Forensic Science International 192 (2009) 94–97

Contents lists available at ScienceDirect

Forensic Science International

journal homepage: www.elsevier.com/locate/forsciint

Voltammetric determination of cocaine in confiscated samples using a cobalt

hexacyanoferrate film-modified electrode
Érica Naomi Oiye a, Natalia Biziak de Figueiredo a, José Fernando de Andrade a,
Heloisa Maria Tristão b, Marcelo Firmino de Oliveira a,*
a
USP – Faculdade de Filosofia Ciências e Letras de Ribeirão Preto – Departamento de Quı´mica – 14040-901 – Ribeirão Preto, São Paulo, Brazil
b
SPTC – Nucleo de Perı´cias Criminalı´sticas de Ribeirão Preto – 14015-040 – Ribeirão Preto, São Paulo, Brazil

A R T I C L E I N F O A B S T R A C T

Article history: In this work, a fast, non destructive voltammetric method for cocaine detection in acetonitrile medium
Received 28 May 2009 using a platinum disk electrode chemically modified with cobalt-hexacyanoferrate (CoHCFe) film is
Received in revised form 4 August 2009 described. The deposition of CoHCFe film at platinum disk (working electrode) was carried out in
Accepted 5 August 2009
aqueous solution containing NaClO4 at 0.1 mol L1 as supporting electrolite. Stability studies of the film
Available online 31 August 2009
and subsequent voltammetric analysis of cocaine were made in acetonitrile medium with NaClO4 at
0.1 mol L1 as supporting electrolite. A reversible interaction between cocaine and CoHCFe at the film
Keywords:
produces a proportional decrease of original peak current, due to the formation of a complex between
Cocaine
Cobalt hexacyanoferrate
cocaine and cobalt ı́ons, with subsequent partial passivation of the film surface, being the intensity of
Forensic chemistry current decrease used as analytical signal for cocaine. A linear dependence of cocaine detection was
Chemically modified electrodes carried out in the range from 2.4  104 to 1.5  103 mol L1, with a linear correlation coefficient of
0.994 and a detection limit of 1.4  104 mol L1. The analysis of confiscated samples by the proposed
method indicated cocaine levels from 37% to 95% (m/m) and these results were validated by comparison
to HPLC technique, being obtained good correlation between both methods.
ß 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Actually, the analysis of apprehended samples of cocaine by

Cocaine [methyl(1R, 2R, 3S, 5S)-3-(benzoyloxy)-8-methyl-8-
azabicyclo [3,2,1] octane-2-carboxylate) (Fig. 1) – the main
alkaloid produced by Erythroxylum coca plant – consists on a
central nervous system stimulant substance and causes pharma-
codependence [1]. Its human consumption can be made by
different ways, such as nasal aspiration of the powder and venous
injection of the aqueous solution (for cocaine chloridrate) or by
inhalation of its vapour form, when it is smoked (for free base
cocaine). The purified form of cocaine consists on a white,
crystalline solid, with aromatic odor and soluble in several organic
solvents. In alcaline pH solutions, cocaine is promptly converted to
benzoylecgonine, which is soluble in aqueous solutions [2].
After the observation of its collateral damages as a pharma-
cological compound along the history, cocaine actually consists on
an illegal substance in most part of the world, making part of the
main constituent of police apprehensions, followed by hemp and
synthetic drugs, resulting in an increased amount of scientific
efforts to analyze this compound quickly and in a simple way [1].
* Corresponding author. Tel.: +55 16 3602 3750.
E-mail address: marcelex@ffclrp.usp.br (M.F.d. Oliveira).
police laboratories is carried out by different methodologies, such
as colorimetric spot tests [3], spectrometric (FTIR, Raman, MS) [4]
and chromatographic techniques (TLC, HPLC and GC [02]). In this
case, an interesting presumptive colorimetric chemical test for
cocaine (Scott test) [5] can be cited: an aqueous solution of cobalt
thiocyanate is added to suspect samples of cocaine and, after
simple pretreatment, if an insoluble blue color mass is observed,
than cocaine can be present. Each cobalt ion can complex two
molecules of cocaine, even though acidic or alcaline pH conditions
are used.
It is important to mention that, in forensic analysis, the
experimental results must prove the nature of the substance, once
these results can absolve or condemn people involved in criminal
processes. In this context, an apprehended substance is commonly
analyzed by different techniques, according to legislation of each
country. As an example, the United States Department of Justice [6]
classifies the chemical analysis techniques in three classes: A, B
and C, according to specificity and less susceptibility to false-
positive or negative errors, in this order. Table 1 indicates some
examples of these techniques. According to these norms, a
conclusive verdict is obtained only if two chemical analyses
(one by class A test and other by A, B or C test) or three exams (two
by class B tests and the third by B or C test) are provided. In fact,

