Ammunition Identification by Means of the Organic Analysis of

Article pubs.acs.org/ac

Ammunition Identification by Means of the Organic Analysis of Gunshot Residues Using Raman Spectroscopy María López-López,†,‡ Juan José Delgado,§ and Carmen García-Ruiz*,†,‡ †

University Institute of Research in Police Sciences and ‡Department of Analytical Chemistry, 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: The ability to unequivocally identify a gunshot residue (GSR) when a firearm is discharged is a very important and crucial part of crime scene investigation. To date, the great majority of the analyses have focused on the inorganic components of GSR, but the introduction of “lead-free” or “nontoxic” ammunitions makes it difficult to prevent false negatives. This study introduces a fast methodology for the organic analysis of GSR using Raman spectroscopy. Six different types of ammunition were fired at short distances into cloth targets, and the Raman spectra produced by the GSR were measured and compared with the spectra from the unfired gunpowder ammunition. The GSR spectrum shows high similarity to the spectrum of the unfired ammunition, allowing the GSR to be traced to the ammunition used. Additionally, other substances that might be found on the victim's, shooter's, or suspect's clothes and might be confused with GSR, such as sand, dried blood, or black ink from a common ballpoint pen, were analyzed to test the screening capability of the Raman technique. The results obtained evidenced that Raman spectroscopy is a useful screening tool when fast analysis is desired and that little sample preparation is required for the analysis of GSR evidence.

G

Basically, chemical analysis of GSR can be divided into two areas, inorganic and organic.1,10 To date, the inorganic analysis of GSR has been much more widely investigated. 10 Components such as antimony and barium from the cartridge primers and lead and copper contained in the bullets with smaller amounts of copper, zinc, and nickel can be accurately revealed by means of many analytical techniques.6,11 Scanning electron microscopy coupled with energy dispersive spectroscopy (SEM−EDS) was acknowledged as the most specific method for the analysis of inorganic GSR.2,8,10,12−16 In fact, SEM is able to identify GSR particles on the bais of their characteristic morphology and elemental composition.11 However, the introduction of “lead-free” or “nontoxic” ammunitions makes it difficult to identify GSR, as the particles found are not unique anymore.7,8,15 To overcome this problem, the forensic and scientific communities have proposed several alternatives. Thus, Távora et al.16 suggested the introduction of high photoluminescent markers in ammunition to allow visual detection of GSR under UV light. Although this solution is inexpensive, it requires a change in the present production of ammunition and proves useless for lead-free ammunition already manufactured. Therefore, the study of the organic components of GSR presents an increasing interest,8 to provide complementary information that may strengthen the evidential

unshot residues (GSRs), also known as cartridge discharge residues or firearm discharge residues, are a set of burned and unburned particles from the propellant, primer components, and metals contained in the projectile (e.g., bullet, bullet jacket, cartridge case) and in the gun barrel when a gun is fired.1−5 Therefore, GSRs are made of a complex mixture of organic compounds (originated from the propellant and the weapon lubricants and some byproduct) as well as inorganic compounds (such as nitrates, nitrites, and metallic particles originated from the primer as well as from the weapon barrel itself).2 When a weapon is fired, some components of the ammunition are vaporized and then solidified into particles with diameter sizes ranging from 0.5 to 100 μm. These particles can be transferred to the shooter, the weapon, the victim, or the surrounding environment.3,6 The use of analytical techniques for the detection and identification of GSR has become a common practice in modern crime scene investigation. GSR detection and identification may provide valuable forensic information to estimate firing distances, to identify bullet holes, and to determine whether a person has been in contact with a surface exposed to GSR and/or has been in the neighborhood during the shooting.6−8 Techniques for examining GSR have evolved from simple color tests (e.g., paraffin cast or dermal nitrate test, sodium rhodizonate test, Walker test, Griess test, Marshall and Tewari test, Lunge reagent, etc.), which are only presumptive in nature and are not specific to GSR,9 to modern analytical methods. Recently, Dalby et al.1 reviewed the scientific literature about GSR formation, distribution, sample collection, and analysis. © 2012 American Chemical Society

