Modeling of radio frequency heating of packed fluid foods moving on a conveyor belt: A case study for tomato puree

In recent years, there is a growing demand by consumers for healthy, high-quality foods which are shelf-stable but also maintain the sensory and nutritional profile of the fresh product (Marszałek, Mitek, & Skąpska, 2015). Conventional in-package heating of fluid foods is done by large baths with hot water or steam in contact with the containers. This process has the advantage of preventing post-process recontamination, but it requires substantial amounts of water that is lost through evaporation, leaks, or impurities. Additionally, the water is heated by boilers that primarily use fossil fuels. Indirect thermal treatments also results in heat losses, because the water needs to be heated above the pasteurization temperature needed for the product (Briggs, Brookes, Stevens, & Boulton, 2004; Ishwarya & Anandharamakrishnan, 2018).

Dielectric heating has the potential to provide higher heating efficiency, while requiring less floor space and water input. During a dielectric heating process, electromagnetic energy is absorbed and converted to heat through bipolar rotation and ionic displacement within the product (Llave & Erdogdu, 2020; Llave, Terada, Fukuoka, & Sakai, 2014).

Dielectric heating is commonly carried out in two different regions of the electromagnetic spectra: microwave (MW) at frequencies ranging from 300 MHz to 300 GHz, and radio frequency (RF) at frequencies lower than 300 MHz. RF technology has the potential to provide better temperature uniformity inside the food material due much longer penetration depths (Llave & Erdogdu, 2020).

MW heating of fluid foods has been extensively studied (Abea et al., 2022). On the contrary, research dealing with RF heating focuses on solid food matrices with a handful of exceptions such as fish soup (Muñoz et al., 2022), liquid egg (Zhu et al., 2021), kiwi puree (Lyu et al., 2018), yogurt (Siefarth, Tran, Mittermaier, Pfeiffer, & Buettner, 2014) and milk (Awuah, Ramaswamy, Economides, & Mallikarjunan, 2005). Understanding heating rates and temperature distribution is necessary for a widespread application of RF technology, however, predictive modeling in liquid foods during dielectric heating is considerably more challenging than in solid matrices due to convection phenomena (Kubo, Curet, Augusto, & Boillereaux, 2019).

Literature shows that temperature distribution during RF heating can be predicted if dielectric properties and equipment dimensions are accurately known. Jiao, Tang, Wang, and Koral (2014) developed a one-dimensional model for the dependency of the heating rate in a parallel-plate free-running oscillator RF system on the dielectric properties of the food load and validated it using a salt solution and peanut butter. Finite-element models have been developed for RF heating of several solid foods such as tuna (Llave, Liu, Fukuoka, & Sakai, 2015), peanut kernels (Zhang, Huang, & Wang, 2017; Zhang, Ramaswamy, & Wang, 2019), pizza (Lan, Qu, Ramaswamy, & Wang, 2020), corn kernels (Wei, Xie, Zheng, & Yang, 2021) and pork (Chen et al., 2023).

Additionally, models for samples moving on a conveyor belt have been studied in wheat using equivalent power absorption (Chen, Huang, Wang, Li, & Wang, 2016) and discrete moving step approaches (Chen, Lau, Chen, Wang, & Subbiah, 2017) and in egg white powder using a moving mesh method (Chen, Lau, Boreddy, & Subbiah, 2019). Existing predictive models, however, have been developed for solid matrices in lab-scale parallel plate systems (through-field configuration) of variable voltage.

Scientific literature regarding other types of electrode configuration such as stray-field and staggered through-field is scarce. In these cases, the electric field is generated by various arrays of rod electrodes (Bernard, Jacomino, & Radoiu, 2014). Bedane, Altin, Erol, Marra, and Erdogdu (2018) studied the electric field distribution in a staggered through-field electrode RF system. According to these authors, it is important to analyze the electric field distribution due to electrode design to determine holding times in industrial processes at different applied voltages and electrode gaps. Particularly, it is necessary to determine the extent to which one-dimensional models can accurately represent heating rates in configurations other than solid matrices between parallel flat plates electrodes.

The development of reliable, non-computationally extensive simulation models can help in the adoption of RF technology for in-package tunnel pasteurization of liquid foods, but the relation between the various model assumptions, the heating behavior and the average temperature must be thoroughly studied. Therefore, the objective of this work was to (I) investigate the electric field distribution on a RF staggered-through field electrode applicator at different nominal voltage and electrode gaps, and then use it to (II) compare the predictive modeling of RF processing of foods moving on a conveyor belt, using a finite-element model and a one-dimensional equation and to (III) study the temperature distribution and average temperatures of bottled fluid foods under RF heating. Mixtures of tomato puree, oil, and salt were used in the validation of both models since juices and purees obtained from the combination of these three ingredients are widely used in a variety of processed foods such as sauces, creams, and soups.

留言 (0)

沒有登入
gif