UCLA

Acetoin, diacetyl, and 2,3-pentanedione are artificial butter flavoring ingredients. Occupational exposures to diacetyl are associated with severe respiratory disease including bronchiolitis obliterans. Dynamic and passive air sampling and analytical methods were developed for simultaneous sampling of these ketones using 10 % (w/w) O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine hydrochloride (PFBHA) on Tenax TA (80/100 mesh). PFBHA O-oximes of the ketones were synthesized with above 95 % purity for standards. Ketone vapors of known concentrations and relative humidities (RHs) were generated in Tedlar gas bags for dynamic sampling. A syringe pump delivered ketones in aqueous solution to an air dilution system and an exposure chamber for passive sampling. The dynamic sampling tubes contained 200 mg of coated solid sorbent in a Pyrex tube. The passive samplers were obtained by compressing 300 mg of coated solid sorbent into a pellet of 30 mm diameter and 0.53 mm thickness. The ketone vapors permeated a silicone membrane and crossed a path length of 3.2 mm to the pellet surface. After sampling, the coated solid sorbent was suspended in water, ultrasonicated, microwaved, extracted with hexane, and centrifuged. An aliquot of the hexane extract was analyzed by capillary gas chromatography-mass spectrometry using selective ion monitoring with the internal standard method. For dynamic sampling, acetoin, diacetyl, and 2,3-pentanedione vapor sampling efficiencies were 90.2 ± 6.9, 92.2 ± 5.9, and 82.5 ± 4.4 % respectively at 5-20 ppb when sampled at 100 mL/min for 8 hours at 25 °C for both 5 and 80 % RHs. Recoveries between 75-125 % were also obtained at 40 ºC at both RHs. For passive sampling, experimental sampling constants for acetoin, diacetyl, and 2,3-pentanedione at 25 ºC for both 5 and 80 % RHs were 59.4 ± 8.5, 55.3 ± 7.6, and 50.0 ± 6.3 mL/min respectively. The experimental sampling constants at 40 ºC were statistically lower at 5 % RH, but had no statistical difference at 80 % RH compared to at 25 ºC. Overall, quantitative, selective, and sensitive dynamic and passive sampling methods were developed around the 2012 ACGIH TLV (10 ppb) for diacetyl in the presence of similar concentrations of acetoin and 2,3-pentanedione.