J Forensic Sci, January 2015, Vol. 60, No. 1 doi: 10.1111/1556-4029.12666 Available online at: onlinelibrary.wiley.com
TECHNICAL NOTE CRIMINALISTICS
Ciro A. F. O. Penido,1 M.Sc.; Marcos Tadeu T. Pacheco,1 Ph.D.; Renato A. Z^ angaro,1 Ph.D.; and Landulfo Silveira, Jr.,1 Ph.D.
Identification of Different Forms of Cocaine and Substances Used in Adulteration Using Near-infrared Raman Spectroscopy and Infrared Absorption Spectroscopy*
ABSTRACT: Identification of cocaine and subsequent quantification immediately after seizure are problems for the police in developing
countries such as Brazil. This work proposes a comparison between the Raman and FT-IR techniques as methods to identify cocaine, the adulterants used to increase volume, and possible degradation products in samples seized by the police. Near-infrared Raman spectra (785 nm excitation, 10 sec exposure time) and FT-IR-ATR spectra were obtained from different samples of street cocaine and some substances commonly used as adulterants. Freebase powder, hydrochloride powder, and crack rock can be distinguished by both Raman and FT-IR spectroscopies, revealing differences in their chemical structure. Most of the samples showed characteristic peaks of degradation products such as benzoylecgonine and benzoic acid, and some presented evidence of adulteration with aluminum sulfate and sodium carbonate. Raman spectroscopy is better than FT-IR for identifying benzoic acid and inorganic adulterants in cocaine.
KEYWORDS: forensic science, cocaine identification, adulterant, degradation, Raman spectroscopy, Fourier transform–infrared spectroscopy, toxicology
Cocaine is an alkaloid stimulant of the central nervous system extracted from native species of South America, Erythroxylon coca and Erythroxylon novogranatense (1). Common forms of cocaine that are used for drug addiction include freebase powder (for inhalation), hydrochloride powder (for venous injection), freebase paste and crack rock (both for smoking). Cocaine abuse is currently a major global issue in terms of public security and health. Numerous health complications are associated with acute and chronic use of cocaine. There are case reports of strokes resulting from overdose, in which cocaine with higher purity is consumed by users unaware of the drug’s actual concentration (2). Furthermore, various adulterants are mixed with cocaine that are used to increase the volume. These adulterants are often deleterious to health and are mainly consumed along with the drug in the crack form. Discriminating the various forms of cocaine (freebase powder or paste, hydrochloride powder, and crack rock), the drug’s true concentration, and the adulterants used to increase volume is of great interest to the forensic toxicology field as a means of conclusively identifying drugs in cases of drug trafficking (3–5).
1
Biomedical Engineering Institute, Universidade Camilo Castelo Branco UNICASTELO, Parque Tecnologico de S~ao Jose dos Campos, Estrada Dr. Altino Bondesan, 500, Eug^enio de Melo, S~ao Jose dos Campos, SP 12247-016, Brazil. *Supported in part by FAPESP (S~ao Paulo Research Foundation) who granted the Raman instrument (Process no. 2009/01788-5). Received 14 Feb. 2013; and in revised form 25 Sept. 2013; accepted 23 Nov. 2013. © 2014 American Academy of Forensic Sciences
Current methods employed in forensic toxicology, such as gas chromatography and flame ionization detection (GC-FID), aim to detect and identify illicit drugs such as cocaine, amphetamines, MDMA (ecstasy), opiates, barbiturates, and benzodiazepines. Despite their high reliability, these methods are destructive, time-consuming, and do not allow reexamination of the evidence (6,7). Nondestructive and rapid analysis of drugs of abuse could be accomplished through the use of vibrational spectroscopic techniques, especially infrared absorption spectroscopy (FT-IR) and Raman spectroscopy (4,8,9). The advantages lie in the possibility of rapid analysis of trace samples from a crime scene and after seizing drugs, without the destruction of evidence. Thus, these techniques may become a powerful instrument in the fight against drug trafficking (1). Raman spectroscopy is a technique based on the inelastic scattering of laser light molecules and has been used for qualitative and quantitative analysis of drugs of abuse and their adulterants. One can perform a molecular analysis in seconds, without the use of toxic chemicals and without direct contact with the sample (1,9–11). Using near-infrared excitation, most sample fluorescence is minimized. Raman bands of cocaine and common adulterants have been reported in the literature (1,12–16). FT-IR is a technique based on absorption in the infrared region by a molecule due to variations in the dipole moment during vibration. Together with an accessory used to perform measurements by attenuated total reflectance (ATR), this method has been employed for the analysis of trace compounds and contaminants (17). Recent studies using FT-IR have included the 171
172
JOURNAL OF FORENSIC SCIENCES
identification of cocaine (18–23), some of them using ATR (18,20). Lu et al. (19) evaluated herbal medicines suspected of adulteration using FT-IR. Maharaj (20) evaluated samples of seized cocaine and correlated quantification by the GC-FID technique with that obtained by FT-IR-ATR, highlighting the advantages of using the latter in routine forensics due to the minimal sample preparation required compared to GC-FID. This work aimed to evaluate the effectiveness of the nearinfrared Raman spectroscopy and FT-IR-ATR spectroscopy techniques in identifying spectral differences between the different forms of cocaine—freebase powder, paste, and crack rock, as well as hydrochloride powder—seized by the Brazilian police and stored in evidence files. These techniques were also used to identify the most important peaks of commonly used adulterants used to increase volume such as aluminum sulfate, sodium carbonate, sodium bicarbonate, magnesium trisilicate, starch, caffeine, benzocaine, and lidocaine. Materials and Methods Samples of cocaine were obtained in different forms as seized by the police, that is, powder, paste, and crack rock, from the evidence files of the Toxicology Laboratory of the Technical Police State of Amapa, Amazon Region, Brazil. One powder sample was submitted to a protocol for purification adopted by the Brazilian Federal Police (Technical Instruction no. 006/2006GAB/DITEC, Federal Police, Brazil), consisting of recrystallization of freebase cocaine and submission of this sample to a series of evaluations by GC-MS using a Cerilliant Analytical Reference Standard, Reference C-008, Lot FC120204-01A, CAS 50-36, UV-Vis and FT-IR spectroscopy and HPLC to ensure purity. This sample was then considered a reference standard. Additionally, certified standards of selected adulterants were obtained, that is, lidocaine hydrochloride, caffeine, benzocaine, aluminum sulfate, sodium carbonate, sodium bicarbonate, magnesium trisilicate, and starch, which are usually found in seizures and commonly used as adulterants. The cocaine samples, cocaine standard, and adulterant samples underwent Raman spectroscopy and FT-IR-ATR spectroscopy. The Raman spectra were obtained with a dispersive Raman spectrometer, described elsewhere (24), composed of a diode laser (model L4830S; Microlaser Systems, Inc., Garden Groove, CA, USA) with an excitation wavelength of 830 nm and 100 mW power, a spectrograph with a grating of 600 grooves/mm (model 250IS; Chromex, Inc., San Jose, CA, USA), and a deep-depletion CCD camera (model LNCCD 1024x256 CCD and model ST130 controller; Princeton Instruments, Inc., Trenton, NJ, USA), producing spectra in the range of 800–1800 cm 1 with a resolution of about 8 cm 1. The exposure time for collecting each spectrum was 10 sec. Prior to spectrum collection, each sample was placed in an aluminum sample holder with wells approximately 5 mm in diameter and 5 mm deep. The same samples of cocaine and adulterants were also subjected to FT-IR-ATR with a diamond crystal (Nicolet IS10; Thermo Scientific, Inc., Tewksbury, MA, USA) and EverGlo IR source in the mid/far infrared range, resulting in a spectral resolution of 4 cm 1 with 16 scans in the range of 600–3500 cm 1. The Raman spectrometer was calibrated using standard procedures described elsewhere (24), using known peaks of naphthalene for wavenumber calibration and the calibrated irradiance pattern of a spectral lamp for intensity response. Each sample’s autofluorescence spectrum was removed by fitting and subtracting a fifth-order polynomial function to the gross spectrum. These
procedures were performed using a routine developed in Matlab 4.2. Cosmic rays were removed manually using Excel 2003 software. The FT-IR spectrometer was calibrated using OMNIC software, which provides automatic control of the spectrometer. The Raman and FT-IR spectra of the different forms of cocaine and adulterants were then plotted in order to make comparisons to the literature and tentatively attribute the most important bands present in the spectra of such substance, in order to find spectral features that can be used to discriminate different forms of cocaine and also identify peaks characteristic of adulterants that may be present in cocaine samples.
