e 7 MINUTES
Figure 8.1 MEKC of bulk heroin, heroin impurities, degradation products and adulterants. Conditions: capillary, 25 cm X 50 mm i.d.; voltage, 20 kV; temperature, 40° C; buffer, 85 m M SDS-8.5 m M phosphate-8.5 m M borate-15% acetonitrile, pH 8.5; detector wavelength, 210 nm. Peaks: b, morphine; c, 3-MAM; d, 6-MAM; e, acetylcodeine; f, heroin; g, phénobarbital; h, noscapine; i, papaverine; j, methaqua-lone. (From Weinberger and Lurie, 1991, with permission.)
migration times based on suitable internal standards (IS) can improve the analytical reproducibility, not only in term of migration times but also quantitatively. Improvements in resolution reportedly resulted from the use of capillaries with smaller diameter (25 /¿m vs. 50 /¿m i.d.), which gave better heat dissipation and lower diffusion of analytes; sensitivity was reduced, however, because optical path length was reduced. Also, fluorescence detection wavelength was tested, and sensitivity was higher (up to 20 times) than that achieved by UV absorption for fluorescent components of the analyzed mixtures.
Another interesting aspect of this study was the comparison of MEKC and HPLC. While in MEKC the drugs represented in Figure 8.1 migrated in the order morphine (1), phenobarbital (2), 6-monoacetylmoprhine (6-MAM) (3), 3-monoacetyl morphine (3-MAM) (4), methaqualone (5), heroin (6), acetylcodeine (7), papaverine (8), and noscapine (9), in reversed-phase HPLC the elution pattern was substantially different and not correlated with MEKC
with, in order, compounds 1, 4, 3, 7, 6, 2, 9, 8, 5). This finding suggests that one technique should be used to confirm the results of the other. In complex mixtures, about twice as many peaks were observed with MEKC than with HPLC, but HPLC proved more sensitive (up to 80 times). Nevertheless, MEKC proved sensitive enough to detect heroin impurities down to 0.2%.
The successful use of MEKC for the analysis of illicit heroin and cocaine was also reported by Staub and Plaut (1994), who used 50 m M SDS in phosphate-borate buffer (10 and 15 m M, respectively) containing 15% of acetonitrile (pH 7.8). In 25 minutes, these authors achieved separation and determination of paracetamol, caffeine, 6-monoacetylmorphine (6-MAM), acetylcodeine, procaine, papaverine, heroin, and noscapine, although with peaks sometimes skewed. On-line recorded UV spectra of the peaks helped the identification of the individual peaks.
Krogh et al. (1994) adopted a similar analytical approach for the separation of test mixtures of the main alkaloids found in illicit heroin and heroin adulterants (14 compounds), as well as of 7 drugs structurally related to amphetamine. These investigators used a 50 cm x 50 /u.m i.d. bare silica capillary and a running buffer of 25 m M SDS in phosphate-borate buffer (10 m M for each salt) pH 9, containing 5% acetonitrile; UV detection occured at 214 nm. Excellent separations were achieved in 15 minutes, with RSDs migration times (relative to crystal violet, the IS) ranging from 0.5 to 1.9% and from 0.89 to 2.23% in within-day and between-day reproducibility tests, respectively. In quantitative studies, standard curves were linear in the range from 0.02 to 0.5 mg/mL, with typical correlation coefficients from 0.997 to 0.999. RSDs were in the range from 2.0 to 4.3% in the analysis of illicit heroin and amphetamine. Reportedly, samples could be injected every 13 minutes, and the fused silica capillary was replaced only after 500 injections to assure good reproducibility. According to the authors, MEKC proved to be a valuable complement to HPLC and GC for the routine analysis of illicit drug preparations.
Trenerry et al. (1994a, 1994b) proposed the use of 50 mM cetyltrimethylam-monium bromide (CTAB), instead of SDS, as micellar agent in MEKC. Also, to overcome the reproducibility problems with late peaks observed by Weinberger and Lurie (1991) and ascribed to inconsistent evaporation of acetonitrile from the buffer vials and to capillary wall fouling, Trenerry et al, suggested the following measures: replacing acetonitrile with the less volatile dimethyl sulfoxide (10%), flushing the capillary with the running buffer between runs, and periodically washing the capillary with 0.1 M NaOH and replacing the running buffer with fresh solution.