Received: December 6, 2011 Accepted: March 15, 2012 Published: March 15, 2012 3581

dx.doi.org/10.1021/ac203237w | Anal. Chem. 2012, 84, 3581−3585

Analytical Chemistry

Article

value of a sample, offering discrimination between GSR and other external residues.10,17 Raman spectroscopy is a noncontact, nondestructive technique able to provide immediate and useful information about the identity of the sample requiring minimal or no sample preparation, features which have fostered the use of this technique in the forensic field over the past few years. In the field of GSR analysis by Raman spectroscopy, literature suggests that the Raman technique has only been used to detect a variety of different metal anions contained in GSR: particles consisting of oxide mixtures, sulfate, and carbonate (along with carbon) such as BaCO3, PbO, PbSO4, and iron oxides.6 Thus, the present scientific research is focused on the identification of the organic part of GSR by Raman spectroscopy, which has never been reported for this technique. This new application of Raman spectroscopy allows additional information to be otained from GSR organic components and provides an analytical tool to deal with the new challenge posed by the incoming lead-free ammunition.

Figure 1. Target cloths after the ammunitions (a) Super X and (b) Rem were fired.

firing distance was set at approximately 1 m. The second cartridge of each batch was kept unfired and was used to collect the unfired propellant. Bullets and propellants from unfired cartridges were removed using a commercial bullet puller carefully cleaned after each use to avoid cross-contamination. Additionally, a sea sand particle (Panreac, Barcelona, Spain), a blood sample collected from a volunteer, and a black ink sample obtained from a BIC crystal medium ballpoint pen sold by Société Bic of Clichy (Clichy, France) were analyzed. Sample Preparation. Approximately 30 mg of unburned propellant powder from the cartridge of each ammunition was dissolved in 1.5 mL of methyl ethyl ketone (samples were placed in an ultrasonic bath for 30 min at 35 °C). Separation of graphite from the remaining components present in the gunpowders was achieved by centrifugation at 3450g for 5 min. Target cloths were placed under a Leica Macroscope Wild 420 M (Leica Microsistemas SLU, Barcelona, Spain), and 5−10 GSR particles were collected using metal tweezers previously cleaned with methyl ethyl ketone. Particles were introduced into a tube, and 5−10 drops of methyl ethyl ketone were added to dissolve the GSR particles. A drop of each of these solutions was placed on microscope slides with central cavity wells, previously cleaned with methyl ethyl ketone. The spectrum was recorded when the complete evaporation of the solvent was achieved. Instrumentation. A Thermo Scientific DXR Raman microscope (Waltham, MA) controlled by the Thermo Scientific Omnic for dispersive Raman 8.1 software was used. Measurements were taken using a laser emitting at 532 nm (the laser power on the sample was 8.0 mW) and a confocal pinhole size of 50 μm. The microscope was set to 50× magnification. Ten spectra of 10 s exposure for both fired and unfired ammunition samples were measured. Standards (diphenylamine, 2-nitrodiphenylamine, 4-nitrodiphenylamine, N-nitrosodiphenylamine, ethyl centralite, methyl centralite, and 2,4dinitrotoluene) were measured for 1 s 10 times. Spectral acquisition times for a sand particle, black ink, nitrocellulose, and dried blood were 10 spectra of 0.5 s, 10 spectra of 2 s, 5 spectra of 60 s, and 10 spectra of 5 s, respectively. Background and fluorescence corrections were applied for all the spectra.

■

EXPERIMENTAL SECTION Standards and Samples. 2,4-Dinitrotoluene (97%, m/m), ethyl centralite (99%, m/m), and methyl centralite (99%, m/ m) were purchased from Sigma-Aldrich (Zwijndrecht, The Netherlands). 2-Nitrodiphenylamine, 4-nitrodiphenylamine, Nnitrosodiphenylamine, and diphenylamine standards were kindly provided by TNO Defence, Safety and Security (Rijswijk, The Netherlands). Nitrocellulose from smokeless gunpowder provided by Acuartelamiento San Juan del Viso (Madrid, Spain) was obtained using an isolation protocol.18 Six different types of ammunition provided by the Ballistic Department of Guardia Civil (Madrid, Spain) were used (see Table 1). Two cartridges from the same ammunition batch were used. One of the cartridges of each batch was fired over white cotton cloths of 20 cm × 20 cm size fixed in cardboard to obtain its GSR (examples of target cloths in Figure 1). Shoots were performed at the shooting range of the Criminalistic Service of Guardia Civil using suitable caliber firearms without previous cleaning to provide a more realistic scenario. The Table 1. Ammunition Analyzed Using Raman Spectroscopy before and after Firing

