Full automation of solid-phase microextraction/on-fiber derivatization for simultaneous determination of endocrine-disrupting chemicals and steroid hormones by gas chromatography–mass spectrometry

A fully automated method using direct immersion solid-phase microextraction (DI-SPME) and headspace on-fiber silylation for simultaneous determinations of exogenous endocrine-disrupting chemicals (EDCs) and endogenous steroid hormones in environmental aqueous and biological samples by gas chromatography–mass spectrometry (GC-MS) was developed and compared to a previously reported manual method. Three EDCs and five endocrine steroid hormones were selected to evaluate this method. The extraction and derivatization time, ion strength, pH, incubation temperature, sample volume, and extraction solvent were optimized. Satisfactory results in pure water were obtained in terms of linearity of calibration curve (R ²=0.9932–1.0000), dynamic range (3 orders of magnitude), precision (4–9% RSD), as well as LOD (0.001–0.124 μg L⁻¹) and LOQ (0.004–0.413 μg L⁻¹), respectively. These results were similar to those obtained using a manual method, and moreover, the precision was improved. This new automated method has been applied to the determinations of target compounds in real samples used in our previous study on a manual SPME method. Exogenous octylphenol (OP), technical grade nonylphenol (t-NP), and diethylstilbestrol (DES) were at 0.13, 5.03, and 0.02 μg L⁻¹ in river water and 3.76, 13.25, and 0.10 μg L⁻¹ in fish serum, respectively. Natural steroid hormones estrone (E₁), 17β-estradiol (E₂), and testosterone (T) were at 0.19, 0.11, and 6.22 μg L⁻¹ in river water; and in female fish serum E₁, E₂, and pregnenolone (PREG) were at 1.37, 1.95, and 6.25 μg L⁻¹, respectively. These results were confirmed by the manual method. The developed fully automated SPME and on-fiber silylation procedures showed satisfactory applications in environmental analysis and the performances show improved precision and a reduced analysis time compared to the manual method.     

Improved extraction of glycol ethers from water by solid-phase micro extraction by carboxen polydimethylsiloxane-coated fiber

Summary 15 glycol ethers can be extracted from water by solidphase microextraction with a carboxen-polydimethyl-siloxane and separated by GC a Carbowax column. Water containing 15 glycol ethers at concentrations 0.1–10 mg.L⁻¹ is saturated at ambient temperature with NaCl. A carboxen-polydimethylsiloxane-coated fiber is then exposed to the liquid for 20 min and then automatically injected into a capillary GC injection port. Calibration curves were linear for different glycol ethers in the rang 0.1–10 mg.L⁻¹ Detection limits of each component of the mixture of glycol ethers between 50–500 μg.L⁻¹. The SPME method with direct immersion in water results in better sensivity than methods based on liquid-liquid extraction.     

Optimization of solid-phase microextraction procedures for the determination of tricyclic antidepressants and anticonvulsants in plasma samples by liquid chromatography

Simple, sensitive, and reproducible off-line solid-phase microextraction and liquid chromatography (SPME/LC) methods are described for the determination of seven anticonvulsants and tricyclic antidepressants in human plasma. Factorial design and simplex methodology were applied in the optimization of the SPME procedure for tricyclic antidepressants analyses. Important factors in the SPME efficiency are discussed, such as the fiber coatings (both lab-made and commercial), extraction time, pH, ionic strength, influence of plasma proteins, and desorption conditions. The development of the lab-made fiber coatings, namely, octadecylsilane, aminosilane, and polyurethane, are further described and applied to anticonvulsants analyses. The investigated plasmatic range for the evaluated anticonvulsants, using CW-TPR fiber, were the following: phenylethylmalonamide (3.00–40.0 μg mL⁻¹), phenobarbital (5.00–40.0 μg mL⁻¹), primidone (3.00–40.0 μg mL⁻¹), carbamazepine and carbamazepine-epoxide (2.00–24.0 μg mL⁻¹), phenytoin (2.00–40.0 μg mL⁻¹), and lamotrigine (0.50–12.0 μg mL⁻¹). The antidepressants’ linear plasmatic concentration ranged from 75.0 to 500 ng mL⁻¹ for imipramine, amitriptyline, and desipramine, and from 50.0 to 500 ng mL⁻¹ for nortriptyline, being in all cases, the limit of quantification represented by the lowest value. The precision (interassays) for all investigated drugs in plasma sample spiked with different concentrations of each analyte and submitted to the described procedures were lower than 15%. The off-line SPME/LC methodologies developed allow anticonvulsants and antidepressants analyses from therapeutic to toxic levels for therapeutic drug monitoring.     

