HighlightsSignificant bioaerosol concentrations were detected in beef slaughter facilitiesAirflow was modeled in beef facilities and potential pathogens were trackedResults indicate transport of bioaerosols toward chiller with final food productNew ventilation designs indicate that plant sanitation can be improvedAdditional studies are required to verify effectiveness of new ventilation Abstract. and Shiga toxin producing (STEC) have long been recognized as pathogens of concern in meat products due to the prevalence of these microorganisms in the gastrointestinal tract and hide of livestock. Bacterial ingestion due to contaminated food products causes a great economic burden from the hospitalization and death of those who become infected. Recently, aerosolized bacteria have been recognized as a threat to human health and shelf life of food. In beef processing facilities, the majority of harmful bacteria are introduced by the cattle. Heating, ventilation, and air conditioning (HVAC) systems can harbor and transport these microscopic organisms. Salmonella and STEC cause 78 billion dollars lost every year due to contaminated food. During the harvesting process, these pathogens may become aerosolized from the carcasses by various mechanisms, including worker activity and airflow from HVAC systems. Although bacteria are robust creatures, environmental conditions including ventilation can be manipulated to disrupt their proliferation. In this study, one rural and one small beef facility were examined. High air volume wetted wall cyclone bioaerosol samplers capable of collecting and concentrating bioaerosols in a liquid effluent were used during the entire processing at bleeding, de-limbing, de-hiding, washing, and chiller locations. Bioaerosols were analyzed using microbial plating, quantitative Polymerase Chain Reaction, and microbiome analysis. Total bacteria counts, STEC, and concentrations were enumerated in the air and critical areas were identified. and STEC were found to increase with each passing day in the facility, as well, total counts and STEC increased between morning and afternoon phases of processing. Significant differences in total counts and temperature were found at different locations in the facilities. Blueprints were obtained from the examined facilities and the cattle processing floors were modeled using computational fluid dynamics. The airflow created from the HVAC systems was found to have a significant effect on the spread of bioaerosols. Similarities were found between the collected concentrations of bioaerosols and particle traces in the modeled facilities. Finally, new ventilation models were generated to significantly increase the sanitation of the beef slaughtering process. Keywords: Airflow pattern, Beef processing facilities, Bioaerosols, Computational fluid dynamics modeling (CFD), Displacement ventilation, Wetted Wall Cyclone (WWC).

A combination of type A (high flow model) or B (low flow model) shrouded probe and appropriate isokinetic air-sampler (IAS) was tested in a wind tunnel that was optimized for high air speed testing using computational flow modeling. Liquid uranine aerosols (LUA) with AED (aerodynamic equivalent diameter) of 10 lm were generated at a constant flow rate using a vibrating orifice aerosol generator. The monodispersed aerosols were introduced into a wind tunnel at speeds of 5, 10, 15 and 20 m/s. The high flow (A) or low flow (B) model shrouded probe and the appropriate isokinetic air-sampler (IAS) was co-located to collect the LUA simultaneously during each treatment. After the test, the LUA deposited on the filters and inside the walls of the two air-samplers were collected and analyzed for fluorescence intensity units to determine the wall loss, transmission and aspiration ratios. While the type B shrouded probe had 20% (at 10 m/s) and 14.3% (at 15 m/s) higher wall loss ratios than model A, it had 16.1% (at 10 m/s) and 11.6% (at 15 m/s) higher transmission ratios compared to model A. Similarly, probe B had 17.6% (at 10 m/s) and 14.6% (at 15 m/s) higher aspiration ratios than probe A at similar air velocities. Overall, the wall loss, transmission and aspiration ratios of 10 mm AED ULA measured with two types of shrouded probes at 5, 10, 15 and 20 m/s air velocities in the optimized wind tunnel had good agreement with the range of standard data.