Results Raman Spectroscopy Figure 1 shows the Raman spectra of various forms of cocaine: paste (Fig. 1A,B), crack (Fig. 1C), hydrochloride powder (Fig. 1D), and freebase reference powder (Fig. 1E). The peak positions of the Raman bands of the different forms of cocaine and the respective attributions of these band vibrations are shown in Table 1. The samples of the powdered forms showed spectra for two different forms of cocaine: the freebase, insoluble in water, and the hydrochloride, the water-soluble form. As seen in Fig. 1, the hydrochloride form (Fig. 1D) presented several peaks at the same positions as those found in the freebase form (Fig. 1E), with observed differences in the peaks at 1036, 1165, 1183, 1319, and 1735 cm 1, which are seen in the freebase and missing or shifted in the hydrochloride. On the other hand, the hydrochloride form showed peaks at 1026 and 1207 cm 1. The peaks at 1605 and 1712 cm 1 shifted their positions in the hydrochloride form: The first one shifted to 1601 cm 1, and the second one, to 1716 cm 1. There were also differences in the intensities of the peaks attributed to the tropanic ring, with the peaks at 848 and 898 cm 1 exhibiting lower intensities in the hydrochloride form. Despite a lower signal-to-noise ratio for the spectrum of the paste (Fig. 1A) due to the strong fluorescence emission from the yellow paste, the spectral profile and peak positions were the same as the spectrum of the reference cocaine (Fig. 1E). Some spectral differences were found related to the presence of a peak at 1639 cm 1 in the samples shown in Fig. 1B,C compared to 1E, suggestive of the presence of benzoic acid (from cocaine degradation) (27). These samples also presented spectral features at 1605 and 1712 cm 1 with different intensities compared to the sample in Fig. 1E, with the intensity increased for the peak at 1605 cm 1 and decreased for the peak at 1712 cm 1, which have been characterized as reflecting cocaine degradation to benzoylecgonine (25). Figure 2 shows the Raman spectra of different compounds commonly found in cocaine seizures and used as adulterants. Table 1 also shows the peak positions of the main Raman bands of these adulterants and their assignments. These bands are seen at characteristic positions that would, in turn, make it possible to identify these compounds in the spectrum of adulterated cocaine. Figure 3 shows the Raman spectra of different samples of freebase cocaine (Fig. 3A–D) and a freebase cocaine reference powder (Fig. 3E). Considerable similarities were observed in the peak positions and intensities of all the samples, in accordance with recent literature (12,14,16,25,26). As observed in Fig. 1B, C, the spectra in Fig. 3A–D showed spectral features suggestive of degradation to benzoic acid and benzoylecgonine (a new peak at 1639 cm 1, an increase in the peak at 1605 cm 1 and a decrease in the peak at 1712 cm 1).
PENIDO ET AL.
.
IDENTIFICATION OF COCAINE AND ADULTERANTS BY RAMAN AND IR
173
TABLE 1––Peak positions of the main Raman and FT-IR bands of different forms of cocaine and adulterants, with their tentative vibration assignments based on recent literature.
Compound Cocaine: freebase, hydrochloridric, paste, and crack
Raman Peak Position (cm 1) 848, 874, 898 1004
1683
C-C stretching (tropane) Symmetric stretching of aromatic ring breathing Asymmetric stretching of aromatic ring C-N stretching C-N stretching C-N stretching C-H twisting Asymmetric CH3 deformation C=C stretching - aromatic ring C=O symmetric stretching - carbonyl C=O asymmetric stretching - carbonyl C-C stretching (tropane), change the polarizability by protonated nitrogen Asymmetric stretching - aromatic ring (freebase at 1036 cm 1) C-N stretching shifted by the protonated nitrogen (freebase at 1183 cm 1) Asymmetric CH3 deformation (freebase at 1453 cm 1) C=C stretching – aromatic ring (freebase at 1605 cm 1) C=C stretching – aromatic ring (freebase at 1605 cm 1) C=O symmetric stretching – carbonyl (freebase at 1712 cm 1) O-C-O asymmetric stretching of sodium carbonate (contaminant or adulterant) C=O stretching – benzoic acid C=O stretching – benzoic acid C=O symmetric stretching – carbonyl C-C stretching C-C stretching C-C stretching – aromatic ring and C-N stretching C-N stretching C-C and C-N stretching and aromatic ring stretching C-N stretching C-N stretching and N-H bending Stretching and bending of aromatic ring C-C stretching H-C=N bending C-N stretching C-N stretching C-N stretching C=C stretching C=O stretching C=O stretching C-O stretching In-plane H-C-H bending C-C, C-N and C-O stretching and aromatic ring stretching Stretching and bending of aromatic ring and NH2 scissoring C=O stretching
1052
Si-O-Si stretching
1069 1076
C-O stretching C-O stretching
1036
Peaks exclusive of hydrochloridric cocaine
1165 1183 1279 1319 1453 1605 1712 1735 848, 898† 1026† †
1207
1462† 1596† 1601† 1716† Peaks exclusive of paste cocaine Peaks exclusive of crack cocaine Lidocaine
1069* 1639* 1639* 1719 954 991 1044 1094 1275
Caffeine
Benzocaine
1387 1448 1596 1073 1242 1287 1330 1363 1602 1658 1700 862 1173 1282 1605
Talc (magnesium trisilicate) Sodium carbonate
Attribution (1,4,15,25,26,29–33)
FT-IR Peak Position (cm 1)
Attribution (4,25,29,33)
726 1040
Out-of-plane C-H deformation C-O and C-N stretching
1110
C-O and C-N stretching
1280 1453 1710‡ 1740‡ 2800-3030‡
C-O and C-N stretching C=C stretching - aromatic ring C=O stretching C=O stretching C-H sp3 and sp2 vibrations
736‡
Out-of-plane C-H deformation (freebase at 726 cm 1)
1030‡
1489‡
C-O and C-N stretching (freebase at 1040 cm 1) C-O and C-N stretching (freebase at 1280 cm 1) C=C stretching of aromatic ring
1713‡
C=O stretching (freebase at 1710 cm 1)
1732‡
C=O stretching (freebase at 1740 cm 1)
‡
1270
2300-2900‡
+
N-H stretching
788 1443 1480
Out-of-plane C-H aromatic ring bending C-N stretching C-N stretching
1548 1657
C=C stretching – aromatic ring C=O stretching – amide I
3390
NH2 stretching
745 1242 1551 1658 1696 2957 3116
Out-of-plane C-H aromatic ring bending C-N stretching C=N stretching C=N stretching C=O stretching C-H stretching =C-H stretching
775 1170 1290
Out-of-plane C-H aromatic ring bending C-H bending C-O stretching
1320
C-N stretching
1600 1640 1683 3226 3347 3428 1017 3680 850 1437
C-C stretching – aromatic ring N-H bending C=O stretching O-H stretching N-H stretching N-H stretching Si-O-Si stretching O-H stretching Out-of-plane CO3 angular deformation Asymmetric C=O stretching
174
JOURNAL OF FORENSIC SCIENCES TABLE 1—Continued.
Compound Sodium bicarbonate
Aluminum sulfate Starch
Raman Peak Position (cm 1) 1046 1268
993 864 940 1127 1262 1343 1385 1462
Attribution (1,4,15,25,26,29–33) C-O stretching O-C-O symmetric stretching
SO4 stretching C-C stretching C-C stretching (carbohydrates) C-C stretching (carbohydrates/proteins) In-plane =C-H deformation (carbohydrates) CH2/CH3 bending (carbohydrates) CH3 bending (carbohydrates) CH2 bending (carbohydrates/proteins)
FT-IR Peak Position (cm 1) 700 834 1000 1300-1400 1619 1076 1105 1670 1010 1660 2940 3300
Attribution (4,25,29,33) CO2 angular deformation Out-of-plane CO3 bending C-OH stretching CO2 symmetric stretching CO2 asymmetric stretching SO4 stretching SO4 stretching O-H stretching C-H bending C-OH stretching C-H stretching C-OH stretching
*Indicates Raman peaks of adulterants or degradation products in the samples of freebase cocaine. † Indicates Raman peaks with differences in the position and intensity caused by protonated nitrogen due to the presence of hydrochloric acid in hydrochloridric cocaine compared to freebase cocaine. ‡ Indicates FT-IR peaks with differences in the position and intensity caused by protonated nitrogen due to the presence of hydrochloric acid in hydrochloridric cocaine compared to freebase cocaine.
Figure 4 shows Raman spectra of different freebase cocaine samples in the paste form (Fig. 4A–D) compared to the freebase cocaine reference powder (Fig. 4E). In these paste samples (4A–D), peaks were observed at 983 and 1069 cm 1, which indicate adulteration by aluminum sulfate and sodium carbonate, respectively. These spectra also presented characteristic peaks of degradation into benzoic acid and benzoylecgonine, represented by a new peak at 1639 cm 1, an increase in the peak at 1605 cm 1 and a decrease in the peak at 1712 cm 1.
forms of cocaine and the respective attributions of these band vibrations are shown in Table 1. The spectral differences in the hydrochloride form (Fig. 5D) compared to the free base form (standard) (Fig. 5E) were mainly in the 2300–2900 cm 1 range, attributable to +N-H stretching resulting from the coordinate
FT-IR Spectroscopy Figure 5 shows FT-IR spectra of cocaine as a powder, paste, and crack. The peak positions of the IR bands of the different
FIG. 1––Raman spectra of cocaine samples in different forms: (A) yellow freebase paste, (B) white freebase paste, (C) crack rock, (D) hydrochloride cocaine powder, and (E) freebase cocaine reference powder, with their characteristic peaks labeled according to Table 1. The presence of the peak at 1639 cm 1 in samples B and C suggests the occurrence of benzoic acid, a degradation product of cocaine. Also, the differences in intensities of the peaks at 1605 and 1712 cm 1 indicate degradation into benzoylecgonine.
