Search

The potential of reversed-phase liquid chromatography (RPLC) with UV and amperometric detection (AD), and of normal-phase liquid chromatography (NPLC) with UV and electron-capture detection (ECD) for the determination of pentachlorophenol (PCP) in wood samples has been studied. When PCP concentrations of at least 1-5 ppm have to be determined, RPLC-UV and RPLC-AD on C18-modified silica are useful techniques, provided a two- or three-step sample-preparation step is used. NPLC-UV on bare silica columns does not offer any advantage over RPLC-UV. NPLC-ECD on bare silica and with an acidified toluene-hexane mixture as eluent offers good selectivity and sensitivity, as well as satisfactory linearity and reproducibility for the determination of PCP in wood samples down to low ppb levels. Use of the two-step clean-up procedure is sufficient, and even a single-step procedure has been utilized. In the latter case, analysis times are longer because of the presence of late-eluting ECD-active interferences. The two-step clean-up procedure generally used involves a liquid-liquid extraction with dichloromethane, and solid-liquid sorption using a Sep Pak C18 cartridge. PCP recovery over the 0.2-10 ppm range is 75-100%. Several wood samples containing 1-50 ppm of PCP have successfully been analyzed, and the good potential of NPLC-ECD for trace-level determination of PCP has been demonstrated.

A new method for on-line sample preparation of surface water extracts for the reversed phase HPLC analysis of the fungicide iprodione at the ppb level is presented. Water samples are extracted with dichloromethane and after concentration and evaporation to dryness, taken up in a mixture of acetonitrile/water. 2-Ml aliquots are injected onto a small precolumn, which is subsequently flushed with 20% acetonitrile in water as a clean-up step. The precolumn is switched on-line with the analytical column, and, using 47.5% acetonitrile in water as the eluent, the concentrated zone of iprodione is desorbed and transported to the analytical column. Detection takes place with UV at 229 nm. The resulting chromatograms are free of interferences. The detection limit of the method in 0.02 ppb. Good reproducibility and linear calibration curves are obtained. The results of the method are in agreement with those of capillary GC analysis with cold-on-column injection. The advantage of the method compared to GC is the lower susceptibility to errors due to the presence of interferences.

A method is described for the post-column reaction detection of organosulphur compounds in liquid chromatography, which is based on a ligand-exchange reaction between the palladium(II)-calcein complex and the sulphur-containing compounds. The release of free calcein provides an indirect measure of the amount of organosulphur compounds via fluorescence detection. The kinetics of the reaction and the signal intensity have been studied as a function of the structure of selected organosulphur compounds and of parameters such as carrier- and reagent-stream composition and temperature. Optimal conditions are as follows: an aqueous mobile phase of pH 5-7, to which may be added up to 20% of methanol; an elevated reaction temperature of around 60 degrees C and a palladium(II)-calcein reagent solution of ca. 10(-5) M, which should contain some zinc(II) to enhance the fluorescence intensity. Detection limits, which are dependent on the structure of the compounds, typically are between 0.5 and 1.0 ng. Good linearity is observed over a two- to three-order concentration range. The method has been applied to the determination of ethylene thiourea in e.g. wash-water of apples and tomatoes, and to the detection of penicillamine in spiked serum and urine.

A liquid chromatographic method for the determination of bromide present in human body fluids at the level of 0.5-5.0 ppm is presented. The method involves liquid--liquid extraction of lipids and other lipophilic compounds, destruction of the aqueous phase and analysis of the residue on an aminopropyl bonded silica column with UV detection at 214 nm. The method was applied to the analysis of 278 samples of Dutch human milk. Comparison of the results obtained with those from a routinely used colorimetric procedure for plasma indicated excellent agreement. The ease of automation of the described procedure and its excellent reproducibility make it a good alternative to existing methods for bromide analysis in body fluids.

A liquid chromatographic method is presented for the determination of the phenylurea herbicide diuron and its major metabolite, 3,4-dichloroaniline in asparagus. The method involves a simple isolation step using dichloromethane extraction, followed by reversed phase chromatography on a C18 column, using a basic aqueous-organic eluent and UV detection at 254 nm. The limit of determination is 0.02 mg/kg for both analytes.

A method is described for the fully automated analysis of large numbers of 1--2 ml serum and plasma or urine samples containing the anti-tumorigenic drugs etoposide and teniposide and their aglycone. The blood samples are hydrolysed by a proteolytic enzyme, subtilisin A, prior to preconcentration on a small precolumn. The hydrolysis step serves both to release the strongly protein-bound drugs and to prevent clogging of the chromatographic system. On-line preconcentration is carried out with precolumns packed with PRP1, a micro-particulate divinylbenzene-styrene copolymeric sorbent. Chromatography takes place, after column switching, in a C18/methanol--water system. After a post-column clean-up step using continuous extraction with dichloroethane in an autoanalyzer system, native fluorescence of these analytes is used for detection of the drugs. Recovery of etoposide and teniposide from spiked serum and plasma samples was 100%. Calibration curves of etoposide and teniposide typically show correlation coefficients of 0.9994 over a two-to-three order linear range. The detection limit of etoposide is approx. 8 ng per sample. Repeatability was found to be excellent. Unattended overnight routine analysis is possible without any problems. This method, considering optimal sample throughput, reliability and selectivity, competes favourably with existing techniques for the analysis of etoposide and teniposide.

Enzymatic hydrolysis of blood samples with subtilisin-A releases protein-bound drugs and permits the repeated (10-50 times) injection of up to 1-ml volumes on short (2-30 mm) precolumns without appreciable build-up of pressure or loss of performance of the precolumn. The principle of fully automated serum and plasma analysis is demonstrated with the drug secoverine as a model compound. After enzymatic hydrolysis of the sample with an equal volume of a 1 mg/ml solution of subtilisin-A for 15 min at 55 degrees C, the model compound is preconcentrated using a microprocessor-controlled column switching unit. Separation occurs in a reversed-phase liquid chromatographic system using a CN-type stationary phase and a buffered aqueous dioxane solution as mobile phase. Detection is done by UV spectrophotometry or fluorometrically after post-column ion-pairing reaction with dimethoxyanthracenesulphonate. The relative standard deviation of the procedure is less than +/- 6% (n = 10).

A sensitive method is described for the simultaneous determination of glycyrrhizin and glycyrrhetic acid in human plasma. Quantitation is by high-performance liquid chromatography using ion-pair chromatography on a reversed-phase column. Variable-wavelength ultraviolet detection is employed. For the extraction of both compounds from plasma, a new solid-phase ion-pair extraction procedure using octadecylsilane columns was developed. Because of the strong forces involved in the protein binding of glycyrrhizin, quantitative extraction of this compound from plasma was possible only after an alkaline pH shift. A considerable improvement in selectivity and sensitivity was obtained by automated column switching involving on-line preseparation of the solid-phase extract on a short precolumn and chromatographic analysis of a heart-cut from the precolumn eluate. The limit of detection of both glycyrrhizin and glycyrrhetic acid was 0.1 mg/l in 0.5 ml of plasma. From a preliminary study in human volunteers, it was concluded that glycyrrhetic acid rather than glycyrrhizin is preferred in a study in human volunteers to assess the zero effect level of glycyrrhizin.