Selected ion flow tube mass spectrometry of exhaled breath condensate headspace

Collection of exhaled breath condensate (EBC) is a relatively simple noninvasive method of breath analysis; however, no data have been reported that would relate concentration of volatile compounds in EBC to their gaseous concentrations in exhaled air. The aim of the study was to investigate which volatile compounds are present in EBC and how their concentrations relate to results of direct breath analysis. Thus, samples of EBC were collected in a standard way from several subjects and absolute levels of several common volatile breath metabolites (ammonia, acetone, ethanol, methanol, propanol, isoprene, hydrogen cyanide, formaldehyde and acetaldehyde) were then determined in their headspace using selected ion flow tube mass spectrometry (SIFT-MS). Results are compared with those from on-line breath analyses carried out immediately before collecting the EBC samples. It has been demonstrated that SIFT-MS can be used to quantify the concentrations of volatiles in EBC samples and that, for methanol, ammonia, ethanol and acetone, the EBC concentrations correlate with the direct breath levels. However, the EBC concentrations of isoprene, formaldehyde, acetaldehyde, hydrogen cyanide and propanol do not correlate with direct breath measurements. Copyright (c) 2008 John Wiley & Sons, Ltd.     

An exploratory comparative study of volatile compounds in exhaled breath and emitted by skin using selected ion flow tube mass spectrometry

Selected ion flow tube mass spectrometry (SIFT-MS) has been used to carry out a pilot parallel study on five volunteers to determine changes occurring in several trace compounds present in exhaled breath and emitted from skin into a collection bag surrounding part of the arm, before and after ingesting 75 g of glucose in the fasting state. SIFT-MS enabled real-time quantification of ammonia, methanol, ethanol, propanol, formaldehyde, acetaldehyde, isoprene and acetone. Following glucose ingestion, blood glucose and trace compound levels were measured every 30 min for 2 h. All the above compounds, except formaldehyde, were detected at the expected levels in exhaled breath of all volunteers; all the above compounds, except isoprene, were detected in the collection bag. Ammonia, methanol and ethanol were present at lower levels in the bag than in the breath. The aldehydes were present at higher levels in the bag than in breath. The blood glucose increased to a peak about 1 h post-ingestion, but this change was not obviously correlated with temporal changes in any of the compounds in breath or emitted by skin, except for acetone. The decrease in breath acetone was closely mirrored by skin-emitted acetone in three volunteers. Breath and skin acetone also clearly change with blood glucose and further work may ultimately enable inferences to be drawn of the blood glucose concentration from skin or breath measurements in type 1 diabetes.     

Alcohol in breath and blood: a selected ion flow tube mass spectrometric study

In this paper we compare the amounts of ethanol in breath and in blood after ingestion of whisky using analysis by selected ion flow tube mass spectrometry (SIFT-MS). Blood ethanol concentrations were also obtained using standard hospital forensic procedures for blood alcohol analyses. We demonstrate the quantitative nature of SIFT-MS analysis by correlating the observed alcohol content of the headspace above 5-mL amounts of venous blood and aqueous solution to which known trace amounts of alcohol have been added. This procedure provides a Henry's Law coefficient for ethanol in aqueous solution at 298 +/- 3 K of 209 +/- 7 mol/kg*bar. We also demonstrate that measurement of the ethanol concentration in the alveolar portion of a single breath using the SIFT-MS technique gives an accurate measure of blood alcohol and could obviate the need for blood samples in forensic processing. The storage performance of breath samples in Mylar bags with a volume greater than 1 L has been shown to maintain the mixture integrity for ethanol but not for some other species.     

Selected ion flow tube mass spectrometry for on-line trace gas analysis in biology and medicine

Selected ion flow tube mass spectrometry, (SIFT-MS), is a technique for simultaneous real-time quantification of several trace gases in air and exhaled breath. It relies on chemical ionization of the trace gas molecules in air/breath samples introduced into helium carrier gas, using H(3)O(+), NO(+) and O(2)(+) reagent (precursor ions). Reactions between the precursor ions and the trace gas molecules proceed for an accurately defined time, the precursor and product ions being detected and counted by a downstream mass spectrometer. Absolute concentrations of trace gases in single breath exhalation can be determined by SIFT-MS down to parts-per-billion (ppb) levels, obviating sample collection into bags or onto traps. Calibration using chemical standards is not required, as the concentrations are calculated using the known reaction rate constants and measured flow rates and pressures. SIFT-MS has been used for many pilot investigations in several areas of research, especially as a non-invasive breath analysis tool to investigate physiological processes in humans and animals, for clinical diagnosis and for therapeutic monitoring. Examples of the results obtained from several such studies are outlined to demonstrate the potential of SIFT-MS for trace gas analysis of air, exhaled breath and the headspace above liquids.     