FIG. 2––Raman spectra of samples of different adulterants: (A) lidocaine, (B) caffeine, (C) benzocaine, (D) talc (magnesium trisilicate), (E) starch (wheat flour), (F) aluminum sulfate, (G) sodium bicarbonate, and (H) sodium carbonate (barilla), with their characteristic peaks labeled according to Table 1.
PENIDO ET AL.
.
IDENTIFICATION OF COCAINE AND ADULTERANTS BY RAMAN AND IR
175
FIG. 3––Raman spectra of freebase cocaine powders from different seizures. The presence of the peak at 1639 cm 1 in samples A to D suggests the occurrence of benzoic acid, a degradation product of cocaine. Also, the differences in the intensities of the peaks at 1605 and 1712 cm 1 indicate degradation into benzoylecgonine. Freebase cocaine reference E does not show such characteristics.
FIG. 5––FT-IR spectra of samples of different forms of cocaine: (A) yellow freebase paste, (B) white freebase paste, (C) crack rock, (D) hydrochloride cocaine powder, and (E) freebase cocaine reference powder, with their characteristic peaks labeled according to Table 1.
FIG. 4––Raman spectra of freebase cocaine in the paste forms (A to D) compared to the spectra of freebase cocaine reference powder (E), with the suggestive presence of adulterants with peaks at 983 and 1069 cm 1. The differences in the intensities of the peaks at 1605 and 1712 cm 1 and the presence of the peak at 1639 cm 1 suggest benzoylecgonine and benzoic acid, respectively.
covalent bond of the cocaine with HCl, the double peak at 1710 and 1740 cm 1, which appeared shifted closer to each other (1713 and 1732 cm 1, respectively), the peak at 1489 cm 1, which showed greater intensity, and peak shifts from 726 to 736 cm 1, 1040 to 1030 cm 1, and 1280 to 1270 cm 1. Several peaks showed reduced intensity in the hydrochloride form, especially in the ranges between 700 and 900 cm 1 and 1100 and 1300 cm 1. Figure 6 shows FT-IR spectra of different compounds commonly found in cocaine seizures and used in its adulteration. Table 1 also shows the peak positions of the main Raman bands of these adulterants and their assignments. Figure 7 shows FT-IR spectra of different samples of freebase cocaine. In these spectra, similarities were found in the peak positions and intensities of all the samples, but in the spectrum in Fig. 7E, the presence of bands attributed to benzoylecgonine was verified, with peaks at 1353, 1403, 1597, and 1720 cm 1 (symmetrical stretching of the carboxyl group at 1353 and
1403 cm 1, asymmetrical stretching of the carboxyl group at 1597 cm 1 and stretching of the carbonyl group at 1720 cm 1) (28). The spectra in Fig. 7B,C appeared with broad bands at around 1650 and 3300 cm 1, which could be attributed to water due to moisture in the samples. The bands of the selected contaminants were not found in the FT-IR spectra of the cocaine samples. Discussion Optical techniques such as Raman and FT-IR have been described for use in criminal forensics for the identification of drugs and added adulterants (1,23), drugs dissolved in alcohol (8), drug residues on clothing, fingerprints, and paper currency (10,11,14), drugs in herbal medicines (19), and quantification of cocaine in mixtures of commonly used adulterants (9,12,14), among others. The advantages and benefits include the lack of sample preparation, assessment at the crime scene, and preservation of the samples. In addition, small amounts of the sample can be used; theoretically, as little as a single crystal of a drug may be evaluated (11). Both spectroscopic techniques presented in this work presented specificity regarding the type of cocaine analyzed (freebase or hydrochloride form). In the case of Raman spectroscopy, the differences between the two spectra have been reported in the literature, especially the peak shifts from 1605 to 1601 cm 1 and from 1712 to 1716 cm 1, and differences in the relative
176
JOURNAL OF FORENSIC SCIENCES
FIG. 6––FT-IR spectra of samples of different adulterants: (A) lidocaine, (B) caffeine, (C) benzocaine, (D) talc (magnesium trisilicate), (E) starch (wheat flour), (F) aluminum sulfate, (G) sodium bicarbonate, and (H) sodium carbonate (barilla), with their characteristic peaks labeled according to Table 1.
intensities of peaks at 848 and 898 cm 1 of the tropanic ring (1,12,23,25,26). The latter are mainly due to the presence of the protonated nitrogen in the tropanic ring, which changes the polarizability of the electron cloud, thereby altering the energy of the bonds near the tropanic ring. The differences in the FT-IR spectra of freebase and hydrochloride cocaine reported here were also in accordance with the literature (25,29,30), particularly the increases in the peaks at 1453 and 1489 cm 1 as well as the peaks in the 2300- to 2900-cm 1 region, and the peak shifts from 726 to 736 cm 1, 1040 to 1030 cm 1, 1270 to 1260 cm 1, and the double peak at 1710 and 1740 to 1713 cm 1 and 1732 cm 1. Several studies have indicated the possibility of identifying adulterants in cocaine using vibrational spectroscopy (1,12,14,23,29–33). In this work, the Raman spectra of different forms of freebase cocaine revealed bands suggestive of the presence of sodium carbonate and aluminum sulfate as adulterants and also showed the suggestive presence of degradation products such as benzoic acid and benzoylecgonine. In degradation by hydrolysis, where benzoylecgonine is formed, cocaine loses a
methyl group and forms a hydroxyl (34). This causes a change in the relative intensities of the Raman bands at 1605 and 1712 cm 1 (25). In the degradation of benzoylecgonine, hydrolysis occurs again, releasing benzoic acid. In the Raman spectra, the peaks from this process were at 1003 and 1604 cm 1 (assigned to the aromatic ring) and at 1639 cm 1 (assigned to carbonyl stretching) (27). These spectral changes suggestive of degradation processes, especially the emergence of the peak at 1639 cm 1 and a decrease in peak intensity at 1712 cm 1, were found in practically all samples, many of them with the benzoylecgonine and benzoic acid features. The presence of moisture is not an important factor for Raman spectroscopy due to the weak Raman scattering of water. With regard to the FT-IR technique, previous studies have employed the ATR method for detecting the presence of cocaine and adulterants (20,21,29,33). In this study, peaks suggestive of the presence of contaminants in cocaine were not observed by FT-IR. Unlike Raman, the FT-IR bands of various adulterants were broad and mostly overlapped, with less discriminatory capability in terms of the adulterants. As with Raman, the FT-IR
PENIDO ET AL.
.
IDENTIFICATION OF COCAINE AND ADULTERANTS BY RAMAN AND IR
177
FIG. 7––FT-IR spectra of freebase cocaine powder from different seizures: Samples A and B are pastes with the suggestive presence of moisture, and sample C is a freebase powder with the suggestive presence of moisture. Samples D and E are crack rocks, with the latter presenting features suggestive of degradation to benzoylecgonine (D). The freebase cocaine reference powder (F) does not show such characteristics.
spectra showed evidence of sample degradation such as benzoylecgonine, with characteristic peaks at 1350, 1400, and 1599 cm 1 and, unlike freebase or hydrochloride cocaine, a single peak at 1721 cm 1 (4,28). These spectral features were observed in the sample in Fig. 7B (in this case, 1353, 1403, 1597, and 1720 cm 1). Unlike Raman, FT-IR did not show characteristic peaks for benzoic acid, which has its most intense peaks at 935, 1296, and 1689 cm 1 (4,27). FT-IR spectra showed the presence of moisture, mainly in samples of the paste (the OH stretching vibration in the 3300–3700-cm 1 region and the angular deformation of HOH at 1640 cm 1). This technique therefore allows the moisture in samples to be measured, which might be important in evaluating the possible degradation of cocaine by hydrolysis. It has been found that spectral differences between freebase and hydrochloride cocaine can be identified using either the Raman or FT-IR techniques. The selection of optical techniques is preferred due to minimal sample preparation; this technique is economically viable for criminology institutes. A major advantage of Raman spectroscopy is that it can be performed in real time at the crime scene. Such field operation of FT-IR is difficult because of the need for humidity control in the environment. It is of the utmost importance in the forensic field that samples should not be submitted to prior preparation, as the crime scene should be preserved, with the measurements being made on site and samples being evaluated without fragmentation. Also, knowledge of the adulterant used to increase volume is important to assess cocaine trafficking routes. The Raman technique may benefit from the surface plasmon resonance effect, such as using SERS (surface-enhanced Raman scattering) on the same spectrometer (35–39) or by measuring
Raman scattering with the use of nano-structured surfaces or colloidal gold or silver to increase the efficiency of the Raman signal and improve the signal-to-noise ratio for the detection of traces of drugs or metabolites. Cocaine and its degradation products have been evaluated by SERS in urine, sweat, and saliva (1,36–39). Portable Raman spectrometers have been developed to investigate drug trafficking through ports and airports (5). Another advantage of Raman spectroscopy is that drugs can be identified even when wrapped in plastic or in glass bottles (8). By operating in the visible to near-infrared range, it allows the use of fiber optics for remote analysis (3). Conclusion In this study, it has been shown that Raman and FT-IR spectroscopy techniques can be used in a routine to identify forms of freebase and hydrochloride cocaine and the associated degradation product of cocaine, benzoylecgonine. These compounds could be detected by both FT-IR and Raman spectroscopy, and benzoic acid was detectable by the Raman technique, by measuring shifts in the positions and differences in the intensities of the major peaks of cocaine samples. The Raman technique was also able to identify certain adulterants in samples of freebase cocaine, mainly sodium carbonate and aluminum sulfate, which are used to increase the volume of the product. References 1. Carter JC, Brewer WE, Angel SM. Raman spectroscopy for the in situ identification of cocaine and selected adulterants. Appl Spectrosc 2000;54:1876–81.
