e. B200-

o, Po, 30

Keywords:AmphetaminesEcstasy

tampowerful stimulants of the central nervous system. Due to a dramatic increase in the abuse of

tives, individually and/or in a mixture, and to the incoherent and contradictory

Some of the psychotropic drugs have been used since pre-history, but in consumption produces a variety of systemic and organ-specic adverse

Bioelectrochemistry 79 (2010) 7783

Contents lists available at ScienceDirect

Bioelectroc

j ourna l homepage: www.e lsevmany countries are proscribed and are consequently subject toclandestine synthesis and illegal trafc world-wide. Amphetamine-type drugs are abused by about 34million people in the world [1]. InPortugal, amphetamine-type drugs, account for a greater part of illicitdrugs seizure and consumption [2].

Amphetamine and amphetamine derivatives have become popularrecreational drugs of abuse because they are powerful stimulants of thecentral nervous system; they increase self-condence and wakefulnessand improve physical performance. In particular, a dramatic increase inthe abuse and recreational use of methylenedioxylated amphetaminederivatives has been detected in many countries, especially among

effects [4]. Consequently,MDMA abuse has the potential to give rise to amajor public health problem.

Ecstasy usually is sold in the form of pills or tablets with distinctivepatterns and shapes. The amount of MDMA varies enormously fromtablet to tablet, but commonly a typical dose is 100 mg MDMA pertablet. However, the composition could differ considerably. In Portugallittle is known about the content and amount of MDMA present inecstasy tablets.

The growing manufacturer's desires to maintain the supply ofecstasy tablets needed to meet consumer demand may have played apart in the increase of ecstasy tablet contamination that occurred in theyoung people. The most important substanmethylenedioxymethamphetamine (MDMA,

Corresponding author. Departamento de EngenhariaEngenharia do Porto, IPP, Rua Dr. Antnio Bernardino dePortugal. Tel.: +351 22 8340500; fax: +351 22 832115

E-mail address: jjg@isep.ipp.pt (J.M.P.J. Garrido).

1567-5394/$ see front matter 2009 Elsevier B.V. Aldoi:10.1016/j.bioelechem.2009.12.002art in the human society.is prevalent, mainly among youth, despite warnings of irreversibledamage to the central nervous system [3]. Furthermore, MDMARecreational drugs have always played a pElectrochemistrySeizure samplesBiological uids

1. Introductionmethamphetamine (MA), methylenedioxyamphetamine (MDA) and methylenedioxymethamphetamine(MDMA) was studied in different buffer systems by cyclic, differential pulse and square-wave voltammetryusing a glassy carbon electrode. A quantitative electroanalytical method was developed and successfullyapplied to the determination of MDMA in seized samples and in human serum. Validation parameters, suchas sensitivity, precision and accuracy, were evaluated. The results found using the developed electro-analytical methodology enabled to gather some information about the content and amount of MDMApresent in ecstasy tablets found in Portugal. Moreover, the data found in this study outlook the possibility ofusing the voltammetric methods to investigate the potential harmful effects of interaction between drugssuch as MDMA and methamphetamine and other substances often used together in ecstasy tablets.

2009 Elsevier B.V. All rights reserved.

ed as illicit substance inmost countries. The recreational use of MDMAce in this group is 3,4-ecstasy) which is classi-

last years. The plargely due to anwith other subssubstances in ecimportant publicsubstances in ecsbe partly responneurotoxicity-rel

Qumica, Instituto Superior deAlmeida 431, 4200-072 Porto,9.

l rights reserved.mplished. The oxidative behaviour of amphetamine (A),Accepted 5 December 2009Available online 11 December 2009interpretation of the electrochemical data available on this subject, a comprehensive study of the redoxproperties of amphetamine-like drugs was accoReceived in revised form 4 December 2009 methylenedioxylated derivaElectrochemical oxidation of amphetaminelectroanalysis of ecstasy in human serum

E.M.P.J. Garrido a, J.M.P.J. Garrido a, , N. Milhazes b, Fa Departamento de Engenharia Qumica, Instituto Superior de Engenharia do Porto, IPP, 4b Instituto Superior de Cincias de Sade-Norte, Gandra, Paredes, Portugalc Departamento de Qumica, Faculdade de Cincias, Universidade do Porto, 4169-007 Portd Departamento de Qumica, Faculdade de Cincias e Tecnologia, Universidade de Coimbra

a b s t r a c ta r t i c l e i n f o

Article history:Received 14 August 2009

Amphetamine and amphe-like drugs and application to

orges c, A.M. Oliveira-Brett d

072 Porto, Portugal

rtugal04-535 Coimbra, Portugal

ine-like drugs are popular recreational drugs of abuse because they are

hemistry

i e r.com/ locate /b ioe lechemercent of pure ecstasy tablets decreased over time,increase in tablets sold with mixtures of MDMA alongtances. The overwhelming presence of non-MDMAstasy tablets in some countries is alarming and hashealth implications. It is possible that non-MDMA

tasy tablets, alone or in combination with MDMAmaysible for the discrepant research ndings regardingated memory problems [5].

