Microbial inactivation and effects of interrelated factors of intense pulsed light (IPL) treatment for Pseudomonas aeruginosa Ji Yoon Yi a , Yu-Kyung Bae a , Chan-Ick Cheigh b , Myong-Soo Chung a, * a Department of Food Science and Engineering, Ewha Womans University, Seoul, 03760, Republic of Korea b Department of Food and Food Service Industry, Kyungpook National University, Sangju, 37224, Republic of Korea article info Article history: Received 9 June 2016 Received in revised form 1 October 2016 Accepted 12 November 2016 Available online 14 November 2016 Keywords: Intense pulsed light (IPL) Microbial inactivation Weibull model Fluence Pseudomonas aeruginosa abstract Intense pulsed light (IPL) inactivation of Pseudomonas aeruginosa for different pulse repetition rates (2 e15 Hz) and widths (0.15e1.5 ms) were described using the double Weibull model and their energy incidents were compared. The values of the regression coefficient (R 2 ), RMSE (root mean sum of squared errors), accuracy factor (A f ), and bias factor (B f ) strongly suggested that the model provided a good fit to the data, and they were coupled with the fluence for the first log reduction (F R ) to compare the energy incidents of different treatments. The incident was higher for a lower pulse repetition rate or a longer pulse width. Moreover, in order to examine the effects of interrelated factors on the IPL fluence in terms of energy efficiency, we proposed several terms: the V F value is defined as the increase in the voltage required for a 1-J/cm 2 increase in the fluence, and the z prr and z pw values are defined as the increases in repetition rate and width of the pulses, respectively, that result in one unit increase in the V F value. By using these terms, the effects of pulse repetition rate and width on the IPL fluence were analyzed and predicted for further investigation. © 2016 Elsevier Ltd. All rights reserved. 1. Introduction Intense pulsed light (IPL) is an effective nonthermal technology for the microbial inactivation of foodborne pathogens (Dunn, Ott, & Clark, 1995; Oms-Oliu, Martín-Belloso, & Soliva-Fortuny, 2010), and its kinetics has been investigated extensively to assess the potential of this method (Bialka, Demirci, & Puri, 2008; Buzrul & Alpas, 2004; Buzrul, Alpas, & Bozoglu, 2005; Chen & Hoover, 2004; Coroller, Legu erinel, Mettler, Savy, & Mafart, 2006; Ross, 1996; Rowan, Valdramidis, & G omez-L opez, 2015). The Weibull model has been suggested to provide a better fit than the traditional log-linear model (Bialka et al., 2008; Buzrul et al., 2005; Chen & Hoover, 2004; 5 msMafart, Couvert, Gaillard, & Legu erinel, 2002), where the factor effects on the inactivation capability could be interpreted using linear regressions (Buzrul et al., 2005; Chen & Hoover, 2004). IPL involves the emission of short-duration, high-energy pulses from a light source such as a xenon lamp, whose broad-spectrum light includes irradiation in the UV-C range of 200e280 nm (Dunn et al., 1995; Oms-Oliu et al., 2010). UV-C plays an important role in the lethality (Wang, MacGregor, Anderson, & Woolsey, 2005), as does the photochemical effect, a formation of dimers that impairs DNA and cell replications (Bolton & Linden, 2003), and the photothermal effect small increases in temperature (Wekhof, 2000). A conventional log-linear model theory of the IPL inactivation curves was based on the assumption that they reflect first-order kinetics (Mafart et al., 2002; Peleg, 1999). However, actual micro- bial populations have not generally shown log-linear relationships, which has prompted several suggestions for modified models (Bialka et al., 2008; Buzrul et al., 2005; Ferrario, Alzamora, & Guerrero, 2013; Rowan et al., 2015). Buzrul et al. (2005) reported that since individual microorganisms rarely have the same sensi- tivity to a lethal source, the inactivation time conforms to a particular distribution, and the Weibull distribution that is based on the engineering principle of failure is widely used (Peleg, 2006). The Weibull model can be used to describe the nonlinear microbial inactivation curve (Bialka et al., 2008; Ferrario et al., 2013; Mafart et al., 2002; Uesugi, Woodling, & Moraru, 2007). This model in- cludes scale and shape parameters, which show different scales and concavity, respectively, according to the combination of the type of microorganism and the lethal source intensity (Buzrul et al., 2005; Rowan et al., 2015). Moreover, Coroller et al. (2006) used two Weibull distributions to produce a double Weibull model that * Corresponding author. E-mail address: mschung@ewha.ac.kr (M.-S. Chung). Contents lists available at ScienceDirect LWT - Food Science and Technology journal homepage: www.elsevier.com/locate/lwt http://dx.doi.org/10.1016/j.lwt.2016.11.030 0023-6438/© 2016 Elsevier Ltd. All rights reserved. LWT - Food Science and Technology 77 (2017) 52e59