178
JOURNAL OF FORENSIC SCIENCES
2. Kondziella D, Danielsen ER, Arlien-Soeborg P. Fatal encephalopathy after an isolated overdose of cocaine. J Neurol Neurosur Ps 2007;78:437–8. 3. Ali EMA, Edwards HGM, Hargreaves MD, Scowen IJ. Raman spectroscopic investigation of cocaine hydrochloride on human nail in a forensic context. Anal Bioanal Chem 2008;390:1159–66. 4. Moffat AC, Osselton MD, Widdop B. Clarke’s analysis of drugs and poisons, 3rd edn. London, U.K.: Pharmaceutical Press, 2004. 5. Hargreaves MD, Page K, Munshi T, Tomsett R, Lynch G, Edwards HGM. Analysis of seized drugs using portable Raman spectroscopy in an airport environment – a proof of principle study. J Raman Spectrosc 2008;39:873–80. 6. Bujan L, Fernandez P, Lafuente N, Aldonza M. Comparison of two chromatographic methods for the determination of cocaine and its metabolites in blood and urine. Anal Let 2001;34:2263–75. 7. Da Silva MJ, Dos Anjos EV, Honorato RS, Pimentel MF, Paim AP. Spectrophotometric cocaine determination in a biphasic medium employing flow-batch sequential injection analysis. Anal Chim Acta 2008;629:98–103. 8. Eliasson C, Macleod NA, Matousek P. Non-invasive detection of cocaine dissolved in beverages using displaced Raman spectroscopy. Anal Chim Acta 2008;607:50–3. 9. Penido CA, Pacheco MTT, Silveira L. Quantification of binary mixtures of cocaine and adulterants using dispersive Raman and FT-IR spectroscopy and principal component regression. Instrum Sci Technol 2012;40:441–56. 10. Ali EMA, Edwards HGM, Hargreaves MD, Scowen IJ. In-situ detection of drugs-of-abuse on clothing using confocal Raman microscopy. Anal Chim Acta 2008;615:63–72. 11. Day JS, Edwards HGM, Dobrowski SA, Voice AM. The detection of drugs of abuse in fingerprints using Raman spectroscopy I: latent fingerprints. Spectrochim Acta 2004;60:563–8. 12. Ryder AG, O’Connor GM, Glynn TJ. Identifications and quantitative measurements of narcotics in solid mixtures using near-IR Raman spectroscopy and multivariate analysis. J Forensic Sci 1999;44:1013–9. 13. Baranska M, Proniewicz LM. Raman mapping of caffeine alkaloid. Vib Spectrosc 2008;48(1):153–7. 14. Noonan KY, Beshire M, Darnell J, Frederick KA. Qualitative and quantitative analysis of illicit drug mixtures on paper currency using Raman microspectroscopy. Appl Spectrosc 2005;59:1493–7. 15. Teng D, Ma J, Huang Y, Zhang X, Zheng R. Investigate seawater and seawater anions’ aqueous mixed solution by laser Raman spectroscopy. In: Chen L, Suzuki H, Rudy PT, Zhu N, editors. Semiconductor lasers and applications III. Proc SPIE; 2007 Nov 11. Beijing, China; Bellingham, WA: SPIE, 2008;68240P. 16. Nadine WSK. Analysis of drug samples using Raman microspectroscopy. J Young Invest 2001;5(1):1 http://legacy.jyi.org/volumes/volume5/issue1/ articles/kam.html (accessed July 10, 2013). 17. Ekgasit S, Pattayakorn N, Tongsakul D, Hammacharoen C, Kongyou T. A novel ATR FT-IR microspectroscopy technique for surface contamination analysis without interference of the substrate. Anal Sci 2007;23:863–8. 18. Koulis CV, Reffner JA, Bibby AM. Comparison of transmission and internal reflectance infrared spectra of cocaine. J Forensic Sci 2001;46:822–9. 19. Lu F, Li S, Le J, Chen G, Cao Y, Qi Y, et al. A new method for testing synthetic drugs adulterated in herbal medicines based on infrared spectroscopy. Anal Chim Acta 2007;589:200–7. 20. Maharaj R. Quantitative analysis of cocaine using Fourier Transform Infrared Spectroscopy – Attenuated Total Reflectance: a preliminary investigation. Internet J Third World Med 2009;7(2):8635 http://doi:10. 5580/15e9 (accessed July 10, 2013). 21. Ng PH, Walker S, Tahtouh M, Reedy B. Detection of illicit substances in fingerprints by infrared spectral imaging. Anal Bioanal Chem 2009;394:2039–48.
22. Goh CY, van Bronswijk W, Priddis C. Rapid nondestructive on-site screening of methylamphetamine seizures by Attenuated Total Reflection Fourier Transform Infrared Spectroscopy. Appl Spectrosc 2008;62: 640–8. 23. Ryder AG, O’Connor GM, Glynn TJ. Quantitative analysis of cocaine in solid mixtures using Raman spectroscopy and chemometric methods. J Raman Spectrosc 2000;31:221–7. 24. Silveira L, Sathaiah S, Z^angaro RA, Pacheco MTT, Chavantes MC, Pasqualucci CAG. Correlation between near-infrared Raman spectroscopy and the histopathological analysis of atherosclerosis in human coronary arteries. Lasers Surg Med 2002;30:290–7. 25. Smith FP. Handbook of forensic drug analysis, 1st edn. New York, NY: Academic Press, 2004. 26. Gamot P, Vergoten G, Fleury G. Etude par spectroscopie Raman du chlorhydrate de cocaine. Talanta 1985;32:363–72. 27. Brittain HG. Vibrational spectroscopic studies of cocrystals and salts. 1. The benzamide-benzoic acid system. Cryst Growth Des 2009;9:2492–9. 28. Sigma-Aldrich. Product information. Benzoylecgonine. Catalog Number B3277; http://www.sigmaaldrich.com/etc./medialib/docs/Sigma/Datasheet/ 7/b3277-61k88051dat.Par.0001.File.tmp/b3277-61k88051dat.pdf (accessed July 10, 2013). 29. Nakamoto K. Infrared and Raman spectra of inorganic and coordination compounds. Part A, 5th edn. New York, NY: John Wiley & Sons, 1997. 30. Smith E, Dent G. Modern Raman spectroscopy: a practical approach, 1st edn. West Sussex, U.K.: John Wiley & Sons, 2005. 31. Kloprogge JT, Frost RL. Raman microscopy study of basic aluminum sulfate. J Mater Sci 1999;34:4199–202. 32. Tavender SM, Johnson SA, Balsom D, Parker AW, Bisby RH. The carbonate, Co3 , in solution studied by resonance Raman spectroscopy. Laser Chem 1999;19:311–6. 33. Mayo DW, Millar FA, Hannah RW. Course notes on the interpretation of infrared and Raman spectra. Hoboken, NJ: Wiley Interscience, 2003. 34. Geissh€usler S, Brenneisen R. The content of psychoactive phenylpropyl and phenylpentenyl khatamines in Catha edulis Forsk of different origin. J Ethnopharmacol 1987;19:269–77. 35. Karch SB. Drug abuse handbook, 1st edn. Boca Raton, FL: CRC Press, 1998. 36. Abalde-Cela S, Aldeanueva-Potel P, Mateo-Mateo C, Rodrıguez-Lorenzo L, Alvarez-Puebla RA, Liz-Marzan LM. Surface-enhanced Raman scattering biomedical applications of plasmonic colloidal particles. J R Soc Interface 2010;7:S435–50. 37. Sobrido MS, Lorenzo LR, Abalde SL, Fernandez AG, Duarte MAC, Puebla RAA, et al. Label-free SERS detection of relevant bioanalytes on silver-coated carbon nanotubes: the case of cocaine. Nanoscale 2009;1:153–8. 38. Shende C, Inscore F, Maksymiuk P, Farquharson S. Ten-minute analysis of drugs and metabolites in saliva by surface-enhanced Raman spectroscopy. In: Cullum BM, Carter JC, editors. Smart medical and biomedical sensor technology III. Proc SPIE; 2005 Oct 23. Boston, MA; Bellingham, WA: SPIE, 2005;60070S. 39. Kim K, Shin KS. Surface-enhanced Raman scattering: a powerful tool for chemical identification. Anal Sci 2011;27:775–83. Additional information and reprint requests: Landulfo Silveira, Jr, Ph.D. Professor Universidade Camilo Castelo Branco – UNICASTELO Parque Tecnologico de S~ao Jose dos Campos Estr. Dr. Altino Bondesan, 500, Eug^enio de Melo S~ao Jose dos Campos, SP, Brazil, 12247-016 E-mail: [email protected]
TECHNICAL NOTE CRIMINALISTICS
Ciro A. F. O. Penido,1 M.Sc.; Marcos Tadeu T. Pacheco,1 Ph.D.; Renato A. Z^ angaro,1 Ph.D.; and Landulfo Silveira, Jr.,1 Ph.D.
