Model Predictions of Occupational Exposures to Diacetyl and 2,3-Pentanedione Emitted From Roasted Whole Bean and Ground Coffee: Influence of Roast Level and Physical Form on Specific Emission Rates

Abstract

Roasted coffee emits hazardous volatile organic compounds including diacetyl and 2,3-pentanedione. Workers in non-flavored coffee roasting and packaging facilities might inhale diacetyl and 2,3-pentanedione from roasted coffee above occupational exposure limits depending on their work activities and proximity to the source of emissions. Objectives of this laboratory study were to: (1) investigate factors affecting specific emission rates (SERs) of diacetyl and 2,3-pentanedione from freshly roasted coffee, (2) explore the effect of time on SERs of coffee stored in sealed bags for 10-days, and (3) predict exposures to workers in hypothetical workplace scenarios. Two roast levels (light and dark) and three physical forms (whole bean, coarse ground, and fine ground) were investigated. Particle size for whole bean and ground coffee were analyzed using geometric mean of Feret diameter. Emitted chemicals were collected on thermal desorption tubes and quantified using mass spectrometry analysis. SERs developed here coupled with information from previous field surveys provided model input to estimate worker exposures during various activities using a probabilistic, near-field/far-field model. For freshly roasted coffee, mean SER of diacetyl and 2,3-pentantedione increased with decreasing particle size of the physical form (whole bean < coarse ground < fine ground) but was not consistent with roast levels. SERs from freshly roasted coffee increased with roast level for diacetyl but did not change for 2,3-pentanedione. Mean SERs were greatest for diacetyl at 3.60 mg kg−1 h−1 for dark, fine ground and for 2,3-pentanedione at 3.88 mg kg−1 h−1 for light, fine ground. For storage, SERs of whole bean remained constant while SERs of dark roast ground coffee decreased and light roast ground coffee increased. Modeling demonstrated that near-field exposures depend on proximity to the source, duration of exposure, and air velocities in the near-field further supporting previously reported chemical air measurements in coffee roasting and packaging facilities. Control of source emissions using local exhaust ventilation especially around grinding activities as well as modification of work practices could be used to reduce exposures in this workforce.

Determinants of Task-Based Exposures to Alpha-Diketones in Coffee Roasting and Packaging Facilities Using a Bayesian Model Averaging Approach

“…A positive association was expected because both diacetyl and 2,3-pentanedione are natural byproducts from coffee roasting and are emitted during roasting, grinding, and packaging whole bean and ground coffee. However, differences in emissions of diacetyl and 2,3-pentanedione from freshly roasted coffee have been observed previously, with increased emissions rates of diacetyl, but not 2,3-pentanedione, observed with increasing roast level ( 52 ). These differences likely contributed to our observation of roast depth associated determinants (average roast length in minutes) as a notable determinant of increased exposure to diacetyl, but not 2,3-pentanedione, during roasting, packaging, and all tasks.…”

Data_Sheet_1.DOCX

“…A positive association was expected because both diacetyl and 2,3pentanedione are natural byproducts from coffee roasting and are emitted during roasting, grinding, and packaging whole bean and ground coffee. However, differences in emissions of diacetyl and 2,3-pentanedione from freshly roasted coffee have been observed previously, with increased emissions rates of diacetyl, but not 2,3-pentanedione, observed with increasing roast level (52). These differences likely contributed to our observation of roast depth associated determinants (average roast length in minutes) as a notable determinant of increased exposure to diacetyl, but not 2,3-pentanedione, during roasting, packaging, and all tasks.…”