Repository logo

Thermal inactivation of Salmonella and Escherichia coli O157:H7 in heat treated, recycled used cooking oil




Murphy, Rachel Lynn, author
Delmore, Robert J., advisor
Woerner, Dale R., committee member
Martin, Jennifer N., committee member
Hess, Ann, committee member
Belk, Keith E., committee member

Journal Title

Journal ISSN

Volume Title


Rendered oils and fats are commonly used in pet food products. Given the implementation of the Food Safety Modernization Act (FSMA), which enforces a zero tolerance policy for Salmonella in all pet foods, the microbiological safety of these products has become an area of focus for the industry. Objectives of this study were to (i) understand lethality of a selective set of strains of Salmonella and E. coli O157:H7 in heat treated, recycled used cooking oil, (ii) estimate D- and z-values for a selective set of strains of Salmonella and E. coli O157:H7 in heat treated, recycled used cooking oil, and (iii) establish time/temperature processing steps targeting a selective set of strains of Salmonella and E. coli O157:H7 for use as preventive controls in a product safety plan. Objectives were addressed by inoculating previously rendered, recycled used cooking oil, with a five-strain mixture of Salmonella or E. coli O157:H7 at a level of approximately 8.0 log CFU/ml. Inoculated samples were exposed to heat at 62, 71, or 82°C using a water bath fitted with a circulator. At predetermined time intervals, individual samples were removed from water bath and immediately chilled in an ice bath. Samples were diluted in 0.1% buffered peptone water supplemented with 1% emulsifier (Tween 80). Samples were plated on selective agars (xylose lysine deoxycholate [XLD, Salmonella]; sorbitol MacConkey [SMAC, E. coli O157:H7]) and a non-selective agar (tryptic soy agar supplemented with sodium pyruvate at a level of 1%). A traditionally used aqueous inoculum would not be adequately distributed throughout an oil matrix and could potentially interfere with heat transfer during heat treatment. Thus, pathogen cells were resuspended in warmed (37°C), sterile vegetable oil (soybean oil). No difference in inoculation level was observed when an aqueous solution and sterile vegetable oil were compared as inoculum matrices. Volumetric measurement of oil samples proved inconsistent and, thus, the mass was used to measure samples for experimentation. Addition of Tween 80 aided in homogenization of sample and diluent during surface plating and eliminated phase separation and errors in dilution (i.e. no ten-fold differences across dilutions). Additionally, a larger width tube (50 ml Falcon Tube) was used for the first dilution blank and vortexing was standardized to 30 s—these components proved to be critical for microbial analysis of high-fat liquids or semi-solids. Six replicates were performed for each pathogen and temperature combination. Data from surviving populations was utilized to generate thermal death curves. Segmented regression was performed using the Proc NLIN function of SAS (v 9.4) and z-values were calculated using linear regression (SAS v 9.4). Non-linear death curves were observed for all pathogen and temperature combinations. This response to heat treatment indicated not only pathogen survival, but also possible heat tolerance among some strains. As expected, D-values were lower as temperature increased and, for all pathogens, ranged from 0.03 to 0.04 min at 82°C, 0.14 to 0.27 min at 71°C, and 0.77 to 1.49 min at 62°C. Z-values ranged from 14.14°C to 27.78°C for all pathogens. Recommended time/temperature processing steps established from these data act as preventive controls in heat treatment of recycled used cooking oil. Processing times range from 7.46 min to 0.20 min.


Rights Access


Escherichia coli O157:H7
used cooking oil


Associated Publications