Published: January 29, 2013 2595

dx.doi.org/10.1021/ac302774w | Anal. Chem. 2013, 85, 2595−2600

Analytical Chemistry

Technical Note

IRMS could assist in the investigation of complex forensic cases by providing a level of discrimination not achievable applying traditional forensic techniques.12 With regard to vibrational techniques, attenuated total reflection (ATR)-Fourier transform infrared (FTIR) spectroscopy was applied for detection of AN and other explosive particles in fingerprints.13 A little longer is the list of articles in which the Raman spectroscopy technique is used for the analysis of compounds present in dynamites. Different applications like identification of AN particles on clothes and human nails,14,15 analysis of AN particles trapped between synthetic fibers and colored textile specimens,16 or the quantitative analysis of AN17 are reported. Raman variants such as deep ultraviolet resonance Raman spectroscopy (RRS)18 and surface enhanced Raman spectroscopy (SERS)19 (using molecularly imprinted Au matrixes) enable the ultratrace detection of AN (ppb) using RRS and NG (20 pM) and EGDN (400 fM) through SERS. Additionally, a hybrid dual Raman-laser-induced breakdown spectroscopy (LIBS) system for the standoff detection of energetic materials was developed and applied for the identification of AN from a dynamite sample.20 In the case of NC, an extensive review focused on the analytical techniques applied to the analysis of NC was recently published.21 The authors reported that most articles published were designed to investigate its polymeric parameters, morphological and thermal characteristics, or the degradation of the molecule. However, the analysis of NC in explosives by Raman spectroscopy has only been performed in fireworks22 and propellants.23 Other techniques such as FTIR, mass spectrometry, IMS, and LC techniques were also used for the determination of NC in explosives, although the use of dynamites as NC sources was never mentioned. The Raman technique was selected in the present scientific research to overcome the challenge of dynamite analysis. Raman spectroscopy is a powerful analytical tool suitable for this purpose thanks to its easy sampling with no or little sample preparation and its unequivocal chemical identification capability. Furthermore, the possibility of dealing with bulk samples while automatically mapping the microscopic component distribution makes Raman spectroscopy the ideal tool for the analysis of the complex samples that are dynamites. As consequence, the aim of this work was to study the use of Raman spectroscopy for the analysis of dynamite samples, introducing to the potential analyst a new analytical approach for dynamite analysis.

Figure 1. Pink (a) and white (b) dynamite samples (dynamites diameter, 26 mm). Spots distinguished with the naked eye and analyzed by Raman spectroscopy are marked at the left of each picture.

methanol (samples were placed in an ultrasonic bath 5 min). Then, sample was centrifuged at 3450g during 5 min, and the supernatant was transferred to the other tube. Over the supernatant, 3 mL of water was added to induce the NC flocculation. Instrumentation. Thermo Scientific DXR Raman microscope (Waltham, MA), with a 900 lines per mm grating and 532 nm excitation wavelength was used. The laser power on the sample was 5.0 mW, and the confocal pinhole size was 25 μm. The microscope was set to 50× magnification. Spectral acquisition times were 10 s × 10 acquisitions (EGDN, dynamite dye, NC, and flour), 1 s × 30 acquisitions (sawdust), and 1 s × 10 acquisitions (AN and CaCO3). Background and fluorescence corrections were applied for all the spectra. Raman spectral mapping was carried out automatically; an area corresponding to approximately 610 μm × 410 μm was defined on the optical image provided by the instrument video camera. Step sizes between two successive measurements were set to 35 μm × 34 μm (247 spectra) and 36 μm × 41 μm (198 spectra) for the pink and the white dynamite samples, respectively. The microscope was set to 50× magnification, and the spectral acquisition times were 5 spectra of 2 s. The wavenumber range measured ranged from 50 to 3500 cm−1. Chemical maps based on an intensity-based color scale where red represents high intensity and blue represents low intensity referenced to a particular Raman shift (650 cm−1 for EGDN and 1641 cm−1 for AN) were obtained using OMNIC Atlμs Imaging Software. Safety Concerns. Under the experimental conditions applied, sample burning was not observed in any of the samples. However, NC is very sensitive to the laser beam and burn easily. Burn spots were observed in the NC standard when the 532 nm laser was impinging on it for more than 10 s at the maximum laser power on the sample (10 mW). NC burns in open air, and it can explode only when it is confined. Therefore, samples should not be confined in cuvettes when analyzing. Dynamites should be handled and stored with care (temperatures should be in a range of −15 °C and +60 °C), ensuring the product is kept away from flames, methane environments, and excessive heat sources.

