Bias arising from 222Rn contamination in standardized methods for biobased content determination and a simple removal method

Abstract In recent years standardized methods for determining biobased content using 14C liquid scintillation spectrometry (LSC) have been developed. However, 222Rn contamination can artificially increase the apparent 14C content and hence the biobased content. In EN 16640:2017 and ASTM D6866–11 methods, which describe the calculation of biobased content using 14C LSC measurements, it is stated that 222Rn can be removed by pumping benzene at dry ice temperatures. In this work we examine this claim by investigating the removal efficiency of 222Rn using the proposed process. First, 222Rn-rich air was injected into benzene, which was then pumped while frozen at dry ice temperatures. Finally, the remaining 222Rn contamination was measured using an ultra-low level liquid scintillation system with a 214Bi/214Po pair counting technique. The results indicated that after a single freeze-pump cycle at dry ice temperatures 12% of 222Rn contamination remained. Taking 222Rn contamination from our earlier works, we estimated that the plausible average error was 0.18% for a typical sample and could reach up to 2% of the biogenic content for the most highly-contaminated natural samples measured in our laboratory. These experiments were repeated several times at different temperatures but the results were not significantly affected by the temperature. Finally, we propose a 222Rn removal method that reduces the contamination by a factor of 103 that can be performed on vacuum lines for benzene production that are used in most liquid scintillation radiocarbon laboratories.