0379-0738/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.forsciint.2009.08.004
 É.N. Oiye et al. / Forensic Science International 192 (2009) 94–97
Fig. 1. Molecular structure of cocaine.
Table 1
Some analytical techniques for drug determination used by USA Justice
Department.
Class A Class B Class C
Infrared spectrometry Capilairy eletroforesis Colorimetric tests
Mass spectrometry Gas chromatography Fluorescence
spectroscopy
Nuclear magnetic Thin layer UV–vis
resonance spectroscopy chromatography spectrophotometry
Raman spectroscopy Liquid chromatography Melting point
instrumental chromatographic and spectrometric techniques
figure as the most employed at forensic analysis of drugs.
However, these systems are not available in all police laboratories,
due to the relatively high cost presented.
Despite its applicability to the analysis of a wide range of
substances of human interest [7,8], electrochemical methods of
analysis do not figure as the most popular techniques employed at
forensic laboratories. The electrochemical analysis of cocaine and
its metabolites for example, is extensively reported in the
literature, including information about mechanisms of reaction
for this substance [9–11] and analytical applications such as
dosage of confiscated samples [12–15], demonstrating the
applicability of these techniques.
As an improvement on electrochemical analysis, the chemical
modification of electrode surfaces using chemical films of interest
consists on the basis for development of electrochemical sensors
[16]. In this context, films of transition metal hexacyanoferrates
such as Prussian Blue and its analogues [17] deserves special
consideration. Such modified electrodes are useful for a variety of
charge-transfer studies including efforts to enhance rates of
electron transfer of solution substrates and the ability of solution
substrates to enter (or exit) the film can be controlled by choice of
suitable derivatising layer. This can be evidenced in an extensive
literature for this family of chemical modifiers [7,18–23].
Considering the importance of electroanalytical methods at the
field of drug analyses and the versatility of electrochemical sensors
based on transition metal hexacyanoferrates, such as the known
chemical affinity between cobalt ions and cocaine, this work aims
to investigate the electrochemical behavior of cocaine in non
aqueous medium using a cobalt hexacyanoferate-chemically
modified platinum electrode and its possible applicability for
cocaine determination in confiscated samples.
2. Experimental
2.1. Apparatus and chemicals
Measurements of cyclic voltammetry and linear sweep voltammetry were
carried out using a mAutoLabIII potentiostat, under the control of a microcomputer.
A 5.0 mL conventional electrochemical cell was employed with a Metrohm
platinum disk as working electrode, a Ag/AgCl (saturated aqueous KCl) reference
electrode and a platinum spiral auxiliary electrode.
Chromatographic analysis of cocaine samples were carried out in a Shimadzu
HPLC apparatus, used in isocratic conditions with a mobile phase consisting of
95
acetonitrile /water (95:5 v/v). A Rheodyne Model 7725 injection valve with a 20 mL
sample loop was used. A reversed-phase column Shimadzu Shim-pack C18
(250  6.0 mm I.D., 5 mm) was applied for all measurements with a guard column
(Shimadzu Shim-pack C18). Prior to use, mobile phase solvents were filtered
through a 0.45 mm filter (Millipore Milex) and degassed by ultrasonic couple into
HPLC system. An isocratic program was used for HPLC analysis were as follows:
wavelength detector set at 224 nm and flow-rate 1.0 mL min1.
All chemical reagents were of analytical grade and all aqueous solutions were
prepared using deionised water. Acetonitrile, cobalt chloride and potassium
hexacyanoferrate from Merck and sodium perchlorate and sodium bicarbonate
from Riedel de Haen were used in this work. Cocaine standard and cocaine samples
were obtained from a partnership between this research group and the criminal
experts of laboratory of toxicological analysis – Institute of Criminalistics – in
Ribeirão Preto city – São Paulo, Brazil.
2.2. Solutions
A 0.1 mol L1 NaClO4 in aqueous solution was prepared in order to work as
supporting-electrolyte for preparation of the CoHCFe film. This aqueous solution of
supporting-electrolite was employed at the preparation of the solutions of CoCl2 in
1.0  102 mol L1 and K3[Fe(CN)6] also in 1.0  102 mol L1. A 0.1 mol L1
NaClO4 in acetonitrile solution was prepared in order to work as supporting-
electrolyte for stability tests and electrochemical behavior of cocaine. This solution
was used for preparation of cocaine standard solution in 1.0  102 mol L1.
Cocaine solutions of confiscated samples were prepared with previous pretreat-
ment of small aliquots of the water-soluble cocaine chloridrate samples with
sodium bicarbonate aqueous solution, in order to remove HCl from its chloridrate
form. So, the cocaine free of chloridrate, which is insoluble in water, was removed
by filtration, rinsed with deionized water, dried and dissolved in acetonitrile
supporting-electrolite solution.
In order to remove electroactive oxygen, all solutions were carried out with a
previous flow of nitrogen gas during 15 min.
2.3. Preparation and application of the chemically modified electrode
