Sampling VOCs with porous suction samplers in the presence of ethanol: How much are we losing?

Porous suction samplers have been widely used to obtain ground water samples from the vadose zone. However, previous studies identified different mechanisms that may compromise the sample's representativeness, such as volatilization and sorption. This issue is particularly important when dealing with volatile organic compounds (VOCs) as in gasoline spills. Ethanol is common in modern fuels and so may be present in ground water contamination from fuel releases. The objective of this work was to evaluate the losses of VOCs in the presence of ethanol when using porous suction samplers. Laboratory experiments were performed using a ceramic porous suction sampler to sample test solution containing benzene, toluene, xylenes, trimethylbenzenes, naphthalene, and different volumetric fractions of ethanol. Significant losses were found up to 30% for ethylbenzene. Ethanol was found to affect the accuracy of the readings by two main mechanisms: first, negatively, by increasing the headspace in the sampling tube, and second, positively, increasing partition to the aqueous phase due to the cosolvent effect and therefore decreasing the mass loss by volatilization. As a consequence, the highest losses of VOCs were found at intermediate ethanol volume fractions: 10% and 20% (v/v). The losses can be anticipated by measuring the ratio of gas to water in the sampling line and then by applying simple partition models considering cosolvency by ethanol. The importance of adequate purging when using porous suction samplers was also shown.