History of SIFT-MS begins with the conception and development of the selected ion flow tube, SIFT, technique, for the study of ion-molecule reactions in the gas phase at thermal energies [1]. SIFT has provided rate coefficients and product ion distributions for thousands of ion molecule reactions [2] including those leading to the observed interstellar molecules. Thus, the idea of absolute measurement is very much embedded in the flow tube approach to the study of ionic reactions. In the SIFT technique, rate coefficients are determined by observing the decay rate of a swarm of mass selected reactant ions, which are being convected along a flow tube by fast flowing helium carrier gas, with reactant molecules that are introduced at a controlled flow rate into the carrier gas. A downstream analytical mass spectrometer/ ion counting system detect both reactant and product ions. SIFT-MS essentially inverts this procedure exploiting the knowledge of rate coefficients of the reactions of the reagent (precursor) ions with the molecules of analytes to determine their concentrations. Reaction times in SIFT-MS are about a millisecond and this allows the real time analysis for example of single breath exhalations.

Chemical ionization in SIFT-MS

SIFT-MS is based on chemical ionisation (CI) using selected reagent ions, H₃O⁺, NO⁺ and O₂⁺, coupled with fast flow tube technology and quantitative mass spectrometry. The reagent ions are chosen so that they do not react rapidly with the major components of air: N₂, O₂ and Ar, and also CO₂ and water vapour, but do react rapidly with trace gases and vapours that are of interest in medical, biological and environmental research. A crucial difference between SIFT-MS and other CI techniques such as proton transfer reaction mass spectrometry, PTR-MS, that exploit only one reagent ion specie, typically H₃O⁺ ions, is that all three reagent ion species can be used to analyse a given sample. Use of several reagent ions allows positive identification of a compound. For example, reaction of acetone with H₃O⁺ proceeds via proton transfer producing CH₃COCH₃.H⁺ at an m/z value of 59, with NO⁺, the resulting product ion is NO⁺.CH₃COCH₃ at m/z 88 and with O₂⁺, charge transfer with partial dissociation results in CH₃COCH₃⁺ (m/z 43) and CH₃CO⁺ (m/z 58) and 43 respectively. Isobaric compounds can thus be distinguished by their different reactivity. For example, H₃O⁺ reactions with isobaric compounds acetone and propanal both result in product ions at m/z 59 but NO⁺ reacts very differently with propanal via hydride ion extraction producing (M-H)⁺ (m/z 57)  O₂⁺ is also a valuable reagent ion that charge transfers with some compounds with which neither H₃O⁺ nor NO⁺ react, a good examples being nitric oxide, NO, and nitrogen dioxide, NO₂. O₂⁺ also reacts with ammonia to produce NH₃⁺, which gives a very valuable check on ammonia quantification obtained using H₃O⁺ reagent ions. The ion chemistry together with its applications in various research fields has been reviewed in [3].

Absolute quantification

Absolute concentrations of trace gases and vapours in air, including volatile organic compounds and water vapour, can be calculated in real time using SIFT-MS, by considering the flow tube geometry, ionic reaction time, measured flow rates and pressure and ion-molecule reaction rate coefficients [4]. A combination of SIFT-MS with GC-MS is also showing some promise for absolute quantification of vapours after their separation in the GC column [5].

Case studies

SIFT-MS has wide-ranging applications in many areas of research including environmental science, animal an human physiology, medicine and cell biology. Outlines of the results obtained from several studies are outlined below to illustrate research carried out using SIFT-MS.

Breath analysis

Distribution in the healthy population.

In order to recognise abnormal levels of breath metabolites that may be indicative of disease [6], it is important to know their levels in the breath of the healthy population. The initial SIFT-MS 30-day study provided distributions of concentrations of the common breath metabolites ammonia, acetone, isoprene, ethanol and acetaldehyde in 5 healthy volunteers [7]. More extensive longitudinal study has been carried out of the breath of 30 healthy volunteers over a 6-month period [8^, 9^, 10] confirmed the general results of the initial study, but the larger amount of data allowed a better description of the distributions as log normal with a geometric standard deviation of typically 1.6. The median concentrations were found as ammonia 833 parts-per-billion, ppb, acetone 477 ppb, methanol 461 ppb, ethanol 112 ppb and isoprene 106 ppb [11]. Recently breath of more diverse volunteer cohorts in the age range from 4 to 83 years was analyzed [12^, 13] with a special focus on determining the breath levels of HCN in population. Of great importance is also understanding of the origin of breath compounds, some are truly systemic released in the alveolae, and some are generated predominantly in the oral cavity [14].

Kinetics after ingestion of nutrients

A valuable feature of SIFT-MS is that analyses can be obtained in real time on time scales of seconds. Thus, it is possible to follow rapid changes in the level of volatile breath compounds and single breath exhalations can be analysed for several compounds simultaneously allowing relationships between compounds to be explored. To illustrate this, a study of the kinetics of ethanol metabolism and the production of acetaldehyde has been carried out following the ingestion of a small amount of ethanol (typically 5 ml in 500 ml of tap water) [15]. Similarly, time evolution of concentrations of breath acetone and ammonia after ingestion of carbohydrate and protein meals was studied [3]

Monitoring of breath during haemodialysis for patients with end-stage renal failure.