Identification of Different Forms of Cocaine and Substances Used in Adulteration Using Near-infrared Raman Spectroscopy and Infrared Absorption Spectroscopy*
ABSTRACT: Identification of cocaine and subsequent quantification immediately after seizure are problems for the police in developing
countries such as Brazil. This work proposes a comparison between the Raman and FT-IR techniques as methods to identify cocaine, the adulterants used to increase volume, and possible degradation products in samples seized by the police. Near-infrared Raman spectra (785 nm excitation, 10 sec exposure time) and FT-IR-ATR spectra were obtained from different samples of street cocaine and some substances commonly used as adulterants. Freebase powder, hydrochloride powder, and crack rock can be distinguished by both Raman and FT-IR spectroscopies, revealing differences in their chemical structure. Most of the samples showed characteristic peaks of degradation products such as benzoylecgonine and benzoic acid, and some presented evidence of adulteration with aluminum sulfate and sodium carbonate. Raman spectroscopy is better than FT-IR for identifying benzoic acid and inorganic adulterants in cocaine.
KEYWORDS: forensic science, cocaine identification, adulterant, degradation, Raman spectroscopy, Fourier transform–infrared spectroscopy, toxicology
Cocaine is an alkaloid stimulant of the central nervous system extracted from native species of South America, Erythroxylon coca and Erythroxylon novogranatense (1). Common forms of cocaine that are used for drug addiction include freebase powder (for inhalation), hydrochloride powder (for venous injection), freebase paste and crack rock (both for smoking). Cocaine abuse is currently a major global issue in terms of public security and health. Numerous health complications are associated with acute and chronic use of cocaine. There are case reports of strokes resulting from overdose, in which cocaine with higher purity is consumed by users unaware of the drug’s actual concentration (2). Furthermore, various adulterants are mixed with cocaine that are used to increase the volume. These adulterants are often deleterious to health and are mainly consumed along with the drug in the crack form. Discriminating the various forms of cocaine (freebase powder or paste, hydrochloride powder, and crack rock), the drug’s true concentration, and the adulterants used to increase volume is of great interest to the forensic toxicology field as a means of conclusively identifying drugs in cases of drug trafficking (3–5).
1
Biomedical Engineering Institute, Universidade Camilo Castelo Branco UNICASTELO, Parque Tecnologico de S~ao Jose dos Campos, Estrada Dr. Altino Bondesan, 500, Eug^enio de Melo, S~ao Jose dos Campos, SP 12247-016, Brazil. *Supported in part by FAPESP (S~ao Paulo Research Foundation) who granted the Raman instrument (Process no. 2009/01788-5). Received 14 Feb. 2013; and in revised form 25 Sept. 2013; accepted 23 Nov. 2013. © 2014 American Academy of Forensic Sciences
Current methods employed in forensic toxicology, such as gas chromatography and flame ionization detection (GC-FID), aim to detect and identify illicit drugs such as cocaine, amphetamines, MDMA (ecstasy), opiates, barbiturates, and benzodiazepines. Despite their high reliability, these methods are destructive, time-consuming, and do not allow reexamination of the evidence (6,7). Nondestructive and rapid analysis of drugs of abuse could be accomplished through the use of vibrational spectroscopic techniques, especially infrared absorption spectroscopy (FT-IR) and Raman spectroscopy (4,8,9). The advantages lie in the possibility of rapid analysis of trace samples from a crime scene and after seizing drugs, without the destruction of evidence. Thus, these techniques may become a powerful instrument in the fight against drug trafficking (1). Raman spectroscopy is a technique based on the inelastic scattering of laser light molecules and has been used for qualitative and quantitative analysis of drugs of abuse and their adulterants. One can perform a molecular analysis in seconds, without the use of toxic chemicals and without direct contact with the sample (1,9–11). Using near-infrared excitation, most sample fluorescence is minimized. Raman bands of cocaine and common adulterants have been reported in the literature (1,12–16). FT-IR is a technique based on absorption in the infrared region by a molecule due to variations in the dipole moment during vibration. Together with an accessory used to perform measurements by attenuated total reflectance (ATR), this method has been employed for the analysis of trace compounds and contaminants (17). Recent studies using FT-IR have included the 171
172
JOURNAL OF FORENSIC SCIENCES
identification of cocaine (18–23), some of them using ATR (18,20). Lu et al. (19) evaluated herbal medicines suspected of adulteration using FT-IR. Maharaj (20) evaluated samples of seized cocaine and correlated quantification by the GC-FID technique with that obtained by FT-IR-ATR, highlighting the advantages of using the latter in routine forensics due to the minimal sample preparation required compared to GC-FID. This work aimed to evaluate the effectiveness of the nearinfrared Raman spectroscopy and FT-IR-ATR spectroscopy techniques in identifying spectral differences between the different forms of cocaine—freebase powder, paste, and crack rock, as well as hydrochloride powder—seized by the Brazilian police and stored in evidence files. These techniques were also used to identify the most important peaks of commonly used adulterants used to increase volume such as aluminum sulfate, sodium carbonate, sodium bicarbonate, magnesium trisilicate, starch, caffeine, benzocaine, and lidocaine. Materials and Methods Samples of cocaine were obtained in different forms as seized by the police, that is, powder, paste, and crack rock, from the evidence files of the Toxicology Laboratory of the Technical Police State of Amapa, Amazon Region, Brazil. One powder sample was submitted to a protocol for purification adopted by the Brazilian Federal Police (Technical Instruction no. 006/2006GAB/DITEC, Federal Police, Brazil), consisting of recrystallization of freebase cocaine and submission of this sample to a series of evaluations by GC-MS using a Cerilliant Analytical Reference Standard, Reference C-008, Lot FC120204-01A, CAS 50-36, UV-Vis and FT-IR spectroscopy and HPLC to ensure purity. This sample was then considered a reference standard. Additionally, certified standards of selected adulterants were obtained, that is, lidocaine hydrochloride, caffeine, benzocaine, aluminum sulfate, sodium carbonate, sodium bicarbonate, magnesium trisilicate, and starch, which are usually found in seizures and commonly used as adulterants. The cocaine samples, cocaine standard, and adulterant samples underwent Raman spectroscopy and FT-IR-ATR spectroscopy. The Raman spectra were obtained with a dispersive Raman spectrometer, described elsewhere (24), composed of a diode laser (model L4830S; Microlaser Systems, Inc., Garden Groove, CA, USA) with an excitation wavelength of 830 nm and 100 mW power, a spectrograph with a grating of 600 grooves/mm (model 250IS; Chromex, Inc., San Jose, CA, USA), and a deep-depletion CCD camera (model LNCCD 1024x256 CCD and model ST130 controller; Princeton Instruments, Inc., Trenton, NJ, USA), producing spectra in the range of 800–1800 cm 1 with a resolution of about 8 cm 1. The exposure time for collecting each spectrum was 10 sec. Prior to spectrum collection, each sample was placed in an aluminum sample holder with wells approximately 5 mm in diameter and 5 mm deep. The same samples of cocaine and adulterants were also subjected to FT-IR-ATR with a diamond crystal (Nicolet IS10; Thermo Scientific, Inc., Tewksbury, MA, USA) and EverGlo IR source in the mid/far infrared range, resulting in a spectral resolution of 4 cm 1 with 16 scans in the range of 600–3500 cm 1. The Raman spectrometer was calibrated using standard procedures described elsewhere (24), using known peaks of naphthalene for wavenumber calibration and the calibrated irradiance pattern of a spectral lamp for intensity response. Each sample’s autofluorescence spectrum was removed by fitting and subtracting a fifth-order polynomial function to the gross spectrum. These
procedures were performed using a routine developed in Matlab 4.2. Cosmic rays were removed manually using Excel 2003 software. The FT-IR spectrometer was calibrated using OMNIC software, which provides automatic control of the spectrometer. The Raman and FT-IR spectra of the different forms of cocaine and adulterants were then plotted in order to make comparisons to the literature and tentatively attribute the most important bands present in the spectra of such substance, in order to find spectral features that can be used to discriminate different forms of cocaine and also identify peaks characteristic of adulterants that may be present in cocaine samples.