Therefore, there is an increasing interest in the study and develop-ment of a rapid, selective and sensitivemethod for the identication andquantication of MDMA in seizure samples and biological uids, so thatan effective control of these drugs can be done.

The identication and quantication of amphetamines in biologicalsamples have been described using a variety of methodologies.Chromatographic techniques, such as high performance liquid chroma-tography (HPLC) or gas chromatography (GC), and capillary electro-phoresis are themost commonly usedmethodologies for the analysis ofecstasy [69].

Electroanalytical techniques have become indispensable tools inmodern analytical chemistry, and electrochemical methods have alsobeen used for determining amphetamine-type drugs [1014]. Despitethese applications, the understanding of the electrochemical oxida-tion mechanism of amphetamine-like compounds is not claried. In

ated to an one-compartment glass electrochemical cell equippedwith a

78 E.M.P.J. Garrido et al. / Bioelectrochemistry 79 (2010) 7783three-electrode system consisting of a glassy carbon working electrode(GCE) (d=2mm), a platinum wire counter electrode and an Ag/AgClsaturated KCl reference electrode. The glassy carbon working electrodewas polished manually with aqueous slurry of alumina powder (BDH)on amicroclothpad and rinsedwithwater before use. The electrodewasthen cycled in the potential region of the voltammetric measurementsuntil a reproducible response was obtained. All measurements weremade at room temperature.

The optimum instrumental parameters used in the square-wave(SW) voltammetry quantication were chosen studying the variationof the peak current (Ip) with the SW frequency (f), pulse amplitude(Es) and ionic strength (I). The system was optimised, particularly inrespect of maximum peak current and reproducibility.

Fig. 1. Chemical structures of (panel A) amphetamine (A), (panel B)methamphetaminefact, after the report, fteen years ago, about the voltammetricbehaviour of 3,4-methylenedioxyamphetamine (MDA) [15] no otherstudy on amphetamine's electrochemical properties has recently beendone. This lack of information led sometimes to incoherent interpre-tation of the data found and to contradictory results [16,17].

In order to clarify the electrochemical oxidation of amphetamine-likedrugs, a comprehensive voltammetric study of amphetamine (A), meth-amphetamine (MA), methylenedioxyamphetamine (MDA) and methy-lenedioxymethamphetamine (MDMA) (Fig. 1) was undertaken, and avoltammetric method was developed for the quantication of MDMA inseizure samples and in human blood serum. As a nal goal, this studyintends to gather some information about the content and amount ofMDMA present in ecstasy tablets found in Portugal.

2. Experimental

2.1. Apparatus

Voltammetric experiments were performed using an AutolabPGSTAT 12 potentiostat/galvanostat (EcoChemie, Netherlands) associ-(MA) and the two methylenedioxy analogues, (panel C) MDA and (panel D) MDMA.A Crison pH-meter with glass electrode was used for the pHmeasurements (Crison, Spain). The HPLC experiments [18] wereperformed using a HPLC/DAD, Shimadzu instrument (pumps modelLC-20AD, Tokyo, Japan), equipped with a commercially prepackedTracer Excel 120 ODS-B analytical column (250 mm4.0 mm, 5m,Teknokroma, Barcelona, Spain) and UV detection (SPD-M20A) at themaximumwavelength determined by the analysis of the UV spectrum(210 nm). The mobile-phase consisted of methanol:ammoniumacetate buffer 50 mM (40:60) containing 0.1% triethylamine, pH 3.9,at a ow rate of 0.8 mL/min at room temperature. The chromato-graphic data was processed in a Siemens computer, tted withLabSolutions software (Shimadzu, Japan).

2.2. Chemicals and standards

Amphetamine sulfate and methamphetamine hydrochloride weresupplied by Sigma-Aldrich Qumica (Sintra, Portugal). MDMA andMDAwere synthesized asdescribed elsewhere [16]. Analytical grade reagentspurchased from Merck (Darmstadt, Germany) were used withoutadditional purication. Seized Ecstasy tabletswere provided by PolciaJudiciria (Lisbon, Portugal).

Deionised water (conductivityb0.1S cm1) was used throughoutthe experiments. Buffer solutions, used as supporting electrolytes, were0.2 M ionic strength and the pH range of 1.212.2.

HPLC-grade methanol was obtained from Merck. Ammoniumacetate buffer 50 mM was prepared in deionised water and the pHwas adjusted to pH 3.9. Prior to use, the solvents were ltered througha 0.45-m lter.