Solid phase microextraction of volatile organic compounds using carboxen-polydimethylsiloxane fibers

Summary A new 80 μm Carboxen-polydimethylsiloxane (PDMS) fiber for solid phase microextraction (SPME) was tested for the enrichment of volatile organic compounds from water and air. Detection limits between 13 ng L⁻¹ (CH₂Cl₂) and 0.1 ng L⁻¹ (CHCl₂Br and CHClBr₂) for the combination: Carboxen-PDMS fiber and GC-ECD and between 35 ng L⁻¹ and 45 ng L⁻¹ (BTEX compounds) for the combination: Carboxen-PDMS and GC-FID using the headspace procedure were determined. Comparisons with the 100 μm PDMS fiber and further coatings show the advantages of the Carboxen-PDMS fiber with respect to extraction efficiency. Disadvantages of the new fiber compared with the 100 μm PDMS fiber are poorer repeatability and prolongation of equilibrium time. Distribution coefficients of the BTEX compounds between aqueous solution and SPME fiber coating were calculated and compared with the results of other researchers and with octanol-water partition coefficients.     

Multiresidue procedures for determination of triazine and organophosphorus pesticides in water by use of large-volume PTV injection in gas chromatography

Summary Two procedures, based on large-volume injection with a programmed-temperature vaporizer (PTV), have been developed for the determination of several triazine and organophosphorus pesticides. The use of PTV for injection in gas chromatography (GC) has enabled the introduction of up to 200 μL sample extract into the GC, thus increasing the sensitivity of the method. PTV injection has been combined off-line with two different microextraction procedures—liquid-liquid partition and solid-phase extraction. A simple and rapid off-line liquid-liquid microextraction procedure (5 mL water/1 mL methyltert-butyl ether) was applied to surface water samples spiked at levels between 0.01 and 5μg L⁻¹. Recoveries of the overall procedure were >80% and the precision was better than 15%. Detection limits were <30 ngL⁻¹ from 200-μL injections in GC-NPD analysis of triazines and GC-FPD analysis of organophosphorus pesticides. Off-line automated solid-phase extraction with C₁₈ cartridges has been applied to water samples (50 mL) spiked at 0.01, 0.1 and 1 μg L⁻¹. The overall procedure was satisfactory (recoveries >80% and coefficients of variation <12%) and the limits of detection ranged from 1 to 9 ng L⁻¹. Finally, several surface water samples were anlysed, and triazine herbicides were detected at concentrations of approx. 0.1–0.2 μg L⁻¹. The results were similar to those obtained by conventional solvent extraction then GC-MSD after splitless injection of 2 μL.     

Quantitative Analysis of aliphatic aldehydes by headspace SPME sampling and ion-trap GC-MS

