JBCS

INTRODUCTION

Milk powder is one of the main ingredients with important technological and economic action for the food industry. It is known that milk powder is the main source of fat and non-fat milk solids such as proteins, carbohydrates, and mineral salts for those products. The creaminess, incorporation of air, melt control, and visual appearance of those products are closely related to the free fat content of milk powder, that is the fat that is not fully protected by the native fat globule membrane or covered by amphiphilic molecules and the fat globule size which must have a larger average particle size.^1,2

The kinetic stabilization of milk, which is one of the technological stages in milk powder manufacturing, is based on homogenization where a pressure of 20 MPa is the most common industrially used.³ This process of homogenization consists of the breakdown, size reduction, and strengthening of the encapsulation of the fat globules by the adsorption of proteins (mainly casein micelles) in their interface.⁴ Due to the increase in kinetic stability, there is a reduction in free fat content in milk powder, which is not favorable for some food manufacturing.³

Homogenization is a technique that has currently been used to produce nanoparticulate products, and this technology results in foods with improved emulsion stability, increased digestibility and nutrient absorption, and also helps to make active compounds available that act more efficiently at specific sites in living organisms.^5-7 Furthermore, this unit operation is widely used in the manufacture of powdered milk with a direct impact on the solubility of the product, with homogenization pressure being a central attribute of this process.⁸

In the work developed by Zacaron et al.,⁸ the objective was to understand the effect of increasing homogenization pressures on the particle size distribution during the manufacture of homogenized whole milk powder and on the rehydration of the final product. The fluid milk was heat treated, homogenized, concentrated by rotary evaporation and then dried using a spray dryer. The results found demonstrate that increasing the homogenization pressure decreases the particle size of the reconstituted milk powder, indicating the potential for future studies on how this phenomenon affects its physicochemical properties and the final product.

In addition to homogenization, another process that can be studied in the manufacture of milk powder is the crystallization of lactose. The crystals of lactose rupture the fat globule membrane, resulting in an increase in the free fat content of the final product.^3,9,10

Homogenization and crystallization cause interactions between the various components in milk that influence milk powder functionality. Several authors have studied the effect of the properties of dehydrated dairy products on the characteristics of a variety of food products, including chocolate, ice cream, fermented milk and yogurt, confectionery, baked goods, soups, and sauces. Functional properties of milk powders looking forward specific end uses to some extent, can be induced by modification of processing techniques including homogenization and controlling lactose crystallization.^11-18

Therefore, the aim of this study was to evaluate the influence of different pressures of homogenization during the production of whole milk powder with crystallized lactose on the free fat content and particle size, to develop a milk powder product with favorable properties for application in the food industry.

 

EXPERIMENTAL

Production process of the analyzed samples

For each treatment, 625 g of pasteurized whole milk was heated to ± 35 ºC in a water bath (Q334M-28, Quimis, Diadema, Brazil). Then, 375 g of whole milk powder was added under stirring, totalizing approximately 1 kg of concentrated milk sample with 46% (w w^-1) total solids. All the concentrated milk samples were heated to 50 ºC and then subjected to the homogenization process. The first treatment (T1, the control) was homogenized (NS2006H, GEA Niro Soavi, Søborg, Denmark) without pressure (0 MPa first stage pressure/0 MPa second stage pressure). The second and third treatments (T2 and T3, respectively) were homogenized at 15/5 MPa and 75/5 MPa, respectively. After the homogenization process, the seeding step was performed for the crystallization of lactose, and 0.5% (w w^-1) of microcrystalline lactose was added to each sample. Thereafter, the concentrated milk samples (T1, T2 and T3) were cooled and maintained at 8 ºC, under stirring, for 4 h. After this period, the milk samples were stored at 8 ºC in a refrigerator for 14 h.