Time-resolved selected ion flow tube mass spectrometric quantification of the volatile compounds generated by E. coli JM109 cultured in two different media

Preliminary measurements have been made of the volatile compounds emitted by the bacterium E. coli JM109 cultured in the commonly used media Dulbecco's modified Eagle's medium (DMEM) and lysogeny broth (LB) using selected ion flow tube mass spectrometry, SIFT-MS, as a step towards the real time, non-invasive monitoring of accidental infections of mammalian cell cultures. In one procedure, the culture medium alone and the E. coli cells/medium combination were held at 37 °C in bottles sealed with septa for a given time period, usually overnight, to allow the bacterium to proliferate, after which the captured headspace was analysed directly by SIFT-MS. Several compounds were seen to be produced by the E. coli cells that depended on the liquid medium used: when cultured in DMEM, copious amounts of ethanol, acetaldehyde and hydrogen sulphide were produced; in LB ammonia is the major volatile product. In a second procedure, to ensure aerobic conditions prevailed in the cell culture, selected volatile compounds were monitored by SIFT-MS in real time for several hours above the open-to-air E. coli/DMEM culture held at close to 37 °C. The temporal variations in the concentrations of some compounds, which reflect their production rates in the culture, indicate maxima. Thus, the maxima in the ethanol and acetaldehyde production are a reflection of the reduction of glucose from the DMEM by the vigorous E. coli cells and the maximum in the hydrogen sulphide level is an indication of the loss of the sulphur-bearing amino acids from the DMEM. Serendipitously, emissions from DMEM inadvertently infected with the bacterium C. testosteroni were observed when large quantities of ammonia were seen to be produced. The results of this preliminary study suggest that monitoring volatile compounds might assist in the early detection of bacterial infection in large-scale bioreactors.     

Analysis of petrol and diesel vapour and vehicle engine exhaust gases using selected ion flow tube mass spectrometry

We have used selected ion flow tube mass spectrometry (SIFT-MS) to analyse the vapours emitted by petrol and diesel fuels and the exhaust gases from petrol (spark ignition) and diesel (compression ignition) engine vehicles fitted with catalytic converters. Only those components of these media that have significant vapour pressures at ambient temperatures were analysed and thus particulates were obviously not detected. These media have been analysed using the full scope of SIFT-MS, i.e., with the three available precursor ions H3O+, NO+ and O2+. The combination of the H3O+ and NO+ analyses is seen to be essential to distinguish between different product ions at the same mass-to-charge ratio (m/z) especially in identifying aldehydes in the exhaust gases. The O2+ precursor ions are used to detect and quantify the large amount of nitric oxide present in the exhaust gases from both engine types. The petrol and diesel vapours consist almost exclusively of aliphatic alkanes, alkenes and alkynes (and dienes) and aromatic hydrocarbons. Some of these compounds appear in the exhaust gases together with several aldehydes, viz. formaldehyde, acetaldehyde, pentanal, pentenal (acrolein), butenal, and also methanol and ethanol. Acetone, nitric oxide and ammonia are also present, acetone and nitric oxide being much more abundant in the diesel exhaust gas than in the petrol exhaust gas. These data were obtained from samples collected into pre-evacuated stainless steel vessels. Trapping of the volatile compounds from the gas samples is not required and analysis was completed a few minutes later. All the above compounds are detected simultaneously, which demonstrates the value of SIFT-MS in this area of research.     

On-line measurement of the absolute humidity of air, breath and liquid headspace samples by selected ion flow tube mass spectrometry

We describe how selected ion flow tube mass spectrometry (SIFT-MS) can be used to determine the absolute humidity of air, breath and liquid headspace samples. This involves the determination of the relative count rates of the H3O+ ions and those H3O+.(H2O)(1,2,3) hydrate ions that inevitably form in the helium carrier gas when humid samples are being analysed by SIFT-MS using H3O+ precursor ions. This requires an understanding of the kinetics of hydrated hydronium ion formation, the involvement of mass discrimination in the analytical quadrupole mass spectrometer and the decreased diffusive loss of the heavier hydrates along the flow tube. Thus, we show that the humidity of breath and liquid headspace samples, typically at the few percent level, can be directly obtained on-line to the SIFT-MS instrument along with the concentrations of trace gases, which are present at much lower levels. We emphasise the value of parallel humidity measurements in ensuring good real-time sampling of breath and liquid headspace and the value of such measurements to trace gas analysis using SIFT-MS.     

