Original papers
Effect of Vegetable Milks on the Physical and Rheological Properties of Ice Cream
2014 Volume 20 Issue 5 Pages 987-996
Details
2014 Volume 20 Issue 5 Pages 987-996
The nutritional properties and health benefits of ice cream can be improved by substituting cow's milk with vegetable milks. In the present study cow's milk in ice cream was replaced by soy, coconut and composite milk (combinations of coconut or cow milks with soy milk). The changes in ice cream eating qualities and physical properties (melting rate, apparent viscosity, hysteresis, fat globule size and its zeta potential and freezing behavior) were evaluated. The use of vegetable milk to replace cow milk increased pH and decreased melting rate. Ice creams containing composite milk have reduced the melting rate, freezable water amount, particle size and total acceptability of ice creams whereas increased viscosity and hysteresis area with increasing soy milk content. The vegetable milks as composite milk can use to replace cow milk without markedly affecting the quality of ice cream.
The demand for alternatives to cow's milk in ice cream making is growing due to problems associated with its fat, cholesterol and lactose (allergenicity) contents and increasing desire for vegetable milk based ice cream. Soy milk is regarded as a suitable choice (Abdullah et al., 2003) because of its high nutritional quality especially with respect to protein content and balance amino acids (Gandhi et al., 2001). Frequent consumption of soy products offers health benefit such as lowering the risk of getting cancers, diseases associated with cardiovascular, hypercholesterolemia, diabetes, bone and kidney (Dervisoglu et al., 2005). Coconut (Cocos nucifera) milk is another vegetable milk that may be used to replace cow milk. This milk is easy to digest and it contains an abundance of minerals (particularly calcium, phosphorus and potassium) and vitamins (B, C and E vitamins) and antioxidant activities. The high oleic and lauric acid content in coconut milk help in preventing arteriosclerosis and related illness. Coconut milk is commonly used by confectioners, bakeries, biscuits and ice cream industries to enhance flavor and taste of various products (Belewu and Belewu, 2007).
Wide use of these two vegetable milks in food preparation can be associated with the vegetable proteins favorable effects on improving physical properties of foods. For instance, the viscosity, melting time and hardness of ice cream samples increased by substituting of skimmed milk powder with soy protein isolate (Akesowan, 2009). The fortification of yogurt ice cream with soy protein improve the texture, firmness and viscosity of the product (Mahdian et al., 2012). Samoto et al. (2007) noted soy lecithin acts as emulsifier and also helps increase the viscosity, stability, texture and extends the melting time of the ice cream. Abdullah et al. (2003) improved the quality of ice cream by using different ratios of skim milk in soy milk blend and found that large quantity of skim milk with soy milk reduces the beany flavor of soy beans and increased the quality of ice cream. Limited numbers of studies have been carried out on the use of coconut milk to replace cow milk in ice cream. In order to attempt this, Kerdchouaym and Surapat (2008) improved the physical properties of low fat coconut milk ice cream by replacing skim milk powder with whey protein concentrate and it has increased ice cream mix viscosity and reduced melting rate of ice cream. Vegetable milks has also reported to improve nutritional content of ice cream. For example, Bisla et al. (2011) showed increased in expressed protein, fat, iron, vitamin C values and overall acceptability of the ice creams in comparison to standard cow's milk ice cream by the using soy milk, watermelon seeds milk and guava pulp.
The main challenge in using coconut or soy milk in ice cream is to obtain a stable colloidal system. For example lecithin in the soy milk is responsible for the formation of hard ice cream resulting in the requirement of about 15 minutes of standing at room temperature to soften before serving (Wangcharoen, 2012). It is important to establish the extent of improvement in the physical properties of ice cream as a result of adding coconut or soy milks. Thus, the aims of this study were to determine the chemical, physical and sensory properties of the ice creams when cow's milk is partially replaced by coconut or soy milk.
Materials Fresh cow milk, coconut, soybean, soy oil, butter and skim milk powder (Dutch lady, Malaysia), sugar and vanilla were purchased from local grocery. Cremodan SE 734 veg (Danisco AS, Copenhagen, Denmark, containing mono- and diacyl-glycerols of fatty acid, cellulose gum, guar gum, carrageenan) was used as stabilizers.