A platinum disk (work electrode) was polished until a mirror-like appearance
with alumina powder, then rinsed with water, cleaned in an ultrasonic aqueous
bath and again rinsed with water before use. Equal volumes of 0.5 mL of CoCl2
1.0  102 mol L1 and K3[Fe(CN)6] 1.0  102 mol L1 solutions were added to the
electrochemical cell containing 4.0 mL of aqueous solution of NaClO4 0.1 mol L1.
The formation of CoHCFe film was obtained by 30 successive potential cycles from
0.1 to 0.7 V (vs. Ag/AgCl) using a scan rate of 100 mV s1. The stability of the
electrodeposited film was tested in acetonitrile solution of NaClO4 0.1 mol L1,
being applied 100 successive potential cycles at the film, with subsequent
registration of the voltammograms. The obtainment of cocaine in different
concentrations at the electrochemical cell, such as the analysis of the confiscated
samples, were provided by standard addition method.
3. Results and discussion
3.1. Formation of the CoHCFe film
The CoHCFe film was successfully electrodeposited at working
electrode surface. Its formation can be evidenced at Fig. 2, due to
the increase of both cathodic and anodic peak currents. The results
showed that the electron transfer between the substrate and the
CoHCFe film is the fast transport process. The peak currents
increase linearly with scan rate and the relation iap/icp is around to
the unity, being consistent with reversible redox system beha-
viour. Once cobalt(II) ions cannot be reduced in this potential
range, the redox peaks correspond to the surface-confined
hexacyanoferrate(II/III) redox couple [11], whose redox process
corresponds to the following reaction:
NaCoII ½FeIII ðCNÞ6  þ Naþ þ e Ð Na2 CoII ½FeII ðCNÞ6  (1)
The formal potential of +0.58 V vs. Ag/AgCl was obtained by
averaging the cathodic and anodic peak potentials.
3.2. Stability studies of the CoHCFe film
After its formation in aqueous medium, the chemically
modified electrode was dried at room temperature and applied
at the organic medium with acetonitrile containing NaClO4
0.1 mol L1 as supporting electrolite. The choice of this solvent
96 É.N. Oiye et al. / Forensic Science International 192 (2009) 94–97
Fig. 2. CoHCFe film formation. Supporting-electrolyte: 0.1 mol L1 NaClO4 solution;
area of the platinum disk electrode: 7 mm2; number of cycles: 30; scan rate:
100 mV s1.
is based on the fact that both forms of apprehended samples of
cocaine (free basis and cocaine chloridrate) present good solubility
in this medium. In aqueous medium it is not possible to
dissolve crack samples, for example, which contains the free basis
species of cocaine. Successive potential scanning in acetonitrile
showed that, after 100 cycles, there was a decrease of only 6% in the
peak current, indicating good stability of the CoHCFe film, as
presented at Fig. 3.
The exchange from aqueous medium to a more resistive organic
medium exhibited peculiar differences in terms of electrochemical
Fig. 3. Study of stability of CoHCFe film in acetonitrile medium after 100
cycles of potential. Supporting-electrolyte: 0.1 mol L1 NaClO4; scan rate:
100 mV s1.
behavior: a considerable increase in the relation iap/icp is observed;
the formal potential for hexacyanoferrate(II/III) redox couple was
obtained at 0.78 V (vs. Ag/AgCl), indicating a shift of 200 mV to
anodic region and a decrease of amperometric sensitivity is
observed in comparison to aqueous system. Additionally, a
difference between Eap and Ecp higher than 100 mV was observed
for this redox couple in acetonitrile, due to the more resistive
medium in relation to water. This behavior is also in accordance to
the electrochemical behavior of Prussian Blue analogues in
acetonitrile [23].
3.3. Electrochemical behavior of cocaine at CoHCFe film
The presence of cocaine species at the electrochemical system
produces a decrease of both peak currents of the CoHCFe film.
Considering the increase in the relation iap/icp for acetonitrile
medium, anodic peak currents were employed in the cocaine
dosage. A reversible interaction between cocaine and the
deposited film is observed, once a restoration of peak current
magnitudes is produced when the modified electrode is cleaned
and applied at a free cocaine-supporting electrolyte solution. The
linearity of this current decrease can be observed at Fig. 4. A linear
correlation coefficient of 0.994 and a standard deviation (SD) value
of 1.8  107 A were observed for this dependence, and its
correspondent equation was:
1
iap ¼ 2:7 mA  3:8  103 mA mol L ½cocaine (2)
These results allowed a detection limit (LOD) of 1.4  104 mol L1
and a quantification limit (LOQ) of 4.7  104 mol L1 using the
relations of 3SD/m and 10SD/m, respectively, where m is the
amperometric sensitivity of the curve [7].
Based on the partial passivation of film surface by cocaine-
cobalt complex, which blocks Na+ mobility for the subsequent
electron transfer of the FeII/FeIII system, the detection of cocaine is
provided without redox stages or degradation of the molecule,
consisting on a non destructive method for cocaine dosage.
Fig. 4. Influence of cocaine concentration at the voltammetric answer of the CoHCFe
film: successive 100 mL additions of cocaine standard solution. v = 100 mV/s.
 É.N. Oiye et al. / Forensic Science International 192 (2009) 94–97 97