The use of the cationic micellar agent CTAB (50 mM) in phosphate-borate buffer (10 m M of each salt), pH 8.6, with 10% acetonitrile was preferred because of a faster separation (about 15 min) of heroin and related substances. Because the cationic surfactant, which coats the capillary silica wall with a positively charged layer, reverses the electroosmotic flow (EOF), the voltage (-15 kV) must be applied with a reversed polarity (with the cathode at the injection point). Detection was by UV absorption at 280 nm.
This method was successfully tested with different illicit heroin preparations, and when the results were compared with those obtained with HPLC, a good quantitative correlation was observed. Precision was slightly less in MEKC than in HPLC, but the resolution of complex samples of illicit heroin was better with MEKC.
The same method, with only slight variations (acetonitrile from 10 to 7.5%, detection wavelength from 280 to 230 nm) was adopted for the analysis of illicit cocaine. Benzoylecgonine, cocaine, and cis- and rram-cinnamoylcocaine, obtained from samples taken from cocaine seizures, were separated. MEKC results correlated well with those from gas chromatography (GC), with similar RSD values, and with HPLC. These authors reported that MEKC methods also proved highly reliable for heroin and cocaine analysis in interlaboratory proficiency tests.
The separation of enantiomers of amphetamine, methamphetamine, ephed-rine, pseudoephedrine, norephedrine, and norpseudoephedrine was reported by Lurie (1992) as a means of investigating the synthesis of illicit drug preparations. A derivatization with a chiral reagent, 2,3,4,6-tetra-(9-acetyl-/3-D-gluco-pyranosyl isothiocyanate, was used, followed by the MEKC separation of the resulting diastereomers. All the enantiomers of the above-mentioned pheneth-ylamines were resolved in a single run, using a bare silica capillary (48 cm X 50 /urn i.d.) and a mixture consisting of 20% methanol and 80% aqueous buffer solution (100 mM SDS, 10 mM phosphate-borate buffer, pH 9.0).
More recently, Lurie et al. (1994) reported the chiral resolution by means of neutral and anionic cyclodextrins (CDs) of a number of basic drugs of forensic interest: amphetamine, methamphetamine, cathinone, methcathi-none, cathine, cocaine, propoxyphene, and various a-hydroxyphenethylam-ines. In this separation, the resolution was optimized by varying the ratio of neutral to anionic CDs. In fact, both types of CD have chiral selectivity, but because of their negative charge, anionic CDs display an electrophoretic countermigration and consequently high retarding effect on analytes.
Chiral resolution was also achieved by means of CE chromatographic techniques, with an enantioselective stationary phase, as reported by Li and Lloyd (1993), who used a,-acid glycoprotein as stationary phase packed in fused silica capillaries of 50 mm i.d. These authors reported the optimization (by varying pH, electrolyte, and organic modifier concentration in the mobile phase) of the separation of the enantiomers of hexobarbital, pentobarbital, isofosfamide, cyclophosphamide, diisopyramide, metoprolol, oxprenolol, al-prenolol, and propranolol.
An excellent review of the applications of HPCE for the analysis of seized preparations of illicit drugs was published by Lurie (1994).
An approach suitable for drug screening alternative to MEKC was proposed by Chee and Wan (1993), who used capillary zone electrophoresis (CZE) with 50 mM phosphate buffer pH 2.35 in a 75 jitm i.d. (60 cm long) bare silica capillary. In 11 minutes, they achieved the separation of 17 basic drugs of potential forensic interest: methapyrilene, brompheniramine, amphetamine, methamphetamine, procaine, tetrahydrozoline, phenmetrazine, butacaine, medazepam, lidocaine, codeine, acepromazine, meclizine, diazepam, doxa-pram, benzocaine, and methaqualone. Detection was by UV absorption at 214 nm.
It was observed that a drug with lower pKa, hence less positive charge, showed higher migration times. Clear correlation between pKa values and migration times, however, could not be obtained because of the influence of other factors (according to the authors: molecular size, tendency to interact with the column, and ability to form doubly charged species).
Reproducibility in migration times was characterized by RSD values better than 1%; peak area RSDs ranged from 1.5 to 4.3%. Worse reproducibility was found for analytes with very slow migration (i.e., with pKa values close to the pH of the background buffer). Notwithstanding an inherent suitability for direct screening of forensic samples, the above-described method was applied only for the analysis of biological fluids. After a simple one-step chloroform-isopropanol (9:1) extraction from plasma and urine samples, previously adjusted to pH 10.5, "clean" electropherograms were obtained, which allowed detection limits of about 0.50 jiig/mL of each drug.
