Abstract
Chromosorb 102 (60/80 mesh) was as good a sampling sorbent for Aroclor 1254 vapor as deactivated Florisil (30/60 mesh), but recoveries were less precise when vapor collection occurred at 90% relative humidity. Tenax-GC (35/60 mesh) and XAD-2 (20/60 mesh) are probably also acceptable based on spiking experiments. The phenomenon of vapor phase enrichment of the most volatile PCBs in Aroclor 1254 was noted and was allowed for by'the Webb-McCall method of quantitation. An inexpensive generator of PCB vapor to produce a 300 µg/m3 volatilization rate over a two-week period also was developed.