Practical Considerations in High-Precision Compound-Specific Radiocarbon Analyses: Eliminating the Effects of Solvent and Sample Cross-Contamination on Accuracy and Precision

Preparative capillary gas chromatography (pcGC) is widely used for the isolation of single compounds for radiocarbon determinations. While being effective at isolating compounds, there are still genuine concerns relating to contamination associated with the isolation procedure, such as incomplete removal of solvent used to recover isolated compounds from the traps and cross-contamination, which can lead to erroneous ¹⁴C determinations. Herein we describe new approaches to identifying and removing these two sources of contamination. First, we replaced the common "U" trap design, which requires recovery of compounds using organic solvent, with a novel solventless trapping system (STS), consisting of a simple glass tube containing a glass wool plug, allowing condensation of a target compound in the wool and its solventless recovery by pushing the glass wool directly into a foil capsule for graphitization. With the STS trap, an average of 95.7% of the target compound was recovered, and contamination from column bleed was reduced. In addition, comparison of ¹⁴C determinations of fatty acid methyl ester (FAME) standards determined offline to those isolated by pcGC in STS traps showed excellent reproducibility and accuracy compared to those isolated using the commercial "U" traps. Second, "coldspots" were identified in the instrument, i.e., the termini of capillaries in the preparative unit, which can be cleaned of compounds condensed from earlier runs using a heat gun. Our new procedure, incorporating these two modifications, was tested on archeological fat hoards, producing ¹⁴C dates on isolated C_16:0 and C_18:0 fatty acids statistically consistent with the bulk dates of the archeological material.