RESUMEN
Consumer growing demands for high-quality and safe food and beverages have stimulated the interest in alternative preservation technologies. Short-wavelength ultraviolet light (UV-C, 254 nm) has proven to be useful for the decontamination of a great variety of clear juices while improving their quality compared to traditional thermal treatments. Suspended solids and coloured compounds in turbid juices, diminish light transmission. The use of UV-C under a hurdle approach, may be a promising strategy for their treatment. The purpose of this study was to analyse Escherichia coli ATCC 25922, Saccharomyces cerevisiae KE 162 and Lactobacillus plantarum ATCC 8014 inactivation in clear pear juice (PJ), turbid orange-tangerine (OT) and orange-banana-mango-kiwi-strawberry (OBMKS) juices processed by single UV-C (390 mJ/cm2, 20 °C) and UV-C assisted by mild heat (UV-C/H, 50 °C) at pilot-scale in a coiled tubing unit and stored under refrigeration (5 °C). Inactivation studies were also conducted in peptone water (PW) and model solution (MS). The adequacy of the Coroller, Weibull and Biphasic Plus Shoulder models was studied. UV-C was highly effective in PW, MS and PJ, achieving up to 5.5-6.3-4.7, 4.8-5.1-4.6 and 4.4-5.5 log reductions for L. plantarum, E. coli,and S. cerevisiae, respectively. Whereas, a moderate inactivation by single UV-C was recorded in the turbid blends, reducing up to 2.4-3.8-1.6 and 3.6-3.7-1.3 log-cycles in OT and OBMKS, respectively. When the UV-C/H treatment was applied, high bacterial inactivation was observed achieving 5.2-5.6, 6.3-6.6 and 5.5-6.7 log reductions in OT, OBMKS and PJ, respectively, while 4.6-4.9 log reductions were determined for the yeast in OBMKS and OT, respectively. Thus, additive inactivation effects between UV-C and H were observed. All the models tested gave useful information regarding the existence of microbial subpopulations with varying resistances. However, the cumulative Weibull distribution function was the most versatile one, fitting inactivation curves with different shapes. Additionally, the frequency distributions of resistances showed that UV-C/H not only increased the UV-C microbicidal effect but changed the distribution of inactivation times. Principal component analysis revealed that UV-C effectiveness was associated to low particle size, aâ°, turbidity and high UV-C transmittance. An increase on the inactivation of treated bacterial populations was recorded along storage, while no yeast recovery was observed, thus emphasizing the contribution of refrigerated storage to microbial inactivation. Microbial inactivation in clear and turbid juices achieved by UV-C (390 mJ/cm2) assisted by mild heat (50 °C) and subsequent refrigerated storage may represent an useful alternative for multiple applications in the juice industry.