Subsequently, the concentrated milk samples were heated to 40 ºC and dried in a mini spray dryer (B-290, Buchi Labortechnik AG, Flawil, Switzerland). In the T3, it was added 10% of the distiller, aiming to enable the atomization, as the sample presented high viscosity. The drying parameters were: airflow (L min^-1): 40 to 50; product flow (L min^-1): 0.45 to 0.95; inlet temperature (ºC): 150 to 160; outlet temperature (ºC): 80 to 90. At the end of the drying process, the powders were collected, packed, and stored in a temperature-controlled (25 ºC) place and protected from light.

Figure 1 schematically summarizes the process flowchart.

 

 

Laboratory analyses performed on the samples

Concentrated milk before drying was analyzed for viscosity, optical microscopy and particle size analysis in solution. Powders were analyzed for pH, total fat and free fat content, particle size in solution, scanning electron microscopy, Raman spectroscopy, zeta potential.

Physicochemical analysis

The viscosity analysis was performed in viscometer (Q860M, Quimis, Diadema, Brazil). For concentrated milk samples of the control and 15/5 MPa treatments, probe 1 was used, while for the concentrated milk samples of the 75/5 MPa treatment, probe 4 was used. The viscosity result (in MPa s^-1) was obtained after reaching about 50% of the motor force. The analysis was performed in duplicate.

After the seeding stage of the concentrated milk, the lactose crystals in the milk were visualized in optical microscope (CX40; Olympus Optical Co., Tokyo, Japan), which has an area of 1.06 × 10¹¹ nm² at the magnification of 0.85 (ocular lens) × 40 (objective lens).¹⁹

The pH of the milk powder samples was obtained by direct reading in the potentiometer (K39-1014B, Kasvi, São José dos Pinhais, Brazil). The powders were reconstituted in water (10% w w^-1 of dry matter) with immersion of the electrode and the potentiometer thermometer in such solution. The analysis was performed in triplicate.

Particle size distribution analysis

The rehydration capacity of the powders was studied by particle size distribution obtained with Beckman Coulter LS 13 320 laser diffraction analyzer (Beckman Coulter, Miami, FL, USA) coupled to the liquid analysis module (aqueous liquid module, Beckman Coulter, Miami, FL, USA). Enough sample to generate turbidity readings was added to the liquid analysis module tank, which contained water at room temperature. Samples were added slowly to prevent the formation of agglomerates. The data were collected in the region of 0.375 to 2000 μm with a time of each collection established in 90 s. Results were calculated with refractive indices of 1.332 for the dispersant (water) and 1.57 for the casein micelles,¹⁵ and 1.47 for the fat globules²⁰ aiming at the observation of total solubility. Data were represented by the percentage of occupied volume by the particles as a function of their size. The analyses were performed over two times (1.5 and 3.0 min) and in duplicate.

Total fat and free fat analysis

The total fat content was determined using the Gerber method, as described by the Adolfo Lutz Institute.²¹ The sample (11 mL of reconstituted milk powder at 12% w w^-1 of dry matter) was added in a butyrometer (Original, São Paulo, Brazil) along with the addition of sulfuric acid (10 mL) and isoamyl alcohol (1 mL). The reading was done on the scale of the butyrometer after centrifugation (ThermoScientific™, Heraeus™, Biofuge™, Stratos™, ThermoFisherScientific, USA) and immersion in a water bath at 65 ºC for 5 min (Q334M-28, Quimis, Diadema, Brazil) with direct reading in butyrometer.

The free fat method is based on the quantification and extraction of free fat through the use of a hydrophobic organic solvent. For that, 10 g of whole milk powder and 50 mL of petroleum ether were added to the beaker under stirring for 5 min. This suspension was filtered then it was added 40 to 50 mL of petroleum ether and allowed to stir again for 5 min. This extraction process was repeated until the maximum free fat recovery was reached. Subsequently, the solvent was distilled off in a rotary evaporator (R-300, Fisher Scientific, Leicestershire, England). After extraction, the sample was submitted to the oven (100 ºC) for 1 h for the complete removal of the solvent. Then it was put in a desiccator for cooling, weighed, and submitted to the oven again for 30 min, with subsequent weighing until it presents constant mass. The free fat was quantified by the gravimetric method (Equation 1) and expressed as a percentage of total fat.²²

where M1 = glass balloon mass + sample; M2 = empty glass balloon mass; GT = total fat. Such analyses were performed in duplicate.

