LWT - Food Science and Technology 154 (2022) 112634 Contents lists available at ScienceDirect LWT journal homepage: www.elsevier.com/locate/lwt Effectiveness of enzymatic treatment for reducing dairy fouling at pilot-plant scale under real cleaning conditions Alfons Eduard Guerrero-Navarro a, Abel Guillermo Ríos-Castillo a, Carolina Ripolles-Avila a, Anna Zamora b, Anne-Sophie Hascoët a, Xavier Felipe c, Manuel Castillo b, José Juan Rodríguez- Jerez a, * a Human Nutrition and Food Science Area, Department of Food and Animal Science, Universitat Autònoma de Barcelona, Travessera dels Turons s/n, Cerdanyola del Vallès, 08193, Spain b Centre d’Innovació, Recerca i Transferència en Tecnologia dels Aliments (CIRTTA), Facultat de Veterinària, Universitat Autònoma de Barcelona, Travessera dels Turons s/n, Cerdanyola del Vallès, 08193, Spain c New Technologies and Food Process, IRTA-Monells, Finca Camps i Armet, Monells, 17121, Spain A R T I C L E I N F O A B S T R A C T Keywords: Conventional cleaning with chemical products uses a substantial amount of water and energy to clean the fouling Cleaning related to dairy production, resulting in significant economic costs. We evaluated an enzymatic treatment based Fouling on the use of protease and amylase to clean dairy fouling generated in an indirect plate heat exchanger and the Enzymatic spray dryer equipment of a pilot plant, representing real cleaning conditions in the dairy industry. The efficacy of Dairy industry Clean-in-place the enzymatic treatment in removing fouling at 50 ◦ C was comparable to that of the clean-in-place method, with alkaline–acid cleaning performed at a maximum temperature of 80 ◦ C. Microbiological analysis showed that the cleaning treatments guaranteed adequate hygienic conditions of the dairy products manufactured. Monitoring fluorescence markers, such as tryptophan, riboflavin, Maillard compounds, and dityrosine could help improve the effectiveness of both alkaline and enzymatic cleaning. The enzymatic treatment fulfills dairy industry ob­ jectives, saving water and energy during washing by reducing chemical product use. Considering that enzymatic cleaning is biodegradable after use and that its economic cost is competitive compared to chemical cleaning, it represents a viable alternative to the chemical cleaning of dairy fouling. 1. Introduction increases their internal pressure and causes obstructions in the circula­ tion of milk (Sadeghinezhad et al., 2015). This issue results in poor heat Every day, the dairy industry processes large quantities of milk for transmission due to the insulating properties of the dairy fouling, and it producing dairy products, increasing year on year (Eurostat, 2020). becomes necessary to apply more heat to the thermal process; thus, During pasteurization, milk is treated at high temperatures to preserve higher surface temperatures during heat treatment of milk result in more its quality and to keep it free of pathogens and most spoilage-causing fouling (Bansal & Chen, 2006; Barish & Goddard, 2013; Hagsten et al., microorganisms. Plate heat exchangers (PHEs) are a common thermal 2016). In addition to the operational effects, there are also health effects. equipment used in the dairy industry for pasteurizing milk. The complex When the cleaning procedures are not effective, the continuous presence biological composition of milk means that when heated it forms of fouling can result in bacterial colonization and the formation of considerable unwanted adhesion of deposits within the PHE, termed biofilms with high resistance to cleaning procedures (Brooks & Flint, dairy fouling (De Jong, 1997; Wallhäußer, Hussein, & Becker, 2012a, 2008; González-Rivas, Ripolles-Avila, Fontecha-Umaña, Ríos-Castillo, & 2012b). Other equipment, such as spray dryers, are also affected by Rodríguez-Jerez, 2018). This implies a reduction in the quality of milk fouling formation when hot air is used in the production of powdered or other processed foods that may result in cross-contamination, milk. The presence of fouling in dairy plants is a major problem because affecting consumer health (Oliver, Jayarao, & Almeida, 2005). of subsequent undesirable effects. The main effect of the formation of Cleaning and disinfection involve the elimination of different types dairy fouling is the coating of the internal surfaces of pipelines, which of microorganisms and the removal of fouling on food contact surfaces * Corresponding author. Universitat Autònoma de Barcelona, Travessera dels Turons s/n, Cerdanyola del Vallès, 08193, Spain. E-mail address: josejuan.rodriguez@uab.cat (J.J. Rodríguez-Jerez). https://doi.org/10.1016/j.lwt.2021.112634 Received 19 March 2021; Received in revised form 8 October 2021; Accepted 13 October 2021 Available online 18 October 2021 0023-6438/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). A.E. Guerrero-Navarro et al. LWT 154 (2022) 112634 (Boyce, Piterina, & Walsh, 2010a, 2010b; Ríos-Castillo, González-Rivas, Table 1 & Rodríguez-Jerez, 2017). In the dairy industry, cleaning-in-place (CIP) Chemical and enzymatic cleaning treatments performed in the plate heat is a widely used cleaning and disinfection method for dairy product exchanger and spray dryer. safety and for restoring PHE characteristics after production. A typical Chemical Enzymatic CIP cycle consists of a pre-rinse step with hot water, followed by an PHE Pre-rinse: Rinse: alkaline and acid step, and a final rinse step. The presence of fouling and Duration: 20 min, 20–22 ◦ C, 500 L Duration: 20 min, 20–22 ◦ C, 500 its consequent cleaning presents immense economic loss because of the tap water/h. L tap water/h. use of cleaners, increased energy consumption, and the periods of Alkaline step: Enzymatic cleaning: 3% Brio Complex (NaOH and 1% (3% Savinase + 3.3% downtime. In addition, the cleaning time is usually longer than is strictly phosphonates) Termamyl Ultra 300 L, necessary due to the lack of reliable monitoring tools (Van Asselt, Van Duration: 20 min, 80 ◦ C, 500 L/h. diluted in 3% buffer solution). Houwelingen, & Te Giffel, 2002). All these economic expenses related to Rinse: Duration: 20 min, 50 ◦ C, 500 L/ the cleaning of fouling are estimated to account for 80% of total pro­ h. duction costs (Bansal & Chen, 2006). In addition, the chemical cleaner Duration: 20 min, 20–22 ◦ C, 500 L Rinse: tap water/h. residues are corrosive, toxic, and non-biodegradable, and their pro­ Acid step: Duration: 20 min, 20–22 ◦ C, 500 longed use has an impact on the environment. Thus, the cleaning of L tap water/h. equipment represents a challenge for dairy plants. 