Preparation of soy milk Soybeans (10 g) were washed three times using tap water, one time rinsing using de-ionized water, followed by soaking in de-ionized water (1 L) for 14 h at room temperature. Excess water was then drained off and the shells were removed. The swollen beans were blended with 250 mL of boiling water in a laboratory blender (Waring, New Hartford, CT, USA) at low speed followed by boiling for 5 min. The blended soybean was then passed through 4 layers of cheesecloth. The soy milk fat content (1.86%) was corrected to 3.4% using 1.54 g soy oil/100 g soy milk. The soy milk was reheated to 80°C for 10 min and immediately chilling (4°C) prior to making ice cream.
Preparation of coconut milk The brown hard coconut shell was cracked open and the white copra was grated followed by mechanical pressing to obtain the milk. To achieve 8% fat coconut milk, 300 g of fresh coconut milk (after sieving with double layers of cheesecloth) was mixed with 700 g of distilled water. The diluted coconut milk was heated at 80°C for 10 min prior to chilling (4°C) and was used within 1 h.
Preparation of ice cream Ice cream was prepared by using various combinations of coconut or cow milks with soy milk. To achieve ice creams with 43% total solids and 10.5% fat for a total batch of 100 g, ice cream mixes were prepared using formula as shown in Table 1. The milk or milk combinations with butter were heated to 50°C prior to mixing with the skim milk powder, sugar and water. This was followed by two stages homogenization (16000 rpm, 70°C, 5 min; Ika Homogenizer T-25 basic Ultra Turrax, Germany). The mixtures were pasteurized (80°C for 10 min in a water bath) followed by cooling to 4°C and aging overnight at 4°C. The ice creams were then frozen in a 1.5 L batch ice cream maker (Baumatic gelato1ss, UK) and packed in 100 mL pre-sterilized plastic cups. The cups were then covered using the lids prior to freezing at −20°C. Three separate batches of ice cream were prepared for each treatment.
| SampleA | Ingredient | |||||||
|---|---|---|---|---|---|---|---|---|
| Milk formula (%) | Butter (%) (Ft = 83.3%) | Skim milk powder (%) | Sugar (%) | Stabilizer-Emulsifier (%) | Vanillin (%) | Water (%) | ||
| W | 55.4 | 10.37 | 7 | 17 | 0.6 | 0.1 | 9.62 | |
| C | 55.4 | 7.31 | 7 | 17 | 0.6 | 0.1 | 9.62 | |
| S | 55.4 | 10.37 | 7 | 17 | 0.6 | 0.1 | 9.62 | |
| SW1 | 55.4 | 10.37 | 7 | 17 | 0.6 | 0.1 | 9.62 | |
| SW2 | 55.4 | 10.37 | 7 | 17 | 0.6 | 0.1 | 9.62 | |
| SW3 | 55.4 | 10.37 | 7 | 17 | 0.6 | 0.1 | 9.62 | |
| SC1 | 55.4 | 9.6 | 7 | 17 | 0.6 | 0.1 | 9.62 | |
| SC2 | 55.4 | 8.84 | 7 | 17 | 0.6 | 0.1 | 9.62 | |
| SC3 | 55.4 | 8.08 | 7 | 17 | 0.6 | 0.1 | 9.62 | |
Chemical analysis The pH of ice cream were measured using digital pH meter whereas titratable acid (TA) was determined by titrating samples (10 g) with sodium hydroxide (0.1 N), using phenolphthalein as an indicator. The total solids were measured by drying samples at 100 ± 1°C for 3.5 h using an air oven (Akin et al., 2007). Fat content was calculated by weight after alkaline hydrolysis coupled with soxhlent extraction (petroleum ether) (AOAC, 2005). All measurements were performed three times.
Melting rate The ice cream melting rate was determined as described by Mahdian et al. (2012). Tempered ice cream samples (spherical shape, −20°C, 30 g) were prepared by scraping the surface of ice cream using a stainless steel table spoon and these were placed on a 0.2 cm wire mesh screen above a beaker at room temperature (25°C). The weight of the melted material was measured after 20 min and declared as percentage weight melted.