Development and validation of an automated monitoring system for oxygenated volatile organic compounds and nitrile compounds in ambient air

Few studies were conducted on oxygenated volatile organic compounds (OVOC) because of problems encountered during the sampling/analyzing steps induced by water in sampled air. Consequently, there is a lack of knowledge of their spatial and temporal trends and their origins in ambient air. In this study, an analyzer consisted of a thermal desorber (TD) interfaced with a gas chromatograph (GC) and a flame ionization detector (FID) was developed for online measurements of 18 OVOC in ambient air including 4 alcohols, 6 aldehydes, 3 ketones, 3 ethers, 2 esters and 4 nitriles. The main difficulty was to overcome the humidity effect without loss of compounds. Water amount in the sampled air was reduced by the trap composition (two hydrophobic graphitized carbons—Carbopack B:Carbopack X), the trap temperature (held at 12.5 °C), by diluting (50:50) the sample with dry air before the preconcentration step and a trap purge with helium. Humidity management allowed the use of a polar CP-Lowox column in order to separate the polar compounds from the hydrocarbon/aromatic matrix. The safe sampling volume for the dual-sorbent trap 75 mg Carbopack X:5 mg Carbopack B was found to 405 mL for ethanol by analyzing a standard mixture at a relative humidity of 80%. Detection limits ranging from 10 ppt for ETBE to 90 ppt for ethanol were obtained for 18 compounds for a sampling volume of 405 mL. Good repeatabilities were obtained at two levels of concentration (relative standard deviation <5%). The calibration (ranging from 0.5 to 10 ppb) was set up at three different levels of relative humidity to test the humidity effect on the response coefficients. Results showed that the response coefficients of all compounds were less affected by humidity except for those of ethanol and acetonitrile (decrease respectively of 30% and 20%). The target compounds analysis shows good reproducibility with response coefficient variability of less then 10% of the mean initial value of calibration for all the compounds. Hourly ambient air measurements were conducted in an urban site in order to test this method. On the basis of these measurements, ethanol, acetone and acetaldehyde have shown the highest concentration levels with an average of 2.10, 1.75 and 1.37 ppb respectively. The daily evolution of some OVOC, namely ethanol and acetaldehyde, was attributed to emissions from motor vehicles while acetone has a different temporal evolution that can be probably associated with remote sources.     

Detection of monobromamine, monochloramine and dichloramine using selected ion flow tube mass spectrometry and their relevance as breath markers

We report a fast, sensitive, real-time method to measure monobromamine, monochloramine and dichloramine using selected ion flow tube mass spectrometry (SIFT-MS). Relative rate coefficients and product distributions are reported for the reagent ions H3O+ and O2 +. Rapid reactions with the haloamines were observed with H3O+ and O2 + but no fast reaction was found with NO+. A slow reaction between NO+ and dichloramine was observed. We demonstrate the feasibility of determining these compounds in a single human breath for which the limit of detection is approaching 10 parts per billion (ppb). We also report preliminary measurements of these compounds in the breath of individuals where the concentrations of bromamine and chloramine ranged from 10 to 150 ppb.     

Quantification of volatile compounds in the headspace of aqueous liquids using selected ion flow tube mass spectrometry

We describe a method by which the concentrations of volatile compounds in the headspace of their dilute aqueous solutions in sealed containers can be determined using on-line selected ion flow tube mass spectrometry (SIFT-MS). Thus, the changing number density of the molecules of the volatile compound in the carrier gas of the SIFT-MS instrument is described in terms of its changing flow rate as the pressure in the sealed container decreases during the sampling procedure. It is shown that the best analytical procedure is to determine the mean concentration of the trace gas in the liquid headspace over a given sampling time and relate this to the required concentration, which is the initial equilibrium concentration established before the pressure in the sealed container reduces significantly. To test the validity of this analytical approach, the headspace concentrations of acetaldehyde, ethanol and acetone above aqueous solutions of known concentrations have been determined. Hence, the Henry's Law constants for these compounds have been determined and found to agree with the published values. The confirmation of the quality of this sampling methodology combined with SIFT-MS for the analysis of volatile compounds in liquid headspace paves the way for the rapid analyses of biological liquids such as urine and serum for clinical diagnosis and physiological monitoring.     