Results Raman Spectroscopy Figure 1 shows the Raman spectra of various forms of cocaine: paste (Fig. 1A,B), crack (Fig. 1C), hydrochloride powder (Fig. 1D), and freebase reference powder (Fig. 1E). The peak positions of the Raman bands of the different forms of cocaine and the respective attributions of these band vibrations are shown in Table 1. The samples of the powdered forms showed spectra for two different forms of cocaine: the freebase, insoluble in water, and the hydrochloride, the water-soluble form. As seen in Fig. 1, the hydrochloride form (Fig. 1D) presented several peaks at the same positions as those found in the freebase form (Fig. 1E), with observed differences in the peaks at 1036, 1165, 1183, 1319, and 1735 cm 1, which are seen in the freebase and missing or shifted in the hydrochloride. On the other hand, the hydrochloride form showed peaks at 1026 and 1207 cm 1. The peaks at 1605 and 1712 cm 1 shifted their positions in the hydrochloride form: The first one shifted to 1601 cm 1, and the second one, to 1716 cm 1. There were also differences in the intensities of the peaks attributed to the tropanic ring, with the peaks at 848 and 898 cm 1 exhibiting lower intensities in the hydrochloride form. Despite a lower signal-to-noise ratio for the spectrum of the paste (Fig. 1A) due to the strong fluorescence emission from the yellow paste, the spectral profile and peak positions were the same as the spectrum of the reference cocaine (Fig. 1E). Some spectral differences were found related to the presence of a peak at 1639 cm 1 in the samples shown in Fig. 1B,C compared to 1E, suggestive of the presence of benzoic acid (from cocaine degradation) (27). These samples also presented spectral features at 1605 and 1712 cm 1 with different intensities compared to the sample in Fig. 1E, with the intensity increased for the peak at 1605 cm 1 and decreased for the peak at 1712 cm 1, which have been characterized as reflecting cocaine degradation to benzoylecgonine (25). Figure 2 shows the Raman spectra of different compounds commonly found in cocaine seizures and used as adulterants. Table 1 also shows the peak positions of the main Raman bands of these adulterants and their assignments. These bands are seen at characteristic positions that would, in turn, make it possible to identify these compounds in the spectrum of adulterated cocaine. Figure 3 shows the Raman spectra of different samples of freebase cocaine (Fig. 3A–D) and a freebase cocaine reference powder (Fig. 3E). Considerable similarities were observed in the peak positions and intensities of all the samples, in accordance with recent literature (12,14,16,25,26). As observed in Fig. 1B, C, the spectra in Fig. 3A–D showed spectral features suggestive of degradation to benzoic acid and benzoylecgonine (a new peak at 1639 cm 1, an increase in the peak at 1605 cm 1 and a decrease in the peak at 1712 cm 1).
PENIDO ET AL.
.
IDENTIFICATION OF COCAINE AND ADULTERANTS BY RAMAN AND IR
173
TABLE 1––Peak positions of the main Raman and FT-IR bands of different forms of cocaine and adulterants, with their tentative vibration assignments based on recent literature.
Compound Cocaine: freebase, hydrochloridric, paste, and crack
Raman Peak Position (cm 1) 848, 874, 898 1004
1683
C-C stretching (tropane) Symmetric stretching of aromatic ring breathing Asymmetric stretching of aromatic ring C-N stretching C-N stretching C-N stretching C-H twisting Asymmetric CH3 deformation C=C stretching - aromatic ring C=O symmetric stretching - carbonyl C=O asymmetric stretching - carbonyl C-C stretching (tropane), change the polarizability by protonated nitrogen Asymmetric stretching - aromatic ring (freebase at 1036 cm 1) C-N stretching shifted by the protonated nitrogen (freebase at 1183 cm 1) Asymmetric CH3 deformation (freebase at 1453 cm 1) C=C stretching – aromatic ring (freebase at 1605 cm 1) C=C stretching – aromatic ring (freebase at 1605 cm 1) C=O symmetric stretching – carbonyl (freebase at 1712 cm 1) O-C-O asymmetric stretching of sodium carbonate (contaminant or adulterant) C=O stretching – benzoic acid C=O stretching – benzoic acid C=O symmetric stretching – carbonyl C-C stretching C-C stretching C-C stretching – aromatic ring and C-N stretching C-N stretching C-C and C-N stretching and aromatic ring stretching C-N stretching C-N stretching and N-H bending Stretching and bending of aromatic ring C-C stretching H-C=N bending C-N stretching C-N stretching C-N stretching C=C stretching C=O stretching C=O stretching C-O stretching In-plane H-C-H bending C-C, C-N and C-O stretching and aromatic ring stretching Stretching and bending of aromatic ring and NH2 scissoring C=O stretching
1052
Si-O-Si stretching
1069 1076
C-O stretching C-O stretching
1036
Peaks exclusive of hydrochloridric cocaine
1165 1183 1279 1319 1453 1605 1712 1735 848, 898† 1026† †
1207
1462† 1596† 1601† 1716† Peaks exclusive of paste cocaine Peaks exclusive of crack cocaine Lidocaine
1069* 1639* 1639* 1719 954 991 1044 1094 1275
Caffeine
Benzocaine
1387 1448 1596 1073 1242 1287 1330 1363 1602 1658 1700 862 1173 1282 1605
Talc (magnesium trisilicate) Sodium carbonate
Attribution (1,4,15,25,26,29–33)
FT-IR Peak Position (cm 1)
Attribution (4,25,29,33)
726 1040
Out-of-plane C-H deformation C-O and C-N stretching
1110
C-O and C-N stretching
1280 1453 1710‡ 1740‡ 2800-3030‡
C-O and C-N stretching C=C stretching - aromatic ring C=O stretching C=O stretching C-H sp3 and sp2 vibrations
736‡
Out-of-plane C-H deformation (freebase at 726 cm 1)
1030‡
1489‡
C-O and C-N stretching (freebase at 1040 cm 1) C-O and C-N stretching (freebase at 1280 cm 1) C=C stretching of aromatic ring
1713‡
C=O stretching (freebase at 1710 cm 1)
1732‡
C=O stretching (freebase at 1740 cm 1)
‡
1270
2300-2900‡
+
N-H stretching
788 1443 1480
Out-of-plane C-H aromatic ring bending C-N stretching C-N stretching
1548 1657
C=C stretching – aromatic ring C=O stretching – amide I
3390
NH2 stretching
745 1242 1551 1658 1696 2957 3116
Out-of-plane C-H aromatic ring bending C-N stretching C=N stretching C=N stretching C=O stretching C-H stretching =C-H stretching
775 1170 1290
Out-of-plane C-H aromatic ring bending C-H bending C-O stretching
1320
C-N stretching
1600 1640 1683 3226 3347 3428 1017 3680 850 1437
C-C stretching – aromatic ring N-H bending C=O stretching O-H stretching N-H stretching N-H stretching Si-O-Si stretching O-H stretching Out-of-plane CO3 angular deformation Asymmetric C=O stretching
174
JOURNAL OF FORENSIC SCIENCES TABLE 1—Continued.
Compound Sodium bicarbonate
Aluminum sulfate Starch
Raman Peak Position (cm 1) 1046 1268
993 864 940 1127 1262 1343 1385 1462
Attribution (1,4,15,25,26,29–33) C-O stretching O-C-O symmetric stretching
SO4 stretching C-C stretching C-C stretching (carbohydrates) C-C stretching (carbohydrates/proteins) In-plane =C-H deformation (carbohydrates) CH2/CH3 bending (carbohydrates) CH3 bending (carbohydrates) CH2 bending (carbohydrates/proteins)
FT-IR Peak Position (cm 1) 700 834 1000 1300-1400 1619 1076 1105 1670 1010 1660 2940 3300
Attribution (4,25,29,33) CO2 angular deformation Out-of-plane CO3 bending C-OH stretching CO2 symmetric stretching CO2 asymmetric stretching SO4 stretching SO4 stretching O-H stretching C-H bending C-OH stretching C-H stretching C-OH stretching
*Indicates Raman peaks of adulterants or degradation products in the samples of freebase cocaine. † Indicates Raman peaks with differences in the position and intensity caused by protonated nitrogen due to the presence of hydrochloric acid in hydrochloridric cocaine compared to freebase cocaine. ‡ Indicates FT-IR peaks with differences in the position and intensity caused by protonated nitrogen due to the presence of hydrochloric acid in hydrochloridric cocaine compared to freebase cocaine.
Figure 4 shows Raman spectra of different freebase cocaine samples in the paste form (Fig. 4A–D) compared to the freebase cocaine reference powder (Fig. 4E). In these paste samples (4A–D), peaks were observed at 983 and 1069 cm 1, which indicate adulteration by aluminum sulfate and sodium carbonate, respectively. These spectra also presented characteristic peaks of degradation into benzoic acid and benzoylecgonine, represented by a new peak at 1639 cm 1, an increase in the peak at 1605 cm 1 and a decrease in the peak at 1712 cm 1.
forms of cocaine and the respective attributions of these band vibrations are shown in Table 1. The spectral differences in the hydrochloride form (Fig. 5D) compared to the free base form (standard) (Fig. 5E) were mainly in the 2300–2900 cm 1 range, attributable to +N-H stretching resulting from the coordinate
FT-IR Spectroscopy Figure 5 shows FT-IR spectra of cocaine as a powder, paste, and crack. The peak positions of the IR bands of the different
FIG. 1––Raman spectra of cocaine samples in different forms: (A) yellow freebase paste, (B) white freebase paste, (C) crack rock, (D) hydrochloride cocaine powder, and (E) freebase cocaine reference powder, with their characteristic peaks labeled according to Table 1. The presence of the peak at 1639 cm 1 in samples B and C suggests the occurrence of benzoic acid, a degradation product of cocaine. Also, the differences in intensities of the peaks at 1605 and 1712 cm 1 indicate degradation into benzoylecgonine.