Rheological measurements Rheological measurements of melted ice cream samples were determined using a Physica MCR 301 rheometer (Anton-Paar GmbH, Graz, Austria) with a concentric cylinder geometry coupled with a circulating cooling bath at 4.0 ± 0.1°C (Rossa et al., 2012). Melted ice creams (around 20 g) were left to equilibrate at 4.0°C for 15 min. The samples flow behavior was generated by linearly increasing the shear rate from 19.6 to 67.3 s−1 in 20 min followed by returning to 19.6 s−1 over a further 20 min.
The hysteresis of ice creams was evaluated by calculating the area between the shear stress/shear rate curves.
The consistency index and the flow behavior were explained by the Power Law model (Eq.1). Apparent viscosity of ice creams was estimated as a function of time under a constant shear rate of 20 s−1.
 |
Where: σ is the shear stress (Pa); K = consistency index (Pa sn); γ = the shear rate (s−1); and n = the flow behavior index (Rossa et al., 2012).
Size and zeta potential The average particle size and zeta potential of fat globules of ice cream mixes were determined by using Zetasizer ZS (Malvern Instruments, UK) at a constant temperature of 25°C. Measurements were carried out with the dilution of the ice cream mixes approximately 1:10000 with deionized water. The zeta potential and size of ice cream mixes were monitored after aging step (Tan and Misran, 2012).
Optical polarizing microscope imaging (OPM) The OPM micrographs of the ice cream mixes (after aging step) were obtained using a Leica Polarizing Microscope equipped with a Leica QWin software. All measurements were carried out at room temperature (25°C) (Tan and Misran, 2012).
Differential scanning calorimetry (DSC) of ice cream The thermal properties of ice cream mixes (after aging step) were measured by a differential scanning calorimeter (DSC) by Mettler Toledo (model DSC822e) according to the method reported by Hwang et al. (2009). Sample of ice cream mixes (about 5 mg) was placed in a pre-weighed aluminum sample pan and the pan was sealed using a Quick Press pan crimper (Perkin Elmer) and the thermal data were recorded from −30°C to +30°C in nitrogen atmosphere with a heating rate of 5°C min−1. An empty pan served as the reference. The flow rates of nitrogen gas for cooling and heating were 110 and 40 cc/min, respectively.
The onset temperatures (T0), peak temperatures (Tp), freezing points (Tf) and enthalpies (ΔHf) of the transitions of ice formation and ice melting were recorded. The onset temperatures are considered as the intersection of the tangent and base line to the left side of the melting peak. Freezing points were calculated by determining the temperature at which the steepest slope was observed (the temperature at maximum slope of the endotherm or the extra-plotted peak onset temperature (T0) of the ice melting (point Tf in Fig.1; Rahman, 2008)). The enthalpy of the phase transition (ΔHf = enthalpy of fusion) was determined by extrapolating the baseline under the peak by connecting the flat baseline before and after the melting peak and integrating the peak above the baseline, as indicated in Fig. 1. The amount of ice formed per gram of sample (freezable water) was determined by integrating the melting curves and dividing the melting enthalpy with the pure ice fusion latent heat (S = 334 J g−1) (Soukoulis et al., 2009).
A typical DSC thermogram to determine the freezing point and ΔHf of ice cream.
Sensory analysis The ice creams were organoleptically evaluated by sixteen untrained panelists (25–30 year; 8 males, 8 females), using a sensory rating scale of 1–10 for taste and flavor, and 1–5 for consistency and 1–5 for appearance and color (Akin et al., 2007). The properties evaluated contained (a) three characteristics for appearance and color (no criticism: 5, dull color: 4-1, unnatural color: 3-1), (b) seven properties for taste and flavor (no criticism: 10, cooked flavor: 9-7, lack of sweetness and too sweet: 9-7, lack of flavor: 8-6, rancid and oxidized:6-1, and other: 5-1) and (c) seven terms describing texture and body (no criticism: 5, coarse: 4-1, crumbly: 4-2, weak: 4-1, fluffy: 3-1, gummy: 4-1, sandy: 2-1).
Statistics The experiments were assayed in triplicates and the results were expressed as mean ± S.E.M (standard mean error) values. The statistical analysis was carried out using SAS statistical software, version 6.12 edition (SAS, 1996) followed by Duncan's multiple range method for mean comparison. The criterion for statistical significance was p < 0.05 (Homayouni et al., 2008).