■

EXPERIMENTAL SECTION Standard and Samples. Two different types of dynamites with known qualitative composition (Figure 1), EGDN, AN, NC, dynamite dye, flour, sawdust, and CaCO3 were kindly provided by the Criminalistic Service of Guardia Civil. According to the information given by the manufacturer, both dynamite samples were basically composed of EGDN and AN, among other components. The first sample (referred to as pink dynamite) also has NC, dynamite dye, sawdust, CaCO3, guar gum, and plasticizers in its composition. The second dynamite (referred as white dynamite) contains NC, flour, CaCO3, guar gum, petro AG, and plasticizers. No quantitative information about the samples was known. Sample Preparation. Dynamite major components were directly measured at the sample, but little sample preparation was needed for the identification of minor components. Approximately 20 mg of dynamite were dissolved in 1 mL of 2596



Technical Note

Figure 2. Dynamite spots identification. Spectral comparison of (a) spot 1 with sawdust standard; (b) spots 2 and 4 with AN standard; (c) spots 3 and 5 with EGDN standard; (d) EGDN reference standard with spot 3 and dynamite dye reference standard (spectrum scaled by a factor of 0.3). Raman conditions: laser at 532 nm, 8.0 mW, 50× magnification objective lens, confocal pinhole size of 25 μm. Spectral acquisition times: 10 s × 10 acquisitions (EGDN and dynamite dye), 1 s × 10 acquisitions (AN), and 1 s × 30 acquisitions (sawdust). Several bands are labeled for clarity.

■

RESULTS AND DISCUSSION Pictures of the two dynamites analyzed in this study are depicted in Figure 1. Although both types of samples, in principle, have similar composition, the first sample of dynamite is pink due to a reddish dye. Dyes could be added during the manufacturing process to identify the type of dynamite. Magnifications of the views are also displayed in order to show the heterogeneity of the samples. Note that the pink dynamite shows yellowish substances (spot 1) distributed throughout the dynamite mass that are not present in the white dynamite. Furthermore, the pink dynamite shows dark pink spots (spot 3) and the white dynamite shows dark gray-green spots (spot 5). The Raman spectra of the five spots easily distinguished with the naked eye (three spots of the first dynamite and the two spots of the second dynamite) were measured by Raman spectroscopy. Figure 2 compares the spots spectra obtained with the spectra of sawdust, AN, and EGDN. As it could be seen in Figure 2a, the spectrum of the yellowish compound present in the pink dynamite (spot 1) corresponds to the sawdust reference spectrum (bands at 2938, 1655, 1595, 1333, 1268, 1125, and 1094 cm−1). The spectra of the light pink part of the pink dynamite sample (spot 2) and the white part of the white dynamite sample (spot 4), both depicted in Figure 2b, match with the reference spectrum of AN (bands at 1042, 714, 167, and 137 cm−1). In the case of the dark pink spot of the pink dynamite sample (spot 3) and the dark gray-

green spot of the white dynamite sample (spot 5), both depicted in Figure 2c, they basically correspond with the EGDN spectrum (bands at 2974, 2619, 1636, 1288, 1114, 854, 560, 470, and 281 cm−1). It should be noted that the spectrum of spot 3 (Figure 2d) has additional bands corresponding to the dynamite dye (bands at 1595, 1482, 1439, 1369, 1338, 1220, 1187, 1150, 1095, and 748 cm−1). Spectra maps were collected for both samples to illustrate the distribution of the two main components (EGDN and AN) in the samples. Many individual sample positions over an approximately 612 μm × 410 μm predefined region were measured. Once the maps were collected, two chemical contour maps (one for EGDN and another for AN) were obtained for each sample. Chemical contour maps based on an intensitybased color scale, where red represents high intensity and blue represents low intensity, were produced taking into account the overall spectral intensity of a certain band. The intense band at 560 cm−1 of EGDN was selected to show its distribution because that band is free of interferences from the spectra of other compounds. In the case of AN, the 1042 cm−1 band (due to NO3− symmetric stretching)20 was selected. The resulting two chemical contour maps for the two dynamites and the video image of them are shown in Figure 3. If the visual images of the mapped area of the samples are compared with the chemical maps obtained, it could be seen that the dark pink and dark gray-green spots mentioned above basically correspond with EGDN, being mainly the rest of the sample AN. 2597



