Effect of Homogenization Pressure and Supplementation with Sucrose Fatty Acid Ester on the Physical Properties of Dairy Cream-based Emulsions

Dairy cream was obtained from Seoul Dairy Cooperative (Korea). Cream was prepared by concentrating the milk by centrifugal separation. It was pasteurized without any mechanical treatment or adding other ingredients, and the final milk fat and protein contents were 38% and 2%, respectively. The SFAE (DK ESTER-F160) supplied by Dai-Ichi Kogyo Seiyaku Co., Ltd. (Japan) is used as an emulsifier. The fatty acids of the SFAE used in this study were composed of palmitic acid and stearic acid, and the HLB value of SFAE was in the range of 15–16. Sodium azide (Sigma-Aldrich Chemical Co., USA) was also used to inhibit the growth of microorganisms during storage.

The emulsions were comprised of 50% (w/w) dairy cream, 0.02% (w/w) sodium azide (as an antimicrobial agent), distilled water, and SFAE (0.08%, 0.16%, and 0.24% w/w). As it was difficult to dissolve the SFAE in water, distilled water was pre-heated in a water bath at 70°C for 30 min. The dairy cream was also pre-heated at 60°C to minimize protein denaturation. SFAE and sodium azide were added to the pre-heated distilled water and mixed using a stirrer (Eurostar 20 High speed digital stirrer, IKA^®, Germany) at 2,000 rpm for 1 min. Next, the pre-heated dairy cream was added slowly into the aqueous phase with SFAE and sodium azide, and the mixture was allowed to mix at 2,000 rpm for 5 min. Homogenization was carried out using a two-stage valve homogenizer (APV-1000, Invensys APV, Denmark) at two different homogenization pressures (10 MPa and 20 MPa), and 20% of the total pressure was maintained in the second stage valve. Finally, the homogenized emulsions were immediately placed in ice water for 30 min and stored overnight at 5°C. The control sample did not contain any SFAE and was not homogenized.

The droplet size distribution of the dairy cream-based emulsions was determined using a laser light scattering droplet size analyzer (Mastersizer 3000, Malvern Instruments Ltd., UK). The emulsions were added to distilled water until an obscuration rate of 5–15% was achieved with stirring at 1,000 rpm. The absorption coefficient was 0.01, and the refractive indexes of milk fat and water were 1.462 and 1.330, respectively. The D[4,3], D[3,2], Dv10, Dv50, and Dv90 values were used to interpret the droplet size distribution, and calculated using Malvern software (version 3.20, Malvern Instruments Ltd., UK). The D[4,3] value is the volume-weighted mean that is defined as the average diameter calculated on a volume basis, and the D[3,2] value is the surface-weighted mean that is defined as the average diameter calculated on a surface basis. Dv10, Dv50, and Dv90 refer to the average droplet sizes corresponding to the cumulative distributions at 10%, 50%, and 90%, respectively.

The rheological properties of the dairy cream-based emulsions were determined using a rheometer (HAKKE Roto Visco-1, Thermo Fisher Scientific, Germany) with a plate-plate system (35 mm in diameter with a gap of 500 μm). Steady shear rheological properties were determined over a shear rate range of 0.4–100 s^–1. To describe the steady shear rheological properties of the emulsions, the data were fitted to the well-known power law model (Eq. (1)).

where σ is the shear stress (Pa), γ. is the shear rate (s^–1), K is the consistency index (Pa sⁿ), and n is the flow behavior index (dimensionless). Using the magnitudes of K and n obtained from the power law model, the apparent viscosity (η_a) was calculated at 10 s^–1.

Dynamic shear data were obtained from frequency sweeps over a range of angular frequencies (0.63–62.8 rad s^–1) at 2% strain. Haake Rheowin software (version 4.41.0000, Thermo Fisher Scientific, Germany) was used to collect the rheological data and to calculate the storage modulus (G') and loss modulus (G"). The G' value is a measure of elastic response that is recoverable, and the G" value is a measure of viscous response that is lost as viscous dissipation. In order to relax the samples prior to the steady and dynamic shear rheological measurements, all samples were allowed to rest on the plate at 4°C for 5 min. All rheological measurements were performed in triplicate at 4°C.