1% Acimix CIP (nitric acid) Enzymatic formulations with proteolytic activity are alternatives to Duration: 20 min, 50 ◦ C, 500 L/h. Rinse chemical agents for removing fouling (Graβhoff, 2002; Lequette, Boels, Duration: 20 min, 20–22 ◦ C, 500 L tap water/h. Clarisse, & Faille, 2010a, 2010b). Among the benefits of using enzymes Spray Pre-rinse: Rinse: in cleaning are the fact that they are biodegradable, they operate with dryer less energy and water consumption, and they generate less residues Duration: 20 min, 20–22 ◦ C, 500 L Duration: 20 min, 20–22 ◦ C, 500 (Boyce et al., 2010a, 2010b; Guerrero-Navarro, Ríos-Castillo, tap water/h. L tap water/h. Alkaline step: Enzymatic cleaning: Ripolles-Avila, Felipe, & Rodríguez-Jerez, 2020). Enzymes such as li­ 3% Jet Foam (H2O2 and 1% (3% Savinase + 3.3% pases and proteases have been studied as cleaning agents for fouling at surfactants). Termamyl Ultra 300 L, laboratory scale, showing high efficiency (Allie, Jacobs, Maartens, & Duration: 20 min, 80 ◦ C, 500 L/h. diluted in 3% buffer solution) + Swart, 2003). Moreover, the increased production of enzymes on a large Rinse: alkaline foaming detergent. Duration: 20 min, 20–22 ◦ C, 500 L Duration: 20 min, 50 ◦ C, 500 L/ scale has reduced the cost of enzyme products with cleaning properties tap water/h. h. to a competitive level (Petrus, Li, Chen, & Norazman, 2008). In contrast Acid step: Rinse: to laboratory-scale enzyme cleaning studies, the activity of enzymes 1% Acid Jet (ortophosphoric acid, Duration: 20 min, 20–22 ◦ C, 500 under real conditions of use is not particularly accurate. The objective of surfactants) L tap water/h. the present study was to evaluate an enzymatic cleaning treatment for Duration: 20 min, 80 ◦ C, 500 L/h. Rinse: eliminating the dairy fouling formed in the production of pasteurized Duration: 20 min, 20–22 ◦ C, 500 L tap water/h. milk and milk powder, to be applied under real conditions of use in the dairy industry. The study was carried out in an experimental dairy production plant, where we monitored the cleaning process, performed evaluated. The dairy fouling cleaning also included microbiological the microbiological analysis, and compared the efficacy of the enzy­ analysis, detection of fluorescence markers, and the identification of matic and chemical treatments. components of fouling adhered in the PHE and the spray dryer once the cleaning had been completed. In addition, we determined the economic 2. Materials and methods costs of the cleaning treatments. The experiments were performed using 4000 L milk, of which 3600 L was used in the PHE, and 400 L was used in 2.1. Location the spray dryer. The raw milk pasteurization was performed to work at 110 ◦ C for 15 min, at a rate of 600 L/h. For each cleaning treatment, The study was performed at the pilot plant of the Institute of Agri­ dairy fouling derived from the spray drying was obtained after 40 min food Research and Technology (IRTA) experimental processing plant using 100 L of milk to produce milk powder. (Monells, Spain). A thermal system using an indirect PHE (HTST Pasteurizer. Inoxpa S.A.U., Banyoles, Spain) was used for the thermal 2.3. Cleaning procedures heat treatment of milk. The PHE thermal treatment section had 18 plates and a total surface of 9 m2. Raw liquid bovine whole milk refrigerated at The chemical and enzymatic cleaning treatments were carried out 5 ◦ C was used to test the cleaning of dairy fouling following milk once the processes in the PHE and the spray dryer had been completed. A pasteurization in the PHE and the production of powdered milk in the schematic procedure of the two treatments is described in Table 1. spray dryer. A fluidized spray dryer Niro FSD-6.3 (GEA, Alcobendas, Spain) was used to powder the milk. The spray dryer has a feed rate of 2.3.1. Chemical cleaning 45 L/h (45% w/w total solid content of the product), inlet air temper­ The chemical treatment of the PHE, based on the previous descrip­ ature of 180 ◦ C and outlet air temperature of 75 ◦ C. A GEA-Niro pilot tion by Boyce et al. (2010a, 2010b) and Lyndgaard, Rasmussen, Engel­ single effect falling-film evaporator composed of 25 evaporation tubes sen, Thaysen, and van den Berg (2014), was carried out in two steps: a was used before the spray dryer to concentrate the milk. Dairy farms in first-step treatment with an alkaline product (5–15% [v/v] sodium hy­ the Baix Empordà region (Girona, Spain) supplied the whole milk (36.2 droxide, phosphonates, pH 13.0 ± 1.0 at 1%; 1.41 ± 0.02 g/cm3 density ± 0.98 g/L of fats, 33.5 ± 1.17 g/L of proteins, 55.31 ± 2.06 g/L of at 20 ◦ C) (Brio Complex; iTram Higiene, Vic, Spain). During this stage, sugars and 124.8 ± 2.3 g/L of total solids). The IRTA plant has a 400 m2 the CIP process began with a pre-rinse using 200 L tap water for 20 min functional surface area for equipment and complies with food safety with a flow rate of 500 L/h at room temperature (20–22 ◦ C). This rinse standards, according to European Union legislation. removes substances that are not highly adhered to surfaces. Similarly, any soluble substance derived from milk thermal processing is elimi­ 2.2. Experimental conditions nated. Alkaline cleaning was then carried out using Brio Complex diluted to 3% v/v in 200 L tap water in recirculation (500 L/h) for 20 The level of effectiveness to clean fouling using an enzymatic min at 80 ◦ C. Once the alkaline cleaning process had been completed, a cleaning solution compared with CIP using chemical cleaners was water rinse (500 L/h) was performed to remove solubilized residues 2 A.E. Guerrero-Navarro et al. LWT 154 (2022) 112634 resulting from this step. Next, the acid stage consisted of cleaning using a composition of proteins, lipids, and sugars was identified using a direct 1% v/v acid product (42% nitric acid, pH 1.1 ± 0.5 at 100%, 1.3 ± 0.02 epifluorescence microscope (DEM) with three selective fluorescent dyes g/cm3 density at 20 ◦ C) (Acimix CIP; iTram Higiene, Spain) in 200 L (fluorescein isothiocyanate [FITC], Nile red [NR], and concanavalin A- recirculated water at 50 ◦ C for 20 min. Last, after this step, a water rinse Alexa Fluor 350 [ConA-350]). This mixture of dyes was used because it was performed. has previously been used for dairy fouling samples (Guerrero-Navarro The chemical cleaning of the spray dryer was performed with a pre- et al., 2020). Before the analysis, all dyes were mixed homogeneously in rinse with tap water at room temperature, followed by the use of 3% v/v 60 μL 0.1 M sodium bicarbonate (Panreac, Castellar del Vallès, Spain) to alkaline product (3–5% hydrogen peroxide and ethoxylated alcohols, < avoid dimerization of the ConA-350 dye. In addition, the fouling residue 5% phosphonates, < 5% anionic surfactants, < 5% nonionic surfactants; samples were analyzed with near-infrared spectrometry (NIRS; NIR pH 13.5 ± 1.0; 1.075 ± 0.025 g/cm3 density at 20 ◦ C) (Jet Foam; iTram 5000, FOSS-NIR Systems Inc., Silver Springs, MD, USA) at 1100–2500 Higiene, Spain) performed at 500 L/h. The treatment was carried out at nm wavelength. Additionally, the samples were carbonized in an oven at 80 ◦ C for 20 min. Subsequently, a second 20-min rinse was performed, 550 ◦ C for 5 h to determine the mineral fraction. and a 1% v/v acid product (34% orthophosphoric acid, < 5% nonionic surfactants; pH 2.25 ± 0.5 at 1%; 1.2 ± 0.025 g/cm3 density at 20 ◦ C) 2.4.3. Microbiological analysis (Acid Jet; iTram Higiene, Spain) was used, using tap water for 20 min at Microbiological analysis of the cleaning was performed after each 50 ◦ C. Last, a 20-min rinse