Pulsed UV-light penetration of characterization and the inactivation of Escherichia coli K12 in solid model systems

With an ever increasing number of foodborne outbreaks in the U.S. tied to produce, novel processing technologies are needed. One such technology is pulsed ultraviolet (UV) light, which provides broadband spectrum light between 100 and 1100 nm, but more than 50% falls into the UV region. However, the efficacy of such a technology is not fully understood, and there is some concern as to the penetrability of pulsed UV-light through opaque materials. Inactivation and energy penetration data obtained from the treatment of agar (clear solid medium) and whey protein (opaque solid medium) gels after treatment with pulsed UV-light were used to construct several models to estimate the amount of energy penetrating the sample at a given depth and the inactivation of E. coli K12. The amount of broadband energy being transmitted through the materials decreased by as much as 60% at a depth of 10 mm. The inactivation curves obtained indicated that the relationship between energy dose and inactivation was non-linear, and the Weibull model was used to estimate microbial inactivations. Three approaches were undertaken to determine the needed model parameters. The first model (Weibull model 1) used a single set of values for the two Weibull distribution parameters. This model was found to accurately estimate the reductions of E. coli K12 based on R ² and RMSE values. The second approach (Weibull model 2) simultaneously solved for the extinction coefficient value in the exponential dose model in addition to the two Weibull distribution parameters. This model was found to increase the R ² values and decrease the RMSE values. However, this model was found to underestimate the dose. The third approach (Weibull model 3) was to determine the two Weibull distribution parameters at each depth. This approach more accurately estimated reductions and dose, but was determined to be overfitted. Each model had its advantages and disadvantages, but in all three it was observed that broadband energy was not a suitable predictor of microbial inactivation. The results presented in this study indicated that pulsed UV-light can penetrate opaque materials up to 10 mm deep with decreasing energy levels, and that the Weibull model can be used to model the inactivation of E. coli K12 with correlation coefficients of 0.75 and 0.79.