Technical Note

Figure 3. Chemical maps of dynamites based on an intensity-based color scale. (a) chemical maps based on the intensity of the spectral band at 560 cm−1 (EGDN); (b) visual images of mapped areas; (c) chemical maps based on the intensity of the spectral band at 1042 cm−1 (AN). Raman conditions: laser at 532 nm, 5.0 mW, 50× magnification objective lens, confocal pinhole size of 25 μm, 2 s × 5 acquisitions. Raman spectral mapping: 610 μm × 410 μm area; step sizes 35 × 34 μm (247 spectra), pink dynamite; step sizes 36 × 41 μm (198 spectra), white dynamite.

Although the two major ingredients present in these types of dynamites (EGDN and AN) are identified directly at the sample, other minor compounds such as NC are present. In order to obtain a clear spectra of NC and other minor components present in the sample that could help to dynamite identification, some sample preparation is suggested in this work. Figure 4 depicts the method proposed. Approximately 20 mg of both dynamites were first dissolved in 1 mL of methanol, and the methanol supernatant (Figure 4(I)) was placed in other tube and a precipitate (Figure 4(II)) was observed in the bottom of the tube. Then, 3 mL of water was added to the supernatant and the NC flocculation was observed. NC from both dynamites were caught and analyzed by Raman spectroscopy. Figure 4a compares the NC spectra from both dynamites

with the reference spectrum of NC (bands at 2974, 2901, 1651, 1454, 1367, 1280, 1149, 1125, 1083, 844, 698, 626, 560, and 201 cm−1), which matches with the NC spectrum referred to in the literature.22,24 In regard to the precipitates obtained, they were analyzed directly at the tube. The pink dynamite precipitate was measured at different sample points, and the spectra obtained corresponded to sawdust (Figure 4b), CaCO3 (Figure 4c; bands at 1083, 714, 282, and 156 cm−1, which is added to the composition to neutralize nitric acid formed by spontaneous decomposition25), and flour (Figure 4d; bands at 2915, 1456, 1134, 1125, 940, 867, and 479 cm−1). In the case of the white dynamite, the two different spectra measured were assigned to CaCO3 (Figure 4c) and flour (Figure 4d). 2598



Technical Note

Figure 4. Procedure proposed for the analysis of dynamite minor components. Spectral comparison of (a) NC isolated from dynamites with an NC standard; (b) sawdust analyzed from pink dynamite with a sawdust standard; (c) CaCO3 from both dynamites with a CaCO3 standard; (d) flour from both dynamites with a flour standard. Raman conditions as in Figure 2. Spectral acquisition times: 10 s × 10 acquisitions (NC and flour), 1 s × 30 acquisitions (sawdust), and 1 s × 10 acquisitions (CaCO3). (∗) 1042 cm−1 band corresponding to AN. Several bands are labeled for clarity.

■

CONCLUSIONS The major components, EGDN and AN, of two types of dynamites were directly identified by confocal Raman spectroscopy. Additionally, the use of Raman mapping to know their distribution in the sample was proposed. Several minor components of the sample (NC, CaCO3, sawdust, and flour) were also identified using this vibrational technique and a previous fast and easy sample preparation. The proposed method has the advantage of fast identification of most compounds present in the dynamites by using only one analytical technique. However, although the proposed approach is useful for identifying several dynamite components, the dynamite stabilizers cannot be identified by this approach. Also

for such heterogeneous solid samples, quantitative compositions will not be easy to determine. The results obtained in this study represent a breakthrough in the chemical identification of dynamites, which could be extremely useful mainly in forensic laboratories for identification purposes as well as in the explosive industry to control the manufacturing process and final products.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: + 34 91 885 64 31. Notes

The authors declare no competing financial interest. 2599



■

Technical Note

ACKNOWLEDGMENTS We thank the Ministry of Science and Innovation for Project CTQ2008-00633-E. M.L.-L. thanks the University of Alcalá for a predoctoral grant.