was performed. cleaning treatment using the water used for the cleaning rinses and the whole milk used for the tests. For this step, 10 mL of each sample was 2.3.2. Enzymatic cleaning transferred to 90 mL buffered peptone water (bioMérieux, Marcy- The enzymatic cleaning in the PHE was carried out with a formula­ l’Étoile, France). The samples were then homogenized on a shaker for tion patented by Rodríguez-Jerez, Ríos-Castillo, and Guerrero-Navarro 30 s, and serial dilutions were made using buffered peptone water (2020), which consists of protease (3.0% [v/v] Savinase® 16 L [16 (bioMérieux, France). From each dilution, Petri dishes were cultured KNPU–S/g]; Novozymes A/Z, Bagsværd, Denmark) and alpha-amylase using five types of culture media: (1) Total viable count (TVC), using the (3.2% [v/v] Termamyl® Ultra 300 L [300 KNU/g]; Novozymes A/Z), automated TEMPO® system (bioMérieux, France), incubated at 30 ◦ C diluted in 3% (v/v) buffer solution (nonylphenoxy poly[ethyleneoxy)] for 48 h (2) Lactic acid bacteria count (LAB) using de Man-Rogosa- ethanol 15EO [10.0% v/v]). The pre-rinse was performed with a flow Sharpe agar (MRS; Oxoid, England), incubated at 30 ◦ C for 48 h (3) rate of 500 L/h using tap water at room temperature for 20 min to Staphylococcus aureus count using Baird-Parker agar with rabbit plasma remove the soluble milk components. Subsequently, a 1% (v/v) enzy­ fibrinogen supplement (BP-RPF; Oxoid, England), incubated at 37 ◦ C for matic formulation in 200 L of tap water in recirculation at 500 L/h and 24 h (4) Enterobacteria count (EC) using violet red bile glucose agar 50 ◦ C for 20 min was used. Last, a 20-min rinse with tap water was (VRBGA; Oxoid, England), incubated at 37 ◦ C for 24 h (5) Yeast and performed at room temperature. The enzymatic cleaning of the spray molds (Y&M), cultured in Sabouraud (SAB) medium with dextrose and dryer was preceded by a first 20-min rinse step with tap water at room chloramphenicol (bioMérieux, France), incubated at 20 ◦ C for 120 h. temperature. Next, a 1% (v/v) enzymatic formula was applied, adding an alkaline foaming detergent (1% v/v) to produce foam and increase 2.4.4. Cleaning monitoring the contact time of the enzymes with the surfaces. The enzymatic The monitoring of the fluorescent markers during the cleaning of the treatment was applied for 20 min at 50 ◦ C. Last, a 20-min rinse with tap spray dryer was carried out by taking samples of the water used, the water was performed at room temperature. cleaning solutions, and the rinse waters (Guerrero-Navarro et al., 2020). The fluorescent analysis was performed with a Cary Eclipse fluorescence 2.3.3. Pre-cleaning of plates before thermal treatment and cleaning spectrophotometer (Agilent Technologies, Madrid, Spain) coupled with Before each cleaning treatment, the PHE unit was opened, separating a 15 W xenon lamp and a front-face geometry accessory (Agilent all the plates, and removing all the incrustations formed. We used a Technologies). The analysis was performed at angles of incidence of 90◦ , Kärcher unit (HD 10/15-4 Cage Food) (Kärcher España, Granollers, and at 35◦ to analyze thinly diluted samples, using Suprasil® quartz Barcelona, Spain) with microfiber tubular browsers (Karcher tubular cuvettes (UV fluorescence cell, Agilent Technologies). Quantification of browser 500) (Kärcher España, Granollers, Barcelona, Spain). The four markers whose fluorescence response depends on the heat treat­ cleaning temperature of the water was 60 ◦ C. Consequently, before each ment of milk (Ayala, Zamora, González, Saldo, & Castillo, 2017a, heat treatment, the PHE was thoroughly cleaned, and any organic res­ 2017b) was performed with standard curves and front-face fluorescence idue present on the inner surface of each plate was removed. The level of of: (1) Dityrosine (Dt), detected at 315 nm excitation wavelength and an organic matter in the internal surface of each plate was the same before emission wavelength of 350–500 nm. (2) Maillard compounds (MC), each treatment. captured with an excitation wavelength of 330 nm and an emission wavelength of 350–500 nm. (3) Tryptophan fluorescence (L-Trp), ob­ 2.4. Cleaning analysis tained with an excitation wavelength of 290 nm and an emission wavelength of 300–450 nm. (4) Riboflavin (Rbf), detected with an 2.4.1. Dry residues excitation wavelength of 450 nm and an emission wavelength of Once the chemical and enzymatic cleaning had been completed, 100 470–570 nm. mL of each cleaning solution were collected in duplicate from the PHE system and spray dryer, and dried. Drying was performed using 20 mL of 2.5. Cost of cleaning treatments each sample in an IDL–FI–80 forced air oven (Labolan, SL, Esparzar de Galar, Spain) at 90 ◦ C for 1–1.5 h. Next, the dried residues from each The average economic costs each cleaning treatment in the year treatment were weighed on a Mettler AE100 analytical balance (Mettler- 2020 were calculated in euros (€) based on the manufacturing prices Toledo S.A.E., Hospitalet del Llobregat, Spain). The average density of provided by the manufacturer. the cleaning products was subsequently discarded, obtaining the amount of milk fouling eliminated during the cleaning. The results were 2.6. Statistical analysis expressed in g/L. The chemical and enzymatic cleaning of the PHE and spray dryer 2.4.2. Fouling components were carried out twice, on separate days. The sampling of each type of The samples to analyze the milk fouling components were obtained cleaning was performed by taking six aliquot samples of cleaning solu­ from the PHE, once the cleaning stages were completed. The tions and rinse waters. All the data collected from the cleaning 3 A.E. Guerrero-Navarro et al. LWT 154 (2022) 112634 fouling than the two-stage chemical method. This may be because the protease and amylase in the enzymatic solution act together to dissolve and degrade the milk organic components. Proteases and amylases are the first and second most used enzymes in the formulation of detergents with enzymatic properties (Mitidieri, Souza Martinelli, Schrank, & Vainstein, 2006). The use of proteases has demonstrated cleaning ben­ efits for the detergent industry. Boyce and Walsh (2012) evaluated fungal-origin proteases on a laboratory scale to determine their potential suitability for cleaning in the dairy industry. Using these proteases achieved adequate industrial-like milk fouling cleaning of stainless-steel surfaces at 40 ◦ C, replacing caustic-based cleaning solutions that act at 70–80 ◦ C. Amylase has been reported in the literature to improve cleaning procedures. In the case of our study, Termamyl® 300 L is a commercial heat-stable alpha-amylase produced by Bacillus licheniformis used to hydrolyze glycosidic linkages, which cleaves starch at 96 ◦ C (Allala et al., 2019; Marasca, Boulos, & Nyström, 2020). In a Fig. 1. Amount of dairy fouling (g/L) removed in each treatment. laboratory-scale study, Guerrero-Navarro, Ríos-Castillo, Ripolles-Avila, Hascoët, Felipe, and Rodríguez-Jerez (2019) observed that a formula­ treatments were processed using R free software (R Development Core tion based on proteases and alpha-amylase (Termamyl®) acting at 55 ◦ C Team). The comparison between the cleaning treatments was performed had 75% efficiency in cleaning milk fouling. Likewise, the authors with the unpaired T-test. A p-value ≤ 0.05 was considered statistically observed that enzymatic cleaning has comparable efficiency to chemical significant. cleaning for eliminating fouling. Besides, Ripolles-Avila, Ríos-Castillo, Fontecha-Umaña, and Rodríguez-Jerez (2020) studied the efficacy of an 3. Results and discussion enzymatic formulation based on protease (5.0%), lipase (0.5%), and amylase (2.5%) as a cleaning treatment to prevent the presence of bio­ 3.1. Fouling cleaning residues films, reporting 34.48–37.5% biofilm detachment when the formula was tested on stainless steel surfaces. The results of the dry residues during the cleaning of the dairy In dairy plants, the principle of the spray drying process performed in fouling are shown in Fig. 1. The dry residue value of the tap water used a drying chamber is the atomization of dairy concentrate in a stream of during cleaning was 0.59 g/L. The dry residues from the cleaning so­ heated air, followed by dehydration, yielding milk powder with a low lutions analyzed for the PHE were calculated as 23.95 g/L for the water content (Moejes & van Boxtel, 2017). Although the cleaning of the chemical treatment, of which 20.61 g/L was obtained during the alka­ spray dryer atomizer can be more efficient because its ease of access line cleaning step and 3.3 g/L during the acid step. By contrast, the ef­ during maintenance allows a deeper clean than for the PHE system, it ficacy of the enzymatic solution for cleaning the dairy fouling showed an has been observed that the accumulation of fouling in this equipment increase of up to 33.21 g/L, which was statistically significant (p < 0.05) could mean a decrease in the quality of the processed foods (Goula & compared to the chemical treatment. However, the results for the dry Adamopoulos, 2004). In the case of powdered milk production, cleaning residues from the spray dryer did not show statistical differences (p > of the dryers by CIP represents 10–26% of the energy costs during this 0.05) between the enzymatic treatment (5.30 g/L) and the chemical stage (Ramírez, Patel, & Blok, 2006). Our results show that the enzy­ treatment (5.10 g/L). matic formulation could be an alternative to chemical products under Chemical cleaning has high efficiency because the alkaline agents real conditions of use for CIP of thermal equipment for pasteurizing milk saponifies fats and degrades fouling proteins, dissolving the residues and producing powdered milk. Graβhoff (2002) demonstrated the effi­ adhered to the surfaces (Ogunbiyi, Miles, & Hilal, 2008). In the present cacy of Savinase® for cleaning a milk pasteurizer, with an optimal use study, sodium hydroxide was used in the alkaline product because of its temperature of 50 ◦ C, followed by a 15-min acid wash. Our results show high pH. Besides, the acid step performs a descaling function, especially efficacy at 50 ◦ C under real conditions to reduce dairy fouling. These on the mineral elements adhered to the surfaces (Graβhoff, 2002). Here, results are comparable with those observed by Guerrero-Navarro, Río­ we found that the enzymatic solution cleaned a greater amount of milk s-Castillo, Ripolles-Avila, et al. (2019), who tested the effectiveness of an Fig. 2. Photographs of PHE after enzymatic cleaning. Red circles indicate fouling residues adhered to stainless-steel surfaces (A) and cleaned surfaces (B). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) 4 A.E. Guerrero-Navarro et al. LWT 154 (2022) 112634 Table 3 Microbiological analysis of milk and rinse waters from chemical and enzymatic treatments performed in the PHE and spray dryer processes (CFU/mLa). Process Sample Total Lactic Enterobacteria Yeast viable acid and count bacteria molds PHE Raw milk 3.2 × 1.9 × 103 4.0 × 103 2.6 × 105 103 Pasteurized 2.4 × 0 0 1.6 × milk 104 103 Post-alkaline 1.0 × 0 0 0 rinse water 10◦ Post-acid rinse 3.4 × 0 0 0 water 101 Post- 1.0 × 0 0 1.0 × enzymatic 10◦ 101 rinse water Spray Powder milk 2.9 × 0 0 0 dryer 102 Powder milk 1.0 × 0 0 0 fouling 102 Fig. 3. Dairy fouling obtained after enzymatic treatment from PHE: (A) dried Post-alkaline 4.0 × 0 0 0 residues with a soft, white appearance. Scale bar: 10 cm. (B) Direct epifluor­ rinse water 10◦ escence microscopy image of fouling stained with fluorescein isothiocyanate Post-acid rinse 2.0 × 0 0 0 [FITC], Nile red [NR], and concanavalin A-Alexa Fluor 350 [ConA-350]), water 10◦ visualized in the green channel. Protein components in yellow-greenish color­ Post- 0 0 0 1.0 × ation. Scale bar: 200 μm. (For interpretation of the references to color in this enzymatic 101 figure legend, the reader is referred to the Web version of this article.) rinse water a Mean values expressed as CFU/mL; n = 4 for every item of data. Table 2 Composition of the residual fouling layer from PHE after enzymatic cleaning action of savinase®, which is an alkaline protease that has an adequate obtained from NIRS and mineral fraction analysis. proteolytic activity at 50 ◦ C and under most washing conditions (Hag­ gag, El-Sayed, & Allam, 2007; Nazari, Montazer, Afzali, & Sheibani, Component Content (% w/w) ± standard deviation 2014). Although the enzymatic activity does not have a direct effect on Mineral 56.6 ± 3.7 the mineral composition, the mineral cleaning may be favored when Protein 23.1 ± 1.6 removed together with proteins such as casein during the enzymatic Fats 15.4 ± 1.9 Water 5.0 ± 0.5 cleaning without increasing the cleaning temperature (50 ◦ C). Other­ Total 100.0 wise, with a heat treatment at a higher temperature, the adherence of minerals to the equipment would have been higher, resulting in a greater energy consumption to remove it (Hagsten et al., 2016; Visser & enzymatic solution at the same temperature to clean milk fouling Jeurnink, 1997). Therefore, the protein composition of milk can also be developed under laboratory conditions. In addition, due to the reduction used to eliminate compounds such as minerals. In this sense, it should be in rinses and washing times when using the enzymatic cleaning, water noted that a high content of mineral and macromineral elements, such as consumption and washing temperature are also decreased, consequently calcium phosphate, calcium phosphocaseinate, and magnesium, are saving energy compared to chemical cleaning. associated with milk proteins (Gaucheron, 2005; Regnault, Dumay, & Cheftel, 2006; Sola-Larrañaga & Navarro-Blasco, 2009). Considering the 3.2. Fouling analysis effectiveness of enzymatic cleaning, it could either be used as a sole cleaning treatment or it could be complemented with acid CIP treat­ After the cleaning treatments, the PHEs were opened, the plates ments to guarantee the total elimination of possible mineral residues. through which the heat-treated milk flowed separated, and the milk Nevertheless, the different characteristics and operating conditions of fouling removal determined. The enzymatic cleaning eliminated the pasteurization and spray drying equipment must be considered. 