To measure the stability of the dairy cream-based emulsions, they were transferred to 50 mL conical tubes and stored at two temperatures (5°C and 40°C). Samples (15 mL) were collected from the top and bottom of the emulsions and their stabilities were evaluated by measuring the droplet size and distribution on day 7, day 15, and day 30.

All results are expressed as the mean±standard deviation. Statistical analysis was performed using one-way ANOVA followed by Duncan’s test with IBM SPSS Statistics 24 (IBM Software, USA). A value of p<0.05 was considered significant.

The effect of homogenization pressure and SFAE concentration on the droplet size distribution of dairy cream-based emulsions is shown in Fig. 1 and Table 1. The D[3,2] and D[4,3] values of the control were 2.32 µm and 3.32 µm, respectively. In contrast, all homogenized emulsions had lower D[3,2] values (0.84–1.15 µm) and D[4,3] values (1.02–1.49 µm). Furthermore, as shown in Fig. 1, all homogenized emulsions had a smaller droplet size and narrower droplet distribution than those of the control. These results were found to be in good agreement with those of previous studies that investigated the effect of homogenization on these parameters (Lee et al., 2009; Heffernan et al., 2009). Our results could be explained by the disruptive forces occurred during homogenization. The intense turbulence and shearing forces, which were generated when the coarse emulsions passed through the interaction chamber in the homogenizer, led to the breaking up of larger droplets into smaller droplets (Floury et al., 2000; Long et al., 2012).

Fig. 1. Effect of homogenization pressure on the droplet size distribution of dairy cream-based emulsions with different sucrose fatty acid ester (SFAE) concentrations. (A) 10 MPa, (B) 20 MPa. Control (non-homogenized and without SFAE).

Download Original Figure

The D[4,3] value demonstrated that the droplet size of the emulsions homogenized at 20 MPa was smaller (1.02–1.28 µm) than that of the emulsions homogenized at 10 MPa (1.35–1.49 µm). However, the emulsion homogenized with 0.08% (w/w) SFAE at 20 MPa showed a wide distribution with a low Dv10 value of 0.53 µm and a high Dv90 value of 2.31 µm. In this case, an asymmetrical droplet size distribution with a shift towards larger droplets was also observed (Fig. 1B), which indicated that some of the milk fat droplets that broke during homogenization had re-flocculated. A high homogenization pressure decreased the droplet size and increased the newly formed surface area. The homogenized small droplets were rapidly re-flocculated or protected from aggregation by absorbing proteins and emulsifiers on the newly formed surface area (Jafari et al., 2004). Consequently, as the homogenization pressure increased and the droplet size decreased, the newly formed surface area became larger and required more proteins and emulsifiers to be absorbed on the droplet surface (Lee et al., 2009; Heffernan et al., 2009). Therefore, in the case of the emulsion homogenized at a high pressure (20 MPa) with low SFAE concentration (0.08% w/w), the partial flocculation of milk fat could be attributed to the lack of proteins and SFAE, which completely covered the large surface area of the newly formed milk fat droplets.

As the concentration of SFAE was increased, the D[4,3] and D[3,2] values decreased from 1.49 to 1.02 µm and from 1.15 to 0.84 µm, respectively (Table 1). This could be explained by the competition between the disruption and formation of fat droplets generated during homogenization. If the timescale of collision between the droplets was longer than the timescale of the adsorption of the emulsifier to the droplet surface, fat droplets would be re-flocculated and larger droplets will be formed (Jafari et al., 2004). Small-molecule emulsifiers such as SFAE would also be quickly absorbed at the newly formed interface, leading to further disruptions by reducing the interfacial tension (Leong et al., 2011; Pichot et al., 2010). Therefore, the addition of SFAE inhibited the re-flocculation of milk fat droplets in dairy cream-based emulsions during homogenization and thus prevented the phase separation of the initial emulsion.