Summary A fast and simple headspace SPME sampling method has been developed for quantification of volatile aliphatic aldehydes in sunflower oil. Analysis has been performed by gas chromatography, on a 30m×0.25 mm i.d. ×0.25 μm CP-Wax 52CB column, with mass spectrometric detection. Carryover from the SPME fiber could be eliminated by heating the fiber in the injection port between runs. Response factors of all the compounds were linear for concentrations up to 100 ng μL⁻¹. The slopes of the calibration curves decrease with the amount of saturation of the aldehydes. The average responses for unsaturated aldehydes were twice as high as those for the saturated variety. Responses for dienes were approximately one order of magnitude higher than for saturated aldehydes. Depletion of the analyte was examined by repeated extraction from the same vial. SPME was optimized—after 30 min extraction most components were found to have reached equilibration. The detection limit for the compounds studied varied between 0.1 and 1 ng μL⁻¹. Distribution constants were determined for ten different aldehydes and Henry's constants were calculated for unsaturated aldehydes. There was a definite relationship between the response factors and the amount of saturation of the aldehydes. Presented at: Balaton Symposium on High-Performance Separation Methods, Siófok, Hungary, September 3–5, 1997     

Application of ethyl chloroformate derivatization for solid-phase microextraction–gas chromatography–mass spectrometric determination of bisphenol-A in water and milk samples

A simple and rapid analytical method based on in-matrix ethyl chloroformate (ECF) derivatization has been developed for the quantitative determination of bisphenol-A (BPA) in milk and water samples. The samples containing BPA were derivatised with ECF in the presence of pyridine for 20 s at room temperature, and the non-polar derivative thus formed was extracted using polydimethylsiloxane solid-phase microextraction (SPME) fibres with thicknesses of 100 μm followed by analysis using gas chromatography–mass spectrometry. Three alkyl chloroformates (methyl, ethyl and isobutyl chloroformate) were tested for optimum derivatisation yields, and ECF has been found to be optimum for the derivatisation of BPA. Several parameters such as amount of ECF, pyridine and reaction time as well as SPME parameters were studied and optimised in the present work. The limit of detection for BPA in milk and water samples was found to be 0.1 and 0.01 μg L⁻¹, respectively, with a signal-to-noise ratio of 3:1. The limit of quantitation for BPA in milk and water was found to be 0.38 and 0.052 μg L⁻¹, respectively, with a signal-to-noise ratio of 10:1. In conclusion, the method developed was found to be rapid, reliable and cost-effective in comparison to silylation and highly suitable for the routine analysis of BPA by various food and environmental laboratories.     

New cloud vapor zone (CVZ) coupled headspace solid-phase microextraction technique

A new cloud vapor zone (CVZ)-based headspace solid-phase microextraction (HS-SPME) technique has been demonstrated with the capability of heating the sample matrix and simultaneously cooling the sampling zone. A bi-temperature-controlled (BTC) system, allowing 10 mL of test sample heating and headspace external-cooling, was employed for the CVZ formation around the SPME-fiber sampling area. In the CVZ procedure, the heated headspace vapor undergoes a sudden cooling near the SPME to form a dense cloud of analyte–water vapor, which is helpful for adsorption or absorption of the analyte. The device was evaluated for the quantitative analysis of aqueous chlorothalonil. Parameters influencing sampling efficiency, e.g., SPME fiber coating, SPME sampling temperature and time, solution modifier, addition of salt, sample pH, and temperature, were investigated and optimized thoroughly. The proposed BTC-HS-SPME method afforded a best extraction efficiency of above 94% accuracy (less than 4.1% RSD, n = 7) by using the PDMS fiber to collect chlorothalonil in the headspace at 5 °C under the optimized condition, i.e., heating sample solution (added as 10% ethylene glycol and 30% NaCl, at pH 7.0) at 130 °C for 15 min. The detection was linear from 0.01 to 80 μg L⁻¹ with a regression coefficient of 0.9998 and had a detection limit of 3.0 ng L⁻¹ based on S/N = 3. Practical application was demonstrated by analyzing chlorothalonil in farm water samples with promising results and recoveries. The approach provided a very simple, fast, sensitive, and solvent-free procedure to collect analytes from aqueous solution. The approach can provide a new platform for other sensitive HS-SPME assays.     