Compositions and physicochemical properties of ice cream The compositions and physicochemical properties of the ice creams are presented in Table 2. The total solid, fat and titratable acidity (TA) were unchanged by partial replacement of cow milk with soy or coconut milks. However, pH and melting rate changed with milk replacement. The pH was found to be the highest in ice creams with C and SC3 ice creams and the lowest in W ice cream. Ice creams showed different melting behavior as a function of milk replacement. While the content of butter used to balance to fat (10.5%) were less in coconut ice cream (7.31 g vs. 10.37 g for ice creams containing cow and ice cream with 100% soy milk), this is regarded to have minor effect on melting behavior. Hyvoen et al. (2003) for instance reported that different types of fat (dairy and vegetable fats) had no significant effect on perceived melting of ice creams, although fat amount did affect melting rate of ice creams. All vegetables and composite milk ice creams (16.27 – 33.36%) showed a slower melting rate than W ice cream (35.88%). The melting rate decreased with increasing soy milk content in ice creams containing composite milk and this presumably can be explained that soy milk proteins is more hydrated and therefore prevent their free movement of water molecules associated with proteins (Akesowan, 2009) which lead to reduced syneresis and increased viscosity (Table 3). The relationship between the increase in viscosity and increase in the resistance of ice cream to melting rate also reported by Kaya and Tekin (2001), Akesowan (2009) and Hermanto and Masdiana (2011). In addition the emulsifying properties of soy lecithin which provides protection for the membrane proteins against damage due to freezing (Samoto et al., 2007) and assists good air distribution and fat structure in the ice cream, also can affect on the increase of the time of melting of the ice cream (Hermanto and Masdiana, 2011).
| SamplesA | Composition | Physico-chemical properties | |||
|---|---|---|---|---|---|
| Total solids (g/100g)B | Fat (g/100g)B | pH (value)B | Titratable acidity (%lactic acid)B | Melting rate (% ice cream melted after 15 min)B | |
| W | 43.91 ± 0.08a | 10.50 ± 0.04a | 6.80 ± 0.01f | 0.158 ± 0.006a | 35.88 ± 10.16a |
| C | 43.16 ± 0.07a | 10.40 ± 0.05a | 7.38 ± 0.01a | 0.164 ± 0.004a | 27.00 ± 4.16bc |
| S | 43.94 ± 0.08a | 10.50 ± 0.02a | 6.93 ± 0.01e | 0.160 ± 0.003a | 16.27 ± 7.00f |
| SW1 | 43.23 ± 0.15a | 10.40 ± 0.04a | 7.04 ± 0.02d | 0.161 ± 0.006a | 22.25 ± 5.50d |
| SW2 | 43.42 ± 0.17a | 10.30 ± 0.05a | 7.08 ± 0.01d | 0.160 ± 0.004a | 30.20 ± 6.70b |
| SW3 | 43.66 ± 0.15a | 10.50 ± 0.02a | 7.14 ± 0.014c | 0.160 ± 0.003a | 33.36 ± 11.10ab |
| SC1 | 43.62 ± 0.10a | 10.30 ± 0.02a | 7.12 ± 0.03c | 0.162 ± 0.009a | 18.11 ± 8.90e |
| SC2 | 42.79 ± 0.12a | 10.50 ± 0.01a | 7.22 ± 0.01b | 0.162 ± 0.008a | 23.50 ± 7.50cd |
| SC3 | 43.21 ± 0.11a | 0.40 ± 0.01a | 7.35 ± 0.01a | 0.160 ± 0.005a | 26.50 ± 10.10c |
| SamplesA | Apparent viscosity (mPa s)b | K (Pa sn)b | nb | R2c |
|---|---|---|---|---|
| upward curves | ||||
| W | 289 ± 0.806h | 0.87 ± 0.01g | 0.65 ± 0.01a | 0.994 |
| 363 ± 1.16g | 1.29 ± 0.01f | 0.56 ± 0.01a | 0.996 | |
| S | 1120 ± 1.06a | 4.67 ± 0.01b | 0.51 ± 0.01a | 0.996 |
| SW1 | 818 ± 1.20c | 3.10 ± 0.01c | 0.55 ± 0.01a | 0.999 |
| SW2 | 488 ± 2.01f | 1.30 ± 0.03f | 0.68 ± 0.01a | 0.998 |
| SW3 | 398 ± 1.01g | 1.18 ± 0.02f | 0.63 ± 0.01a | 0.997 |
| SC1 | 982 ± 1.30b | 4.81 ± 0.01a | 0.47 ± 0.01a | 0.999 |
| SC2 | 739 ± 0.91d | 2.89 ± 0.03d | 0.55 ± 0.01a | 0.999 |
| SC3 | 603 ± 1.80e | 2.17 ± 0.02e | 0.59 ± 0.01a | 0.993 |