On-line analysis of diesel engine exhaust gases by selected ion flow tube mass spectrometry

Selected ion flow tube mass spectrometry (SIFT-MS) has been used to analyse on-line and in real time the exhaust gas emissions from a Caterpillar 3304 diesel engine under different conditions of load (idle and 50% of rated load) and speed (910, 1500 and 2200 rpm) using three types of fuel: an ultra-low-sulphur diesel, a rapeseed methyl ester and gas oil. SIFT-MS analyses of the alkanes, alkenes and aromatic hydrocarbons in the headspace of these fuels were also performed, but the headspace of the rapeseed methyl ester consists mainly of methanol and a compound with the molecular formula C4H8O. The exhaust gases were analysed for NO and NO2 using O2+* reagent ions and for HNO2 using H3O+ reagent ions. The following aldehydes and ketones in the exhaust gases were quantified by using the combination of H3O+ and NO+ reagent ions: formaldehyde, acetaldehyde, propenal, propanal, acetone, butanal, pentanal, butanone and pentanone. Formaldehyde, acetaldehyde and pentenal, all known respiratory irritants associated with sensitisation to asthma of workers exposed to diesel exhaust, are variously present within the range 100-2000 ppb. Hydrocarbons in the exhaust gases accessible to SIFT-MS analyses were also quantified as total concentrations of the various isomers of C3H4, C3H6, C4H6, C5H8, C5H10, C6H8, C6H10, C7H14, C6H6, C7H8, C8H10 and C9H12.     

Influence of weakly bound adduct ions on breath trace gas analysis by selected ion flow tube mass spectrometry (SIFT-MS)

This paper describes how weakly bound adduct ions form when the precursor ions used in selected ion flow mass spectrometry, SIFT-MS, analyses, viz. H₃O⁺, NO⁺ and O₂⁺, associate with the major components of air and exhaled breath, N₂, O₂ and CO₂. These adduct ions, which include H₃O⁺N₂, H₃O⁺CO₂, NO⁺CO₂, O₂⁺O₂ and O₂⁺CO₂, are clearly seen when dry air containing 5% CO₂ (typical of that in exhaled breath) is analysed using SIFT-MS. These adduct ions must not be misinterpreted as characteristic product ions of trace gases; if so, serious analytical errors can result. However, when exhaled breath is analysed these adduct ions are partly removed by ligand switching reactions with the abundant water molecules and the problems they represent are alleviated. But the small fractions of the adduct ions that remain in the SIFT-MS spectra, and especially when they are isobaric with genuine characteristic product ion of breath trace gases, can result in erroneous quantifications; such is the case for H₃O⁺N₂ interfering with breath ethanol analysis and H₃O⁺CO₂ with breath acetaldehyde analysis. However, these difficulties can be overcome when the isobaric adduct ions are properly recognised and excluded from the analyses; then these two important compounds can be properly quantified in breath. The presence of O₂⁺CO₂ in the product ion spectra interferes with the analysis of CS₂ present at low levels in exhaled breath. It is likely that similar problems will occur as other trace compounds are detected in exhaled breath when consideration will have to be given to the possibility of overlapping between their characteristic product ions and ions produced by hitherto unknown reactions. Similar problems are evident in other systems; for example, H₃O⁺CH₄ adduct ions are observed in both SIFT-MS analyses of methane rich mixtures like biologically generated waste gases and in model planetary atmospheres.     

Influence of water vapour on selected ion flow tube mass spectrometric analyses of trace gases in humid air and breath

Selected ion flow tube mass spectrometry (SIFT-MS) detects and quantifies in real time the trace gases, M, in air/breath samples introduced directly into a flow tube. Inevitably, relatively large partial pressures of water vapour are introduced with the sample and the water molecules become involved in the ion chemistry on which this analytical technique depends. When H(3)O(+) ions are used as the precursors for chemical ionisation and SIFT mass spectrometric analyses of M, they generally result in the formation of MH(+) ions. Also, when water vapour is present the H(3)O(+) ions are partially converted to hydrated hydronium ions, H(3)O(+).(H(2)O)(1,2,3). The latter may act as precursor ions and produce new product ions like MH(+).(H(2)O)(1,2,3) via ligand switching and association reactions. This ion chemistry and the product ions that result from it must be accounted for in accurate analyses by SIFT-MS. In this paper we describe the results of a detailed SIFT study of the reactions involved in the quantification of acetone, ethyl acetate, diethyl ether, methanol, ethanol, ammonia and methyl cyanide by SIFT-MS in the presence of water vapour. This study was undertaken to provide the essential data that allows more accurate analyses of moist air and breath by SIFT-MS to be achieved. It is shown using our standard analysis procedure that the error of SIFT-MS quantification caused by the presence of water vapour is typically 15%. An improved analysis procedure is then presented that is shown to reduce this error to typically 2%. Additionally, some fundamental data have been obtained on the association reactions of protonated organic molecules, MH(+) ions, with water molecules forming MH(+).H(2)O monohydrate ions. For some types of M, reaction sequences occur that lead to the formation of dihydrate and trihydrate ions.     