The shear stress (σ) versus shear rate (γ γ.) data for dairy cream-based emulsions with different SFAE concentrations (0.08%, 0.16%, and 0.24% w/w) and homogenization pressures (10 MPa and 20 MPa) are shown in Fig. 2 and Table 2. Experimental data of σ and γ γ. were well-fitted to the power law model with high determination coefficients (r²=0.96–0.98) (Table 2). All emulsions exhibited a high shear-thinning behavior with flow behavior index (n) values that were lower than 1 (n=0.21–0.46). These results were consistent with those of previous studies (Leong et al., 2011; Long et al., 2012; Zhao et al., 2014). At low shear rates, the shear force was insufficient to deform the flocculated droplets with a fixed size and shape, resulting in high viscosity. However, at high shear rates, the shear force proved to be sufficient for the deformation and breaking up of the flocculated droplets, resulting in low viscosity (Derkach, 2009; Floury et al., 2000). Therefore, a decrease in the homogenization pressure and SFAE concentration led to a lower n value because of the re-flocculation of milk fat (Table 2). Consequently, the shear-thinning behavior of emulsions could be explained by the structural breakdown of flocculated droplets. The breakup of the droplets during shear could have a significant effect on the flow behavior of emulsions (Long et al., 2012).

Fig. 2. Shear stress-shear rate plots for dairy cream-based emulsions with different sucrose fatty acid ester (SFAE) concentrations and homogenization pressures. ○: Control (non-homogenized and without SFAE), ■: 10 MPa - 0.08%, ▲: 10 MPa - 0.16%, ♦: 10 MPa - 0.24%, □: 20 MPa - 0.08%, □: 20 MPa - 0.16%, ◊: 20 MPa - 0.24%.

Download Original Figure

The η_a,10 values (0.10–0.21 Pa s) of the emulsions homogenized with SFAE were lower than that of the control (0.23 Pa s), and they decreased with increasing SFAE concentration and homogenization pressure. All emulsions had a lower K (0.35–1.06 Pa sⁿ) than that of the control (1.16 Pa sⁿ), except for the emulsion homogenized with 0.08% (w/w) SFAE at 10 MPa (K=1.28 Pa sⁿ). At the same SFAE concentration, emulsions with a low homogenization pressure (10 MPa) had higher η_a,10 and K values than those of the emulsions with a high homogenization pressure (20 MPa). This result could be attributed to the lack of proteins and SFAE on the droplet surface. At a low homogenization pressure, the fat droplets are large and the surface of the newly formed droplets is small. Consequently, the relative concentrations of the aqueous proteins and SFAE increased at a low homogenization pressure and these emulsifiers formed micelles that led to an increase in the viscosity (Granger et al., 2005; Zhao et al., 2014).

Fig. 3 shows the changes in the storage modulus (G') and loss modulus (G") as a function of frequency (ω) for the dairy cream-based emulsions with different homogenization pressures (10 MPa and 20 MPa) and SFAE concentrations (0.08%, 0.16%, and 0.24% w/w). The G' and G" values increased with increasing ω, except for the G" of the emulsion homogenized with 0.08% (w/w) SFAE at 10 MPa. As the ω increased, the G' values increased more sharply than the G" values, indicating an increase in the elastic properties at a high frequency. The G' and G" values of the control were higher than those of all emulsions homogenized with SFAE. This is in good agreement with the results of Hussain et al. (2017); these authors found that the dynamic moduli (G' and G") values of the commercially available creams stabilized by sodium caseinate and Tween 80 decreased with the increasing homogenization pressure because of the differences in the coating layers of milk fat droplets. According to Derkach (2009), the rheological properties of the emulsions are affected by the surface properties of the fat droplets. During homogenization, the surface of the milk fat droplets is covered by proteins such as casein, and the adsorbed protein can form a casein gel matrix, thus resisting against deformation (Hussain et al., 2017; Murray, 2002). Accordingly, the decrease in G' and G" for emulsions homogenized with SFAE can be attributed to these emulsifiers being absorbed on the milk fat surface, which could result in changes in the properties of the interfacial surface. From these observations, it was concluded that SFAE concentration and homogenization pressure affect the rheological properties of dairy cream-based emulsions.

Fig. 3. Plots of log G' (storage modulus) and G" (loss modulus) versus log ω of dairy cream-based emulsions with different sucrose fatty acid ester (SFAE) concentrations and homogenization pressures. ○: Control (non-homogenized and without SFAE), ■: 10 MPa - 0.08%, ▲: 10 MPa - 0.16%, ♦: 10 MPa - 0.24%, □: 20 MPa - 0.08%, □: 20 MPa - 0.16%, ◊: 20 MPa - 0.24%.