An early SIFT-MS study was to carried out to measure the levels of common breath metabolites in the breath of several patients before and during haemodialysis sessions [16]. Levels of breath ammonia prior to dialysis were found to often exceed10 parts-per-million, ppm, as compared to a typical value for the healthy population. Acetone levels in the breath of the diabetics were about ten times greater than those of the healthy population. The levels of both these metabolites decreased toward normal levels during dialysis, but in some cases the breath isoprene levels actually increased.

Hydrogen cyanide in breath of children with cystic fibrosis.

One of the serious complications in cystic fibrosis, CF, is infection by Psedomonas aeruginosa. Volatile emissions from bacterial cultures grown from cough swabs from children with CF have been investigated [17] with the startling result that HCN is seen to be emitted specifically from Pseudomonas aeruginosa, PA.  Very recently SIFT-MS demonstrated that HCN is indeed elevated in breath of children with CF and a use of this breath test could greatly assist treatment.

Volatile compounds in headspace of urine and cell cultures

Volatile markers of infection and tumours in urinary headspace.

Trace gas analysis has a potential to assist in the early detection of tumours. Formaldehyde was found to be significantly increased in urine from patients with prostate and bladder cancer in comparison with the headspace of urine from healthy controls [18]. It was also observed that nitric oxide gas were present above the acidified urine from some patients with urinary bacterial infection [19]. Recently SIFT-MS methodology and ion chemistry was advanced to allow quantification of trace levels of the potential cancer biomarkers formaldehyde, acetaldehyde and propanol at much lower levels [20].

Acetaldehyde released by cancer cells in-vitro.

As an extension to the search for volatile biomarkers of tumours, SIFT-MS studies have been carried out on the emissions from lung cancer cell lines [21]. These were grown in calf serum and the headspace concentrations of several volatiles were measured for varying numbers of cells in the medium. A wide range of organic species were seen to be present in the headspace, including methanol, ethanol and acetone, but the important observation was that acetaldehyde was present at a concentration in close proportion to the number of cells in the medium. The number of acetaldehyde molecules generated was of order of 10⁶ per cell per minute. The next phase of this work (and that reported in 3.7) is to search for these aldehydes and other compounds in the exhaled breath of patients suffering from lung cancer using on line SIFT-MS breath analysis.

Ovulation.

Volatile compounds in the headspace of urine from a normally ovulating volunteer were studied over three complete menstrual periods and concurrent with the time of ovulation a 3-12 fold increase in the urine headspace acetone was observed [22]. This phenomenon has subsequently been observed in seven normally ovulating volunteers; significantly, it is not seen in post-menopausal volunteers [23].

Aldehydes and other compounds in exhaust gases.

SIFT-MS was used to research concentrations of compounds in petrol and diesel engine exhaust gases relevant to health and safety concerns [24^, 25]. SIFT-MS was connected on line to analyse exhaust a large diesel engine using three types of fuel, viz. ultra-low sulphur diesel, rapeseed methyl ester and gas oil. Compounds detected and quantified included aldehydes that are know respiratory irritants causing asthma and also aliphatic and aromatic hydrocarbons, alcohols and others. The combined use of H₃O⁺ and NO⁺ was essential to distinguish between isobaric compounds, especially to positively identify of various aldehydes including acrolein. The O₂⁺ reagent ions were used to quantifying NO and NO₂ present.

Volatile compounds in rumen gas.

This research was carried out in collaboration with a group concerned with the welfare of diary cows [26]. Samples were taken into bags from the headspace of the rumen liquor of three lactating cows prepared with rumen vistulae. H₂S, (CH₃)₂S, CH₃SH, ammonia and volatile fatty acids were the dominant compounds analysed using SIFT-MS as a function of the period after feeding. Relatively large amounts of sulphur containing compounds released by cattle may be a significant contribution to atmospheric burden when considering the chemistry of global climate change.

Detection of explosives and products of explosions

Resent results demonstrating use of SIFT-MS to analyse industrial explosives and the byproducts of their explosion are detailed in a contribution by Kseniya Dryahina at this meeting.

Thermal decomposition and combustion products of disposable polyethylene terephthalate (PET)

Samples of PET material were combusted by free burning and also in a standardized furnace at controlled temperatures of 500C, 800C. The gaseous products were then analysed using three different analytical methods: high resolution Fourier transform infrared spectroscopy (FTIR), selected ion flow tube mass spectrometry (SIFT-MS) and gas chromatography mass spectrometry (GC-MS). [27]

Flowing afterglow mass spectrometry FA-MS for quantification of deuterium in water vapour and the measurement of total body water, TBW.