FIG. 2––Raman spectra of samples of different adulterants: (A) lidocaine, (B) caffeine, (C) benzocaine, (D) talc (magnesium trisilicate), (E) starch (wheat flour), (F) aluminum sulfate, (G) sodium bicarbonate, and (H) sodium carbonate (barilla), with their characteristic peaks labeled according to Table 1.
PENIDO ET AL.
.
IDENTIFICATION OF COCAINE AND ADULTERANTS BY RAMAN AND IR
175
FIG. 3––Raman spectra of freebase cocaine powders from different seizures. The presence of the peak at 1639 cm 1 in samples A to D suggests the occurrence of benzoic acid, a degradation product of cocaine. Also, the differences in the intensities of the peaks at 1605 and 1712 cm 1 indicate degradation into benzoylecgonine. Freebase cocaine reference E does not show such characteristics.
FIG. 5––FT-IR spectra of samples of different forms of cocaine: (A) yellow freebase paste, (B) white freebase paste, (C) crack rock, (D) hydrochloride cocaine powder, and (E) freebase cocaine reference powder, with their characteristic peaks labeled according to Table 1.
FIG. 4––Raman spectra of freebase cocaine in the paste forms (A to D) compared to the spectra of freebase cocaine reference powder (E), with the suggestive presence of adulterants with peaks at 983 and 1069 cm 1. The differences in the intensities of the peaks at 1605 and 1712 cm 1 and the presence of the peak at 1639 cm 1 suggest benzoylecgonine and benzoic acid, respectively.
covalent bond of the cocaine with HCl, the double peak at 1710 and 1740 cm 1, which appeared shifted closer to each other (1713 and 1732 cm 1, respectively), the peak at 1489 cm 1, which showed greater intensity, and peak shifts from 726 to 736 cm 1, 1040 to 1030 cm 1, and 1280 to 1270 cm 1. Several peaks showed reduced intensity in the hydrochloride form, especially in the ranges between 700 and 900 cm 1 and 1100 and 1300 cm 1. Figure 6 shows FT-IR spectra of different compounds commonly found in cocaine seizures and used in its adulteration. Table 1 also shows the peak positions of the main Raman bands of these adulterants and their assignments. Figure 7 shows FT-IR spectra of different samples of freebase cocaine. In these spectra, similarities were found in the peak positions and intensities of all the samples, but in the spectrum in Fig. 7E, the presence of bands attributed to benzoylecgonine was verified, with peaks at 1353, 1403, 1597, and 1720 cm 1 (symmetrical stretching of the carboxyl group at 1353 and
1403 cm 1, asymmetrical stretching of the carboxyl group at 1597 cm 1 and stretching of the carbonyl group at 1720 cm 1) (28). The spectra in Fig. 7B,C appeared with broad bands at around 1650 and 3300 cm 1, which could be attributed to water due to moisture in the samples. The bands of the selected contaminants were not found in the FT-IR spectra of the cocaine samples. Discussion Optical techniques such as Raman and FT-IR have been described for use in criminal forensics for the identification of drugs and added adulterants (1,23), drugs dissolved in alcohol (8), drug residues on clothing, fingerprints, and paper currency (10,11,14), drugs in herbal medicines (19), and quantification of cocaine in mixtures of commonly used adulterants (9,12,14), among others. The advantages and benefits include the lack of sample preparation, assessment at the crime scene, and preservation of the samples. In addition, small amounts of the sample can be used; theoretically, as little as a single crystal of a drug may be evaluated (11). Both spectroscopic techniques presented in this work presented specificity regarding the type of cocaine analyzed (freebase or hydrochloride form). In the case of Raman spectroscopy, the differences between the two spectra have been reported in the literature, especially the peak shifts from 1605 to 1601 cm 1 and from 1712 to 1716 cm 1, and differences in the relative
176
JOURNAL OF FORENSIC SCIENCES
FIG. 6––FT-IR spectra of samples of different adulterants: (A) lidocaine, (B) caffeine, (C) benzocaine, (D) talc (magnesium trisilicate), (E) starch (wheat flour), (F) aluminum sulfate, (G) sodium bicarbonate, and (H) sodium carbonate (barilla), with their characteristic peaks labeled according to Table 1.
intensities of peaks at 848 and 898 cm 1 of the tropanic ring (1,12,23,25,26). The latter are mainly due to the presence of the protonated nitrogen in the tropanic ring, which changes the polarizability of the electron cloud, thereby altering the energy of the bonds near the tropanic ring. The differences in the FT-IR spectra of freebase and hydrochloride cocaine reported here were also in accordance with the literature (25,29,30), particularly the increases in the peaks at 1453 and 1489 cm 1 as well as the peaks in the 2300- to 2900-cm 1 region, and the peak shifts from 726 to 736 cm 1, 1040 to 1030 cm 1, 1270 to 1260 cm 1, and the double peak at 1710 and 1740 to 1713 cm 1 and 1732 cm 1. Several studies have indicated the possibility of identifying adulterants in cocaine using vibrational spectroscopy (1,12,14,23,29–33). In this work, the Raman spectra of different forms of freebase cocaine revealed bands suggestive of the presence of sodium carbonate and aluminum sulfate as adulterants and also showed the suggestive presence of degradation products such as benzoic acid and benzoylecgonine. In degradation by hydrolysis, where benzoylecgonine is formed, cocaine loses a
methyl group and forms a hydroxyl (34). This causes a change in the relative intensities of the Raman bands at 1605 and 1712 cm 1 (25). In the degradation of benzoylecgonine, hydrolysis occurs again, releasing benzoic acid. In the Raman spectra, the peaks from this process were at 1003 and 1604 cm 1 (assigned to the aromatic ring) and at 1639 cm 1 (assigned to carbonyl stretching) (27). These spectral changes suggestive of degradation processes, especially the emergence of the peak at 1639 cm 1 and a decrease in peak intensity at 1712 cm 1, were found in practically all samples, many of them with the benzoylecgonine and benzoic acid features. The presence of moisture is not an important factor for Raman spectroscopy due to the weak Raman scattering of water. With regard to the FT-IR technique, previous studies have employed the ATR method for detecting the presence of cocaine and adulterants (20,21,29,33). In this study, peaks suggestive of the presence of contaminants in cocaine were not observed by FT-IR. Unlike Raman, the FT-IR bands of various adulterants were broad and mostly overlapped, with less discriminatory capability in terms of the adulterants. As with Raman, the FT-IR
PENIDO ET AL.
.
IDENTIFICATION OF COCAINE AND ADULTERANTS BY RAMAN AND IR
177
FIG. 7––FT-IR spectra of freebase cocaine powder from different seizures: Samples A and B are pastes with the suggestive presence of moisture, and sample C is a freebase powder with the suggestive presence of moisture. Samples D and E are crack rocks, with the latter presenting features suggestive of degradation to benzoylecgonine (D). The freebase cocaine reference powder (F) does not show such characteristics.
spectra showed evidence of sample degradation such as benzoylecgonine, with characteristic peaks at 1350, 1400, and 1599 cm 1 and, unlike freebase or hydrochloride cocaine, a single peak at 1721 cm 1 (4,28). These spectral features were observed in the sample in Fig. 7B (in this case, 1353, 1403, 1597, and 1720 cm 1). Unlike Raman, FT-IR did not show characteristic peaks for benzoic acid, which has its most intense peaks at 935, 1296, and 1689 cm 1 (4,27). FT-IR spectra showed the presence of moisture, mainly in samples of the paste (the OH stretching vibration in the 3300–3700-cm 1 region and the angular deformation of HOH at 1640 cm 1). This technique therefore allows the moisture in samples to be measured, which might be important in evaluating the possible degradation of cocaine by hydrolysis. It has been found that spectral differences between freebase and hydrochloride cocaine can be identified using either the Raman or FT-IR techniques. The selection of optical techniques is preferred due to minimal sample preparation; this technique is economically viable for criminology institutes. A major advantage of Raman spectroscopy is that it can be performed in real time at the crime scene. Such field operation of FT-IR is difficult because of the need for humidity control in the environment. It is of the utmost importance in the forensic field that samples should not be submitted to prior preparation, as the crime scene should be preserved, with the measurements being made on site and samples being evaluated without fragmentation. Also, knowledge of the adulterant used to increase volume is important to assess cocaine trafficking routes. The Raman technique may benefit from the surface plasmon resonance effect, such as using SERS (surface-enhanced Raman scattering) on the same spectrometer (35–39) or by measuring
Raman scattering with the use of nano-structured surfaces or colloidal gold or silver to increase the efficiency of the Raman signal and improve the signal-to-noise ratio for the detection of traces of drugs or metabolites. Cocaine and its degradation products have been evaluated by SERS in urine, sweat, and saliva (1,36–39). Portable Raman spectrometers have been developed to investigate drug trafficking through ports and airports (5). Another advantage of Raman spectroscopy is that drugs can be identified even when wrapped in plastic or in glass bottles (8). By operating in the visible to near-infrared range, it allows the use of fiber optics for remote analysis (3). Conclusion In this study, it has been shown that Raman and FT-IR spectroscopy techniques can be used in a routine to identify forms of freebase and hydrochloride cocaine and the associated degradation product of cocaine, benzoylecgonine. These compounds could be detected by both FT-IR and Raman spectroscopy, and benzoic acid was detectable by the Raman technique, by measuring shifts in the positions and differences in the intensities of the major peaks of cocaine samples. The Raman technique was also able to identify certain adulterants in samples of freebase cocaine, mainly sodium carbonate and aluminum sulfate, which are used to increase the volume of the product. References 1. Carter JC, Brewer WE, Angel SM. Raman spectroscopy for the in situ identification of cocaine and selected adulterants. Appl Spectrosc 2000;54:1876–81.