■

RESULTS AND DISCUSSION

GSRs are particles originated from materials from all components of the ammunition, but the major contribution of the organic components lies in the components of the propellant present in the original cartridge. Nitrocellulose is the main constituent of smokeless gunpowders used as a propellant in ammunition. Nitrocellulose is the unique active component of single-base gunpowders, whereas it is mixed with nitroglycerin in double-base gunpowders. Nitroguanidine can also 3582



Article

Figure 2. Raman spectra of the gunpowders from two different unfired cartridges. One of them has diphenylamine (DPA) and the other one ethyl centralite (EC) as the stabilizer. Raman spectra of stabilizers are depicted on the right: DPA, N-nitroso-DPA, 2-nitro-DPA, 4-nitro-DPA, and EC. Raman conditions: laser at 532 nm, 8.0 mW, 50× magnification objective lens, confocal pinhole size of 50 μm. Spectral acquisition times: 10 s × 10 acquisition (propellant), 1 s × 10 acquisition (standards).

be added to nitrocellulose and nitroglycerin, although this formulation, characteristic of triple-base gunpowders, is used primarily in rockets and large-caliber military-grade weapons, which are difficult to obtain in the open market.19 Stabilizers such as diphenylamine and its derivatives (e.g., 2-nitrodiphenylamine, 4-nitrodiphenylamine, N-nitrosodiphenylamine, etc.) or centralites, as well as other components as flash inhibitors, such as 2,4-dinitrotoluene, can also be found in the gunpowder composition. As recently reported, the identification of the stabilizer diphenylamine in gunpowder formulations can be performed by Raman spectroscopy.20 Figure 2 compares the Raman spectra of two propellants from two unfired ammunitions; one sample includes diphenylamine in its primary composition as the stabilizer, while the other has ethyl centralite as the stabilizer. Comparison of the Raman spectra of the two gunpowders clearly shows that the band at about 1342 cm−1 is characteristic of gunpowders with diphenylamine in its primary composition. The right part of Figure 2 depicts the spectra obtained for diphenylamine, Nnitrosodiphenylamine, 2-nitrodiphenylamine, 4-nitrodiphenylamine, and ethyl centralite, showing that the band at about 1342 cm−1 is basically due to 2-nitrodiphenylamine. Close-range shots over cloths were fired to obtain small amounts of GSR. As an example, Figure 1 shows the pictures of two pieces of cloths after ammunition was fired over them. Then, although good spectra can be obtained by measuring only one particle directly, to obtain a representative Raman spectrum of the fired propellant for each cartridge, 5−10 particles of GSR were picked with tweezers and dissolved in 5− 10 drops of methyl ethyl ketone. Then the Raman spectrum of the previously dissolved propellant from the unfired cartridge

was measured. A solution of both types of samples was prepared in an attempt to homogenize the sample, allowing in the case of unfired propellants removal of the hardly graphitized surface present in some samples, which makes the measurement of Raman spectra difficult. A measurement of the GSR and gunpowder spectra obtained from different cartridges before and after firing was taken to establish if it was possible to obtain additional information from the GSR organic components and to explore the possibility of linking GSR with suspect ammunition by means of Raman spectroscopy. Figure 3 shows the Raman spectra of six different propellant ammunitions (GFL 38 sp, SB 96+, Super X, RP 45 AUTO, SBT 93+, and Rem) and their respective GSRs. Intensity differences in some bands (∼3100−2800, ∼1200−1000, and ∼700−400 cm−1) were observed between the spectra of the GSRs and their corresponding unfired propellants and also among the different spectra from the same sample. These intensity differences could be due to different parameters that influence the measured Raman spectra, such as the GSR sample collection criteria, the extent of particle burning affecting the composition, or a lack of homogeneity in the propellant grains. In spite of the effects of these uncontrolled parameters on Raman spectra, it should be emphasized that when the GSR spectrum is compared with the gunpowder spectrum of the unfired ammunition, both spectra show similar profile spectra. These findings indicate that when a cartridge is shot, it is possible to find unburned and/or partially burned gunpowder particles with composition very similar to that of the nonfired unburned particles, and a preliminary differentiation of the type of ammunition could be done by analyzing the GSR. Therefor, ammunitions GFL 38 sp, SB 96 +, and Super X can be easily 3583