选择离子流动管质谱及其在痕量气体分析中的应用

选择离子流动管质谱(SIFT-MS)结合流动管技术、化学电离和质谱,有选择地使用F13O^ 、NO^ 和O2^ 初始离子,可在几秒之内对空气、呼吸气体和液表蒸气中的痕量气(如乙醇、乙醛、丙酮、氨和2-甲基丁二烯等,行多组分实时在线分析。介绍了选择离子流动管(SIFT)技术、SIFT-MS的分析方法及其物理和离子化学基础、SIFT-MS在不同领域的痕量气体分析中的应用。     

Identification of volatile lung cancer markers by gas chromatography–mass spectrometry: comparison with discrimination by canines

In this work, a chromatographic method for identification of volatile organic compounds was compared with canine recognition. Gas chromatography and mass spectrometry (GC-TOF MS) were used for determination of concentrations of trace gases present in human breath. The technique enables rapid determination of compounds in human breath, at the parts per billion level. Linear correlations were from 0.83-234.05 ppb, the limit of detection was the range 0.31-0.75 ppb, and precision, expressed as relative standard deviation (RSD), was less than 10.00 %. Moreover, trained dogs are able to discriminate breath samples of patients with diagnosed cancer. We found a positive correlation between dog indications and the ethyl acetate and 2-pentanone content of breath (r = 0.85 and r = 0.97, respectively). The methods presented for detection of lung cancer markers in exhaled air could be used as a potential non-invasive tool for screening. In addition, the canine method is relatively simple and inexpensive in comparison with chromatography.     

Kinetics and isotope patterns of ethanol and acetaldehyde emissions from yeast fermentations of glucose and glucose-6,6-d2 using selected ion flow tube mass spectrometry: a case study.

As a prelude to investigations of the emission of metabolites from human cell lines in vitro, we have conducted a study using selected ion flow tube mass spectrometry (SIFT-MS) of the acetaldehyde and ethanol that appear in the headspace above a fermenting yeast/glucose/water mixture in sealed glass bottles at a temperature of 30 degrees C. A fixed quantity of yeast (10 mg) and varying amounts (2, 4, 8 and 16 mg) of both non-deuterated glucose and glucose-6,6-d2 in 5 mL of water were used and the emission of the acetaldehyde and the ethanol were observed as a function of time. The ethanol and acetaldehyde concentrations in the headspace were obtained from the magnitudes of their characteristic ions on the accumulated SIFT mass spectra and, when the deuterated glucose was used, characteristic singly and doubly deuterated ions were obvious. This study indicates, as expected, that ethanol is the major species generated and that acetaldehyde is a relatively minor component of the headspace and a very minor component of the liquid phase. We estimate that about 10(8) ethanol molecules are produced per minute per cell in this yeast fermentation process. The distribution of the non-deuterated and partially deuterated ethanol under these fermentation conditions is observed to be C2H5OH (66 +/- 4)%, C2H4DOH(6 +/- 1)%, C2H3D2OH(28 +/- 4)%, and the analogous distribution for the acetaldehyde is the same, within error. These results indicate that the D atoms in the glucose-6,6-d2 are mostly retained by the 6-C atom, but the appearance of the singly deuterated ethanol and acetaldehyde indicates that some D/H mixing must be occurring in the enzymatic reactions. The results of this study illustrate the potential and power of on-line SIFT-MS analysis in this area of research.     

