Model based optimization of transflection near infrared spectroscopy as a process analytical tool in a continuous flash pasteurizer

Abstract

Near infrared spectroscopy in combination with a transflection probe was investigated as inline measurement in a continuous flash pasteurizer system with a sugar–water model solution. Robustness and reproducibility of fluctuations of recorded spectra as well as trueness of the chemometric analysis were compared under different process parameter settings. Variable parameters were the flow rate (from laminar flow at 30 L/h to turbulent flow at 90 L/h), temperature (20 to 100 °C) and the path length of the transflection probe (2 and 4 mm) while the pressure was kept constant at 2.5 bar. Temperature and path length were identified as the most affecting parameters, in case of homogenous test medium. In case of particle containing systems, the flow rate could have an impact as well. However, the application of a PLS model, which includes a broad temperature range, and the correction of prediction results by applying a polynomial regression function for prediction errors, was able to compensate these effects. Also, a path length of 2 mm leads to a higher accuracy. The applied strategy shows that by the identification of relevant process parameters and settings as well as the establishment of a compensation strategy, near infrared spectroscopy is a powerful process analytical tool for continuous flash pasteurization systems.

Near‐infrared spectroscopy for the inline classification and characterization of fruit juices for a product‐customized flash pasteurization

“…Processing of fruit juice samples and near‐infrared inline measurements were carried out in a laboratory system for HTST treatment of liquid foodstuff type HT220 (OMVE). The experimental setup has been described in detail previously (Weishaupt et al., 2020 ) so that only a brief description of the general setup is given here. The inline NIR measurements were conducted under constant process conditions with flow rate of 90 L/h, temperature of 20°C, and pressure of 3 bar.…”

Heat impact control in flash pasteurization by estimation of applied pasteurization units using near infrared spectroscopy

“…In particular, near infrared (NIR) spectroscopy 11 data collected from consumables are used to accomplish several tasks, such as modeling the correlation between the spectra and composition of a beverage (e.g., the sugar amount) via regression 12–14 and microbial behavior analysis. 15,16 In a recent study, the robustness and reproducibility of inline NIR spectroscopy measurements in flash pasteurizer systems were shown 17 ; the divergences in spectra caused by changes in flow rate and temperature were successfully extracted, modeled, and therefore eliminated.…”

“…Based on observations during the pasteurization process using a container, regression models were trained to accurately represent the relationship between the NIR spectra and the PU values, which was calculated using the formulas. Based on the proven robustness of NIR spectroscopy data 17 it is argued that given the spectra obtained by an integrated NIR spectroscopy sensor, these models can be used to accurately estimate the actual PU values in flash pasteurizer systems. Several regression models and their success in predicting the actual PU values are discussed, as well as their superiority compared to the classical formula-based calculation of PU currently being applied in the industry.…”

“…An online, real-time method to determine PU values during pasteurization is yet to be implemented. In this regard, a recent study 17 showed that near infrared (NIR) spectroscopy can be used as an inline process analytical method for flash pasteurizers. This enables the idea of optimizing the control process with real-time prediction of actual PU using ML models fed with NIR spectroscopy data.…”