Chee and Wan (1993) believe that CZE offers some advantages over MEKC for drug screening, particularly, simple background electrolyte preparation and shorter analysis times. They see the main limitation as the inability to analyze acidic, neutral, and basic drugs together. An additional advantage offered by CZE is that a peculiar separation mechanism, poorly correlated with MEKC, allows the use of this technique in parallel with MEKC for confirmation purposes, with the same instrumental hardware.
MEKC is almost universally adopted for illicit drug analysis not only in pharmaceutical or illicit formulations, but also in biological samples.
Wernly and Thormann (1991) used a phosphate-borate buffer pH 9.1 with 75 mM SDS for the qualitative determination in urine of many drugs of abuse (and their metabolites), including benzoylecgonine, morphine, heroin, 6-monoacetylmorphine, methamphetamine, codeine, amphetamine, cocaine-, methadone, methaqualone, and benzodiazepines.
Sample purification and concentration used "double-mechanism" (cation exchange and reversed phase) solid phase extraction cartridges. The extracts from 5 mL of urine were dried, redissolved with 100 jiiL of running buffer, and injected, to achieve a detection limit of 100 ng/mL. For peak identification, not only the retention times were used, but also the on-line-recorded UV spectra of the peaks, which were compared to computer-stored models.
Because of a sensitivity comparable to that obtainable from nonisotopic immunoassays, as well as low running costs and possibility of automation, Wernly and Thormann proposed MEKC for confirmation testing, following immunometric screenings.
An MEKC system quite similar to that described earlier (50 mM SDS in phosphate-borate buffer, pH 7.8) allowed, also, a high resolution separation of barbiturates, including barbital, allobarbital, phenobarbital, butalbital, thio-
pental, amobarbital, and pentobarbital (Thormann et al., 1991). Again, on-column multiwavelength detection helped peak identification. Sensitivity was in the order of the low micrograms per milliliter. It is interesting to note that while urine samples needed extraction prior to injection, human serum can be injected directly because some barbiturates, including phénobarbital, elute in an interference-free window of the electropherogram.
Another MEKC separation (75 m M SDS, phosphate-borate buffer, pH 9.1) was reported for the determination of ll-nor-A-9-tetrahydrocannabinol-9-carboxylic acid, the major metabolite of A-9-tetrahydrocannabinol present in urine (Wernly and Thormann, 1992a). Sample treatment included basic hydrolysis of urine (5 mL), solid phase extraction, and concentration. The resulting sensitivity was 10 to 30 ng/mL (i.e., comparable to the cutoffs of immunoassays). Again, detection was by on-line recording of peak spectra, by means of fast-scanning UV detector.
Wernly et al. (1993) reported also an attempt to use MEKC, without hydrolysis, to determine morphine-3-glucuronide, the major metabolite of morphine (and heroin) in urine. The sensitivity limit (about 1 ¿¿g/mL after solid-phase extraction and concentration) was unsatisfactory for confirmation of the results of the usual enzyme immunoassays, but improvements were deemed to be achievable.
Quite recently, Schafroth et al. (1994) determined the major urinary compounds of eight common benzodiazepines (flunitrazepam, diazepam, midazolam, clonazepam, bromazepam, temazepam, oxazepam, and lorazepam). The used MEKC with 75 m M SDS in a phosphate-borate buffer (pH 9.3). After enzymatic hydrolysis and extraction with commercial "double-mechanism" cartridges, the sensitivity was reportedly better than that obtained with the common immunoassay EMIT (enzyme-multiplied immunoassay technique).
The relatively poor sensitivity of HPCE in terms of concentration requires that the sample be concentrated severalfold before injection, if an acceptable degree of sample sensitivity is to be achieved. This concentration is possible provided the biological material is cleansed of all the interfering substances, first of all organic and inorganic ions, that, if present at high concentrations in the injected solution, would alter completely the separation (causing peak distortion and loss of efficiency). Wernly and Thormann (1992b) described a stepwise solid phase extraction for human urine preliminary to MEKC, using commercial "double-mechanism" cartridges exhibiting hydrophobic and ionexchange interactions. This extraction produced "clean" electropherograms, even injecting extracts from a urine sample concentrated 50 times (Fig. 8.2).
In the same paper, these authors also observed that MEKC with surfactants in plain aqueous buffers failed to resolve the highly hydrophobic analytes (e.g., amphetamine, methamphetamine, methadone, benzodiazepines). Resolution of these compounds could be obtained by adding low percentages of acetonitrile (5-10%), thus bringing about MEKC conditions resembling those reported by Weinberger and Lurie (1991).
Was this article helpful?