Standard stock solutions of A, MA, MDA and MDMA (2.5 mM)were prepared in deionised water. In a typical run, 10 mL ofsupporting electrolyte was transferred to the electrochemical celland the required volume of the standard stock solution was added tothe electrolyte, in order to get a nal concentration of 0.1 mM. Forcalibration curves, standard solutions were prepared in the voltam-metric cell adding accurate volumes of the stock standard solution ofMDMA to the selected phosphate pH 7.3 supporting electrolyte inorder to obtain concentrations between 8 and 45 M. The calibrationcurve for SW voltammetry was constructed by plotting the peakcurrent against the MDMA concentrations (8.7, 17.4, 26.1, 34.7 and43.4 M). The limit of quantication (LOQ) and the limit of detection(LOD) were calculated according to USP guidelines [19]. A S/N ratio often and three was used respectively (LOD=3S.D.(sensitivity)1;LOQ=10S.D.(sensitivity)1). The method precision was checkedat different days, within day (n=5) and between days (n=5) forthree different concentrations. The accuracy of the proposed methodwas determined by comparing the results with those obtained using apreviously published HPLCmethod [18]. The precision and accuracy ofthe analytical method was also described by the use of relative errors(Bias%).

2.3. Analysis of seized ecstasy tablets

A total of twelve tablets containing MDMA were analysed in thisstudy. For MDMA determination each tablet was weighed and groundto a homogeneous ne powder in a mortar. An accurately weighedportion equivalent to a stock solution of a concentration about 2.5 mMwas transferred to a volumetric ask and dissolved with deionisedwater. The mixture was sonicated for 5 min to attain completedissolution and ltered to remove any remaining insoluble matter.Working solutions of the seized ecstasy tablets were prepared exactlyas the standard solutions.

2.4. Analysis of spiked serum samples

Human serum samples were collected from healthy volunteers

and stored frozen until the assay. An aliquot volume of sample was

fortied with MDMA and diluted in deionised water to achieve thenal concentration of 2.5 mM. Themixture was treated with 0.4 mL ofmethanol as serum denaturating and precipitating agent, and then thevolume was completed to 1.5 mL with the same serum denaturedsample. The tubes were vortexed for 5 min and then centrifuged for10 min at 4000 rpm for removing of protein residues. The supernatantwas taken carefully. Appropriate volumes of this solution were addedto phosphate pH 7.3 supporting electrolyte and the voltammogramswere then recorded.

The concentration ofMDMA in the human serum samples varied inthe range of 12 to 45 M. For the recovery studies the amount ofMDMA in spiked human serum samples was calculated from therelated calibration equation.

Methanol and acetonitrile, in different amounts, were tested asserum protein precipitating agents. The best results were obtainedusing 0.4 mL of methanol.

3. Results and discussion

The alarming increase in the recreational use of amphetamine-likedrugs, in particular MDMA, and the multitude of adverse effectsresulting from its misuse require a complete understanding of thepharmacological and toxicological prole of these amphetaminederivatives. To date, there is a general consensus that MDMAmetabolism, resulting in the formation of highly redox activemetabolites, is required for the expression of MDMA-induced toxicity[20,21].

To gain insight into the possible formation of these metabolitesand to provide additional information about the mechanism by whichthey could exert their toxic effects, the electrochemical behaviour ofA, MA, MDA and MDMA (Fig. 1) was studied over the pH interval 1.2to 12.2, at a glassy carbon electrode using different voltammetrictechniques.

3.1. Electrochemical oxidation

Differential pulse (DP) voltammetry of A showed, as expected, thatit is not electroactive over the entire pH range studied, since Apossesses in its molecular structure only a single functional group, analiphatic primary amine. Primary amines usually oxidise at potentialshigher than those allowed by the potential window of the glassycarbon electrodes.

For MA, a single anodic wave can be observed above pH 9, Ep=+0.92 V (Fig. 2A), corresponding to the oxidation of the secondaryamine present in the MA molecule. The slope of the dotted line(Fig. 2B), ca. 60 mV per pH unit, shows that the mechanism of thisoxidation process in aqueous media involves the same number ofelectrons and protons. The appearance of a peak at this potential hasalso been described in the literature for other secondary aliphaticamines [2224].

The rst step in the anodic oxidation of aliphatic amines inaqueous solution is considered to be the abstraction of an electronfrom the lone-pair of electrons on the amino-nitrogen. In fact the MAsecondary amine group is more easily oxidised in basic solution than

79E.M.P.J. Garrido et al. / Bioelectrochemistry 79 (2010) 7783Fig. 2. A) 3D plot and B) plot of Ep vs. pH from DP voltammograms of 0.1 mM solutions1of MA in different buffer electrolytes as a function of pH. Scan rate 5 mV s .Fig. 3. A) 3D plot and B) plot of Ep vs. pH from DP voltammograms of 0.1 mM solutions1of MDA in different buffer electrolytes as a function of pH. Scan rate 5 mV s .

in acidic solution, and this data correlates well with the pKa=10.1 ofthe amine function of MA [25]. Cyclic voltammograms obtainedshowed the irreversible oxidation of this compound, as one well-dened anodic peak and no reduction peak was observed.

