Manuel Castillo

Most researchers who have studied enzymatic milk coagulation in depth recognize that this process is composed of two overlapping phases: an enzymatic phase (first) and a non-enzymatic phase (second). The latter is composed of two consecutive sub-phases that are also partially overlapping: the first is called micellar aggregation, and the second is called the gel hardening phase by many researchers. This final phase, which consists of the rearrangement of the three-dimensional para-κ-casein network, is one of the most complex and least studied milk coagulation stages. A review conducted prior to this publication confirmed that existing models describe micellar aggregation up to network formation but cannot characterize the subsequent network rearrangement and gel hardening process. The objective of this review is to examine the main models that describe the gel hardening phase, evaluating each one to highlight its advantages, disadvantages, and main limitations. Among the models reviewed, those based on light scattering stand out, as they are non-invasive and allow for the prediction of gel firmness in real time. This review is expected to provide greater support to the cheese industry in achieving better control of enzymatic milk coagulation.

Whey is widely used across various food industries, and its functional properties are critical for product quality and consumer acceptance. The current study aimed to assess the feasibility of using front-face fluorescence spectroscopy (FFF) as a nondestructive technique to predict the technological functionality of whey, particularly its foaming, gel-forming, and emulsifying capabilities. Heat treatments were applied on reconstituted skim milk enriched with whey protein isolated (80°C with 7 holding times) to induce variability in all the studied functional properties. Then, whey samples were obtained by isoelectric precipitation of caseins from enriched milk. Synchronous FFF of whey was measured at an excitation of 250 to 550 nm (λ₀ = 20 nm and Δλ = 10 nm, where λ₀ = initial delta marks the first emission scan at a specific excitation; because the excitation was set at 250 nm, the first emission was at 270 nm. Δλ = delta increment is the change in wavelength for each subsequent scan). Excitation and emission wavelengths and maximum intensities were collected for 10 areas. These FFF parameters and their mathematical transformations were used in the maximum R² procedure of SAS software (version 9.4, SAS Institute Inc.) to select the predictors and obtain the best models, which were cross-validated. Excellent prediction models (R² > 0.95, ratio of performance deviation [RPD] >4) were obtained for gel-forming and emulsifying properties. However, for foam properties, the obtained models were successful for foam overrun (0.90 ≤ R² ≤ 0.95, 3 ≤ RPD ≤4) and moderately successful (0.80 ≤ R² < 0.90, 2.25 ≤ RPD <3) for foam stability. This research demonstrated that FFF spectroscopy has the potential to become a method for assessing whey functional properties, leading to improved industry process control and new products.

Monitoring pH and acidity during yoghurt fermentation is essential for product quality and process efficiency. Conventional measurement methods, however, are invasive and labour-intensive. This study developed artificial neural network (ANN) models to predict pH and titratable acidity during yoghurt fermentation using CIELAB colour parameters (L, a*, b*). Reconstituted milk powder with 12% total solids was prepared with varying protein levels (4.2–4.8%), inoculum concentrations (1–3%), and fermentation temperatures (36–44 °C). Data were collected every 10 min until pH 4.6 was reached. Forty models were trained for each output variable, using 90% of the data for training and 10% for validation. The first two phases of the fermentation process were clearly distinguishable, lasting between 4.5 and 7 h and exceeding 0.6% lactic acid in all treatments evaluated. The best pH model used two hidden layers with 28 neurons (R² = 0.969; RMSE = 0.007), while the optimal acidity model had four hidden layers with 32 neurons (R² = 0.868; RMSE = 0.002). The strong correlation between colour and physicochemical changes confirms the feasibility of this non-destructive approach. Integrating ANN models and colourimetry offers a practical solution for real-time monitoring, helping improve process control in industrial yoghurt production.

Current systems that allow inline pH control in the fermented dairy industry have drawbacks, such as protein adhesion on the non-glass pH probes, measurement distortion, frequent recalibration needs, and sensitivity to extreme pH conditions encountered during clean-in-place operations. Therefore, the objective of this study was to validate the feasibility of estimating the pH of milk during the yogurt making process by using a NIR light backscatter sensor measuring under different fermentation temperatures and milk protein concentrations using a mathematical model that correlates the light scatter signal with pH. Three replications of the experiment with two protein concentrations (3.5 and 4.0%) and two fermentation temperatures (43 and 46 °C) were used to validate this inline pH prediction model. Continuous and discontinuous measurements of pH were collected as a reference during fermentation, simultaneously with the light backscatter data acquisition. Also, the effect of adjusting the initial voltage gain of the light scatter device on the accuracy of the pH prediction model was evaluated. Temperature and initial voltage were the main factors affecting the fitting accuracy of the model. The adjustment of the initial voltage gain improved the pH prediction model fit. The model has been successfully validated for both continuous and discontinuous measurements of pH, with SEP values < 0.09 pH units and CV < 1.78%. The proposed optical inline and non-destructive method was feasible for inline pH monitoring of milk fermentation, avoiding traditional manual pH measurement.