| Downward curves | ||||
| W | 287 ± 1.07h | 0.71 ± 0.02f | 0.69 ± 0.01a | 0.997 |
| C | 294 ± 1.16h | 0.76 ± 0.01f | 0.68 ± 0.01a | 0.998 |
| S | 1012 ± 0.91a | 3.61 ± 0.01a | 0.57 ± 0.01a | 0.997 |
| SW1 | 784 ± 1.11c | 2.66 ± 0.01b | 0.58 ± 0.01a | 0.996 |
| SW2 | 536 ± 0.87f | 1.83 ± 0.03c | 0.58 ± 0.01a | 0.997 |
| SW3 | 391 ± 0.96g | 1.22 ± 0.01e | 0.62 ± 0.01a | 0.998 |
| SC1 | 817 ± 1.09b | 2.43 ± 0.01b | 0.63 ± 0.01a | 0.995 |
| SC2 | 667 ± 1.03d | 1.87 ± 0.02c | 0.65 ± 0.01a | 0.996 |
| SC3 | 577 ± 2.06e | 1.62 ± 0.01d | 0.647 ± 0.01a | 0.996 |
Ice creams containing cow milk had a higher melting rate than ice creams containing coconut milk. Melting rate can be influenced by the differences in freezing points and viscosity of recipes (Salem et al., 2005). However, in the present study no differences (p > 0.05) were observed in the freezing points amongst ice creams containing composite milk (Table 6). However they have noticeable differences in freezable water and the enthalpy of fusion (Table 6) due to their kind of proteins and their hydration tendency (Alvarez et al., 2005) which affect on serum concentration and freezable water in the ice creams, hence their fusion enthalpies. Ice crystallisation is strongly dependent on the extent of freezing point and the percentage of bound water (unfrozen water) (Soukoulis et al., 2009). Whey protein and casein isolates have a higher amount of aspartic and glutamic acids (negative charge) than coconut protein, as well as a higher proportion of lysine and arginine (positive charge). The value of zeta potential is higher in whey protein than in coconut protein whereas the surface activity is said to be higher in whey protein than in coconut protein (Onsaard et al., 2006). The coconut proteins are generally known for having poor solubility in water (Tangsuphoom, 2008), therefor it contributes to increase in the percentage of unbound water (freezable water) in ice creams. Therefore, the freezable water amount was not the factor for the reduction in melting rate of ice creams containing composite milk, because melting rate increased with increasing freezable water (Hwang et al., 2009).
Another effective factor on melting rate of ice creams is their differences in apparent viscosity. Ice creams containing coconut milk had higher melting rate because they had higher apparent viscosity than ice creams containing cow milk. On the other hand, our results show the major contribution to the difference in melting rate can be attributed to the differences in apparent viscosity of the ice creams. This is because, ice creams containing higher amount of soy milk has the lowest melting rate, and highest apparent viscosity.
Rheological measurements The apparent viscosity, flow behavior index and consistency index of the melted ice creams made with different milk are shown in Table 3. W and C (289 and 363 mPa s respectively which are ice creams without soy milk) melted ice creams had lower apparent viscosity than those containing soy milk. The highest apparent viscosity was in S ice cream (1120 mPa s), followed by SC1 and SW1 (982 and 818 mPa s respectively; Table 3). This could be explained by soy protein properties which able to provide several functionalities such as water holding and emulsifying properties (Akesowan, 2009). Hence soy proteins form a stable network like a gel structure which create greater resistance to flow (Batista et al., 2005). This is agreement to previous studies which showed grape wine less (Hwang et al., 2009) and inulin (Pinto et al., 2012) water retention effects and subsequent increase apparent viscosity of ice cream.