A single anodic peak over the entire pH range examined, Ep=+1.17 V at pH 2 (Fig. 3A), was observed for MDA, corresponding tothe removal of one electron from the aromatic nucleus present in themolecule and subsequent formation of a radical cation. The anodicpeak potential was shifted to more negative values on increasing thesolution pH, Fig. 3B. The EppH plot shows that the electrode processis pH-dependent, the slope of the dotted line (Fig. 3B), ca. 30 mV perpH unit, shows that the mechanism of this oxidation process inaqueous media involves a number of electrons that is double thenumber of protons.

Cyclic voltammograms recorded also showed one irreversibleanodic peak. This result indicates that a very fast subsequent chemicalreaction of the radical cation generated on the anodic oxidation takesplace, so that the reduction of the radical cation will not occur in thetime scale of the experiment. This is well documented for relatedcompounds [15,2628]. However, at high scan rates this oxidationwave can exhibit partial reversibility, as was demonstrated using lowtemperature ESR spectroscopy [29].

Excluding the peak observed at+0.35 V for pH3 and4 that is relatedwith the GCE background current, the DP voltammetry of MDMA

showed the rst anodic peak, P1 at Ep=+1.18 V, starting at pH 2

Fig. 5. DP voltammograms of 0.1 mM solutions of () MDMA, ( ) MDA and1

Scheme 1. Proposedmechanism for the electrochemical oxidation of MDMA. P1, P2 andP3 correspond to the voltammetric oxidation peaks observed for MDMA.

80 E.M.P.J. Garrido et al. / Bioelectrochemistry 79 (2010) 7783Fig. 4. A) 3D plot and B) plots of Ep (lled symbols) and Ip (open symbols) vs. pH fromDP voltammograms of 0.1 mM solutions of MDMA in different buffer electrolytes as afunction of pH. () rst peak, P1; () second peak, P2; () third peak, P3. Scan rate

15 mV s .(Fig. 4A) corresponding to an oxidation on the aromatic nucleus of themolecule leading to the formation of a radical cation (Scheme 1). TheEppH plot (Fig. 4B) shows that peak P1 electrode process is pH-dependent over the whole pH range; Ep decreases linearly withincreasing pH. The slope of the dotted line, ca. 30 mV per pH unit,shows that the mechanism of this oxidation process, in aqueous mediainvolve a number of electrons that is double the number of protons.

The second anodic peak, P2 at Ep=+1.31 V starts appearing at pH 4(Fig. 4A), and is related to the oxidation of a species formed bydimerization of the initial radical cation (Scheme 1). The currentobserved is in good agreement with the formation of a dimeric product[26,27]. The slope of the dotted line, ca. 30 mV per pH unit, shows thatthemechanism of this oxidation process, P2, in aqueousmedia involvesa number of electrons that is double the number of protons.

Above pH 9 another anodic peak, P3 at Ep=+0.86 V at pH 9(Fig. 4A), corresponding to the oxidation of the secondary aminepresent in the MDMA molecule (Scheme 1) is observed. Theappearance of this peak is related with the acidbase properties ofMDMA. In fact, at this pH deprotonation of the secondary amine groupoccurs and, as a result, an oxidative process can take place due to theexistence of an electron lone-pair. This data not only correlates wellwith the EppH behaviour observed for the other anodic waves but isalso consistent with the voltammetric prole of methamphetaminepreviously described. The values of Ep for peak P3 are pH-dependent,(Fig. 4B), and the slope of the dotted line, ca. 60 mV per pH unit, showsthat the mechanism of P3 oxidation process in aqueous mediainvolves the same number of electrons and protons (Scheme 1).() MA in pH 9 buffer electrolyte. Scan rate 5 mV s .

Based on the data obtained for all the amphetamine compoundsstudied an oxidation mechanism is proposed for MDMA (Fig. 5,Scheme 1). In fact, the only structural difference between MDMA andMDA lies in the amine group, MDMA has a secondary and MDA aprimary amine group. Thus, the peak P3 observed for MDMA startingat pH 9 is related with the oxidation of the secondary amine group.This pattern is consistent with the voltammetric behaviour described,for A and MA, molecules containing also primary and secondaryamines, respectively.