Scanning electron microscopy analysis

The powders were analyzed by scanning electron microscopy (Hitachi TM 3000, Hitachi Ltd., Tokyo, Japan) at magnitude of 250×.

To evaluate the length of the particles obtained from the SEM images, the ImageJ software, version 1.54k,²³ was used.^24,25 For this, the images were analyzed in duplicate at magnitude of 850×. With such an evaluation of the particle length, the SPAN value was obtained. SPAN value corresponds to the width or interval of the size distribution based on the volume, being calculated by Equation 2:

where: Dv90, Dv10 and Dv50 are, respectively, 90, 10 and 50% of the particles with values equal to or less than the result found.

The SPAN value was used to calculate the polydispersivity index (PDI) value. The PDI refers to the degree of nonuniformity of the particle size distribution. A monodisperse sample shows a high degree of uniformity (PDI value less than 0.4). A polydispersed sample has a low degree of uniformity (PDI value greater than 0.4).

Raman spectroscopy analysis

For Raman spectroscopy, it was used a Raman spectrometer (RFS 100, FT-Raman Bruker, Massachusetts, USA) equipped with a Ge detector, using liquid nitrogen as the cooling fluid, and excited at 1064 nm from a Nd:YAG beam. The samples were placed in aluminum capsules, irradiated with a laser pulse with power of 200 mW, and the scattered radiation was collected at 180º. For each spectrum, an average of 512 scans were collected at 4 cm^-1 resolution across 3200 to 200 cm^-1. All spectra were obtained in duplicate.

Zeta potential analysis

The surface charge of the powder particles during the rehydration process was obtained using the Zetasizer equipment (Nano ZS90, Malvern Instruments Ltd, Worcestershire, UK). The powders were diluted (1:25) and the analysis was performed in triplicate. The zeta potential of the powder particles was measured at 23 ± 2 ºC.

Statistical analysis

The R software,²⁶ version 4.3.1, was used for statistical analysis of the data, performing the Tukey test (5% probability level) for the particle size distribution (mean particle size, Dv10, Dv50, Dv90, % volume), zeta potential, pH, SPAN, viscosity. Excel software,²⁷ version 2019, was used to create bar graphs with the respective error bars for particle size and SPAN analyses. For distribution of particle size, statistical analyses were conducted with the equipment software (version 5.03). OPUS software,²⁸ version 6.0, was used to collect the analytical data from the Raman spectroscopy.

 

RESULTS AND DISCUSSION

The average particle size, as well as the number of particles in the region below < 1 μm, were statistically different (p < 0.05) between the control and the 15/5 MPa treatment samples of concentrated milk. However, there was no significant difference by the Tukey test in the particle size distribution of the concentrated milk samples homogenized at 15/5 MPa and 75/5 MPa (Table 1).

 

 

The particle size distribution curve (Figure 2) exposes a population of casein micelles (150-200 nm) and a population corresponding to fat globules and other larger particles of difficult rehydration (1 to 4 μm).²⁹

 

 

With the results obtained from the particle size analysis (Table 1 and Figure 2), it was verified that the control treatment of the concentrated milk had a small number of particles in the region < 1 μm. After homogenization at 15/5 MPa, almost all particles were reduced to diameters smaller than 2 μm. The results for treatment at 75/5 MPa showed the diameter of the particles in the solution decreased further (0.25 μm), shifting to the nanometric region. Thus, it can be seen that the increase in the homogenization pressure led to an improvement in the distribution of the emulsion particles. Nonetheless, as the determination of the particle size distribution was performed soon after the homogenization process, it cannot be said that the emulsion remained stable for longer periods.