■

REFERENCES

(1) Alfred Bernhard Nobel. In Encyclopedia Britannica. Retrieved from http://www.britannica.com/EBchecked/topic/416842/AlfredBernhard-Nobel. (2) Zukas, J. A.; Walters, W. Explosive Effects and Applications; Springer: New York, 1998. (3) Saiz, J.; Bravo, J. C.; Avila, E.; Torre, M.; Garcia-Ruiz, C. J. Sep. Sci. 2011, 34, 3353−3358. (4) Zhao, X.; Yinon, J. J. Chromatogr., A 2002, 977, 59−68. (5) Holmgren, E.; Ek, E.; Colmsjö, A. J. Chromatogr., A 2012, 1222, 109−115. (6) Piccin, E.; Dossi, N.; Cagan, A.; Carrilho, E.; Wang, J. Analyst 2009, 134, 528−532. (7) Hutchinson, J. P.; Evenhuis, C. J.; Johns, C.; Kazarian, A. A.; Breadmore, M. C.; Macka, M.; Hilder, E. F.; Guijt, R. M.; Dicinoski, G. W.; Haddad, P. R. Anal. Chem. 2007, 79, 7005−7013. (8) Hilton, C. K.; Krueger, C. A.; Midey, A. J.; Osgood, M.; Wu, J.; Wu, C. Int. J. Mass Spectrom. 2010, 298, 64−71. (9) Chen, C.; Dong, C.; Du, Y.; Cheng, S.; Han, F.; Li, L.; Wang, W.; Hou, K.; Li, H. Anal. Chem. 2010, 82, 4151−4157. (10) Oxley, J. C.; Smith, J. L.; Kirschenbaum, L. J.; Marimganti, S.; Vadlamannati, S. J. Forensic Sci. 2008, 53, 690−693. (11) Mattarozzi, M.; Bianchi, F.; Bisceglie, F.; Careri, M.; Mangia, A.; Mori, G.; Gregori, A. Anal. Bioanal. Chem. 2011, 399, 2741−2746. (12) Benson, S. J.; Lennard, C. J.; Maynad, P.; Hill, D. M.; Andrew, A. S.; Roux, C. Sci. Justice 2009, 49, 73−80. (13) Mou, Y.; Rabalais, J. W. J. Forensic Sci. 2009, 54, 846−850. (14) Ali, E. M. A.; Edwards, H. G. M.; Hargreaves, M. D.; Scowen, I. J. J. Raman Spectrosc. 2009, 40, 144−149. (15) Ali, E. M. A.; Edwards, H. G. M.; Scowen, I. J. Talanta 2009, 78, 1201−1203. (16) Ali, E. M. A.; Edwards, H. G. M.; Scowen, I. J. J. Raman Spectrosc. 2009, 40, 2009−2014. (17) Zachhuber, B.; Ramer, G.; Hobro, A.; Chrysostom, E.; Lendl, B. Anal. Bioanal. Chem. 2011, 400, 2439−2447. (18) Tuschel, D. D.; Mikhonin, A. V.; Lemoff, B. E.; Asher, S. A. Appl. Spectrosc. 2010, 64, 425−432. (19) Riskin, M.; Ben-Amram, Y.; Tel-Vered, R.; Chegel, V.; Almong, J.; Willner, I. Anal. Chem. 2011, 83, 3082−3088. (20) Moros, J.; Lorenzo, J. A.; Lucena, P.; Tobaria, M.; Laserna, J. J. Anal. Chem. 2010, 82, 1389−1400. (21) Fernández, M. A.; López-López, M.; Torre, M.; García-Ruiz, C. Trend Anal. Chem. 2011, 30, 1740−1755. (22) Castro, K.; Fdez-Ortiz de Vallejuelo, S.; Astondoa, I.; Goñi, F. M.; Madariaga, J. M. J. Raman Spectrosc. 2011, 42, 2000−2005. (23) López-López, M.; Ferrando, J. L.; García-Ruiz, C. Anal. Chim. Acta 2012, 717, 92−99. (24) Moore, D. S.; McGrane, S. D. J. Mol. Struct. 2003, 661−662, 561−566. (25) Patnaik, P. A Comprehensive Guide to the Hazardous Properties of Chemical Substances; John Wiley & Sons: Hoboken, NJ, 2007.

2600


Dynamite Analysis by Raman Spectroscopy As a Unique Analytical Tool

Recommend Documents