Melted ice creams containing coconut milk had a higher apparent viscosity than ice creams containing cow milk. This could be due to the higher particle size of ice creams containing coconut milk because coconut proteins have poor emulsifying properties (Tangsuphoom and Coupland, 2009). This is in contrast to the effects of the milk protein concentrates in ice cream which increase the viscosities due to the increased voluminosity of the dispersed particles as described by the Eilers equation (Alvarez et al., 2005).
The ice creams rheological behavior after the reduction of shear rate (downward curves) is as shown in Table 3. All ice creams demonstrated non-Newtonian behavior, i.e. their viscosity decreases with increasing shear rate (Fig. 2). The viscosity reduction is known to be dependent on the aggregation of fat globules which decrease in size during shearing (Rossa et al., 2012). The K (consistency index) varied from 0.87 to 4.81 Pa s−1 (Table 4). SC1, S and SW1 ice creams had the highest consistency indexes.
Effect of shear rate on the apparent viscosity of ice creams.
| Samples | Hysteresis (Pa)a |
|---|---|
| W | 23.93 ± 0.96f |
| C | 36.19 ± 1.14e |
| S | 45.69 ± 2.03d |
| SW1 | 45.20 ± 1.51d |
| SW2 | 28.69 ± 1.30f |
| SW3 | 2.70 ± 1.81g |
| SC1 | 100.41 ± 1.42a |
| SC2 | 60.00 ± 1.61b |
| SC3 | 55.33 ± 1.59c |
The highest K values were related to ice creams containing soy milk and also increased with increasing soy milk content which again demonstrates that the addition of soy milk increased the resistance to structural breakdown due to aggregation of soy proteins which resulted in gel formation and subsequent increase in water retention (Zayas, 1997).
The flow behavior index (n), which reflects the degree of pseudoplasticity of a fluid, ranged from 0.47 to 0.68, but the differences were not significant (p > 0.05).
The n values of upward curve were lesser than those of the downward curve, indicating a decrease in the pseudoplastic properties as the shear rate decreased. The decrease in K and increase in n can be ascribed to the structural rupture of the protein network of the ice cream because of shearing, which favors this behavior (Rossa et al., 2012).
The formation of hysteresis (Table 4) is an important feature of the shear stress versus shear rate results. The fluid viscosity (regarding area formed between the curves of upward and downward) is time dependent (Rossa et al., 2012). González-Thomás et al. (2008) and Karaca et al. (2009) noted the presence of hysteresis in their studies on ice cream. The addition of soy milk increased ice cream hysteresis areas in ice creams containing coconut milk larger than ones containing cow milk. It is probably due to poor emulsifying properties of coconut proteins (Tangsuphoom and Coupland, 2009) and thus a higher particle size of ice creams containing of coconut milk which lead to a higher apparent viscosity ice creams containing coconut milk. Tárrega et al. (2004) suggested that a high-viscosity thixotropic fluid may indicate a larger hysteresis area than a lower viscose, even if the latter undergoes a more accentuated destruction of the structure. An increase in hysteresis as an outcome of higher viscosity was also reported by Debon et al. (2010) for a dairy product with inulin and Pinto et al. (2012) for frozen yogurt containing microencapsulated Bifidobacterium Bb-12. The SW3 ice cream showed the lowest hysteresis area, and SC1 ice cream also showed the largest hysteresis area. Hence, SC1 ice cream provided a firmer product because more energy is required to break the ice cream structure due to their protein networks (Rossa et al., 2012).