Despite the statistical difference between the viscosity of the control treatment and concentrated milk homogenized at 75/5 MPa (Table 1), the largest crystals were found in these samples, which can be confirmed by the higher hydrodynamic value (Dv90). The lower viscosity of the concentrated milk homogenized at 15/5 MPa, although not presenting a statistical difference when compared to the control treatment, was enough to favor the formation of smaller lactose crystals. Mercan et al.³⁰ studied the effect of the high-pressure homogenization (HPH) on the viscosity of concentrated whole milk and observed that, up to a pressure of 15/5 MPa, there was a decrease in product viscosity when compared to concentrated milk without homogenization. At pressures up to 35 MPa, the viscosity increased in proportion to the homogenization pressure level applied. Finally, pressures above 40 MPa resulted in a large viscosity increase, which may be related to changes in the structure and equilibria properties of casein micelles.³¹

Concentrated milk in the control and 75/5 MPa treatments showed the highest SPAN values (more polydisperse), resulting in a lower distribution uniformity of the lactose crystals when compared to the concentrated milk homogenized at 15/5 MPa.

According to the results presented in Table 1, there was no significant difference in particle size distribution between the powders of the treatments T1 and T3 (mean and Dv90). The milk powder homogenized at 15/5 MPa showed the lowest average particle diameter and the lowest Dv90. Regarding the distribution of particles in the region < 1 μm, the milk powder homogenized at 15/5 MPa presented the greatest number of particles in this region, followed by the milk powder homogenized at 75/5 MPa and the milk powder homogenized at 0/0 MPa.

It is known the high homogenization pressure (75/5 MPa) results in particles with smaller diameters.³² Nevertheless, due to the large rupture of fat globules at this pressure, there is not enough protein to cover the surface of the membrane of all the globules. Thus, many fat globules are unprotected, the encapsulation coalesce weakens, and the aggregation of fat globules can increase.^33,34 This may explain the fact that milk powder homogenized at 75/5 MPa presented less number of particles in the region < 1 μm when compared to milk powder homogenized at 15/5 MPa. The study by Mercan et al.²⁹ analyzed concentrated whole milk treated with HPH and had similar results, where was observed reduction of the particle diameter when homogenized at pressures up to 25 MPa; at pressures higher than that, the particle diameter increased.

The obtained milk powders presented different profiles of particle distribution curves, as shown in Figure 2. The control treatment milk powder presented a curve of a dispersion form, with a population of particles in the region of 0.1 to 0.2 μm (region of casein micelles) and another population in the region of 1 to 5 μm (region of fat globules). This powder presented the worst rehydration characteristic compared to the other treatments, confirmed by the lower number of particles in the region < 1 μm and the higher Dv90.

For the milk powder homogenized at 15/5 MPa, a curve with better dispersion was observed, where a large amount of the particles was reduced to diameters smaller than 2 μm, presenting a modal distribution in the region of 1 μm attributed to the homogenization. An increase in the percentage of the volume occupied in the region between 0.2 and 0.3 μm, which may refer to the population of casein micelles, was also observed (Figure 2). Therefore, the rehydration capacity of the milk powder homogenized at 15/5 MPa improved, confirmed by the higher percentage of particles in the region < 1 μm and lower value of Dv90, as shown in Table 1.

In contrast, the milk powder homogenized at 75/5 MPa showed a worsening in the rehydration capacity, when compared to the milk powder homogenized at 15/5 MPa (lower percentage of particles in the region < 1 μm). This can be explained by the weakening of the encapsulation of fat globules and by the high free fat content in the powder, which leads to the increase of the particle size.³ It is inferred that the homogenization at 75/5 MPa caused a great rupture in the fat globules and there was not enough protein to constitute the globule membrane of casein micelles.³³

In the present study, the higher free fat content directly affected the rehydration of the powder, as the milk powders homogenized at 0/0, 15/5, and 75/5 MPa had a free fat content of 19.39 ± 0.16%, 5.03 ± 0.17%, and 13.78 ± 0.14%, respectively. In other words, the powder that presented the best rehydration capacity (T2) was the one with the lowest free fat content. The milk powder that exhibited intermediate rehydration capacity (T3) had an intermediate free fat content, while the milk powder that had the worst rehydration capacity (T1) exhibited the highest free fat content.

