Effect of standard phase differences between gas and liquid and the resulting experimental bias in the analysis of gaseous volatile organic compounds

Liquid- or gas-phase standards can be used for the analysis of VOCs in air. Once the accuracy is secured in the standard preparation stage, the use of gas-phase standard should be more reliable with the least matrix effect. However, it is not difficult to find that the liquid-phase standard is used more preferably in many laboratories for several reasons (e.g., low expense, easy handling, etc.). As such, one needs to accurately evaluate any possible bias stemming from the use of different standard phases. To this end, standards for 8 VOCs consisting of 4 aromatic compounds (benzene (B), toluene (T), styrene (S) and p-xylene (p-X)) and 4 others (methyl ethyl ketone (MEK), methyl isobutyl ketone (MIBK), butyl acetate (BuAc), and isobutyl alcohol (i-BuAl)) were prepared in both liquid and gas phases. Each standard was analyzed by the initial collection on the adsorption tube and by the combined application of thermal-desorption–gas chromatography–mass spectrometry (TD/GC/MS). The results indicated that experimental bias between the two phases, if expressed in terms of percent difference (PD), was very low in many target VOCs (B (1.09%), T (2.41%), p-X (3.64%), MEK (6.76%), and MIBK (0.17%)), while it was not in some targets (e.g., >10%: e.g., S, i-BuAl, and BuAc). In an ancillary experiment, biases were evaluated further by (1) calibrating gaseous samples against liquid phase standard and via (2) comparison between two different types of gas phase standards. In conclusion, treatment of different standards (e.g., between the same or different phases) will inevitably induce biases in most VOCs, although certain volatiles (e.g., benzene, MIBK, etc.) are virtually unaffected by such variables in a practical sense.

Comparison of experimental bias (expressed in percent difference (PD)) against liquid phase standard: (1) gas-phase standard and (2) gaseous samples prepared at two concentration levels of low (L) and high (H).Figure optionsDownload as PowerPoint slideHighlights
► Experimental bias often becomes inevitable due to phase differences between standard and sample.
► The use of gas-phase standard is ideal to analyze gas-phase samples with the least matrix effect.
► The experimental bias in VOC analysis was assessed in relation to phase differences in standards.
► The bias can be reduced significantly for many VOCs with the proper control of experimental conditions.