An important development that stemmed from SIFT-MS is flowing afterglow mass spectrometry, FA-MS [28], which is exploited for the on-line, real time analysis of the deuterium content of breath water vapour and the headspace of aqueous liquids. This technique relies on the accurate measurement of the of the water cluster ions signal ratio H₃O⁺(H₂O)₂HDO/H₃O⁺(H₂O)₂H₂¹⁸O as generated in an afterglow plasma from H₃O⁺ precursor ions and the mixture of H₂O and HDO molecules present in a breath/headspace introduced into the plasma. Thus, following a known dose of D₂O, the deuterium disperses as HDO throughout the TBW and, at equilibrium, a measurement of the deuterium content of single breath exhalations provides a measure of the TBW, a parameter so important in body composition and renal studies [29^, 30].

Concluding remarks

SIFT-MS has already been used with success in various interdisciplinary fields. However, it is also been continuously improved both from the point of view of sensitivity and performance, where the fundamental understanding of the processes in the plasma of the microwave discharge ion source is vital [31] and from the point of view of understanding the fine details of ion chemistry relevant to analysis at sub ppb levels [32].

References

[1]     N.G. Adams, D. Smith, Int J Mass Spectrom Ion Physics 21 (1976) 349.

[2]     V.G. Anicich, An index of the literature for bimolecular gas phase cation-molecule reaction kinetics, JPL Publication 03–19. NASA, Pasadena, 2003.

[3]     D. Smith, P. Španel, Mass Spectrom Reviews 24 (2005) 661.

[4]     P. Spanel, K. Dryahina, D. Smith, Int. J. Mass Spectrom 249/250 (2006) 230.

[5]     J. Kubista, P. Spanel, K. Dryahina, C. Workmanm, D. Smith, Rapid Commun Mass Spectrom 20 (2006) 563.

[6]     A. Amman, D. Smith, Breath Analysis for Clinical Diagnosis and Therapeutic Monitoring, World Scientific, Singapore, 2005.

[7]     A.M. Diskin, P. Španel, D. Smith, Physiol Meas 24 (2003) 107.

[8]     C. Turner, P. Španel, D. Smith, Physiol Meas 27 (2006) 637.

[9]     C. Turner, P. Španel P, D. Smith , Physiol Meas 27 (2006) 321.

[10]   C. Turner, P. Španel P, D. Smith , Rapid Commun Mass Spectrom, 20 (2006) 61.

[11]   C. Turner, P. Španel P, D. Smith , Physiol. Meas, 27 (2006) 13.

[12]   P. Španel, K. Dryahina, D. Smith, J. Breath Res. 1 (2007) 011001.

[13]   P. Španel, K. Dryahina, D. Smith, J. Breath Res. 1 (2007) 026001.

[14]   A Pysanenko, P Španel, D Smith J. Breath Res. 2 (2008) 046004 (13pp)

[15]   D. Smith, T.S. Wang and P. Španel, Physiol. Meas 23 (2002) 477-489.

[16]   S. Davies, P. Španel, D. Smith, Kidney Int, 52 (1997) 223.

[17]   W. Carroll, W. Lenney, T.S. Wang, P. Španel, A. Alcock, D. Smith, Pediatric Pulmonology, 39 (2005) 452.

[18]   P. Španel, D. Smith, T.A. Holland, W. Al Singary, J.B. Elder, Rapid Commun Mass Spectrom, 13 (1999) 1354.

[19]   D. Smith, P. Španel, T.A. Holland, W. Al Singari and J.B. Elder, Rapid Commun Mass Spectrom, 13 (1999) 724.

[20]   P. Spanel, D. Smith, J. Breath Res. 2  (2008) 046003

[21]   D. Smith, T.S. Wang, J. Sule-Suso, P. Španel, A. El Haj Rapid Commun Mass Spectrom, 17(2003) 845.

[22]   A.M. Diskin, P. Španel, D. Smith Physiol Meas 24 (2003) 191.

[23]   D. Smith, K.M.K. Ismail, A.M. Diskin, G. Chapman, J.L. Magnay, P.Španel, S. O’brien, Acta Obstetricia et Gynecologica 85 (2006) 1008.

[24]   D. Smith, P. Cheng, P. Španel, Rapid Comm Mass Spectrom 16, (2002), 1124.

[25]   D. Smith, P. Španel, D. Dabill, J. Cocker, B. Rajan, Rapid Comm Mass Spectrom 18 (2004) 2830.

[26]   R.J. Dewhurst, R.T. Evans, P. Španel, D. Smith, Dairy Sci 84 (2001) 1438.

[27]   K. Sovova et al. Mol. Phys. 106 (2008) 1205.

[28]   D. Smith and P. Španel, Rapid Commun Mass Spectrom15 (2001) 25.

[29]   R.B. Asghar, A.M. Diskin, P. Španel, D. Smith, S. Davies Kidney Int 64 (2003) 1911.

[30]   P. Španel, T.S. Wang, D. Smith, Physiol Meas 26 (2005) 447.

[31]   P. Španel, K. Dryahina, D. Smith Int. J. Mass Spectrom. 267 (2007) 117-124.

[32]   P. Španel,  D. Smith Influence of weakly bound adduct ions on breath trace gas analysis by selected ion flow tube mass spectrometry (SIFT-MS) Int. J. Mass Spectrom. (2009) in press, corrected proof.