Particle size analysis supports the above results, showing a higher content of fat globule particles in the region between 1 and 5 μm for the milk powder homogenized at 0/0 MPa, which may have contributed to the higher free fat content when compared to the other treatments. On the other hand, the milk powder homogenized at 15/5 MPa presented smaller fat globules and greater kinetic stability due to homogenization, resulting in a lower free fat content. Finally, milk powder homogenized at 75/5 MPa showed a higher free fat content than the milk powder homogenized at 15/5 MPa, which can be explained by the weakening of the kinetic stability of the fat globule, because of the lack of phospholipids and proteins to adsorb on the surface of the micelle membrane, thereby forming a more exposed fat interface.^2,3,35 The rehydration of the powders is directly associated with the free fat content, which significantly differed (p < 0.05) among them according to Tukey's test. It is noteworthy that the high amount of free fat obtained in this study can be related to the pre-crystallization stage of lactose.

It is known that a cluster of smaller particles promotes a greater amount of air within them, which allows a better rehydration of the powders. Notwithstanding, some food manufacturing like chocolate, ice cream, and yogurt,^3,16,18,36 milk powder with such added air particles may rupture and raise the viscosity of the product. This occurs by increasing the interfacial area, resulting in the need for greater addition of fat, such as cocoa butter, to reduce such viscosity. This addition of cocoa butter raises the cost of chocolate.¹²

Therefore, a milk powder with desirable characteristics for chocolate, yogurt, and ice cream production should contain lactose crystals (to increase free fat content), absence of homogenization (to increase the free fat content and average particle size), or homogenization at high pressures (to increase the free fat content and average particle size, when compared to the conventional homogenization at 15/5 MPa).

Figure 3 presents the results of the hydrodynamic diameter and SPAN value of the milk powders. The milk powder homogenized at 75/5 MPa presented the lowest SPAN value, followed by the milk powder homogenized at 15/5 MPa and the milk powder homogenized at 0/0 MPa. This suggests that the milk powder with higher homogenization pressure had a more uniform (more polydispersed) particle distribution. Scanning electron microscopy images (Figure 4) confirmed this behavior. That is, the powders homogenized at 0/0 MPa and at 15/5 MPa presented more agglomerates, when compared with the powder homogenized at 75/5 MPa. However, even if the uniformity of the particles is an effect of increasing the homogenization pressure, as for the milk of powder homogenized at 75/5 MPa, the microscopic profile of the same can also be correlated with the drying process by spray drying, since it was necessary to add distilled water to reduce the viscosity and favor the drying process.

 

 

 

 

In Figure 4, it can be observed well-defined bands in the regions at 3000, 1650, 1450, 1350, 1150, 900, and 350 cm^-1, which indicate crystallization of lactose.^33,34 The Raman spectroscopy results confirm that the lactose crystals produced in the concentrated milk remained in the dehydrated products. Since Raman spectroscopy is a non-destructive, rapid, and flow-through analysis tool, the bands indicating the crystallization of lactose can be used as marker for crystallized milk powders directed to the production of food products like chocolate, ice cream, and yogurt.^3,16,18,36

From the results presented in Table 1, it was concluded that there was no significant difference (p > 0.05) in the zeta potential and pH between the different treatments. The milk powders presented a negative zeta potential due to milk proteins, mainly caseins (80% of total bovine milk protein), which are negatively charged at the natural pH of the milk (6.7).³⁷

The analysis of zeta potential is a determinant factor for several studies as it evaluates intermolecular interactions and predicts the stability of colloidal systems. Particles with higher zeta potential, which means a higher surface charge (in modulus), usually promote greater stability to colloidal suspensions. To predict the stability of samples, a difference of at least 10 mV in the zeta potential is required.³⁸ However, as zeta potential for T1, T2, and T3 presented similar modulus values, it is expected emulsions with the three powder milk should present similar colloidal stabilities.