Table 2 Acknowledgements
A comparison between the results obtained in triplicate by voltammetric and
chromatographic methods for cocaine determination in confiscated samples (LSV,
linear sweep voltammetry; HPLC, high performance liquid chromatography; SD, The authors acknowledge to Fundação de Amparo à Pesquisa do
standard deviation; CV, coefficient of variation). Estado de São Paulo (FAPESP) and Conselho Nacional de Desenvol-
vimento Cientı́fico e Tecnológico (CNPq) for the financial support.
Sample LSV HPLC

Cocaine SD CV (%) Cocaine SD CV (%) References

content (%) content (%)

1 94.3 3.1 3.3 95.6 2.9
2 84.8 4.3 5.1 86.0 3.1
3 65.1 2.9 4.5 65.6 3.2
4 36.7 2.5 6.8 37.4 2.1
3.4. Quantitative assay of cocaine in confiscated samples
The voltammetric determination of cocaine in the prepared
solutions of confiscated samples was carried out in triplicate.
Using the standard addition method, it was possible to find
cocaine theors from 37% to 95% (m/m) at the apprehended
samples. The chromatographic analysis of cocaine indicated a
well defined peak for this species at a retention time of
3.55 min. The recovery measurements for analytical curve
were situated between 95% and 101%, indicating a good precision
of this technique.
Both voltammetric and chromatographic results were reported
at Table 2.
These results were validated by comparison to HPLC technique,
being obtained good correlation between both methods. In this
context, is important to mention that the chromatographic method
presented lower LOD and LOQ values when compared to direct
linear scan voltammetry technique. In this study, HPLC technique
presented a LOD of 2.5  106 mol L1 and a LOQ of
8.3  106 mol L1. However, the voltammetric method presented
an analytical frequency of 50 h1, five times higher than the
chromatographic method.
4. Conclusions
The usage of CoHCFe film as a chemical modifier for
platinum electrode allowed the development of a new voltam-
metric method for cocaine detection in confiscated samples.
Considering the good correlation between the proposed
method and a conventional chromatographic analysis by HPLC
technique, it is possible to conclude that this chemically
modified electrode can be employed as an alternative instru-
mental technique at routine analysis in forensic laboratories,
presenting an analytical frequency which attends the diary
demand and presents some advantages in relation to colori-
metric conventional methods of cocaine detection. Additionally,
quantitative assay of cocaine can be provided by analytical curve
method, offering an important comparison parameter for
forensic investigation.
3.0 [1] H.L. Schlesinger, Topincs in the chemistry of cocaine, United Nations Office on
3.6 Drugs (1985) 63–78.
4.9 [2] P.S. Cardona, A.K. Chaturvedi, J.W. Soper, D.V. Canfield, Simultaneous analyses of
5.6 cocaine, cocaethylene, and their possible metabolic and pyrolytic products,
Forensic Sci. Int. 157 (2006) 46–56.
[3] J. Swiatko, P.R. De Forest, M.S. Zedeck, Further studies on spot test and micro-
crystal tests for identification of cocaine, J. Forensic Sci. 48 (2003) 581–585.
[4] K.M. Clauwaert, J.F. Van Bocxlaer, W.E. Lambert, A.P. De Leenheer, Segmental
analysis for cocaine and metabolites by HPLC in hair of suspected drug overdose
cases, Forensic Sci. Int. 110 (2000) 157–166.
[5] Y. Tsumura, T. Mitome, S. Kimoto, False positives and false negatives with a