Milk is subjected to different industrial processes, provoking significant physicochemical modifications that impact milk’s functional properties. As a rapid and in-line method, front-face fluorescence can be used to characterize milk instead of conventional analytical tests. However, when applying fluorescence spectroscopy for any application, it is not always necessary to determine which compound is responsible for each fluorescent response. In complex matrixes such as milk where several variables are interdependent, the unique identification of compounds can be challenging. Thus, few efforts have been made on the chemical characterization of milk’ fluorescent spectrum and the current information is dispersed. This review aims to organize research findings by dividing the milk spectra into areas and concatenating each area with at least one fluorophore. Designations are discussed by providing specific information on the fluorescent properties of each compound. In addition, a summary table of all fluorophores and references cited in this work by area is provided. This review provides a solid foundation for further research and could serve as a central reference.

Heat treatments play a critical role in ensuring the safety and preservation of milk, but it can affect its nutritional and sensory properties. The present paper proposes the use of a portable system based on fluorescence spectroscopy as an alternative method for the quantification of four thermal damage markers at once (hydroxymethylfurfural, sulfhydryl groups, ascorbic acid, and riboflavin). The obtained prediction models using autofluorescent compounds (tryptophan, dityrosine, Maillard compounds, and riboflavin), validated with skimmed milk processed under several industrial conditions, granted the development of a portable and/or online system, allowing for the real-time monitoring of thermal damage and control of the heat treatment process. The results of this study will certainly contribute to the development of new process analytical technologies for the dairy industry, enabling quality control and adjustment of the manufacturing process to ensure safe and high-quality products.

From a state-of-the-art point of view, it is currently possible to optically monitor the enzymatic coagulation of milk for real time estimation of the elastic modulus to cut the gel at optimum gel firmness. However, European cheese industry produces a wide variety of cheeses, many of them artisanal, and has a very fragmented productive structure with many small-, medium-sized companies. Therefore, if the technology is to be successfully uptake, it must be not only accurate but very low-cost. The objective of this work was to evaluate a low-cost commercial multifiber probe, for inline optical determination of curd firmness during cheese making. Preliminary tests were carried out to select the most appropriate fiber core size and wavelength and after that coagulation trials were performed following a fully randomized factorial design with two factors, i.e., concentration of protein (3.2, 3.6 and 4.0%) and added calcium (150, 200 and 250 mg L⁻¹), with three replicates. The observed linear increase of the least square means of the initial voltage with the protein content (V₀ = 0.15[P, %] + 0.88; R² = 0.999), will be likely synergistic with the elastic modulus prediction if the model needs to be corrected for protein. Finally, the multifiber probe allowed predicting curd firmness using the proposed model with SEP values < 7 Pa. The present work has proven that a low-cost multifiber probe is suitable for accurate, real-time prediction of curd firmness during cheese manufacture.

This study aims to provide the dairy industry with a direct control model focused on milk coagulation by using multifiber probes to determine parameters in the curding process, such as cutting time, at a lower cost. The main objective of the research is to confirm that a multifiber NIR light scattering probe can be used to predict the elastic modulus of curd during milk coagulation in cheese production. Two randomized complete block designs were used with a 3 × 3 factorial arrangement of three protein levels (3%, 3.5% and 4%) and three wavelengths (870 nm, 880 nm and 890 nm). Using a multifiber probe at a wavelength of 880 nm allowed obtaining a better optical response of the sensor during enzymatic milk coagulation than the 870 nm. It showed greater sensitivity to variations in the protein content of the milk and lower variation in the response. The multifiber probe at a wavelength of 880 nm generated a NIR light backscatter profile like those obtained with other systems. The results showed that the prediction model parameters had a variation as a function of the protein content, which opens the possibility of improving the prediction model’s performance substantially. Furthermore, the initial voltage obtained with the probe responded linearly to the different protein levels in milk. This fact would make it possible, at least theoretically, to estimate protein concentration with the same inline probe for G’ determination, facilitating the incorporation of a corrective protein factor in the prediction models using a single instrument.

From a state-of-the-art point of view, it is currently possible to optically monitor the enzymatic coagulation of milk for real time estimation of the elastic modulus to cut the gel at optimum gel firmness. However, European cheese industry produces a wide variety of cheeses, many of them artisanal, and has a very fragmented productive structure with many small-, medium-sized companies. Therefore, if the technology is to be successfully uptake, it must be not only accurate but very low-cost. The objective of this work was to evaluate a low-cost commercial multifiber probe, for inline optical determination of curd firmness during cheese making. Preliminary tests were carried out to select the most appropriate fiber core size and wavelength and after that coagulation trials were performed following a fully randomized factorial design with two factors, i.e., concentration of protein (3.2, 3.6 and 4.0%) and added calcium (150, 200 and 250 mg L^-1), with three replicates. The observed linear increase of the least square means of the initial voltage with the protein content (V₀ = 0.15[P, %] + 0.88; R² = 0.999), will be likely synergistic with the elastic modulus prediction, if the model needs to be corrected for protein. Finally, the multifiber probe allowed predicting curd firmness using the proposed model with SEP values < 7 Pa. The present work has proven that a low-cost multifiber probe is suitable for accurate, real-time prediction of curd firmness during cheese manufacture.