Optimisation of secondary electrospray ionisation (SESI) for the trace determination of gas-phase volatile organic compounds

Analyst. 2010 Feb;135(2):306-14. doi: 10.1039/b918899a. Epub 2010 Jan 7.


An electrospray ionisation triple quadrupole mass spectrometer (Varian 1200 L) was modified to accept nitrogen samples containing low concentrations of volatile organic compounds. Six candidate probe compounds, methyl decanoate, octan-3-one, 2-ethylhexanoic acid, 1,4-diaminobutane, dimethyl methylphosphonate, and 2,3-butanediol, at concentrations below 50 ppb(v) were generated with permeation tubes in a test atmosphere generator. The concept of using a set of molecular probes to evaluate gas-phase electrospray ionisation of volatile analytes was assessed and the feasibility of adopting a unified ionisation approach for gas and liquid contamination of exobiotic environments established. 450 experiments were run in a five-replicate, fifteen-level, three-factor, central-composite-design with exponential dilution for each of the six probe compounds studied. The three factors studied were ionisation voltage, drying-gas flow and nebulising-gas flow. Parametric modelling by regression analysis enabled the differences in the ionisation behaviours of the probe compounds to be described by the optimisation models. Regression coefficients were in the range 0.91 to 0.99, indicating satisfactory levels of precision in the optimisation models. A wide range in ionisation efficiency was observed, with different optimised conditions required for the probe compounds. It was evident that no one factor appeared to dominate the response and the different factors produced different effects on the responses for the different molecules. 1,4-Butanediamine and dimethyl methylphosphonate required significantly lower ionisation voltages (1.2 kV) than the other four, which achieved optimised sensitivity towards the maximum voltage used in this design (5 to 6 kV). Drying-gas flow rates were found to be more important than nebulising-gas flow rates. However, variations in the constant term B(0) in the optimisation models indicated that other factors, not included in this study, were also likely to be involved in the ionisation process. Electrolyte-flow rate and ionisation temperature were proposed for follow up studies. Exponential dilution data indicated sensitive and analytically useful responses in the target range of 5 to 50 ppb(v) for all six compounds. Significantly, responses were seen at concentrations significantly below 5 ppb(v), with sub ppt(v) responses observed for 1,4-butanediamine, 2-ethylhexanoic acid, dimethylmethylphosphonate, and 1,3-butanediol. Responses in the ppt(v) to ppb(v) range were observed for the remaining two compounds. The observations from this study demonstrated the utility of adopting a set of probe compounds to evaluate electrospray ionisation performance for volatile organic compound based assays; indicated the existence of multiple ionisation mechanisms; and revealed potential sensitivity at the parts per quadrillion level ppq(v).