Determination of glyphosate and AMPA in surface and waste water using high-performance ion chromatography coupled to inductively coupled plasma dynamic reaction cell mass spectrometry (HPIC–ICP–DRC–MS)

A novel method employing high-performance cation chromatography in combination with inductively coupled plasma dynamic reaction cell mass spectrometry (ICP–DRC–MS) for the simultaneous determination of the herbicide glyphosate (N-phosphonomethylglycine) and its main metabolite aminomethyl phosphonic acid (AMPA) is presented. P was measured as ³¹P¹⁶O⁺ using oxygen as reaction gas. For monitoring the stringent target value of 0.1 μg L⁻¹ for glyphosate, applicable for drinking and surface water within the EU, a two-step enrichment procedure employing Chelex 100 and AG1-X8 resins was applied prior to HPIC–ICP–MS analysis. The presented approach was validated for surface water, revealing concentrations of 0.67 μg L⁻¹ glyphosate and 2.8 μg L⁻¹ AMPA in selected Austrian river water samples. Moreover, investigations at three waste water-treatment plants showed that elimination of the compounds at the present concentration levels was not straightforward. On the contrary, all investigated plant effluents showed significant amounts of both compounds. Concentration levels ranged from 0.5–2 μg L⁻¹ and 4–14 μg L⁻¹ for glyphosate and AMPA, respectively.     

Solid-phase microextraction and gas chromatography-mass spectrometry for determination of monoaromatic hydrocarbons in blood and urine: Application to people exposed to air pollutants

Summary To assess individual exposure to monoaromatic hydrocarbons (benzene, toluene, ethylbenzene and xylenes-BTEX) in biological fluids, a GC-MS method was developed. Headspace sampling of BTEX was by solidphase microextraction (SPME) with a 75 μm Carboxenpolydimethylsiloxane (PDMS) fiber. Linearity was established for concentrations up to 50 μg L⁻¹. Detection limits, calculated both in human blood and urine, ranged 5–10 ng L⁻¹. Repeatability was in the range 6.5–9.2% for all compounds. The method was applied to the evaluation of the internal dose of BTEX in a group of cyclists running for 2 h within city routes. Benzene and toluene in blood, and toluene and xylenes in urine significantly increased after exercise as compared to prerun values, such changes being consistent with airborne concentrations. The combination of SPME with GC-MS seems to represent an appropriate analytical approach to detect changes in the concentration of monoaromatic hydrocarbons in biological media resulting from exposure to environmental pollution.     

Trace analysis of sulfonylurea herbicides and their metabolites in water using a combination of off-line or on-line solid-phase extraction and liquid chromatography–tandem mass spectrometry

Two alternatives for the rapid simultaneous quantification of six sulfonylurea herbicides and five of their main degradation products in natural water are proposed. For concentration, the compounds were extracted on a polystyrene–divinylbenzene solid phase under pH and elution conditions that suppressed any hydrolysis. The eluates were analysed by liquid chromatography coupled to electrospray tandem mass spectrometry within 20 min. The whole method was validated and shown to give no hydrolysis artefacts. The application of off-line and on-line SPE of sulfonylureas enabled the 0.1 μg L⁻¹ and 1 ng L⁻¹ LOQ levels to be reached, respectively. The on-line SPE–LC–MS–MS method allowed the accurate quantitation of all sulfonylureas and three degradation products at 0.1 μg L⁻¹ or below in natural water, with an average repeatability of 8%.     

Determination of Trihalomethanes in Drinking Water by Dispersive Liquid–Liquid Microextraction then Gas Chromatography with Electron-Capture Detection