Article

Figure 3. Comparison among Raman spectra of propellants and their respective GSRs from six different unfired cartridges. Raman conditions as in Figure 2. Acquisition time: 10 spectra of 10 s exposure each.

of a victim, the shooter, or a suspect) and could be confused with GSR. For that purpose a sand particle, dried blood, black ink from a common ballpoint pen, and a GSR particle were directly measured by Raman spectroscopy, placing a white cloth stained with the samples in the Raman equipment. The Raman spectra for the four above-mentioned samples and the spectrum of nitrocellulose are depicted in Figure 4. Remarkably, the spectrum of the GSR is easily distinguished from the other three samples analyzed and shows a strong similarity to the nitrocellulose spectrum; specifically, the bands at around 1665 cm−1 (antisymmetric NO2 stretching), 1370 cm−1 (probably related to the cellulose ring), 1287 cm−1 (symmetric NO2 stretching), and 860 cm−1 (NO stretching mode) are present in both spectra.21

differentiated from ammunitions RP 45 AUTO, SB-T 93+, and Rem (see Figure 3). In the first ammunition set (GFL 38 sp, SB 96 +, and Super X), a characteristic band at about 1340 cm−1 is observed. It can be attributed to the combination of the peaks at 1325 and 1346 cm−1 corresponding to the 2-nitrodiphenylamine and to the peak at 1320 cm−1 present in Nnitrosodiphenylamine and 4-nitrodiphenylamine spectra (shown in Figure 2). In the second ammunition set (RP 45 AUTO, SB-T 93+, and Rem), the characteristic band at about 1340 cm−1 is negligible but the band at about 2970 cm−1 characteristic of ethyl centralite (also displayed in Figure 2) is more marked than in the other ammunition group. To take into account the possible barrel contamination problems present in real cases, fired gun barrels were not cleaned in this study to emulate a more realistic scenario. The results obtained show that particles from previous shots had no significant influence. The screening capability of Raman spectroscopy for the analysis of GSR was also tested by analyzing other substances that could be present at the crime scene (found in the clothes

■

CONCLUSIONS The results obtained in the current study revealed that Raman spectroscopy is a useful screening tool when a fast analysis of GSR evidence is required. The comparison between the 3584



Article

Carol-Visser and TNO (Rijswijk, The Netherlands) for providing standards. We are also grateful for the help provided by the personnel of the Ballistic Department of Guardia Civil in Madrid (Spain) responsible for the execution of the shootings. We thank the Ministry of Science and Innovation for Project CTQ2008-00633-E. M.L.-L. thanks the University of Alcalá for a grant.

■

Figure 4. Raman spectra of a sand particle, dried blood, ink from a black BIC crystal medium ballpoint pen, GSR, and nitrocellulose. Raman conditions as in Figure 2. Spectral acquisition times: 0.5 s × 10 acquisition (sea sand), 5 s × 10 acquisition (dried blood), 2 s × 10 acquisition (black ink), 1 s × 10 acquisition (GSR), and 60 s × 5 acquisition (nitrocellulose).