85–95% of fouling. The presence of fouling was particularly evident in the areas of the plates where the temperature increased the most during 3.3. Microbiological analysis the thermal pasteurization (Fig. 2A) and was soft and whitish in color (Fig. 3A). The other areas did not present adhered fouling (Fig. 2B). The As expected, the microbiological analysis of the pasteurized milk remaining fouling was extracted, and NIRS (organic fraction) and showed that the highest microbial load was obtained from raw milk carbonization (mineral fraction) evaluation of its composition showed a (Table 3). By contrast, S. aureus was not detected in raw milk or in any high percentage of mineral and protein contents (Table 2). Microscopy cleaning solutions and rinse waters (data not shown). After pasteuriza­ also revealed a high protein content (Fig. 3B). Our results agree with tion, LAB and EC were not detected, the presence of TVC was reduced, those of Visser and Jeurnink (1997), who indicated that proteins and and the Y&M count remained at 101 CFU/mL. During the alkaline and minerals are the main components of the fouling produced during the acid cleaning steps, no microbiological load was detected, possibly due heat treatment of milk. Guerrero-Navarro et al. (2020) showed that, to the effect of the cleaning products during washing, although the post- under laboratory conditions, the proportion of protein analyzed by treatment rinse showed TVC counts that were higher with the acid step confocal laser scanning microscopy (CLSM) was 39.4% prior to a compared to the alkaline one. This observation may have been due to the cleaning treatment. Although in our study we did not initially analyze presence of microorganisms adhered to the surfaces prior to cleaning or the protein content of the fouling, once the enzymatic cleaning was from milk in a state which, due to the effect of chemical treatments, was completed under real conditions, we found a low protein content viable but non-culturable (VBNC) (Ayrapetyan & Oliver, 2016), and (23.1%). By contrast, a higher mineral content was found (56.6%). The were only detected from rinses after chemical treatments. In this regard, enzymatic cleaning of proteins can be attributed to the proteolytic an adequate post-cleaning rinse is essential to eliminate the injured 5 A.E. Guerrero-Navarro et al. LWT 154 (2022) 112634 Table 4 environmental yeast recognized as an emerging pathogen and, because Concentrations of fluorescence markers in the different stages of spray dryer this genus produces extracellular proteases and lipases, it is likely to cleaning. contribute to the spoilage of dairy products (Wirth & Goldani, 2012). Sample Concentrationa ± standard deviation L-Trp (mg/ Rbf (mg/ MC (mL/ Dt (mL/L) 3.4. Fluorescence markers L) L) L) The results of the front-face analysis of the markers of the fouling Tap water 0.05 ± 0.00 0.25 ± 0.23 ± 0.04 0.05 0.06 obtained from the spray dryer are shown in Table 4. For chemical Chemical cleaning cleaning, it was observed that the L-Trp, MC, and Dt values were similar Pre-rinse 0.06 ± 1.02 ± 0.30 ± 0.30 ± between tap water and pre-rinse, with the exception of Rbf (0.0 mg/L in 0.04 0.50 0.10 0.10 tap water and 1.02 mg/L in pre-rinse). The alkaline solution showed the Alkaline solution 0.22 ± 2.36 ± 0.76 ± 0.97 ± 0.10 1.00 0.30 0.40 highest values of the markers (LTrp, Rbf, MC, and Dt), which were Rinse after alkaline 0.05 ± 0.37 ± 0.28 ± 0.27 ± reduced with the rinse after alkaline treatment. In this sense, an treatment 0.04 0.04 0.10 increased reduction was observed with the Rbf marker (2.36 mg/L to Acid solution 0.05 ± 0.85 ± 0.31 ± 0.28 ± 0.37 mg/L). In the case of the acid step of the chemical cleaning, these 0.04 0.45 0.10 values were found to be similar when comparing the acid solution and Rinse after acid treatment 0.06 ± 0.95 ± 0.29 ± 0.31 ± 0.04 0.40 0.12 the rinse after acid treatment in all the markers tested. During the Enzymatic cleaning enzymatic cleaning, we observed a reduction in the presence of L-Trp Enzymatic solution 0.23 ± o.d.b 2.41 ± 2.73 ± (0.05 mg/L), MC (0.28 mL/L), and Dt (0.27 mL/L) in the samples ob­ 0.10 tained from the rinse after enzymatic treatment compared with the Rinse after enzymatic 0.05 ± 0.98 ± 0.28 ± 0.27 ± treatment 0.04 0.45 0.12 0.10 enzymatic solution. In the case of the Rbf results, the changes could not be compared because although the result of the rinse after the enzymatic Tryptophan (L-Trp), Riboflavin (Rbf), Maillard compounds (MC), Dityrosine treatment was 0.98 mg/L, the results for this marker were out of the (Dt). a detection range in the enzymatic solution. Mean values expressed as mg/L or mL/L; n = 4 for every item of data. b Changes in fluorescence intensity, particularly during the alkaline Out of detection range. step of the chemical cleaning and the enzymatic cleaning, may be due to the use of Jet Foam alkaline cleaning agent based on hydrogen peroxide microorganisms that can be reactivated after cleaning treatments. In the and enzymes, which contain fluorescent compounds. In addition, the dairy industry, chemical cleaning with alkaline products such as sodium presence of compounds released from the milk fouling due to the use of hydroxide reduces the presence of bacterial microorganisms and im­ cleaning agents must also have influenced the changes in the fluorescent proves the removal of dairy biofilms and fouling (Bremer, Fillery, & marker intensities. Because there were changes in the intensity of the McQuillan, 2006). In our work, the combination of NaOH and phos­ markers, front-face measurement of the concentration of fluorophores phonates reduced the presence of total aerobic bacteria, observed as a could be used as a method for evaluating the cleaning efficacy of alka­ reduction in their presence in the alkaline post-cleaning rinse. Although line and enzymatic solutions for dairy fouling. Given that acid treatment acid cleaning reduces bacterial load to a lesser extent than alkaline mostly descales inorganic matter, analysis with these markers is not cleaning, its activity also has an effect on bacterial growth. Unlike the highly recommended. Previous authors have traced the cleaning process chemical cleaning, the presence of TVC was observed during the enzy­ using fluorescent measurements. Berg, Ottosen, van den Berg, and Ipsen matic cleaning. This can be explained by the fact that enzymes do not (2017) observed that inline ultraviolet–visible (UV–Vis) spectroscopy have a direct antimicrobial effect on cells, but have activity to descale could be used to optimize the processing time, energy, and chemicals them when they are attached to surfaces (Lequette et al., 2010a, 2010b; required for cleaning in place of membrane filtration plants. Lyndgaard Ripolles-Avila et al., 2020). The effect of detaching microorganisms by et al. (2014) demonstrated that UV spectroscopy in combination with an the enzymatic cleaning solution was evidenced in