Identification and measurement of diacetyl substitutes in dry bakery mix production

J Occup Environ Hyg. 2011 Feb;8(2):93-103. doi: 10.1080/15459624.2011.547148.


In 2008, a company using multiple buttermilk flavorings in the production of dry bakery mixes replaced one liquid flavoring containing 15-20% diacetyl with a proprietary substitute meant to lower occupational risk for diacetyl-related bronchiolitis obliterans. Subsequently, the National Institute for Occupational Safety and Health (NIOSH) evaluated buttermilk flavoring-related exposures at this company's facility, with a focus on measuring ketones by several methods. Volatile organic compounds (VOCs) were evaluated in the headspaces of six bulk flavorings samples, including the substitute buttermilk flavoring. Ketones were evaluated in workplace air via area and personal samples collected during batch preparation of the substitute buttermilk flavoring and production of a bakery mix containing the same flavoring. Air samples were evaluated using five different methods: NIOSH 2549, Modified OSHA PV2118, OSHA 1013, NIOSH Draft Procedure SMP2, and evacuated canisters. Of five buttermilk flavorings from five different flavorings manufacturers, diacetyl was present in four, including the substitute flavoring; acetoin in two; 2,3-pentanedione in four; 2,3-hexanedione in one; and 2,3-heptanedione in three. Among material safety data sheets (MSDS) for four flavorings, only one listed a hazardous ingredient, which was acetoin. The predominant flavoring ingredient identified in the headspace of the substitute flavoring was 2,3-pentanedione; all other chemicals noted above were also present. Diacetyl and 2,3-pentanedione were measured in workplace air via evacuated canisters. In one area and one personal air sample, 2,3-pentanedione was measured by OSHA Method 1013 at concentrations of 78 and 91 ppb, respectively. Without their or the employer's knowledge, workers who used buttermilk flavorings were exposed to substitute ketones from many flavorings manufacturers. Because 2,3-pentanedione, 2,3-hexanedione, and 2,3-heptanedione all share the same functional α-diketone group as diacetyl, these compounds also may share diacetyl's mechanism of toxicity. Until more is known about 2,3-pentanedione and other α-diketone compounds, they should not be assumed to be safe. Companies using artificial buttermilk flavorings should use a precautionary approach that assumes these flavorings pose a health risk and limit exposures through engineering and administrative controls and use of personal protective equipment.

MeSH terms

  • Air Pollutants, Occupational / analysis*
  • Diacetyl / analogs & derivatives*
  • Diacetyl / analysis*
  • Flavoring Agents / analysis
  • Food Handling*
  • Humans
  • Ketones / analysis
  • National Institute for Occupational Safety and Health, U.S.
  • Occupational Exposure / analysis*
  • Occupational Health
  • United States
  • Volatile Organic Compounds / analysis*


  • Air Pollutants, Occupational
  • Flavoring Agents
  • Ketones
  • Volatile Organic Compounds
  • Diacetyl