spectrum of the ammunition propellant and the spectrum of the GSR obtained after firing proves that additional information can be obtained from a GSR particle through Raman spectroscopy. Using this vibrational technique, the particles of fired GSR were easily detected and distinguished from other external residues that can be found at crime scenes. Interestingly, discrimination between two types of ammunition can be easily performed on the basis of their stabilizers. These findings demonstrated the potential of Raman spectroscopy as a complementary analytical technique to SEM-EDS for the forensic analysis of GSR especially to deal with the new challenge posed by the incoming lead-free ammunition. Additionally, the combination of the information obtained with both analytical techniques (after macroscopic observation of the sample) can be very useful in crime scene investigation involving firearms. Although the results obtained in this study represent a breakthrough in GSR detection and identification, further research should be performed in this field. Unequivocal identification of GSR by using chemometrics should be studied by taking into account the parameters which can have an influence on the Raman spectra, such as criteria for the GSR collection, burned particle distribution over cloths according to the shooting distance and/or ammunition type, heterogeneity in ammunition composition, etc. Additionally, a library containing the Raman spectra of the most common ammunition types would be very useful for firearm identification purposes in the field of crime scene investigation.

■

REFERENCES

(1) Dalby, O.; Butler, D.; Birkett, J. W. J. Forensic Sci. 2010, 55, 924− 943. (2) Brozek-Mucha, Z. X-Ray Spectrom. 2007, 36, 398−407. (3) Romolo, F. S.; Margot, P. Forensic Sci. Int. 2001, 119, 195−211. (4) Speers, S. J.; Doolan, K.; McQuillan, J.; Wallace, J. S. J. Chromatogr., A 1994, 674, 319−327. (5) Pun, K. M.; Gallusser, A. Forensic Sci. Int. 2008, 175, 179−185. (6) Stich, S.; Bard, D.; Gros, L.; Wenz, H. W.; Yarwood, J.; Williams, K. J. Raman Spectrosc. 1998, 29, 787−790. (7) Meng, H. H.; Lee, H. C. Forensic Sci. J. 2007, 6, 39−54. (8) Laza, D.; Nys, B.; De Kinder, J.; Kirsch-De Mesmaeker, A.; Moucheron, C. J. Forensic Sci. 2007, 52, 842−850. (9) Schoeble, A. J.; Exline, D. L. Current Methods in Forensic Gunshot Residue Analysis; CRC Press LLC: Boca Raton, FL, 2000. (10) Dalby, O.; Birkett, J. W. J. Chromatogr., A 2010, 1217, 7183− 7188. (11) Brandone, A.; de Ferrari, F.; Pelliza, P.; Signori, M. Forensic Sci. Int. 1999, 47, 289−295. (12) Garofano, L.; Capra, M.; Ferrari, F.; Bizzaro, G. P.; Di Tullio, D.; Dell’Olio, M.; Ghitti, A. Forensic Sci. Int. 1999, 103, 1−21. (13) Steffen, S.; Otto, M.; Niewoehner, L.; Barth, M.; Brożek-Mucha, Z.; Biegstraaten, J.; Horváth, R. Spectrochim. Acta, Part B 2007, 62, 1028−1036. (14) Sharma, S. P.; Lahiri, S. C. Sci. Justice 2009, 49, 197−204. (15) Burleson, G. L.; Gonzalez, B.; Simons, K.; Yu, J. C. J. Chromatogr., A 2009, 1216, 4679−4683. (16) Távora, I.; Geber, A. J.; de Melo, M. A.; Oliveira, M.; Alves, S. Jr. Anal. Chem. 2011, 83, 4720−4723. (17) Wheals, B. B.; Ellison, M. J. Forensic Sci. Int. 1989, 41, 147−162. (18) López-López, M.; Fernández, M. A.; Sáiz, J.; Ferrando, J. L.; Vega, A.; Torre, M.; García-Ruiz, C. Talanta 2010, 81 (2010), 1742− 1749. (19) Scherperel, G.; Reid, G.; Smith, R. Anal. Bioanal. Chem. 2009, 394, 2019−2028. (20) López-López, M.; Ferrando, J. L.; García-Ruiz, C. Anal. Chim. Acta 2012, 717, 92−9. (21) Paris, C.; Coupry, C. J. Raman Spectrosc. 2005, 36, 77−82.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: + 34 91 885 64 31. Web site: www.inquifor.com. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We express our gratitude to Jose Luis Acosta Martinez and Juan Miguel Gonzalez Dávila for their assistance, as well as to Jeroen 3585


Ammunition Identification by Means of the Organic Analysis of

Recommend Documents