the reduction of TVC explorative multivariate classification method such as principal on analyzing the rinse after the enzymatic treatment. These results component analysis can be used as a method for monitoring the per­ indicate an effective microbiological cleaning of the installation, and formance of the cleaning procedures. In addition, previous studies have thus removal of fouling deposit, using the chemical and enzymatic demonstrated the efficacy of measuring fluorophore compounds for solutions. determining the compounds in milk subjected to heat treatments. Ayala, Microbiological values of the powdered milk fouling obtained during Zamora, Rinnan, Saldo, and Castillo (2020) found that tryptophan the cleaning of the spray dryer were low. A TVC count of no more than fluorescence allowed sample classification by PCA analysis according to 102 CFU/mL was found, and the presence of yeasts and molds was only heat load, i.e., temperature and time. In another study, the same authors evidenced in the rinse water after the enzymatic cleaning. The low evaluated the correlation of lactulose concentration with fluorescence microorganism count after the spray dryer cleaning can also be related markers such as tryptophan, dityrosine, Maillard intermediate com­ to the ease of cleaning the fouling compared to the PHE. However, the pounds, and riboflavin to obtain prediction models of lactulose con­ formation of fouling facilitates the development of biofilms; therefore, centration in heat-treated milk based on front-face fluorescence (Ayala proper cleaning should ensure complete elimination of fouling in dairy et al., 2017a, 2017b). Liu, Zamora, Castillo, and Saldo (2018) and plants with spray drying systems for producing milk powder (Walmsley, Alvarado, Zamora, Liu, Saldo, and Castillo (2019) studied the potential Walmsley, Atkins, Neale, & Tarighaleslami, 2015). One of the bacterial of predicting retinol loss and riboflavin, respectively, in milk during microorganisms isolated most often from the spray dryer was Bacillus thermal processing using front-face fluorescence spectroscopy, amyloliquefaciens, which was biochemically identified with the API 50 concluding that this technique has the potential to replace existing CBH system (bioMérieux, France). Although B. amyloliquefaciens has not conventional analytical techniques for heat-treated milk. been implicated in food poisoning cases, it has the ability to produce highly thermoresistant spores (surviving 125 ◦ C, 30 min), and has pro­ 3.5. Costs teolytic activity that affects the quality of dairy products (Lücking, Stoeckel, Atamer, Hinrichs, & Ehling-Schulz, 2013). In the case of yeast The analysis of the costs derived from the cleaning in the pilot plant and molds, there were low counts of these microorganisms, with Rho­ shows that the alkaline treatment costs 9.36 €/200 L, whereas the dotorula spp. one of the more frequently detected microorganisms in the combined alkaline–acid treatment costs 12.20 €. The enzymatic treat­ rinse water after milk pasteurization. Rhodotorula is a common ment (9.00 €) together with the expense of the buffer for the treatment 6 A.E. Guerrero-Navarro et al. LWT 154 (2022) 112634 Table 5 Investigation, Writing – review & editing, Visualization, Supervision, Product costs (€) of each cleaning treatment. Project administration, Funding acquisition. Product Main Concentration (% Kg/ €/Kg Total component v/v) 200 L cost Acknowledgements Alkaline NaOH 3% 6.00 1.56 9.40 cleaner The authors thank Sarah Davies for the English grammar review. Acid cleaner HNO3 1% 2.00 1.42 2.80 Enzymes Enzymes and 1% 2.00 4.50 9.00 References surfactants Enzymatic Buffer 3% 6.00 1.50 9.00 Allala, F., Bouacem, K., Boucherba, N., Azzouz, Z., Mechri, S., Sahnoun, M., et al. (2019). buffer Purification, biochemical, and molecular characterization of a novel extracellular thermostable and alkaline α-amylase from Tepidimonas fonticaldi strain HB23. International Journal of Biological Macromolecules, 132, 558–574. https://doi.org/ (9.00 €) costs 18.00 € (Table 5). The higher cost of the enzymatic 10.1016/j.ijbiomac.2019.03.201 treatment than the chemical treatment is due to the high cost of the Allie, Z., Jacobs, E. P., Maartens, A., & Swart, P. (2003). Enzymatic cleaning of ultrafiltration membranes fouled by abattoir effluent. Journal of Membrane Science, buffer used. However, the buffer can be diluted up to 1/10 v/v as a 218(1–2), 107–116. https://doi.org/10.1016/S0376-7388(03)00145-5 savings strategy for commercial use. This concentration reduction in Alvarado, U., Zamora, A., Liu, J., Saldo, J., & Castillo, M. (2019). Rapid quantification of buffer usage could represent an enzyme treatment cost of 9.90 €, which riboflavin in milk by front-face fluorescence spectroscopy: A preliminary study. Foods, 9(1), 6. https://doi.org/10.3390/foods9010006 would make it more competitive against current cleaning treatments, Ayala, N., Zamora, A., González, C., Saldo, J., & Castillo, M. (2017a). Predicting lactulose with a 25% cost reduction. In addition, the advantages of enzymatic concentration in heat-treated reconstituted skim milk powder using front-face cleaning compared to chemical cleaning go beyond the immediate costs, fluorescence. Food Control, 73, 110–116. https://doi.org/10.1016/j. foodcont.2016.09.040 as it does not corrode the processing equipment and is biodegradable in Ayala, N., Zamora, A., Rinnan, Å., Saldo, J., & Castillo, M. (2020). The effect of heat the environment after use. Other works have also reported methods treatment on the front-face fluorescence spectrum of tryptophan in skim milk. aimed at reducing the costs of cleaning fouling, including the applica­ Journal of Food Composition and Analysis, 92, 103569. https://doi.org/10.1016/j. tion of ultrasound and nanofiltration procedures to recover and reuse jfca.2020.103569 Ayrapetyan, M., & Oliver, J. D. (2016). The viable but non-culturable state and its cleaning products (Fernández, Riera, Álvarez, & Álvarez, 2010; Luján-­ relevance in food safety. Current Opinion in Food Science, 8, 127–133. https://doi. Facundo, Mendoza-Roca, Cuartas-Uribe, & Álvarez-Blanco, 2016). org/10.1016/j.cofs.2016.04.010 Bansal, B., & Chen, X. D. (2006). A critical review of milk fouling in heat exchangers. Comprehensive Reviews in Food Science and Food Safety, 5(2), 27–33. https://doi.org/ 4. Conclusions 10.1111/j.1541-4337.2006.tb00080.x Barish, J. A., & Goddard, J. M. (2013). Anti-fouling surface modified stainless steel for The enzymatic cleaning based on the use of protease and amylase food processing. Food and Bioproducts Processing, 91(4), 352–361. https://doi.org/ 10.1016/j.fbp.2013.01.003 demonstrated comparable efficacy to the alkaline–acid chemical CIP for Berg, T. H. A., Ottosen, N., van den Berg, F., & Ipsen, R. (2017). Inline UV-Vis cleaning the milk fouling generated in a pilot-scale plate heat exchanger spectroscopy to monitor and optimize cleaning-in-place (CIP) of whey filtration and spray dryer operating under real cleaning conditions in the dairy plants. Lebensmittel-Wissenschaft und -Technologie- Food Science and Technology, 75, 164–170. https://doi.org/10.1016/j.lwt.2016.08.014 industry. The enzymatic formulation may ensure an efficient cleaning by Boyce, A., Piterina, A. V., & Walsh, G. (2010a). Assessment of the potential suitability of removing microorganisms present in the equipment, thus meeting the selected commercially available enzymes for cleaning-in-place (CIP) in the dairy hygienic conditions necessary for producing dairy products. Monitoring industry. Biofouling, 26(7), 837–850. https://doi.org/10.1080/ 08927014.2010.522705 fluorescence markers during washes can be a method for improving the Boyce, A., & Walsh, G. (2012). Identification of fungal proteases potentially suitable for efficiency of alkaline and enzymatic cleaning procedures. The enzymatic environmentally friendly cleaning-in-place in the dairy industry. Chemosphere, 88(2), formulation tested fulfills the objectives of the dairy industry, including 211–218. https://doi.org/10.1016/j.chemosphere.2012.03.022 Bremer, P. J., Fillery, S., & McQuillan, A. J. (2006). Laboratory scale Clean-In-Place (CIP) saving water and energy consumption, by reducing the use of chemical studies on the effectiveness of different caustic and acid wash steps on the removal of products. Considering that the enzymatic cleaning treatment is biode­ dairy biofilms. International Journal of Food Microbiology, 106(3), 254–262. https:// gradable and non-corrosive and is competitively priced compared to doi.org/10.1016/j.ijfoodmicro.2005.07.004 Brooks, J. D., & Flint, S. H. (2008). Biofilms in the food industry: Problems and potential traditional chemical cleaning, it could be used in a complementary solutions. International Journal of Food Science and Technology, 43(12), 2163–2176. manner or may be an alternative to the use of chemical products for the https://doi.org/10.1111/j.1365-2621.2008.01839.x cleaning of milk fouling. De Jong, P. (1997). Impact and control of fouling in milk processing. Trends in Food Science & Technology, 8, 401–405. https://doi.org/10.1016/S0924-2244(97)01089-3 Eurostat. (2020). Milk and milk product statistics. Retrieved March 1, 2021, from https Funding ://ec.europa.eu/eurostat/statistics-explained/pdfscache/29568.pdf. Fernández, P., Riera, F. A., Álvarez, R., & Álvarez, S. (2010). Nanofiltration regeneration of contaminated single-phase detergents used in the dairy industry. Journal of Food This work was supported by the European Union project Engineering, 97(3), 319–328. https://doi.org/10.1016/j.jfoodeng.2009.10.023 ‘ENTHALPY’ [grant number FP7-KBBE-2013-7]. Gaucheron, F. (2005). The minerals of milk. Reproduction Nutrition Development, 45, 473–483. https://doi.org/10.1051/rnd:2005030 González-Rivas, F., Ripolles-Avila, C., Fontecha-Umaña, F., Ríos-Castillo, A. G., & CRediT authorship contribution statement Rodríguez-Jerez, J. J. (2018). Biofilms in the spotlight: Detection, quantification, and removal methods. Comprehensive Reviews in Food Science and Food Safety, 17(5), Alfons Eduard Guerrero-Navarro: Methodology, Formal analysis, 1261–1276. https://doi.org/10.1111/1541-4337.12378 Goula, A. M., & Adamopoulos, K. G. (2004). Influence of spray drying conditions on Investigation, Writing – original draft. Abel Guillermo Ríos-Castillo: residue accumulation - simulation using CFD. Drying Technology, 22(5), 1107–1128. Conceptualization, Methodology, Software, Validation, Resources, Data https://doi.org/10.1081/DRT-120038583 curation, Writing – original draft, Writing – review & editing. Carolina Graßhoff, A. (2002). Enzymatic cleaning of milk pasteurizers. Food and Bioproducts Processing: Transactions of the Institution of of Chemical Engineers, Part C, 80(4), Ripolles-Avila: Methodology, Validation, Investigation, Resources, 247–252. https://doi.org/10.1205/096030802321154736 Data curation, Writing – original draft, Writing – review & editing. Anna Guerrero-Navarro, A. E., Ríos-Castillo, A. G., Avila, C. R., Hascoët, A. S., Felipe, X., & Zamora: Methodology, Validation, Formal analysis, Data curation, Rodríguez Jerez, J. J. (2019). Development of a dairy fouling model to assess the Writing – review & editing. Anne-Sophie Hascoët: Formal analysis, efficacy of cleaning procedures using alkaline and enzymatic products. Lebensmittel- Wissenschaft und -Technologie- Food Science and Technology, 106, 44–49. https://doi. Investigation, Resources, Writing – original draft, Writing. Xavier org/10.1016/j.lwt.2019.02.057 Felipe: Conceptualization, Validation, Investigation, Resources, Writing Guerrero-Navarro, A. E., Ríos-Castillo, A. G., Ripolles-Avila, C., Felipe, X., & Rodríguez- – review & editing, Visualization, Funding acquisition. Manuel Cas­ Jerez, J. J. (2020). Microscopic analysis and microstructural characterization of the organic and inorganic components of dairy fouling during the cleaning process. tillo: Conceptualization, Validation, Writing – review & editing, Visu­ alization. José Juan Rodríguez-Jerez: Conceptualization, Validation, 7 A.E. Guerrero-Navarro et al. LWT 154 (2022) 112634 Journal of Dairy Science, 103(3), 2117–2127. https://doi.org/10.3168/jds.2019- sakazakii biofilms from food contact surfaces through enzymatic catalysis. Journal of 16957 Food Safety, 40(2), Article e12755. https://doi.org/10.1111/jfs.12755 Haggag, K., El-Sayed, H., & Allam, O. G. (2007). Degumming of silk using microwave- Rodríguez-Jerez, J. J., Ríos-Castillo, A. G., & Guerrero-Navarro, A. E. (2020). Process for assisted treatments. Journal of Natural Fibers, 4(3), 1–22. https://doi.org/10.1300/ removing the fouling deposited in a milk processor unit and a cleaning solution used J395v04n03_01 therein. Retrieved March 1, 2021, from US Patensts website https://patents.google. Hagsten, C., Altskär, A., Gustafsson, S., Lorén, N., Hamberg, L., Innings, F., et al. (2016). com/patent/WO2017216062A1/en. Composition and structure of high temperature dairy fouling. Food Structure, 7, Sadeghinezhad, E., Kazi, S. N., Dahari, M., Safaei, M. R., Sadri, R., & Badarudin, A. 13–20. https://doi.org/10.1016/j.foostr.2015.12.002 (2015). A comprehensive review of milk fouling on heated surfaces. Critical Reviews Lequette, Y., Boels, G., Clarisse, M., & Faille, C. (2010a). Using enzymes to remove in Food Science and Nutrition, 55(12), 1724–1743. https://doi.org/10.1080/ biofilms of bacterial isolates sampled in the food-industry. Biofouling, 26(4), 10408398.2012.752343 421–431. https://doi.org/10.1080/08927011003699535 Sola-Larrañaga, C., & Navarro-Blasco, I. (2009). Chemometric analysis of minerals and Liu, J., Zamora, A., Castillo, M., & Saldo, J. (2018). Using front-face fluorescence trace elements in raw cow milk from the community of Navarra, Spain. Food spectroscopy for prediction of retinol loss in milk during thermal processing. Chemistry, 112(1), 189–196. https://doi.org/10.1016/j.foodchem.2008.05.062 Lebensmittel-Wissenschaft und -Technologie- Food Science and Technology, 87, 151–157. Van Asselt, A. J., Van Houwelingen, G., & Te Giffel, M. C. (2002). Monitoring system for https://doi.org/10.1016/j.lwt.2017.08.073 improving cleaning