Effect of milk replacement on droplets suspension Measurements of zeta potential (the electrical charge of the droplets) along with particle size can be used to predict the stability of ice cream emulsions. Theoretically, a high negative zeta potential prevents aggregation of the emulsion droplets and increases stability through electrostatic repulsion (Achouri et al., 2012). The zeta potential of fat globules was higher (more negative) (p < 0.05) in ice creams containing cow milk (−26.40 to −37.60 mV) compared to ones containing coconut milk (−26.7 to −34.30 mV) and fat globule size of ice cream containing cow milk (810 – 900 nm) was lower than others containing coconut milk (1567 – 2541 nm) (Table 5 and Fig. 3). The bigger fat globule size for coconut milk ice cream can be attributed to the less surface activity of the coconut proteins than whey proteins (Tangsuphoom and Coupland, 2009) and thus coconut proteins are not particularly effective in preventing droplet aggregation and also creating small droplets during or after homogenization (Onsaard et al., 2006). This makes ice creams containing cow milk being more stable than ice creams containing coconut milk. This is supported by smaller hysteresis areas in the ice creams containing cow milk than ice creams containing coconut milk which indicate higher ability for cow milk ice cream to recover their structure and viscosity (Lopez and Sepulveda, 2012).
| Samples | Particle size (nm) | Zeta potential (mV) |
|---|---|---|
| W | 900f | −36.56cd |
| C | 1736b | −30.70b |
| S | 1604d | −35.50cd |
| SW1 | 810h | −36.87d |
| SW2 | 820gh | −37.60d |
| SW3 | 835g | −26.40a |
| SC1 | 1567e | −33.20bc |
| SC2 | 1677c | −34.30dc |
| SC3 | 2541a | −26.70a |
Micrographs (×50 magnification) of ice cream mixes with different milk: W: ice cream with 100% cow milk; C: ice cream with 100% coconut milk; S: ice cream with 100% soy milk; SW1: ice cream with 75% soy+25% cow milk; SW2: ice cream with 50% soy+50% cow milk; SW3: ice cream with 25% soy+75% cow milk; SC1: ice cream with 75% soy+25% coconut milk; SC2: ice cream with 50% soy+50% coconut milk; SC3: ice cream with 25% soy+75% coconut milk.
Data from rheological studies showed increased ice creams viscosity with increasing amount of soy milk in ice cream made with composite milk. This can be attributed to the change in microstructure whereas the reduction in the fat particle diameters (Fig. 3) related in an increase in consistency index (K value; Table 3) and thus the increased product stability (Chiewchan et al., 2006).
The thermal properties of ice creams with different milks The thermal properties associated with ice crystal-melting of ice creams with different milk (Fig. 4) were measured by differential scanning calorimetry (DSC). There is no significant difference in the peak temperature (Tp) and in the freezing point (Tf) between the ice creams. However, there is some variation in the ice cream containing purely (100%) individual milk in their onset temperature (T0). They showed the highest of T0 in C ice cream and the lowest in S and W ice creams. The enthalpy for the ice crystal melting of W, S, C, SW1, SW2, SW3, SC1, SC2 and SC3 was 108.57, 118.74, 132.29, 105.81, 115.65, 128.96, 113.79, 115.72 and 125.16 J g−1, respectively. The enthalpy values associated with the ice melting transition decreased with the addition of soy milk in ice creams containing composite milks. Two possible factors affect enthalpy value i.e. the final moisture content and the amount of freezable water in the sample (Hwang et al., 2009). The moisture content was highly likely not the factor for the reduction in the enthalpy in the present study because all ice creams had the same moisture content (total solid ice creams ∼ 43 – 44). This makes the freezable water amount as the most probable reason (Table 6). Increasing soy milk proportion and hence soy protein in ice creams made with composite milks could have increased water retention (Akesowan, 2009) and subsequently a decrease in amount of freezable water and thus the melting rate. A positive relationship between the enthalpy of ice-melting transition and the amount of freezable water have been previously reported in wheat- and soy-containing breads (Vittadini and Vodovotz, 2003) and ice cream containing grape wine lees (Hwang et al., 2009).
