Ultraviolet resonance Raman spectroscopy for the detection of cocaine in oral fluid

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 188 (2018) 338–340

Contents lists available at ScienceDirect

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa

Short Communication

Ultraviolet resonance Raman spectroscopy for the detection of cocaine in oral fluid Valentina D'Elia a, Gemma Montalvo a, Carmen García Ruiz a, Vladimir V. Ermolenkov b, Yasmine Ahmed b, Igor K. Lednev b,⁎ a Department of Analytical Chemistry, Physical Chemistry and Chemical Engineering and University Institute of Research in Police Sciences (IUICP), University of Alcalá, Ctra. Madrid-Barcelona Km. 33.6, 28871 Alcalá de Henares (Madrid), Spain b Department of Chemistry, University at Albany, State University of New York, 1400 Washington Avenue, Albany, NY 12222, USA

a r t i c l e

i n f o

Article history: Received 7 April 2017 Received in revised form 19 June 2017 Accepted 11 July 2017 Available online 15 July 2017 Keywords: Raman spectroscopy Ultraviolet Raman spectroscopy Cocaine Oral fluid Forensic

a b s t r a c t Detecting and quantifying cocaine in oral fluid is of significant importance for practical forensics. Up to date, mainly destructive methods or biochemical tests have been used, while spectroscopic methods were only applied to pretreated samples. In this work, the possibility of using resonance Raman spectroscopy to detect cocaine in oral fluid without pretreating samples was tested. It was found that ultraviolet resonance Raman spectroscopy with 239-nm excitation allows for the detection of cocaine in oral fluid at 10 μg/mL level. Further method development will be needed for reaching the practically useful levels of cocaine detection. © 2017 Elsevier B.V. All rights reserved.

1. Introduction The detection of cocaine (COC) in oral fluid (OF) is of great importance for practical forensics due to a high worldwide COC consumption and the ease of obtaining OF samples. Up to date, the majority of the methods used to detect COC in OF are based on immunological procedures or chromatographic techniques coupled with mass spectrometry or tandem mass spectrometry [1–6]. However, enzyme immunoassays involve a significant probability of false negative and false positive results. Consequently, these methods are only used for preliminary screening followed by further analyses to confirm the results. Hyphenated chromatographic techniques are used as confirmatory tests, but they are destructive and require complex sample pretreatment, which makes the analysis costly and time consuming [4–6]. Most recently, spectroscopic techniques have been utilized to carry out rapid and confirmatory analyses using small sample quantity; however, due to the strong interference from OF, a prior extraction and/or preconcentration Abbreviations: COC, cocaine; IR, infrared; OF, oral fluid; RRS, resonance Raman spectroscopy; RS, Raman spectroscopy; UV, ultraviolet; UVRRS, ultraviolet resonance Raman spectroscopy. ⁎ Corresponding author. E-mail addresses: [email protected] (V. D'Elia), [email protected] (G. Montalvo), [email protected] (C.G. Ruiz), [email protected] (V.V. Ermolenkov), [email protected] (Y. Ahmed), [email protected], http://www.sites.google.com/site/lednevlab/, http://www.inquifor.com (I.K. Lednev).

http://dx.doi.org/10.1016/j.saa.2017.07.010 1386-1425/© 2017 Elsevier B.V. All rights reserved.

pretreatment was needed before the spectroscopic analysis could be conducted [7]. Raman spectroscopy (RS) allows for confirmatory identification of cocaine and other illegal drugs of abuse for forensic purposes [8]. Typically, near-IR light is used for exciting Raman scattering to combat fluorescence interference and sample damage. Sands et al. [9] demonstrated that COC as well as other narcotics and explosives could be efficiently identified using resonance Raman spectroscopy (RRS) with ultraviolet (UV) 244-nm excitation. They reported RS and RRS spectra of pure COC powder as well as that mixed with a “scouring” compound. Resonance excitation allowed avoiding the fluorescence interference and identifying COC in both samples [9]. In the present preliminary study, the capability of RRS with UV excitation to detect COC in liquid OF, without any sample pretreatment, was evaluated. COC UV–visible absorption spectrum exhibits two electronic transitions at ~ 200 nm and ~ 230 nm. Ultraviolet resonance Raman spectroscopy (UVRRS) with 239-nm excitation was utilized for characterizing liquid OF samples doped with various concentrations of COC and the detection limit was estimated. 2. Material and Methods 2.1. Chemicals and Sample Preparation Standard COC used to prepare samples was purchased from Cerilliant (Round Rock, TX, USA) as 1 mg/mL acetonitrile solution. OF