Dispersive liquid–liquid microextraction (DLLME) has been used for preconcentration of trihalomethanes (THMs) in drinking water. In DLLME an appropriate mixture of an extraction solvent (20.0 μL carbon disulfide) and a disperser solvent (0.50 mL acetone) was used to form a cloudy solution from a 5.00-mL aqueous sample containing the analytes. After phase separation by centrifugation the enriched analytes in the settled phase (6.5 ± 0.3 μL) were determined by gas chromatography with electron-capture detection (GC–ECD). Different experimental conditions, for example type and volume of extraction solvent, type and volume of disperser solvent, extraction time, and use of salt, were investigated. After optimization of the conditions the enrichment factor ranged from 116 to 355 and the limit of detection from 0.005 to 0.040 μg L⁻¹. The linear range was 0.01–50 μg L⁻¹ (more than three orders of magnitude). Relative standard deviations (RSDs) for 2.00 μg L⁻¹ THMs in water, with internal standard, were in the range 1.3–5.9% (n = 5); without internal standard they were in the range 3.7–8.6% (n = 5). The method was successfully used for extraction and determination of THMs in drinking water. The results showed that total concentrations of THMs in drinking water from two areas of Tehran, Iran, were approximately 10.9 and 14.1 μg L⁻¹. Relative recoveries from samples of drinking water spiked at levels of 2.00 and 5.00 μg L⁻¹ were 95.0–107.8 and 92.2–100.9%, respectively. Comparison of this method with other methods indicates DLLME is a very simple and rapid (less than 2 min) method which requires a small volume of sample (5 mL).     

Determination of a wide range of volatile and semivolatile organic compounds in snow by use of solid-phase micro-extraction (SPME)

Quantification and transformation of organic compounds are pivotal in understanding atmospheric processes, because such compounds contribute to the oxidative capacity of the atmosphere and drive climate change. It has recently been recognized that chemical reactions in snow play a role in the production or destruction of photolabile volatile organic compounds (VOC). We present an environmentally friendly method for determination of VOC and semi-VOC in snow collected at three sites—remote, urban, and (sub-)arctic. A solid-phase micro-extraction (SPME) procedure was developed and (semi-)VOC were identified by gas chromatography with mass spectrometric detection (GC–MS). A broad spectrum of (semi-)VOC was found in snow samples, including aldehydes, and aromatic and halogenated compounds. Quantification was performed for 12 aromatic and/or oxygenated compounds frequently observed in snow by use of neat standard solutions. The concentrations detected were between 0.12 (styrene and ethylbenzene) and 316 μg L⁻¹ (toluene) and limits of detection varied between 0.11 (styrene) and 1.93 μg L⁻¹ (benzaldehyde). These results indicate that the SPME technique presented is a broad but selective, versatile, solvent-free, ecological, economical, and facile method of analysis for (semi-)VOC in natural snow samples.     

Sensitive measurement of ultratrace phenols in natural water by purge-and-trap with in situ acetylation coupled with gas chromatography-mass spectrometry

A novel purge-and-trap method coupled with gas chromatography-mass spectrometry (GC-MS) is developed for the analysis of trace and ultratrace phenols based on their derivatization with acetic anhydride. Parameters affecting the extraction efficiency, such as purge temperature, concentration of sodium chloride, purge time, and volume of derivatization reagent, were investigated. The optimized conditions were addition of 150 μL acetic anhydride, purge time of 25 min at the purge temperature of 60 °C with 30% NaCl. The linear range was 0.2–100 μg L⁻¹ for phenols. The limits of detection (LODs) ranged from 0.08 to 0.15 μg L⁻¹ and the relative standard deviations (RSDs) for most of the phenols at the 10 μg L⁻¹level were below 10%. Natural water samples collected from a pool were successfully analyzed using the proposed method. The recovery of spiked water samples was 72.9–84.2%.     

Characterization of volatile organic compounds and odors by in-vivo sampling of beef cattle rumen gas, by solid-phase microextraction, and gas chromatography–mass spectrometry–olfactometry

