Cocaine countries clarified

Skip to Navigation


  • Published: Jul 7, 2008
  • Author: Steve Down
  • Channels: Gas Chromatography
thumbnail image: Cocaine countries clarified

Cocaine use in Europe remains at an alarmingly high level, according to the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). Globally, it is the second most trafficked drug, behind cannabis, and the main European entry point is the Iberian Peninsular. In its latest annual report, the EMCDDA claims that the increase in the number of seizures and the quantities of the drug "confirms its status as Europe's stimulant drug of choice." It is estimated that around 12 million Europeans (4% of all adults) have tried cocaine at some point.

The most prolific country supplying the drug is Colombia, followed by Peru and Bolivia and this fits in with European observations that the majority of the European seizures originate from South America. The geographical origin of the drug can provide useful criminal intelligence, by linking one seizure to another, for instance, and is a useful tool for investigators.

In practice, one characteristic of the bulk drug is surprisingly useful in determining the country of origin. Crystals of cocaine hydrochloride, which are produced from the free base form, contain small but detectable amounts of solvents trapped within the crystals that can be exploited by analysts.

The solvents can be analysed by static headspace-gas chromatography (HS-GC). Samples are heated in a sealed environment and any volatile components, including the trapped solvents, are released into a loop for subsequent analysis. This technique has been adopted by scientists at the National Police Scientific Institute at Ecully, France, who used a factorial design to optimise the procedure.

Laurence Dujourdy and Fabrice Besacier studied the effects of five variables on method performance by measuring the solvent GC peak areas. Cocaine.HCl was dissolved in water saturated with a sulphate salt and the released solvents were analysed by GC with flame ionisation detection. A fractional factorial design reduced the number of comparative experiments from the maximum of 32 (for five variables at two levels each) to 8.

These studies revealed that salt type, stirring speed and time and loop equilibration time influenced peak areas but the fourth variable, pressure on the sample, did not. The optimum conditions required the addition of 22% sodium sulphate in the cocaine.HCl solution which was stirred at a high level for 30 min at 8 psi, with a loop equilibration time of 0.03 min.

These conditions were used to study the residual solvents in 177 cocaine samples. A total of 18 solvents were deemed to be significant and their variations, expressed as relative standard deviations, were in the ranges 7-22% for linked samples and 79-621% for non-linked samples. So, a clear differentiation appeared possible based on this solvent set.

The overall aim was to compare the trapped solvent profile of a new sample with that in a database, so Dujourdy and Besacier developed a system built using PHP software and a MySQL open source database. A cosine function was employed to compare the unknown to the database and provide a similarity value on a scale of 0-100, where 0 corresponds to a perfect non-match and 100 to a perfect match. For non-linked samples, the values were in the range 0-99.35 with a mean of 32.63 and the matched samples fitted in the extremely narrow range 99.37-100. The researchers selected a threshold value of 99.4.

A total of 2863 cocaine samples were analysed over 6 years for the 18 solvents. Of these, 7 appeared in more than 60% of the cases, namely acetone, diethyl ether, isobutanol, propyl acetate, toluene, isopropyl acetate/benzene and hexane/chloroform/ethyl acetate. In the latter two cases, it was not possible to distinguish the respective components.

When samples from Bolivia, Colombia and Peru were compared, their solvent profiles were quite different. Rather than relying on a visual comparison of the chromatograms, the data were subjected to a hierarchical cluster analysis and the results were plotted for easy interpretation.

The Peruvian and Colombian samples could be differentiated clearly, but a few Bolivian samples were clustered close to those of the other two countries. Nevertheless, the discrimination was regarded as promising and now needs to be validated using certified authentic samples.

The method was partially successful in comparing samples which were a mixture of cocaine base and cocaine.HCl, but only for cocaine base contents up to 30%. Above that level, the peak intensities of the solvent peaks dropped sharply and the cosine function values fell below 99.4, reducing more rapidly as the percentage of cocaine base rose.

This new optimised method should provide good criminal intelligence by pinpointing the origin of seized cocaine samples and comparing one batch

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.


Social Links

Share This Links

Bookmark and Share


Suppliers Selection
Societies Selection

Banner Ad

Click here to see
all job opportunities

Most Viewed

Copyright Information

Interested in separation science? Visit our sister site

Copyright © 2018 John Wiley & Sons, Inc. All Rights Reserved