V. D'Elia et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 188 (2018) 338–340

samples were purchased from Bioreclamation, Inc. (Hicksville, NY, USA). All aqueous solutions were prepared in distilled water using a procedure previously described by D'Elia et al. where the homogeneity of the sample mixtures was confirmed [10]. Since standard COC was initially dissolved in acetonitrile, various amounts of the standard COC solution were evaporated under vacuum for 30 min at 25 °C to remove the organic solvent and then redissolved in the same volumes of OF. The concentration of COC in the prepared samples of OF varied from 1 mg/mL to 1 μg/mL. At these concentrations, COC was readily soluble in aqueous solution [11]. A reference solution of COC in water (1 mg/mL) was also prepared using the same procedure. Liquid samples were then transferred to quartz NMR tubes and directly analyzed. 2.2. Instrumental Resonance Raman (RR) spectra were recorded using a homebuilt Raman spectrograph equipped with a liquid nitrogen cooled CCD camera (Roper Scientific, Inc., Sarasota, FL, USA) [12]. Pulsed laser radiation (pulse duration 6–7 ns, repetition rate 50 Hz) at 239 nm (the first antiStokes component of the Stimulated Raman Scattering of the fourth harmonic of the Nd:YAG laser) was used for excitation. The radiation power on the sample was 10 mW. The spectra were recorded using the WinSpec 32 software (Roper Scientific, Inc., Sarasota, FL, USA) in the range of 500–1900 cm−1. Each spectrum was obtained by averaging 20 accumulations with 30 s acquisition time for every accumulation. 2.3. Spectral Treatment The collected Raman spectra were processed using Thermo Scientific OMNIC™ for dispersive Raman 8.3.103 software (Waltham, MA, USA): the fluorescence contribution was corrected using a 3rd-order polynomial baseline correction and the noise was reduced by an 11-point smoothing. In order to estimate the sensitivity of the method, the reference spectrum of pure OF was subtracted from the COC-doped OF spectra, using the automatic subtraction algorithm of the software employed. 3. Results and Discussions Liquid samples of OF doped with COC concentrations between 1 mg/mL and 1 μg/mL were analyzed in NMR quartz tubes at 239-nm

339

excitation. An aqueous solution of pure COC and a sample of pure OF were also analyzed for comparison. A liquid sample was constantly stirred with a magnetic bar to avoid photodamage. Each UV Raman spectrum represented an average of 20 short (30 s) accumulations. The individually accumulated spectra were compared prior to averaging to check for possible photodamage. The spectra did not change with consecutive accumulations, indicating that no apparent photodegradation occurred under the used experimental conditions. UV–Vis spectra of both COC in water and pure OF were also collected (Fig. S1 in Supplementary material). At 239 nm wavelength, the Raman signal of COC can be resonantly enhanced although the absorption of OF is almost insignificant. In Fig. 1, the normalized RR spectra are shown for 1 mg/mL COC in acetonitrile, 1 mg/mL COC in water, a pure OF sample, and an OF sample doped with 100 μg/mL of COC. The spectrum of pure OF was subtracted from that of a 100 μg/mL COC-doped OF sample, and the resulting difference spectrum is also shown in Fig. 1. One can appreciate that both COC and OF have characteristic Raman bands, which allow for their differentiation. The spectra of COC in aqueous solution and in OF (difference spectrum in Fig. 1) are close to each other and resemble well the Raman spectrum of solid cocaine hydrochloride obtained at 244-nm excitation [9]. The Raman band at 998 cm−1 in the COC spectra can be tentatively attributed to the symmetric breathing mode of the phenyl ring in the COC molecule [13,14]. The OF spectrum also has a weak band near this position, which can be tentatively assigned to the phenyl ring. The Raman peak at 1607 cm−1 and a broad band centered at 1714 cm− 1 in the COC spectra can be tentatively assigned to the trigonal phenyl ring breathing mode and the ester carbonyl C_O stretch, respectively [9,13]. Bands at 1281 and 1178 cm− 1 could be assigned to C\\N stretching in the amine group in the COC molecule [15]. Finally, the OF band at 1617 cm−1 could be attributed to phenyl ring C\\C stretching in the side chains of proteins commonly present in that matrix [16, 17]. Table S1 outlining these vibrational mode assignments appears in Supplementary material. In order to determine the detection limit of the method in detecting COC in OF, OF samples doped with lower concentrations of COC, between 50 μg/mL and 1 μg/mL, were analyzed. Fig. 2 shows spectra obtained after the subtraction of the pure OF spectrum from the spectra of the samples doped with the different concentrations of COC. The limit of detection could be estimated by monitoring the gradual disappearance of the characteristic bands of COC in the difference spectra

Fig. 1. Resonance Raman spectra of 1 mg/mL COC in acetonitrile (ACN), 1 mg/mL COC in water, 100 μg/mL COC in oral fluid, pure oral fluid, and the difference spectrum between those of COC in oral fluid and the pure oral fluid. Excitation wavelength was 239 nm and excitation power was 10 mW.