efficiency of cleaning-in-place processes in dairy environments. Lücking, G., Stoeckel, M., Atamer, Z., Hinrichs, J., & Ehling-Schulz, M. (2013). Food and Bioproducts Processing: Transactions of the Institution of of Chemical Engineers, Characterization of aerobic spore-forming bacteria associated with industrial dairy Part C, 80(4), 276–280. https://doi.org/10.1205/096030802321154772 processing environments and product spoilage. International Journal of Food Visser, J., & Jeurnink, T. J. M. (1997). Fouling of heat exchangers in the dairy industry. Microbiology, 166(2), 270–279. https://doi.org/10.1016/j.ijfoodmicro.2013.07.004 Experimental Thermal and Fluid Science, 14(4), 407–424. https://doi.org/10.1016/ Luján-Facundo, M. J., Mendoza-Roca, J. A., Cuartas-Uribe, B., & Álvarez-Blanco, S. S0894-1777(96)00142-2 (2016). Cleaning efficiency enhancement by ultrasounds for membranes used in Wallhäußer, E., Hussein, M. A., & Becker, T. (2012). Detection methods of fouling in heat dairy industries. Ultrasonics Sonochemistry, 33, 18–25. https://doi.org/10.1016/j. exchangers in the food industry. Food Control, 27, 1–10. https://doi.org/10.1016/j. ultsonch.2016.04.018 foodcont.2012.02.033 Lyndgaard, C. B., Rasmussen, M. A., Engelsen, S. B., Thaysen, D., & van den Berg, F. Walmsley, T. G., Walmsley, M. R. W., Atkins, M. J., Neale, J. R., & Tarighaleslami, A. H. (2014). Moving from recipe-driven to measurement-based cleaning procedures: (2015). Thermo-economic optimisation of industrial milk spray dryer exhaust to Monitoring the Cleaning-In-Place process of whey filtration units by ultraviolet inlet air heat recovery. Energy, 90, 95–104. https://doi.org/10.1016/j. spectroscopy and chemometrics. Journal of Food Engineering, 126, 82–88. https://doi. energy.2015.03.102 org/10.1016/j.jfoodeng.2013.10.037 Wirth, F., & Goldani, L. Z. (2012). Epidemiology of rhodotorula: An emerging pathogen. Marasca, E., Boulos, S., & Nyström, L. (2020). Bile acid-retention by native and modified Interdisciplinary Perspectives on Infectious Diseases, 465717. https://doi.org/10.1155/ oat and barley β-glucan. Carbohydrate Polymers, 236, 116034. https://doi.org/ 2012/465717 10.1016/j.carbpol.2020.116034 Mitidieri, S., Souza Martinelli, A. H., Schrank, A., & Vainstein, M. H. (2006). Enzymatic detergent formulation containing amylase from Aspergillus Niger: A comparative Key references study with commercial detergent formulations. Bioresource Technology, 97(10), 1217–1224. https://doi.org/10.1016/j.biortech.2005.05.022 Ayala, N., Zamora, A., González, C., Saldo, J., & Castillo, M. (2017b). Predicting lactulose Moejes, S. N., & van Boxtel, A. J. B. (2017). Energy saving potential of emerging concentration in heat-treated reconstituted skim milk powder using front-face technologies in milk powder production. Trends in Food Science & Technology, 60, fluorescence. Food Control, 73, 110–116. https://doi.org/10.1016/j. 31–42. https://doi.org/10.1016/j.tifs.2016.10.023 foodcont.2016.09.040This work evaluated the concentration of lactulose in heat- Nazari, A., Montazer, M., Afzali, F., & Sheibani, A. (2014). Optimization of proteases treated milk with fluorescence markers such as tryptophan, dityrosine, Maillard pretreatment on natural dyeing of wool using response surface methodology. Clean compounds, and riboflavin. In our study, we evaluated these markers to monitor the Technologies and Environmental Policy, 16(6), 1081–1093. https://doi.org/10.1007/ fouling cleaning process. s10098-013-0709-0 Boyce, A., Piterina, A. V., & Walsh, G. (2010b). Assessment of the potential suitability of Ogunbiyi, O. O., Miles, N. J., & Hilal, N. (2008). The effects of performance and cleaning selected commercially available enzymes for cleaning-in-place (CIP) in the dairy cycles of new tubular ceramic microfiltration membrane fouled with a model yeast industry. Biofouling, 26(7), 837–850. https://doi.org/10.1080/ suspension. Desalination, 220(1–3), 273–289. https://doi.org/10.1016/j. 08927014.2010.522705This study reported the efficacy of lipase and protease for desal.2007.01.034 CIP application on laboratory scales that we used to determine the efficacy of Oliver, S. P., Jayarao, B. M., & Almeida, R. A. (2005). Foodborne pathogens in milk and enzymatic cleaning performed on real scales at a dairy pilot plant. the dairy farm environment: Food safety and public health implications. Foodborne Guerrero-Navarro, A. E., Ríos-Castillo, A. G., Ripolles-Avila, C. R., Hascoët, A. S., Pathogens and Disease, 2, 115–129. https://doi.org/10.1089/fpd.2005.2.115 Felipe, X., & Rodriguez Jerez, J. J. (2019). Development of a dairy fouling model to Petrus, H. B., Li, H., Chen, V., & Norazman, N. (2008). Enzymatic cleaning of assess the efficacy of cleaning procedures using alkaline and enzymatic products. ultrafiltration membranes fouled by protein mixture solutions. Journal of Membrane Lebensmittel-Wissenschaft und -Technologie- Food Science and Technology, 106, 44–49. Science, 325(2), 783–792. https://doi.org/10.1016/j.memsci.2008.09.004 https://doi.org/10.1016/j.lwt.2019.02.057We used this study, performed under Ramírez, C. A., Patel, M., & Blok, K. (2006). From fluid milk to milk powder: Energy use laboratory conditions, as a base for our study to evaluate dairy fouling formed under and energy efficiency in the European dairy industry. Energy, 31(12), 1984. https:// real conditions during the milk pasteurization and spray drying processes. doi.org/10.1016/j.energy.2005.10.014, 2004. Lequette, Y., Boels, G., Clarisse, M., & Faille, C. (2010b). Using enzymes to remove Regnault, S., Dumay, E., & Cheftel, J. C. (2006). Pressurisation of raw skim milk and of a biofilms of bacterial isolates sampled in the food-industry. Biofouling, 26(4), dispersion of phosphocaseinate at 9◦ C or 20◦ C: Effects on the distribution of minerals 421–431. https://doi.org/10.1080/08927011003699535We used this study as a and proteins between colloidal and soluble phases. Journal of Dairy Research, 73(1), base for our work selecting cleaning agents that are active against microorganisms 91–100. https://doi.org/10.1017/S0022029905001494 because it reported enzymatic treatments that are capable of removing bacterial Ríos-Castillo, A. G., González-Rivas, F., & Rodríguez-Jerez, J. J. (2017). Bactericidal biofilms from samples in the food industry. efficacy of hydrogen peroxide-based disinfectants against gram-positive and gram- Wallhäußer, E., Hussein, M. A., & Becker, T. (2012). Detection methods of fouling in heat negative bacteria on stainless steel surfaces. Journal of Food Science, 82(10), exchangers in the food industry. Food Control, 27, 1–10. https://doi.org/10.1016/j. 2351–2356. https://doi.org/10.1111/1750-3841.13790 foodcont.2012.02.033This review shows that fouling formation on heat transfer Ripolles-Avila, C., Ríos-Castillo, A. G., Fontecha-Umaña, F., & Rodríguez-Jerez, J. J. surfaces is a major problem in the food and dairy industry, limiting processing times (2020). Removal of Salmonella enterica serovar Typhimurium and Cronobacter and increasing plant downtime and economic costs. 8