 

CONCLUSIONS

The results of this study evidenced that no homogenization, as well as the crystallization of lactose, gives higher free fat content and achievement of a larger average particle size of milk powder. Despite these results should be undesirable in the production of milk powder for direct consumption, as they impair rehydration, for use in some food industries such as chocolate, ice cream, and yogurt factories are desired properties for use of powder milk as an ingredient. Therefore, for these industries, it can be differentiated manufacture of milk powder to the conventional one, in absence of homogenization and with the addition of microcrystalline lactose so that this variation of milk powder can contribute to the reduction of costs and improvement of product soft. The results obtained in this study show that milk powder with no homogenization (0/0 MPa) was the one that presented the most suitable characteristics for those industries, followed by the one homogenized at 75/5 MPa.

 

ACKNOWLEDGMENTS

The authors gratefully acknowledge the following funding sources: National Council for Scientific and Technological Development (CNPq) with support grants 303569/2022-0 (Luiz Fernando Cappa De Oliveira), 307334/2020-1 (Rodrigo Stephani), 312284/2020-9 (Antonio de Fernandes de Carvalho), 317190/2021-0 (Ítalo Tuler Perrone). Like this, the authors gratefully support grants from the academic master's and doctorate of innovation (403602/2020-3 e MAI-DAI Program, CNPq); Coordination for the Improvement of Higher Education Personnel (CAPES), grant number 001.

 

REFERENCES

1. Holm, G. E.; Greenbank, G. R.; Deysher, E. F.; J. Dairy Sci. 1925, 8, 515. [Crossref]

2. Vignolles, M. L.; Jeantet, R.; Lopez, C.; Schuck, P.; Le Lait 2007, 87, 187. [Crossref]

3. Vignolles, M. L.; Lopez, C.; Madec, M. N.; Ehrhardt, J. J.; Méjean, S.; Schuck, P.; Jeantet, R.; J. Dairy Sci. 2009, 92, 58. [Crossref]

4. Walstra, P.; Wouters, J. T. M.; Geurts, T. J.; Dairy Technology: Principles of Milk Properties and Processes; Taylor & Francis Group: Florida, USA, 2006.

5. Baars, A.; Oosting, A.; Engels, E.; Kegler, D.; Kodde, A.; Schipper, L.; Verkade, H. J.; van der Beek, E. M.; Br. J. Nutr. 2016, 115, 1930. [Crossref]

6. Calvo-Lerma, J.; Fornés-Ferrer, V.; Heredia, A.; Andrés, A.; J.Food Sci. 2018, 83, 2629. [Crossref]

7. Singh, H.; Gallier, S.; Food Hydrocolloids 2017, 68, 81. [Crossref]

8. Zacaron, T. M.; Francisquini, J. d'A.; Perrone, Í. T.; Stephani, R.; J. Dairy Res. 2023, 90, 299. [Crossref]

9. Aguilar, C. A.; Ziegler, G. R.; J. Dairy Sci. 1994, 77, 1198. [Crossref]

10. Fäldt, P.; Bergenståhl, B.; Food Hydrocolloids 1996, 10, 431. [Crossref]

11. Koc, A. B.; Heinemann, P. H.; Ziegler, G. R.; J.Food Sci. 2003, 68, 210. [Crossref]

12. Liang, B.; Hartel, R. W.; J. Dairy Sci. 2004, 87, 20. [Crossref]

13. Franke, K.; Heinzelmann, K.; Int. Dairy J. 2008, 18, 928. [Crossref]

14. Beckett, S. T. In Encyclopedia of Dairy Sciences; Fuquay, J. W., ed.; Elsevier Ltd.: New York, 2011, p. 856.

15. Glicerina, V.; Balestra, F.; Dalla Rosa, M.; Romani, S.; J. Food Eng. 2015, 145, 45. [Crossref]

16. Syed, Q. A.; Anwar, S.; Shukat, R.; Zahoor, T.; Journal of Nutritional Health & Food Engineering 2018, 8, 422. [Crossref]