cocaine-specific field test and modification of test protocol to reduce false
decision, Forensic Sci. Int. 155 (2005) 158–164.
[6] Scientific working group for the analysis of seized drugs (SWGDRUG) recom-
mendations. United States Department of Justice – Drug Enforcement Adminis-
tration. 3rd Ed., pp.14–16, 2007.
[7] N.R. Stradiotto, H. Yamanaka, M.V.B. Zanoni, Electrochemical sensors: a powerful
tool in analytical chemistry, J. Bras. Chem. Soc. 14 (2003) 159–173.
[8] S.S.L. Castro, R.J. Mortimer, M.F. de Oliveira, N.R. Stradiotto, Electrooxidation and
determination of dopamine using a Nafion1-cobalt hexacyanoferrate film mod-
ified electrode, Sensors 8 (2008) 1950–1959.
[9] V. Pavlova, V. Mirceski, et al., Studying electrode mechanism and analytical
determination of cocaine and its metabolites at the mercury electrode using
square-wave voltammetry, Anal. Chim. Acta 512 (2004) 49–56.
[10] P.G. Greco, P.A. Garris, In vivo interaction of cocaine with the dopamine trans-
porter as measured by voltammetry, European J. Pharmacol. 479 (2003) 117–125.
[11] P.A. Broderick, In vivo voltammetric studies on release mechanisms for cocaine
with gamma-butyrolactone, Pharmacol. Biochem. Behav. 40 (1991) 969–975.
[12] S. Komorsky-Lovric, S. Gagic, R. Penovsky, Voltammetric determination of cocaine
microparticles, Electroanalysis 11 (1999) 120–123.
[13] M.T. Fernandez-Abedul, J.R.B. Rodriguez, Voltammetric determination of cocaine
in confiscated samples, Electroanalysis 3 (1991) 409–412.
[14] M.T. Fernandez-Abedul, A. Costa-Garcia, Flow injection analysis with amperometric
detection of cocaine in confiscated samples, Anal. Chim. Acta 328 (1996) 67–71.
[15] S. Komorsky-Lovric, S. Gagic, R. Penovsky, Voltammetric determination of ben-
zoylecgonine, Anal. Chim. Acta 389 (1999) 219–223.
[16] J.A. Cox, R.K. Jaworski, P.J. Kulesza, Electroanalysis with electrodes modified by
inorganic films, Electroanalysis 3 (1991) 869–877.
[17] K. Itaya, I. Uchida, V.D. Neff, Electrochemistry of polynuclear transition-metal
cyanides-prussian blue and its analogues, Acc. Chem. Res 19 (1986) 162–168.
[18] G.Y. Shi, J.X. Lu, F. Xu, W.L. Sun, L.T. Jin, K. Yamamoto, S.G. Tao, J.Y. Jin, Determina-
tion of glutathione in vivo by microdialysis using liquid chromatography with a
cobalt hexacyanoferrate chemically modified electrode, Anal. Chim. Acta 391
(1999) 307–313.
[19] T.R.I. Cataldi, G. De Benedetto, A. Bianchini, Enhanced stability and electrocata-
lytic activity of a ruthenium-modified cobalt-hexacyanoferrate film electrode, J.
Electroanal. Chem. 471 (1999) 42–47.
[20] M.F. de Oliveira, R.J. Mortimer, N.R. Stradiotto, Voltammetric determination of
persulfate anions using an electrode modified with a Prussian blue film, Micro-
chem. J. 64 (2000) 155–159.
[21] M.F. de Oliveira, A.A. Saczk, J.A.G. Neto, P.S. Roldan, N.R. Stradiotto, Flow injection
amperometric determination of persulfate in cosmetic products using a Prussian
Blue film-modified electrode, Sensors 3 (2003) 371–380.
[22] P.J. Kulesza, M.A. Malik, K. Miecznikowski, A. Wolkiewicz, S. Zamponi, M. Ber-
rettoni, R. Marassi, Countercation-sensitive electrochromism of cobalt hexacya-
noferrate films, J. Electrochem. Soc. 143 (1996) L10–L12.
[23] M. Jayalakshmi, H. Gomathi, G.P. Rao, Investigations on the electrochemical
behaviour of prussian blue films in acetonitrile, Solar Energy Mater. Solar Cells
45 (1997) 201–209.