178
JOURNAL OF FORENSIC SCIENCES
2. Kondziella D, Danielsen ER, Arlien-Soeborg P. Fatal encephalopathy after an isolated overdose of cocaine. J Neurol Neurosur Ps 2007;78:437–8. 3. Ali EMA, Edwards HGM, Hargreaves MD, Scowen IJ. Raman spectroscopic investigation of cocaine hydrochloride on human nail in a forensic context. Anal Bioanal Chem 2008;390:1159–66. 4. Moffat AC, Osselton MD, Widdop B. Clarke’s analysis of drugs and poisons, 3rd edn. London, U.K.: Pharmaceutical Press, 2004. 5. Hargreaves MD, Page K, Munshi T, Tomsett R, Lynch G, Edwards HGM. Analysis of seized drugs using portable Raman spectroscopy in an airport environment – a proof of principle study. J Raman Spectrosc 2008;39:873–80. 6. Bujan L, Fernandez P, Lafuente N, Aldonza M. Comparison of two chromatographic methods for the determination of cocaine and its metabolites in blood and urine. Anal Let 2001;34:2263–75. 7. Da Silva MJ, Dos Anjos EV, Honorato RS, Pimentel MF, Paim AP. Spectrophotometric cocaine determination in a biphasic medium employing flow-batch sequential injection analysis. Anal Chim Acta 2008;629:98–103. 8. Eliasson C, Macleod NA, Matousek P. Non-invasive detection of cocaine dissolved in beverages using displaced Raman spectroscopy. Anal Chim Acta 2008;607:50–3. 9. Penido CA, Pacheco MTT, Silveira L. Quantification of binary mixtures of cocaine and adulterants using dispersive Raman and FT-IR spectroscopy and principal component regression. Instrum Sci Technol 2012;40:441–56. 10. Ali EMA, Edwards HGM, Hargreaves MD, Scowen IJ. In-situ detection of drugs-of-abuse on clothing using confocal Raman microscopy. Anal Chim Acta 2008;615:63–72. 11. Day JS, Edwards HGM, Dobrowski SA, Voice AM. The detection of drugs of abuse in fingerprints using Raman spectroscopy I: latent fingerprints. Spectrochim Acta 2004;60:563–8. 12. Ryder AG, O’Connor GM, Glynn TJ. Identifications and quantitative measurements of narcotics in solid mixtures using near-IR Raman spectroscopy and multivariate analysis. J Forensic Sci 1999;44:1013–9. 13. Baranska M, Proniewicz LM. Raman mapping of caffeine alkaloid. Vib Spectrosc 2008;48(1):153–7. 14. Noonan KY, Beshire M, Darnell J, Frederick KA. Qualitative and quantitative analysis of illicit drug mixtures on paper currency using Raman microspectroscopy. Appl Spectrosc 2005;59:1493–7. 15. Teng D, Ma J, Huang Y, Zhang X, Zheng R. Investigate seawater and seawater anions’ aqueous mixed solution by laser Raman spectroscopy. In: Chen L, Suzuki H, Rudy PT, Zhu N, editors. Semiconductor lasers and applications III. Proc SPIE; 2007 Nov 11. Beijing, China; Bellingham, WA: SPIE, 2008;68240P. 16. Nadine WSK. Analysis of drug samples using Raman microspectroscopy. J Young Invest 2001;5(1):1 http://legacy.jyi.org/volumes/volume5/issue1/ articles/kam.html (accessed July 10, 2013). 17. Ekgasit S, Pattayakorn N, Tongsakul D, Hammacharoen C, Kongyou T. A novel ATR FT-IR microspectroscopy technique for surface contamination analysis without interference of the substrate. Anal Sci 2007;23:863–8. 18. Koulis CV, Reffner JA, Bibby AM. Comparison of transmission and internal reflectance infrared spectra of cocaine. J Forensic Sci 2001;46:822–9. 19. Lu F, Li S, Le J, Chen G, Cao Y, Qi Y, et al. A new method for testing synthetic drugs adulterated in herbal medicines based on infrared spectroscopy. Anal Chim Acta 2007;589:200–7. 20. Maharaj R. Quantitative analysis of cocaine using Fourier Transform Infrared Spectroscopy – Attenuated Total Reflectance: a preliminary investigation. Internet J Third World Med 2009;7(2):8635 http://doi:10. 5580/15e9 (accessed July 10, 2013). 21. Ng PH, Walker S, Tahtouh M, Reedy B. Detection of illicit substances in fingerprints by infrared spectral imaging. Anal Bioanal Chem 2009;394:2039–48.
22. Goh CY, van Bronswijk W, Priddis C. Rapid nondestructive on-site screening of methylamphetamine seizures by Attenuated Total Reflection Fourier Transform Infrared Spectroscopy. Appl Spectrosc 2008;62: 640–8. 23. Ryder AG, O’Connor GM, Glynn TJ. Quantitative analysis of cocaine in solid mixtures using Raman spectroscopy and chemometric methods. J Raman Spectrosc 2000;31:221–7. 24. Silveira L, Sathaiah S, Z^angaro RA, Pacheco MTT, Chavantes MC, Pasqualucci CAG. Correlation between near-infrared Raman spectroscopy and the histopathological analysis of atherosclerosis in human coronary arteries. Lasers Surg Med 2002;30:290–7. 25. Smith FP. Handbook of forensic drug analysis, 1st edn. New York, NY: Academic Press, 2004. 26. Gamot P, Vergoten G, Fleury G. Etude par spectroscopie Raman du chlorhydrate de cocaine. Talanta 1985;32:363–72. 27. Brittain HG. Vibrational spectroscopic studies of cocrystals and salts. 1. The benzamide-benzoic acid system. Cryst Growth Des 2009;9:2492–9. 28. Sigma-Aldrich. Product information. Benzoylecgonine. Catalog Number B3277; http://www.sigmaaldrich.com/etc./medialib/docs/Sigma/Datasheet/ 7/b3277-61k88051dat.Par.0001.File.tmp/b3277-61k88051dat.pdf (accessed July 10, 2013). 29. Nakamoto K. Infrared and Raman spectra of inorganic and coordination compounds. Part A, 5th edn. New York, NY: John Wiley & Sons, 1997. 30. Smith E, Dent G. Modern Raman spectroscopy: a practical approach, 1st edn. West Sussex, U.K.: John Wiley & Sons, 2005. 31. Kloprogge JT, Frost RL. Raman microscopy study of basic aluminum sulfate. J Mater Sci 1999;34:4199–202. 32. Tavender SM, Johnson SA, Balsom D, Parker AW, Bisby RH. The carbonate, Co3 , in solution studied by resonance Raman spectroscopy. Laser Chem 1999;19:311–6. 33. Mayo DW, Millar FA, Hannah RW. Course notes on the interpretation of infrared and Raman spectra. Hoboken, NJ: Wiley Interscience, 2003. 34. Geissh€usler S, Brenneisen R. The content of psychoactive phenylpropyl and phenylpentenyl khatamines in Catha edulis Forsk of different origin. J Ethnopharmacol 1987;19:269–77. 35. Karch SB. Drug abuse handbook, 1st edn. Boca Raton, FL: CRC Press, 1998. 36. Abalde-Cela S, Aldeanueva-Potel P, Mateo-Mateo C, Rodrıguez-Lorenzo L, Alvarez-Puebla RA, Liz-Marzan LM. Surface-enhanced Raman scattering biomedical applications of plasmonic colloidal particles. J R Soc Interface 2010;7:S435–50. 37. Sobrido MS, Lorenzo LR, Abalde SL, Fernandez AG, Duarte MAC, Puebla RAA, et al. Label-free SERS detection of relevant bioanalytes on silver-coated carbon nanotubes: the case of cocaine. Nanoscale 2009;1:153–8. 38. Shende C, Inscore F, Maksymiuk P, Farquharson S. Ten-minute analysis of drugs and metabolites in saliva by surface-enhanced Raman spectroscopy. In: Cullum BM, Carter JC, editors. Smart medical and biomedical sensor technology III. Proc SPIE; 2005 Oct 23. Boston, MA; Bellingham, WA: SPIE, 2005;60070S. 39. Kim K, Shin KS. Surface-enhanced Raman scattering: a powerful tool for chemical identification. Anal Sci 2011;27:775–83. Additional information and reprint requests: Landulfo Silveira, Jr, Ph.D. Professor Universidade Camilo Castelo Branco – UNICASTELO Parque Tecnologico de S~ao Jose dos Campos Estr. Dr. Altino Bondesan, 500, Eug^enio de Melo S~ao Jose dos Campos, SP, Brazil, 12247-016 E-mail: [email protected]
