Search

A simple, but sensitive and specific, high-performance liquid chromatographic assay for cocaine with direct fluorimetric detection, particularly intended for the routine analysis of hair and blood samples, is described. Benzoylecgonine, eluting before cocaine in a completely resolved peak, is also detectable. Detection is based on the weak native fluorescence of cocaine and benzoylecgonine, depending on the benzene ring present in both molecules. Hair samples (20-200 mg) were incubated overnight in 2 ml of 0.25 M HCl at 45 degrees C and extracted with a commercial liquid-liquid method; the dried residue reconstituted with 500 microliters of 0.05 M NaH2 PO4 (PH 5.2) was injected. Blood plasma samples (200 microliters) were mixed with 150 microliters of 0.1 M Na2 HPO4 (pH 8.9) and extracted with 5 ml of chloroform-2-propanol (9:1); the organic phase was evaporated and the residue dissolved and injected as above. Isocratic reversed-phase liquid chromatography was carried out on a column (150 x 4.6 mm I.D.) packed with spherical 5-microns poly(styrene-divinylbenzene) particles; the mobile phase was 0.1 M potassium phosphate (pH 3)-methanol-tetrahydrofuran (70:25:5). The excitation and emission wavelengths were set at 230 and 315 nm, respectively. Under the described conditions, cocaine eluted in a symmetrical peak with a capacity factor of about 5. The limit of detection was about 1 ng/ml (0.2 ng injected), with a signal-to-noise ratio of 3.(ABSTRACT TRUNCATED AT 250 WORDS)

Free solution capillary electrophoresis and high performance liquid chromatography were applied to the analysis of cocaine in hair. Capillary electrophoresis conditions were as follows. Background electrolyte: 0.050 M borate, pH 9.2; capillary: bare silica, 40 cm long, 50 micrograms i.d.; potential: 15 kV; detection: UV at 238 nm wavelength. In addition, the separation was accomplished in a 50 cm x 75 microns capillary with an instrument with a photodiode array detector, recording on-line UV spectra of peaks from 200 to 400 nm. The isocratic high performance liquid chromatographic method with fluorimetric detection used 230 nm (ex.) and 315 nm (em.) wavelengths. Cocaine separation was carried out under conditions summarized as follows: column packed with spherical polystyrene-divinylbenzene 5 microns particles, mobile phase 0.1 M potassium phosphate (pH 3.0)/methanol/THF (70:25:5) at a flow rate of 0.5 ml/min and at room temperature. Liquid-liquid and solid-liquid sample preparation methods were used. Typically, levels of cocaine in hair as low as 0.15-0.30 ng/mg were detected by capillary electrophoresis, while HPLC allowed the determination of concentrations lower by one order of magnitude (0.015 ng/mg). Intra- and inter-assay precision data of the two techniques were comparable and characterized by relative standard deviations in the range from 3 to 7%. On the other hand, the sample throughput of capillary electrophoresis (7-10 injections/h) was higher than high performance liquid chromatography (2 injections/h). A good correlation of the results (r2 = 0.92) between the two techniques was observed in the assay of real cases.

A highly sensitive reversed-phase high-performance liquid chromatographic assay for ethanol and methanol in plasma, using a post-column enzymic reactor with electrochemical detection, has been developed. The alcohols, separated on the column, were converted by immobilized alcohol oxidase into their respective aldehydes with formation of stoichiometric amounts of hydrogen peroxide, detected via oxidation at a platinum electrode. As the chromatographic column, two glass cartridges (150 mm x 3 mm I.D.) in series, packed with 10 microns HEMA-S 1000 packing, were used. Alcohol oxidase from Candida boidinii was immobilized onto HEMA-BIO 1000 VS-L (10 microns), packed in a 30 mm x 3 mm I.D. glass cartridge. The reaction product, hydrogen peroxide, was detected with an amperometric detector with a platinum electrode, operated at +500 mV vs. an Ag/AgCl reference electrode. A 20-microliters volume of ten-fold diluted plasma was injected without any pre-treatment. Under the described conditions, methanol and ethanol were well resolved from each other and from the "front" of the chromatogram. The limit of detection was ca. 2.5 nmol for ethanol and 0.6 nmol for methanol in plasma, at a signal-to-noise ratio of 3. Excellent linearity was observed for ethanol, in the range 0.125-4 micrograms injected (r = 0.9999). In contrast, the response for methanol was markedly non-linear above 500 micrograms injected, presumably owing to progressive saturation of the reactor. The precision and accuracy of the assay were satisfactory, as was the reactor life (one month).