Download Original Figure

The D[4,3] value calculated from the volume distribution is more suitable than the D[3,2] value for representing the average droplet size with a higher volume and flocculation of the fat droplets (Ariyaprakai et al., 2013). Therefore, the D[4,3] value was used to examine the physical stability of the dairy cream-based emulsions during storage at different temperatures (5°C and 40°C). The D[4,3] values of the top and bottom emulsions with different homogenization pressures (10 MPa and 20 MPa) and SFAE concentrations (0.08%, 0.16%, and 0.24% w/w) are shown in Table 3. In the control, phase separation was observed after storage at 40°C for 7 d. In the case of the control sample stored at 5°C, the droplet size of the top emulsion (10.5 µm) was larger than that of the bottom emulsion (3.16 µm) on day 7, and phase separation was observed on day 15. It has been previously shown that thermodynamically unstable emulsions are easily flocculated by droplet-droplet interactions during storage (Cheng et al., 2016). Therefore, the phase separation of the control could be attributed to the flocculation of milk fat droplets during storage, owing to the unstable emulsions that were non-homogenized or without additives.

The dairy cream-based emulsions homogenized with SFAE did not show any significant change when stored at 5°C for 30 d. However, in the case of storage at 40°C, the droplet size of the top emulsion became larger than that of the bottom emulsion over the storage period (Table 3). This result implied that the flocculation of milk fat droplets had occurred in the emulsions. According to Stokes’s law, the creaming rate increases with an increase in the diameter of the droplets (Long et al., 2012). Flocculated fat could float upward during storage owing to its low density, resulting in a difference in the droplet size between the top and bottom emulsions (McCrae et al., 1999). The droplet size distribution data for the top of the dairy cream-based emulsions stored at 40°C for 30 d are shown in Fig. 4. All emulsions had a secondary peak in the larger size region during storage at 40°C for 30 d, indicating the formation of large droplets. The emulsion homogenized with 0.08% (w/w) SFAE at 20 MPa had a peak with a large tail in the larger size region after storage for 7 d. This peak was divided into two on day 15, and phase separation was observed on day 30. This could be explained by considering the formation of unstable emulsions with a low concentration of SFAE, which were unable to completely cover the newly formed droplet surface during homogenization at high pressures (Floury et al., 2000; Heffernan et al., 2009). Therefore, the milk fat droplets of emulsions stored at 40°C flocculated more rapidly than those of emulsions stored at 5°C, resulting in phase separation.

Fig. 4. Effect of storage time on the droplet size distribution and stability of dairy cream-based emulsions stored at 40°C. (A) 10 MPa - 0.08%, (B) 20 MPa - 0.08%, (C) 10 MPa - 0.16%, (D) 20 MPa - 0.16%, (E) 10 MPa - 0.24%, (F) 20 MPa - 0.24%.

Download Original Figure

When stored at 40°C for 30 d, the droplet size of the top emulsion increased with the increasing SFAE concentration at the same homogenization pressure, indicating that with a higher SFAE concentration, the emulsion would be more unstable at high temperatures. These findings could be attributed to the protein-surfactant interactions, which affected the stability of the emulsions because of competitive adsorption and displacement (Rouimi et al., 2005). In the presence of an emulsifier, the interfacial strength was reduced as a result of competitive adsorption with proteins, and consequently, the proteins initially adsorbed at the fat droplet interface were released into the aqueous phase (Granger et al., 2005). This produced a thinner and more fragile membrane on the fat droplets, and the emulsion became more susceptible to partial flocculation. Similarly, in the case of emulsions stabilized by milk protein and SFAE, it was found that SFAE weakened the interfacial layer of fat droplets in the emulsions by displacing the proteins (Cheng et al., 2016; Tual et al., 2006). Therefore, the increase in droplet size with an increase in SFAE concentration was attributed to the SFAE adsorbed on the fat droplet, which contributed to the instability of the emulsions at a high temperature (40°C) because of fragile membranes.