Freezing to preserve groundwater samples and improve headspace quantification limits of water-soluble organic contaminants for carbon isotope analysis

Few systematic investigations have addressed the use of freezing for applications in analytical chemistry. Here, we tested its potential to preserve groundwater samples and to improve headspace quantification limits for compound-specific isotope analysis. Analysis of compound concentrations, as well as stable carbon isotope ratios, confirmed that trichloroethene was preserved in frozen suspensions of nanoscale zerovalent iron. In contrast, storage at 7 °C was ineffective, and complete degradation of TCE occurred in 4 weeks. Hence, freezing may stop even abiotic chemical reactions that would not be prevented by cooling or traditional preservation agents. In the absence of iron, we found that headspace concentrations of 14 organic contaminants were considerably higher over frozen solutions than at 25 °C, likely reflecting a freezing-out effect governed by Raoult's law. The observed enhancement depended on the salinity of the samples and was strongest for water-soluble, volatile compounds (values in brackets indicate the minimum observed effect out of six replicates): tert-butyl alcohol (TBA, 35-fold), methyl tert-butyl ether (MTBE, 14-fold), 1,2-dichloroethane (10-fold), or benzene (7-fold). In contrast, little enhancement was observed for less water-soluble compounds, such as tetrachloroethene. Although standard deviations of the measurements were too high for the method to be used for quantitative analysis of total compound concentrations, since we found that freezing introduces no measurable carbon isotope effect for TBA, MTBE, 1,2-dichloroethane, and benzene, the method is an effective way of increasing the sensitivity of compound-specific isotope analysis, particularly of water-soluble organic contaminants.