The appearanceof thepeakP2 atpH4, forMDMAis clariedbelow. Itwas expected that a similar peak would be present for MDAvoltammetric prole since this oxidative signal is attributed to oxidationof a species ensuing from the formation and dimerization of the initialradical cation, occurring both in MDMA andMDA. The use of controlledpotential electrolysis (CPE) was considered to clarify this process;nevertheless it was apparent from the literature that no signicantconclusion would result from this study. Preparative electrolysis at axed potential was done for similar compounds in order to understandthe mechanism of this type of oxidations [26,27]. The oxidation

enhanced resolution.The optimum instrumental parameters used in the SW voltam-

metry quantication were chosen studying the variation of the peakcurrent (It) with the SW frequency (f), pulse amplitude (Es) and ionicstrength (I). The system was optimised, particularly in respect tomaximum peak current and reproducibility. Optimization of the SWvoltammetry parameters regarding peak denition and current wasaccomplished using f=100 Hz, Es=50 mV and I=0.2 M. Under thedescribed experimental parameters, SW voltammograms recordedwith increasing amounts of MDMA (8 to 45 M) showed that the peakcurrents increased linearly with increasing concentration. Electroan-alytical data for the calibration: limit of detection (LOD), limit ofquantication (LOQ), obtained after ve runs, and precision of themethod, evaluated by repeatedly (n=5) measuring MDMA, at threelevels of concentration (12, 30 and 45 M) within a day and over veconsecutive days, are summarized in Table 1.

3.3. Electroanalysis of seized ecstasy tablets

Considering the lack of information regarding the amount of MDMA

81E.M.P.J. Garrido et al. / Bioelectrochemistry 79 (2010) 7783potentials of the products obtained were characterized at the end ofthe electrolysis and none of them corresponded to the potentialobserved for P2. It was concluded that the dimeric species responsiblefor this oxidation peak was an intermediate formed in route to the nalproducts [26,27].

To clarify the process occurring at peak P2 in MDMA the followingwas investigated. It seemed reasonable that the alkylic chain,containing the amine group, attached to the aromatic nucleuswould have some inuence on the stabilization of the initial radicalcation. If that occurs then this could be the main reason for theunexpected disappearance of the wave in MDA. In order to test thishypothesis, the voltammetric behaviour of 1,2-(methylenedioxy)benzene and 3,4-(methylenedioxy)toluene was studied at pH 7, andthe same pattern was observed for these two model compounds(Fig. 6). The introduction of a methyl group as a substituent in thearomatic nucleus seems to affect signicantly the stabilization of theradical cation formed since it produced an anodic peak potential shiftto more negative values. This lower potential value results from theincreased stabilization of the radical cation originated by the electron-donating effect of the methyl group. This explains the apparentdifference in the peak potentials between MDMA and MDA.

In view to an electroanalytical application, the inuence of the pHon the MDMA peak current at a glassy carbon electrode was alsoinvestigated, using DP voltammetry. The plot of Ip vs. pH indicatesthat the peak current reaches a maximum for pH 7 (Fig. 4B). Theexperimental results also showed that the DP voltammograms arebetter dened in pH 7 phosphate buffer as compared to Britton

Fig. 6. DP voltammograms of 0.1 mM solutions of () 1,2-(methylenedioxy)benzene1and () 3,4-(methylenedioxy) toluene in pH 7 buffer electrolyte. Scan rate 5 mV s .Robinson buffer. Consequently, this buffer was chosen as the bestsupporting electrolyte and was used throughout the electroanalyticalstudy.

Cyclic voltammograms were also recorded at different sweep ratesat pH 7. Two well-dened anodic peaks, P1 and P2, were observed forMDMA (Fig. 7). No cathodic peak was observed for the scan ratesinvestigated (from 10 mVs1 to 500 mVs1).

The effect of the potential scan rate (10500 mV s1) on the peakcurrent of MDMA was evaluated, according to the equation Ip=Avx.The x values of 0.5 and 1.0 are expected for diffusion and adsorptioncontrolled process [30]. A plot of the logarithm of peak current versuslogarithm of scan rate yielded a straight line of slope 0.52, indicatingdiffusion controlled process [30]. The peak potential was shifted tomore positive values on increasing the scan rate, which conrms theirreversible nature of the oxidation process.

3.2. Electroanalytical applications

Based on the voltammetric behaviour of MDMA, a quantitativeelectroanalytical method was developed to determine the drugcontent in different types of samples. To select the best electrochem-ical method for this purpose, the anodic peaks obtained by DP and SWvoltammetry were compared. SW voltammetry was found to be thefaster method, giving the best ratio of peak-to-background currentand providing sharper and better dened peaks, leading to an

Fig. 7. Cyclic voltammogram for 0.1 mM solution of MDMA in pH 7 phosphate buffer.Scan rate 20 mV s1.present in seized ecstasy tablets in Portugal, andbased on the developed

Table 1Regression data of the calibration lines for quantitative determination of MDMA usingthe SW voltammetry method.