Effect of the replacement of cow milk with coconut and soy milks on the ice crystal-melting of ice creams measured by differential scanning calorimetry: A) ice creams containing coconut milk, B) ice creams containing cow milk.
| Samples | Peak temperature (°C) | Onset temperature (°C) | Freezing point (°C) | Freezable water (%) | ΔHf (J/g) |
|---|---|---|---|---|---|
| W | −3.82 ± 0.15a | −8.77 ± 0.11c | −5.52 ± 0.09a | 32.50 ± 1.18e | 108.57 ± 4.10d |
| C | −3.17 ± 0.10a | −6.93 ± 0.12a | −4.53 ± 0.14a | 39.61 ± 1.21a | 132.29 ± 5.20a |
| S | −3.90 ± 0.13a | −8.50 ± 0.11c | −5.21 ± 0.12a | 31.91 ± 1.40d | 106.57 ± 3.10d |
| SW1 | −3.44 ± 0.21a | −7.77 ± 0.19b | −5.00 ± 0.10a | 31.68 ± 1.03d | 105.81 ± 6.00d |
| SW2 | −3.75 ± 0.14 | −7.91 ± 0.13b | −5.31 ± 0.17a | 34.62 ± 1.05c | 115.65 ± 4.80c |
| SW3 | −3.68 ± 0.22a | −7.86 ± 0.21b | −5.01 ± 0.21a | 38.61 ± 2.11ab | 128.96 ± 5.30ab |
| SC1 | −3.70 ± 0.11a | −7.86 ± 0.10b | −5.06 ± 0.16a | 34.07 ± 1.07c | 113.79 ± 5.60c |
| SC2 | −3.72 ± 0.16a | −7.91 ± 0.18b | −5.48 ± 0.11a | 34.64 ± 1.04c | 115.72 ± 6.20c |
| SC3 | −3.70 ± 0.12a | −7.40 ± 0.11b | −4.94 ± 0.19a | 37.47 ± 1.09c | 125.16 ± 6.10c |
ΔHf = Enthalpy of fusion
Sensory evaluation Mean scores of flavor, body-texture and taste and color of the samples are shown in Table 7. That the replacement of milk by vegetable milks decreased the body-texture, color and taste. The creaminess, structure, aroma, color and flavor of the products decreased with increasing amount of soy milk (p < 0.05). The total acceptability decreased with increasing soy milk content in ice creams because of soy milk woody or beany off flavors (Abdullah et al., 2003).
Ice creams containing cow milk had the highest total acceptability than ice creams containing coconut milk. The highest total acceptability was found in W and SW3 ice creams and the lowest in S, SC1 and SC2 ice creams. None of the ice-creams were judged to be weak, crumbly, sandy or fluffy.
| Samples | Colour and Appearance (1–5) | Body and Texture (1–5) | Flavor and Taste (1–10) | Total (1–20) |
|---|---|---|---|---|
| W | 4.08 ± 0.05ab | 4.04 ± 0.04a | 7.92 ± 0.05ab | 16.04 ± 0.05a |
| C | 3.25 ± 0.04c | 3.21 ± 0.05b | 6.50 ± 0.06dc | 12.96 ± 0.03c |
| S | 3.12 ± 0.04dc | 3.00 ± 0.05bc | 5.08 ± 0.03e | 11.20 ± 0.02d |
| SW1 | 3.71 ± 0.05b | 3.70 ± 0.03a | 6.42 ± 0.04a | 13.83 ± 0.05bc |
| SW2 | 4.12 ± 0.06a | 4.08 ± 0.02a | 7.17 ± 0.02bc | 15.37 ± 0.06ab |
| SW3 | 4.17 ± 0.07a | 3.91 ± 0.06a | 8.42 ± 0.05a | 16.50 ± 0.04a |
| SC1 | 2.83 ± 0.07d | 2.62 ± 0.07c | 5.58 ± 0.03de | 11.03 ± 0.07d |
| SC2 | 3.04 ± 0.06dc | 2.85 ± 0.04bc | 5.34 ± 0.05e | 11.23 ± 0.05d |
| SC3 | 3.12 ± 0.04dc | 2.79 ± 0.05bc | 6.00 ± 0.03de | 11.91 ± 0.05dc |
In conclusion, examination of selected physical properties showed significant differences among ice creams containing vegetable milks compared with the ice cream containing 100% cow milk. The addition of soy milk in ice creams containing cow and coconut milk improved their physical (viscosity, melting rate and freezable water) properties. The total acceptability and hard texture of S ice cream were improved by the addition of cow milk. The vegetable milks can be used in the production of functional ice creams with significant nutritional and therapeutic properties and also with high physical quality and overall acceptability.
Acknowledgement We acknowledge the financial support of University of Malaya Research Grant (PV113-2012A)