In situ analysis of solvents on breath and blood: a selected ion flow tube mass spectrometric study

We report measurements of residual vapour levels of xylenes and trimethylbenzenes, present following a floor re-surfacing procedure, using the technique of selected ion flow tube mass spectrometry (SIFT-MS). A subject exposed to controlled amounts of xylene and mesitylene was monitored by direct breath exhalation over a 4-hour period after exposure to the volatile organic compounds (VOCs) had stopped. The headspace gases above 5 mL blood samples taken over this period were also monitored. The decays of the solvent levels with time were fitted to a two-compartment model with residence times for xylene and mesitylene of 0.37 h and 0.38 h, respectively (compartment one) and 2.5 h and 2.8 h, respectively (compartment two).     

Non-invasive spatial visualization system of exhaled ethanol for real-time analysis of ALDH2 related alcohol metabolism

A novel imaging system of ethanol in exhaled breath induced by acetaldehyde dehydrogenase (ALDH2)-related alcohol metabolism has been developed. The system provides an image of ethanol distribution as chemiluminescence (CL) on an enzyme-immobilized support. The spatiotemporal change of CL generated by ethanol in exhaled breath after oral administration of ethanol was detected by employing an electron multiplier CCD (EM-CCD) camera, illustrated and analyzed. Prior to measurement of standard gaseous ethanol and ethanol in exhaled breath, the system was optimized by investigating the enzyme-immobilized supports, concentration of substrate and pH condition of Tris-HCl buffer solution. The ethanol skin patch test, a simple method as an indicator of ALDH2, was performed on healthy volunteers. Breath samples of 5 volunteers with ALDH2 (+) and 5 volunteers with ALDH2 (-) were used for exhaled ethanol analysis. Concentration-time profiles of exhaled ethanol obtained from all volunteers were analyzed over a period of 120 min after oral administration of ethanol (0.4 g per kg body weight) in the form of beer which contains 5% of alcohol. The results obtained from the system showed that the peaks of exhaled ethanol concentrations appeared at 30 min, which was considered as a rapid ethanol absorption phase following first-order kinetics. Exhaled ethanol concentrations of volunteers with ALDH2 (+) were lower than volunteers with ALDH2 (-) and the digestion of ethanol in volunteers with ALDH2 (+) was faster than in volunteers with ALDH2 (-). The eliminations were analyzed to follow zero-order kinetics with a rate constant for each group.     

Experimental approach to assess sorptive loss properties of volatile organic compounds in the sampling bag system

In this study, the sorptive loss patterns for volatile organic compounds were evaluated by gaseous standards containing 13 compounds (benzene, toluene, styrene, p‐xylene, methyl ethyl ketone, methyl isobutyl ketone, isobutyl alcohol, butyl acetate, acetaldehyde, propionaldehyde, butyraldehyde, isovaleraldehyde, and valeraldehyde). The gaseous standards, prepared initially at two contrasting concentration levels (40 and 4000 ppb) in a polyester aluminum bag, were measured after two consecutive transfers into empty bags. It indicates that the percent loss patterns, if assessed for all 13 target compounds, are affected most sensitively by the initial concentration levels of samples to yield 2.62 ± 2.22% (at 40 ppb) and 9.57 ± 6.74% (at 4000 ppb). Moreover, the sorptive loss patterns at high concentration samples (4000 ppb) tend to increase in relation with increasing molecular weight of target compounds, although such pattern disappears in low concentration samples (40 ppb). The observed loss patterns, if evaluated in relation to some key parameters like concentration or compound type, suggest the possibility that the sorptive loss of target compounds in storage media can occur in a predictable manner.     

A simple method for the trace determination of methanol, ethanol, acetone and pentane in human breath and in the ambient air by preconcentration on solid sorbents followed by gas chromatography

A simple technique has been developed for preconcentration of gaseous trace organic compounds on solid sorbents, followed by gas chromatography. The sorbent is packed in a cartridge from a syringe needle placed in the gas chromatographic injector and the analytes previously adsorbed are thermally desorbed at the injector temperature and then directly swept by the carrier gas into the column. The system has been tested for a charcoal-based adsorbent and silica gel, with pentane, methanol, ethanol and acetone as the model analytes. The procedure is rapid, the detection limits vary from a few nmol l(-1) to values below 0.1 nmol l(-1) (i.e., a few ppb), the linear dynamic range amounts to at least five concentration decades and a typical relative standard deviation is 10% at the nmol l(-1) concentrations. It has been shown that the method is readily applicable to determination of instantaneous concentrations of the analytes in natural and industrial atmosphere and to their monitoring in human breath which is important for medical and hygienic practice. In general, the procedure is applicable to low-molecular volatile organic compounds.