SWV

Supporting electrolyte Serum

Linearity range (M) 845 1245

Calibration graph (n=6)a

Intercept (A) 0.023 0.035Slope (mA M1) 30.0 33.9Correlation coefcient 0.999 0.996Standard error of intercept 0.031 0.028Standard error of slope 0.76 1.02

Intra-day precision (n=5)b

82 E.M.P.J. Garrido et al. / Bioelectrochemistry 79 (2010) 7783method, SW voltammetry was applied to the direct determination ofMDMA using a fast and simple, ltration and adequate dilution, samplepreparation procedure.

The electroanalysis of MDMA in seized ecstasy tablets showed anaverage content of 120 mg per tablet (corresponding to an averagecontent of ecstasy of 30% (w/w)), Table 2.

Notwithstanding the size of the sample used (due to difculties forsample collection related to the illicit nature of this drug only twelvetablets were available), that cannot be considered enough for deniteconclusions, the average content found follow the trend of seizedtablets composition noted in other European countries [31].

12 M 1.9 2.830 M 1.7 2.045 M 1.3 2.2

Inter-day precision (n=5)b

12 M 2.3 2.630 M 2.0 1.945 M 2.1 2.0LOD (M) 1.2 2.4LOQ (M) 3.7 8.3

a Regression equation Ip (A)=a[MDMA] (M)+b, where a is the slope and b theintercept of the calibration plot.

b Precision as coefcient of variation (CV, %)=standard deviation divided by meanmeasured concentration 100.The accuracy of the voltammetric method proposed was assessedby comparing the results with those obtained using an establishedchromatographic procedure [18], and for the SW voltammetry andHPLC, ve different independent measurements were carried out. Theresults, Table 2, show that the SW voltammetry data tted well withthose obtained by HPLC.

The F- and t-tests were carried out on the data and statisticallyexamined the validity of the obtained results. For a condence level of95%, the values of t- and F-tests were less than the theoretical values,showing that there is no signicant difference between the proposedSW voltammetry and the HPLC control methods.

Recent ecstasy tablet purity surveys in Europe suggested thatMDMA purity rates are around 90100% [31,32]. In fact, the majority

Table 2Results for MDMA analysis in seized tablets and spiked samples.

SWV HPLC

Amount founda (mg) 120.4 119.7RSD (%) 1.8 1.2tb 0.8Fc 2.5Amount added (mg) 25.00Amount founda (mg) 24.77Recovery (%) 99.1RSD (%) 1.3Bias (%) 0.92

a Mean of ve experiments.b Tabulated t-value for P=0.05 and eight degrees of freedom is 2.31.c Tabulated F-value for P=0.05 and f1= f2=4 is 6.39.of the seized or collected tablets contain only MDMA with no otheractive substance added and only excipients, mainly sugars: glucose,lactose, sorbitol, saccharose, and fatty acids. Occasionally found non-MDMA active substances were usually other amphetamine-likemolecules: MDEA, A, MDA, and MA. The blended psychoactivesubstance most frequently found in the tablets containing MDMA iscaffeine [31]. None of these active substances were detected by theSW voltammetry and HPLC methodologies used in the present study.

To conrm the accuracy of the proposed method and to evaluatethe possible interaction with excipients, recovery experiments werecarried out after the addition of known amounts of the pure drug toMDMA seized tablets. The results showed that the excipients areelectrochemically inactive, have no interference effect on the analysis,Table 2, and both methodologies applied had a comparable precisionand accuracy.

3.4. Determination of MDMA in serum samples

In order to apply the proposed method to biological samples, thequantication of MDMA in human serum was tested. Calibrationequation parameters and validation data are summarized in Table 1.Serum samples were spiked with MDMA in order to achieve nalconcentrations of 15, 30 and 45 M. SW voltammograms obtained forthe determination of MDMA in human serum samples are in Fig. 8.

Fig. 8. DP voltammograms obtained from the determination of MDMA in spiked humanserum samples. () blank serum; () 15 M; ( ) 30 Mand () 45 MMDMA in human serum samples.The recoveries obtained were good and are shown in Table 3.

4. Conclusions

The voltammetric behaviour of several amphetamine-like drugs, A,MA, MDA and MDMA was investigated. The analysis performedshowed that, with the exception of A, all the amphetamines studiedare electroactive and their oxidation mechanism is related to anoxidation process taking place on the aromatic nucleus and/or on thesecondary amine group present in the molecules. The data found alsocontributed to clarify the oxidative and metabolic prole of

Table 3Results for MDMA analysis in spiked human plasma using SW voltammetry.