Volatile organic compounds (VOCs) and odors in cattle rumen gas have been characterized by in-vivo headspace sampling by solid-phase microextraction (SPME) and analysis by gas chromatography–mass spectrometry–olfactometry (GC–MS–O). A novel device enabling headspace SPME (HS-SPME) sampling through a cannula was designed, refined, and used to collect rumen gas samples from steers. A Carboxen–polydimethylsiloxane (PDMS) fiber (85 μm) was used for SPME sampling. Fifty VOCs from ten chemical groups were identified in the rumen headspace. The VOCs identified had a wide range of molecular weight (MW) (34 to 184), boiling point (−63.3 to 292 °C), vapor pressure (1.05 × 10⁻⁵ to 1.17 × 10² Pa), and water solubility (0.66 to 1 × 10⁶ mg L⁻¹). Twenty-two of the compounds have a published odor detection thresholds (ODT) of less than 1 ppm. More than half of the compounds identified are reactive and have an estimated atmospheric lifetime of <24 h. The amounts of VFAs, sulfide compounds, phenolic compounds, and skatole, and the odor intensity of VFAs and sulfide compounds in the rumen gas were all higher after feeding than before feeding. These results indicate that rumen gases can be an important potential source of aerial emissions of reactive VOCs and odor. In-vivo sampling by SPME then GC–MS–O analysis can be a useful tool for qualitative characterization of rumen gases, digestion, and its relationship to odor and VOC formation. Figure Modified cannula for rumen gas sampling with SPME     

Determination of triazine herbicides: development of an electroanalytical method utilizing a solid amalgam electrode that minimizes toxic waste residues, and a comparative study between voltammetric and chromatographic techniques

The use of a copper solid amalgam electrode (CuSAE) for the analytical determination of triazine herbicides (atrazine and ametryne) instead of the conventional hanging mercury drop electrode (HMDE) is reported. The results obtained using electroanalytical methods utilizing each of these electrodes were also compared with those provided by the HPLC technique. The results indicated that the CuSAE electrode can be used to detect the herbicides studied, since the detection limits reached using the electrode (3.06 μg L⁻¹ and 3.78 μg L⁻¹ for atrazine and ametryne, respectively) are lower than the maximum values permitted by CONAMA (Brazilian National Council for the Environment) for wastewaters (50 μg L⁻¹) and by the US EPA (Environmental Protection Agency of the United States) in natural water samples (10.00 μg L⁻¹). An electroanalytical methodology employing CuSAE and square wave voltammetry (SWV) was successfully applied to the determination of atrazine and ametryne in natural water samples, yielding good recoveries (70.30%–79.40%). This indicates that the CuSAE provides a convenient substitute for the HMDE, particularly since the CuSAE minimizes the toxic waste residues produced by the use of mercury in HDME-based analyses.     

Data correlation in on-line solid-phase extraction-gas chromatography-atomic emission/mass spectrometric detection of unknown microcontaminants

Summary A procedure is described for the (non-target) screening of hetero-atom-containing compounds in tap and waste water by correlating data obtained by gas chromatography (GC) using atomic emission (AED) and mass selective (MS) detection. Solid-phase extraction (SPE) was coupled on-line to both GC systems to enable the determination of microcontaminants at the 0.02–1 μg L⁻¹ level in 7–50 mL of aqueous sample. The screening was limited to compounds present in at least one heteroatom-selective GC-AED trace above a predetermined concentration level. These compounds were identified by their partial formulae (AED) and the corresponding mass spectra, which were obtained from the GC-MS chromatogram via the retention index concept. The potential of the approach was demonstrated by the identification of target compounds as well as all unknowns present in tap and waste water above the predetermined threshold of 0.05 μg L⁻¹ (tap water) or 0.5 μg L⁻¹ (waste water).     

CE Analysis of Cephalosporins in Environmental Waters

This study investigates an off-line solid phase extraction (SPE) for improving the sensitivity in the capillary electrophoretic (CE) analysis of four cephalosporins. Two sorbents—LiChrolut-C₁₈ and Oasis HLB—were used in a SPE process to detect cephalosporins in natural waters (tap, river and hospital sewage) and their performances were compared. By using Oasis HLB sorbent higher recoveries for river water were obtained (94–107% when 500 mL of sample were analyzed). The off-line SPE–CZE method was validated for river water with good detection limits (3 μg L⁻¹) and the linearity ranged between 5 and 200 μg L⁻¹.     

