Determination of low concentrations of benzene in urine using multi-dimensional gas chromatography.

A method for the determination of benzene in urine of occupationally or environmentally exposed persons was developed. The method was based on dynamic headspace, preconcentration on a solid sorbent, followed by thermal desorption and gas chromatographic determination. To achieve sufficient selectivity, we used multi-dimensional gas chromatography in combination with the inexpensive and robust flame ionisation detector. The limit of detection was 7 ng l-1 and the limit of quantification was 23 ng l-1. The linearity was good (correlation coefficient 0.999) in the range examined (20-4000 ng l-1) and the repeatability was 9%. The average recovery at low concentrations (20-400 ng l-1) was 86%. Analysis of a certified reference material of benzene in water, traceable to NIST, did not differ significantly from the certified value. Samples, frozen (-20 degrees C) in glass bottles sealed with Teflon-silicon septa, were stable for 1 year and refrigerated samples (4 degrees C) for at least 1 week. Loss of benzene during the collection and transfer of urine was investigated and found to be acceptable. The method is a cost effective and robust alternative to GC-MS and permits reliable quantification of occupational exposure and, in most cases, also of urine concentrations that can be expected from environmental exposure.

Supercritical fluid extraction of the pesticides carbosulfan and imidacloprid from process dust waste.

Large amounts of contaminated process dust remain from the procedure of pesticide treatments applied to seed pellets. A pilot study in analytical-scale supercritical fluid extraction (SFE) was performed to determine the possibility of using supercritical carbon dioxide for the extraction of the nonpolar insecticide carbosulfan and the more polar insecticide imidacloprid present in contaminated dust waste, at concentrations of up to 20% (w/w). The effects of various experimental conditions, such as temperature, flow rate, and addition of modifier, on the recovery of the analytes were evaluated by extracting the pesticides both from spiked support material and from real dust samples. It was found that carbosulfan could easily be extracted from the dust waste within 30 min at 138 bar and 40 degrees C with a recovery of 98.9% (RSD = 2.3%, n = 10), compared to values obtained with a validated liquid extraction method. A sufficient removal of the more polar substance imidacloprid required the addition of a modifier, and the results showed a strong dependence of the extraction efficiency on the choice of modifier. Extractions at 276 bar and 80 degrees C with a solvent consisting of supercritical carbon dioxide modified with methanol (5%) gave a recovery of 97.0% (RSD = 3.6%, n = 10) using a 40 min extraction time. The results indicate that it seems to be possible to use process-scale SFE for the decontamination of pesticides from dust waste. The conditions outlined also permit analytical determinations of the two insecticides based on a combination of SFE and liquid chromatography.

Specific determination of benzene in urine using dynamic headspace and mass-selective detection.

A method for the determination of benzene in urine was developed, based on dynamic headspace and preconcentration of the analyte on a solid sorbent. The subsequent analysis by thermal desorption of the sorbent, capillary gas chromatography and mass-selective detection ascertained a low limit of detection (6.5 ng/l) and a highly specific determination. The limit of detection is an order of magnitude lower than that reported earlier and allows reliable quantitation of occupational exposure and of most environmental exposures. Samples could be stored frozen for at least a month without significant loss.

Supported liquid membrane coupled on-line to potentiometric stripping analysis at a mercury-coated reticulated vitreous carbon electrode for trace metal determinations in urine.

A method for trace metal determinations in complex matrices is presented. The method combines supported liquid membrane (SLM) sample clean-up and enrichment with potentiometric stripping analysis (PSA) in a flow system using reticulated vitreous carbon (RVC) as the electrode material. The membrane contained 40% m/m di-2-ethylhexylphosphoric acid dissolved in kerosene. Lead was used as a model substance in high-purity water and urine samples. The samples were enriched after a simple pH adjustment. The SLM enrichment time was 10 min when the last 5 min electrodeposition on the RVC electrode at -1.0 V (versus Ag/AgCl) was performed simultaneously. The influence of various experimental parameters such as deposition time, deposition potential and flow rate on the lead signal was investigated. With a 10 min SLM enrichment including a 5 min deposition time, the detection limit for lead was 0.3 microgram l-1. The relative standard deviation for lead concentrations in the range 4-20 micrograms l-1 was 0.05%. The overall SLM-PSA system was found to be stable for at least 100 urine analyses. The method was validated by running a reference urine sample. The result obtained (five replicates) was 9.7 micrograms l-1 (standard deviation 1.8 micrograms l-1) which is within the recommended range of 9.2-10.8 micrograms l-1.

On-line microporous membrane liquid-liquid extraction for sample pretreatment combined with capillary gas chromatography applied to local anaesthetics in blood plasma.

A new automated procedure for analyzing complex samples has been developed utilizing microporous membrane liquid-liquid extraction (MMLLE) combined with capillary gas chromatography. Some local anaesthetics were used as model compounds in aqueous solution as well as in blood plasma. The MMLLE procedure was performed in a flow system with the sample fed to the donor side of the hydrophobic microporous membrane and with an organic solvent (hexane) in the pores and as the acceptor solution. The analytes in a small volume of sample (< 1 mL) were extracted into the organic acceptor phase which was transferred into the gas chromatographic system by utilizing a loop-type interface compatible with large-volume (300 microL) injection. High selectivity and low carry-over effects were obtained with the system. The detection limits were 0.5-1 ng/mL using 0.5 mL of human plasma, and the precision was approximately 5%. The effects of pH, flow rates, and adsorption of the analytes were evaluated.

Sample preparation using a miniaturized supported liquid membrane device connected on-line to packed capillary liquid chromatography.