MDMA added(M)

MDMA found(M)

Average recoverya

(%)RSD(%)

Bias(%)

15 14.93 99.5 1.4 0.530 30.17 100.6 0.9 0.645 45.08 100.2 1.1 0.2a Mean of three determinations.

amphetamine-like-drugs. A SW voltammetry methodology wasdeveloped for the determination of MDMA in seized tablet samplesand biological samples. The results showed that the electroanalyticalmethod is sensitive and does not lead to statistically signicantdifferences when compared to the established HPLC methodology.Despite the limited sample size used, it was found that the ecstasytablets analysed have MDMA purity rates of ~100% and the dosagelevels of MDMA in seized tablets were ~120 mg.

Althoughmost ecstasy users are aware of the drug's associated risks,theymay still be unaware of the risks they face from adulterated tablets.The reported results in this study outlook the possibility of using thevoltammetric methods to investigate the potential harmful effects ofinteraction between drugs such as MDMA and methamphetamine andother substances often used together in ecstasy tablets.

Acknowledgements

[13] T. Keller, M. Mutz, R. Aderjan, H.P. Latscha, Study of the electrochemical behaviourof amphetamine and its derivatives in aqueous solution, Fresenius' J. Anal. Chem.363 (1999) 270276.

[14] R. Gulaboski, M.N.D.S. Cordeiro, N. Milhazes, J. Garrido, F. Borges, M. Jorge, C.M.Pereira, I. Bogeski, A.H. Morales, B. Naumoski, A.F. Silva, Evaluation of the lipophilicproperties of opioids, amphetamine-like drugs, and metabolites throughelectrochemical studies at the interface between two immiscible solutions, Anal.Biochem. 361 (2007) 236243.

[15] J.A. Squella, B.K. Cassels, M. Arata, M.P. Bavestrello, L.J. Nuez-Vergara,Electrochemical oxidation of methylenedioxyamphetamines, Talanta 40 (1993)13791384.

[16] N. Milhazes, P. Martins, E. Uriarte, J. Garrido, R. Calheiros, M.P.M. Marques, F.Borges, Electrochemical and spectroscopic characterisation of amphetamine-likedrugs: application to the screening of 3, 4-methylenedioxymethamphetamine(MDMA) and its synthetic precursors, Anal. Chim. Acta 596 (2007) 231241.

[17] C. Macedo, P.S. Branco, L.M. Ferreira, A.M. Lobo, J.P. Capela, E. Fernandes, M.L.Bastos, F. Carvalho, Synthesis and cyclic voltammetry studies of 3, 4-methylene-dioxymethamphetamine (MDMA) human metabolites, J. Health Sci. 53 (2007)3142.

[18] M.E. Soares, M. Carvalho, H. Carmo, F. Remio, F. Carvalho, M.L. Bastos,Simultaneous determination of amphetamine derivatives in human urine afterSPE extraction and HPLC-UV analysis, Biomed. Chromatogr. 18 (2004) 125131.

[19] U.S. Pharmacopeia, United States Pharmacopeial Convention, Rockville28th ed,2005.

83E.M.P.J. Garrido et al. / Bioelectrochemistry 79 (2010) 7783Financial support from Fundao para a Cincia e Tecnologia (FCT),project PTDC/QUI/65255/2006, POCI (co-nanced by the EuropeanCommunity Fund FEDER), and CEMUC-R (Research Unit 285), isgratefully acknowledged.

References

[1] United Nations Ofce on Drug and Crime, World Drug Report, United NationsPublications, New York, 2005.

[2] Instituto da Droga e da Toxicodependncia, Relatrio Anual 2006: A Situao doPas em Matria de Drogas e Toxicodependncias, Lisbon, 2007.

[3] H. Kalant, The pharmacology and toxicology of ecstasy (MDMA) and relateddrugs, Can. Med. Assoc. J. 165 (2001) 917928.

[4] A.R. Green, A.O. Mechan, J.M. Elliott, E. O'Shea, M.I. Colado, The pharmacology andclinical pharmacologyof 3, 4-methylenedioxymethamphetamine (MDMA, Ecstasy),Pharmacol. Rev. 55 (2003) 463508.

[5] E.E. Tanner-Smith, Pharmacological content of tablets sold as ecstasy: resultsfrom an online testing service, Drug Alcohol Depend. 83 (2006) 247254.

[6] D. Butler, G.G. Guilbault, Analytical techniques for ecstasy, Anal. Lett. 37 (2004)20032030.

[7] S. Chinaka, R. Iio, N. Takayama, S. Kodama, K. Hayakawa, Chiral capillaryelectrophoresis of amphetamine-type stimulants, J. Health Sci. 52 (2006) 649654.

[8] M. Kumihashi, K. Ameno, T. Shibayam, K. Suga, H. Miyauchi, M. Jamal, W. Wang, I.Uekita, I. Ijiri, Simultaneous determination of methamphetamine and itsmetabolite, amphetamine, in urine using a high performance liquid chromatog-raphy column-switching method, J. Chromatogr. B 845 (2007) 180183.