340

V. D'Elia et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 188 (2018) 338–340

shorter wavelength electronic transition. This study is currently underway in our laboratory. Acknowledgments The authors acknowledge the University of Alcalá for the Project CCG2015/EXP-028. V. D'Elia also thanks the University of Alcalá for her predoctoral grant and for the mobility scholarship offered. The authors are grateful to Antigone McKenna, DVM for technical assistance. This project was supported in part by Award No. 2014-DN-BX-K016 awarded by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice (I.K.L.). The opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect those of the U.S. Department of Justice. Fig. 2. Difference Raman spectra obtained after the subtraction of the oral fluid spectrum from the spectra of the oral fluid samples doped with COC at indicated concentrations. Excitation wavelength was 239 nm and the excitation power was 10 mW.

until the strongest band (1607 cm−1) reached a signal/noise ratio of approximately 3. As evident from Fig. 2, COC was sufficiently detectable until the concentration of 10 μg/mL. In fact, for a sample doped with a 1 μg/mL concentration of COC, no bands were observed in the difference spectrum. 4. Conclusions In this work, the potential of UVRRS for the detection of COC in OF was investigated. The current methods used to detect COC in OF require typically a complex pretreatment of the sample, which complicates its practical application for on-site analyses. In this study, instead, untreated liquid COC-doped OF samples were directly analyzed via UVRRS using a laser excitation at 239 nm. At this wavelength, the absorption of OF is almost insignificant, so that only the Raman signal of COC could be resonantly enhanced, and the direct analysis of doped OF samples proved to be possible. By varying the concentration of COC in OF, a concentration cut-off value of 10 μg/mL was estimated. This value is significantly larger than the cut-off value of 8 ng/mL known for current methods used for COC detection for forensic purposes [18]. Further improvement in the sensitivity is necessary for practical forensic application of this methodology. In particular, the excitation with 200 nm light could result in more significant enhancement of COC Raman signal due to the presence of a

Appendix A. Supplementary Data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.saa.2017.07.010. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]

H.H. Maurer, Anal. Bioanal. Chem. 381 (2005) 110–118. N. Samyn, M. Laloup, G. De Boeck, Anal. Bioanal. Chem. 388 (2007) 1437–1453. L. Caporossi, A. Santoro, B. Papaleo, Biomarkers 15 (2010) 475–487. M. Janicka, A. Kot-Wasik, J. Namieśnik, TrAC Trends Anal. Chem. 29 (2010) 209–224. M. Barroso, E. Gallardo, D.N. Vieira, J.A. Queiroz, M. López-Rivadulla, Anal. Bioanal. Chem. 400 (2011) 1665–1690. M. Meyer, H. Maurer, Anal. Bioanal. Chem. 402 (2012) 195–208. V. D'Elia, G. Montalvo, C. García Ruiz, Appl. Spectrosc. Rev. 50 (2015) 775–796. C.A.F. de O. Penido, M.T.T. Pacheco, I.K. Lednev, L. Silveira Jr., J. Raman Spectrosc. 47 (2016) 28–38. H. Sands, I. Hayward, T. Kirkbride, R. Bennett, R. Lacey, D. Batchelder, J. Forensic Sci. 43 (1998) 509–513. V. D'Elia, M. Calcerrada, G. Montalvo, C. García Ruiz, Electrophoresis 38 (2017) 1217–1223. F. Inscore, C. Shende, A. Sengupta, H. Huang, S. Farquharson, Appl. Spectrosc. 65 (2011) 1004–1008. I.K. Lednev, V.V. Ermolenkov, W. He, M. Xu, Anal. Bioanal. Chem. 381 (2005) 431–437. V. D'Elia, G. Montalvo, C. García Ruiz, Talanta 154 (2016) 367–373. J.C. Carter, W.I.E.B. Er, S.M.I. Angel, Appl. Spectrosc. 54 (2000). C.A.F. de O. Penido, M.T.T. Pacheco, R.A. Zangaro, L. Silveira Jr., J. Forensic Sci. 60 (2015) 171–178. Z. Chi, X.G. Chen, J.S.W. Holtz, S.A. Asher, Biochemistry 37 (1998) 2854–2864. B. Hernández, Y.M. Coïc, F. Pflüger, S.G. Kruglik, M. Ghomi, J. Raman, Spectroscopy 47 (2016) 210–220. Substance Abuse Mental Health Services Administration, Fed. Regist. 80 (2015) 28063.