A miniaturized supported liquid membrane device has been developed for sample preparation and connected on-line to a packed capillary liquid chromatograph. The device consists of hydrophobic polypropylene hollow fiber, inserted and fastened in a cylindrical channel in a Kel-F piece. The pores of the fiber are filled with an organic solvent, in this study 6-undecanone, thus forming a liquid membrane. The sample is pumped on the outside of the hollow fiber (donor), and the analytes are selectively enriched and trapped in the fiber lumen (acceptor). With this approach, the volume of the acceptor solution can be kept as low as 1-2 µL. This stagnant acceptor solution is then transferred through capillaries attached to the fiber ends to the LC system. The system was tested with a secondary amine (bambuterol), as a model substance in aqueous standard solutions as well as in plasma. The best extraction efficiency in aqueous solution, with an acceptor volume of 1.9 µL, was 32.5% at a donor flow rate of 2.5 µL/min. At flow rates above 20 µL/min, the concentration enrichment per time unit was approximately constant, at 0.9 times/min, i.e., 9 times enrichment in about 10 min. The overall repeatability (RSD) for spiked plasma samples was ∼4% (n = 12). Linear calibration curves of peak area versus bambuterol concentration were obtained for both aqueous standard solutions and spiked plasma samples. The detection limit for bambuterol in plasma, after 10 min of extraction at a flow rate of 24 µL/min, was 80 nM.

Liquid membrane work-up of blood plasma samples applied to gas chromatographic determination of aliphatic amines.

A technique for sample work-up and enrichment using a supported liquid membrane in an automated flow system, connected to a gas chromatograph, was used for the determination of aliphatic amines in human blood plasma. The amines studied were N,N-dimethylethylamine, triethylamine, N-methylmorpholine, cyclohexylamine and N,N-dimethylcyclohexylamine. An efficient clean-up of the complex plasma matrix was achieved, resulting in identical blank chromatograms for plasma samples and aqueous solutions. Different parameters influencing the efficiency and selectivity of the extraction procedure were experimentally studied and theoretically explained. The detection limit depends on the extraction flow-rate and the available sample volume. With 1 ml of sample and a flow-rate giving an extraction time of 16 min, the detection limit was ca. 5 ppb (5 micrograms/l); with 4 ml of sample and a lower flow-rate, sub-ppb detection limits could be reached in ca. 3 h. Linear calibration curves up to 500 ppb were obtained. Blood plasma samples from volunteers exposed to N,N-dimethylethylamine in air were analysed, and the results compared favourably with independent measurements by another method.

Trace analysis of airborne 1,6-hexamethylenediisocyanate and the related aminoisocyanate and diamine by glass capillary gas chromatography.

A capillary gas chromatographic method was developed for the analysis of complex air mixtures of 1,6-hexamethylenediisocyanate, 1,6-hexamethyleneaminoisocyanate and 1,6-hexamethylenediamine. The method is based on derivatization in the sampling step of the reactive isocyanate groups to corresponding urethane groups by the alkaline ethanolic solvent and a subsequent derivatization of remaining amino groups to amide groups with heptafluorobutyric acid anhydride. The overall procedure, including sampling, gave a linear response at air concentrations of 3-300 micrograms/m3 for 1,6-hexamethylenediisocyanate with a precision of ca. 4% at 15 micrograms/m3 and a detection limit of ca. 0.2 microgram/m3 using nitrogen selective detection. In a field measurement of air concentrations in welding work on lacquered metal parts at a motor-car workshop, concentrations of 1,6-hexamethylenediisocyanate above 600 micrograms/m3 were found. Also 1,6-hexamethyleneaminoisocyanate and 1,6-hexamethylenediamine were found at concentrations of the order of 15% of the 1,6-hexamethylenediisocyanate concentration.

Determination of piperazine in working atmosphere and in human urine using derivatization and capillary gas chromatography with nitrogen- and mass-selective detection.

A reliable routine method is presented for the determination of piperazine down to the sub-ppm level in aqueous solutions and in urine. The method includes a two-phase derivatization procedure with ethyl- or isobutyl chloroformate as the reagent, followed by a capillary gas chromatographic determination using nitrogen- or mass selective detection. The addition of ammonia ensured a quantitative recovery. Detection limits for piperazine in urine were ca. 20 ng/ml using nitrogen-selective and ca. 1 ng/ml with mass-selective detection. The calibration plots were linear in the investigated range, 100-10,000 ng/ml with nitrogen-selective and 30-3000 ng/ml with mass-selective detection. The precision was ca. 6% at a concentration of 300 ng/ml. Acid anhydrides were investigated as alternative reagents in the two-phase derivatization procedure, and heptafluorobutyric acid anhydride in aqueous solutions gave approximately 100% recovery. However, in urine the recoveries of the investigated acid anhydride derivatives were unsatisfactory.

Simultaneous determination of amines and isocyanates in working atmospheres by gas-liquid chromatography.

A gas-liquid chromatographic method has been developed which permits trace analysis in a single run of amines and isocyanates occurring in relative concentrations of the order of 10,000:1. The amines were determined as free amines after alkalization of the sample and an extractive enrichment into toluene. The isocyanates were hydrolyzed to the corresponding amines during sampling in dilute sulphuric acid. This method gives the total isocyanate concentration. The use of a nitrogen sensitive detector and an alkali-treated packing, Pennwalt 223 with 4% KOH, are important prerequisites. Chromatographic parameters such as column liquid loading, injector impregnation and injector temperature were examined. Good separation between 2,4- and 2,6-toluenediamine isomers was obtained. The enrichment step was studied with respect to the degree of alkalization and to interferences from another amine present in large excess. The method was tested in the working atmospheres of two polyurethane foam factories. The substances determined were N-methylmorpholine, 1,4-diazabicyclo[2,2,2]octane and 2,4- and 2,6-toluenediisocyanates.