[9] J.L. da Costa, E.R. Pintao, C.M.C. Corrigliano, O.N. Neto, Determinao de 3, 4-metilenodioximetanfetamina (MDMA) em comprimidos de Ecstasy por cromato-graa lquida de alta ecincia com deteco por uorescncia (CLAE-DF), Quim.Nova 32 (2009) 965969.

[10] R.E. Michel, A.B. Rege, W.J. George, High-pressure liquid chromatography/electrochemical detection method for monitoring MDA and MDMA in wholeblood and other biological tissues, J. Neurosci. Methods 50 (1993) 6166.

[11] N.A. Santagati, G. Ferrara, A. Marrazzo, G. Ronsisvalle, Simultaneous determinationof amphetamine and one of its metabolites by HPLC with electrochemicaldetection, J. Pharm. Biomed. Anal. 30 (2002) 247255.

[12] A. Domnech, R. Aucejo, J. Alarcn, P. Navarro, Electrocatalysis of the oxidation ofmethylenedioxyamphetamines at electrodesmodiedwith cerium-doped zirconias,Electrochem. Commun. 6 (2004) 719723.[20] T.J. Monks, D.C. Jones, F. Bai, S.S. Lau, The role of metabolism in 3, 4-()-methylenedioxyamphetamine and 3, 4-()-methylenedioxymethamphetamine(Ecstasy) toxicity, Ther. Drug Monit. 26 (2004) 132136.

[21] A. Felim, A. Urios, A. Neudrffer, G. Herrera, M. Blanco, M. Largeron, Bacterial plateassays and electrochemical methods: an efcient tandem for evaluating the abilityof catechol - thioether metabolites of MDMA (Ecstasy) to induce toxic effectsthrough redox-cycling, Chem. Res. Toxicol. 20 (2007) 685693.

[22] M. Masui, H. Sayo, Y. Tsuda, Anodic oxidation of amines. Part I. Cyclic voltammetryof aliphatic amines at a stationary glassy-carbon electrode, J. Chem. Soc. B (1968)973976.

[23] A. Adenier, M.M. Chehimi, I. Gallardo, J. Pinson, N. Vil, Electrochemical oxidationof aliphatic amines and their attachment to carbon and metal surfaces, Langmuir20 (2004) 82438253.

[24] J.M.P.J. Garrido, C. Delerue-Matos, F. Borges, T.R.A. Macedo, A.M. Oliveira-Brett,Voltammetric oxidation of drugs of abuse III. Heroin and metabolites, Electro-analysis 16 (2004) 14971502.

[25] Clarke's Analysis of Drugs and Poisons, London: Pharmaceutical Press. Electronicversion, 2006.

[26] A.H. Said, F.M. Mhalla, C. Amatore, J.-N. Verpeaux, Mechanistic investigation of theanodic oxidation of p-methoxytoluene in dry and wet acetonitrile, J. Electroanal.Chem. 464 (1999) 8592.

[27] A.H. Said, F.M. Mhalla, C. Amatore, L. Thouin, C. Pebay, J.-N. Verpeaux,Mechanistic investigation of the anodic oxidation of 3, 4, 5-trimethoxytoluenein acetonitrile, J. Electroanal. Chem. 537 (2002) 3946.

[28] E.M.P.J. Garrido, J.M.P.J. Garrido, M. Esteves, A. Santos-Silva, M.P.M. Marques, F.Borges, Voltammetric and DFT studies on viloxazine: analytical application topharmaceuticals and biological uids, Electroanalysis 20 (2008) 14541462.

[29] C.J. Schlesener, J.K. Kochi, Stoichiometry and kinetics of p-methoxytolueneoxidation by electron transfer. Mechanistic dichotomy between side chain andnuclear substitution, J. Org. Chem. 49 (1984) 31423150.

[30] D.K. Gosser, Cyclic Voltammetry. Simulation and Analysis of Reaction Mechanisms,Wiley, New York, 1993 pp. 2770.

[31] A.C. Parrott, Is ecstasy MDMA? A review of the proportion of ecstasy tabletscontaining MDMA, their dosage levels, and the changing perceptions of purity,Psychopharmacology 173 (2004) 234241.

[32] I. Giraudon, P.-Y. Bello, Monitoring ecstasy content in France: results from thenational surveillance system 19992004, Subst. UseMisuse 42 (2007) 15671578.

Electrochemical oxidation of amphetamine-like drugs and application to electroanalysis of ecsta.....IntroductionExperimentalApparatusChemicals and standardsAnalysis of seized ecstasy tabletsAnalysis of spiked serum samples

Results and discussionElectrochemical oxidationElectroanalytical applicationsElectroanalysis of seized ecstasy tabletsDetermination of MDMA in serum samples

ConclusionsAcknowledgementsReferences