17. Vásquez, C.; Henríquez, G.; López, J. V.; Penott-Chang, E. K.; Sandoval, A. J.; Müller, A. J.; LWT--Food Sci. Technol. 2019, 111, 744. [Crossref]

18. de Souza, W. F. C.; do Amaral, C. R. S.; Bernardino, P. D. L. S.; International Journal of Gastronomy and Food Science 2021, 24, 100349. [Crossref]

19. Mimouni, A.; Deeth, H. C.; Whittaker, A. K.; Gidley, M. J.; Bhandari, B. R.; Food Hydrocolloids 2009, 23, 1958. [Crossref]

20. Michalski, M.-C.; Briard, V.; Michel, F.; Le Lait 2001, 81, 787. [Crossref]

21. Zenebon, O.; Pascuet, N.; Tiglea, N. S.; Normas Analíticas do Instituto Adolfo Lutz: Métodos Físico-Químicos para Análises de Alimentos, 1ª ed. digital; Instituto Adolfo Lutz: São Paulo, SP, 2008. [Link] accessed in January 2025

22. Schuck, P.; Jeantet, R.; Dolivet, A.; Analytical Methods for Food and Dairy Powders; John Wiley & Sons: Chichester, 2012.

23. ImageJ, version 1.54k; National Institutes of Health, Bethesda, Maryland, 2024.

24. Abramoff, M. D.; Magalhães, P. J.; Ram, S. J.; Biophotonics International 2004, 11, 36. [Link] accessed in January 2025

25. Marcomini, R. F.; de Souza, D. M. P. F.; Ceramica 2011, 57, 100. [Crossref]

26. R Project for Statistical Computing, version 4.3.1; Department of Statistics, University of Auckland, New Zealand, 2023.

27. Excel, version 2019; Microsoft Office, Seattle, Washington, USA, 2019.

28. OPUS, version 6.0; Bruker Optik, Ettlingen, Germany, 2008.

29. Francisquini, J. d'A.; Martins, E.; Renhe, I. R. T.; de Oliveira, L. F. C.; Stephani, R.; Perrone, Í. T.; de Carvalho, A. F.; Quim. Nova 2020, 43, 226. [Crossref]

30. Mercan, E.; Sert, D.; Akın, N.; Int. Dairy J. 2018, 87, 93. [Crossref]

31. Mohan, M. S.; Ye, R.; Harte, F.; Int. Dairy J. 2016, 55, 52. [Crossref]

32. Dodds, J. In Handbook of Food Powders: Processes and Properties; Bhandari, B.; Bansal, N.; Zhang, M.; Schuck, P., eds.; Elsevier Inc., Woodhead Publishing Limited, 2013, p. 309.

33. Hayes, M. G.; Fox, P. F.; Kelly, A. L.; J. Dairy Res. 2005, 72, 25. [Crossref]

34. Martínez-Monteagudo, S. I.; Kamat, S.; Patel, N.; Konuklar, G.; Rangavajla, N.; Balasubramaniam, V. M.; J. Food Eng. 2017, 193, 42. [Crossref]

35. Keogh, M. K.; O'Kennedy, B. T.; Int. Dairy J. 1999, 9, 657. [Crossref]

36. Sharma, A.; Jana, A. H.; Chavan, R. S.; Compr. Rev. Food Sci. Food Saf. 2012, 11, 518. [Crossref]

37. Philippe, M.; Le Graët, T. Y.; Gaucheron, F.; Food Chem. 2005, 90, 673. [Crossref]

38. Roland, I.; Piel, G.; Delattre, L.; Evrard, B.; Int. J. Pharm. 2003, 263, 85. [Crossref]

 

Associate Editor handled this article: Marcela M. Oliveira