A novel kinetic-spectrophotometric method for determination of nitrites in water

A simple, selective and sensitive kinetic method for the determination of nitrite in water was developed. The method is based on the catalytic effect of nitrite on the oxidation of methylene blue (MB) with bromate in a sulfuric acid medium. During the oxidation process, absorbance of the reaction mixture decreases with the increasing time, inversely proportional to the nitrite concentration. The reaction rate was monitored spectrophotometrically at λ = 666 nm within 30 s of mixing. Linear calibration graph was obtained in the range of 0.005–0.5 μg mL⁻¹ with a relative standard deviation of 2.09 % for six measurements at 0.5 μg mL⁻¹. The detection limit was found to be 0.0015 μg mL⁻¹. The effect of different factors such as acidity, time, bromate concentration, MB concentration, ionic strength, and order of reactants additions is reported. Interference of the most common foreign ions was also investigated. The optimum experimental conditions were: 0.38 mol L⁻¹ H₂SO₄, 5 × 10.4 mol L⁻¹ KBrO₃, 1.25 × 10.5 mol L⁻¹ MB, 0.3 mol L⁻¹ sodium nitrate, and 25°C. The proposed method was conveniently applied for the determination of nitrite in spiked drinking water samples.     

Improvement of HS-SPME for analysis of volatile organic compounds (VOC) in water samples by simultaneous direct fiber cooling and freezing of analyte solution

The sensitivity and precision of headspace solid-phase micro extraction (HS-SPME) at an analyte solution temperature (T _as) of +35 °C and a fiber temperature (T _fiber) of +5 °C were compared with those for HS-SPME at T _as and T _fiber of −20 °C for analysis of the volatile organic compounds benzene, 1,1,1-trichloroethane, trichloroethylene, toluene, o-xylene, ethylbenzene, m/p-xylene, and tetrachloroethylene in water samples. The effect of simultaneous fiber cooling and analyte solution freezing during extraction was studied. The compounds are of different hydrophobicity, with octanol/water partition coefficients (Kow) ranging from 126 and 2511. During a first set of experiments the polydimethylsiloxane (PDMS) SPME fiber was cooled to +5 °C with simultaneous heating of the aqueous analyte solution to +35 °C. During a second set of experiments, both SPME fiber holder and samples were placed in a deep freezer maintained at −20 °C for a total extraction time of 30 min. After approximately 2 min the analyte solution in the vial began to freeze from the side inwards and from the bottom upwards. After approximately 30 min the solution was completely frozen. Analysis of VOC was performed by coupling HS-SPME to gas chromatography-mass spectrometry (GC-MS). In general, i.e. except for tetrachloroethylene, the sensitivity of HS-SPME increased with increasing compound hydrophobicity at both analyte solution and fiber temperatures. At T _as of +35 °C and T _fiber of +5 °C detection limits of HS-SPME were 0.5 μg L⁻¹ for benzene, 1,1,1-trichloroethane, trichloroethylene, and tetrachloroethylene, 0.125 μg L⁻¹ for toluene, and 0.025 μg L⁻¹ for ethylbenzene, m/p-xylene, and o-xylene. In the experiments with T _as and T _fiber of −20 °C, detection limits were reduced for compounds of low hydrophobicity (Kow<501), for example benzene, toluene, 1,1,1-trichloroethane, and trichloroethylene. In the concentration range 0.5–62.5 μg L⁻¹, the sensitivity of HS-SPME was enhanced by a factor of approximately two for all compounds by performing the extraction at −20 °C. A possible explanation is that freezing of the water sample results in higher concentration of the target compounds in the residual liquid phase and gas phase (freezing-out), combined with enhanced adsorption of the compounds by the cooled fiber. The precision of HS-SPME, expressed as the relative standard deviation and the linearity of the regression lines, is increased for more hydrophobic compounds (Kow>501) by simultaneous direct fiber cooling and freezing of analyte solution. Background contamination during analysis is reduced significantly by avoiding the use of organic solvents.