Colloids and Surfaces B: Biointerfaces 42 (2005) 18

Melting characteristics of fat present on the surface of industrial spray-dried dairy powders
Esther H. -J. Kim a , Xiao Dong Chen a, , David Pearce b
a

Department of Chemical and Materials Engineering, The University of Auckland, Private Bag 92019, Auckland, New Zealand b Fonterra Research Centre, Palmerston North, New Zealand Received 29 November 2004; accepted 19 January 2005

Abstract The melting characteristics of the fat present on the surface (surface free-fat) of two industrial spray-dried dairy powders (cream powder and whole milk powder) were investigated in comparison with those of other milk fat fractions present in the powder, such as free-fat from the interior of the powder particle (inner free-fat) and encapsulated fat. The melting characteristics of the milk fat fractions were studied by fatty acid composition, melting prole and solid fat content prole. The results indicated that all milk fat fractions including surface free-fat contained various triglycerides with melting points ranging from 40 to +40 C. However, some fractionation was observed among the different milk fat fractions. The free-fat fractions (surface free-fat and inner free-fat) had a greater proportion of high-melting triglyceride species than the encapsulated fat. Furthermore, the high-melting triglyceride species present in the free-fat fractions were slightly accumulated at the surface of powder. This phenomenon was observed in both cream powder and whole milk powder and its effect on wetting time was established. This indicates that manipulation of the surface fat content during drying operation may hold the key to functionality improvement. 2005 Elsevier B.V. All rights reserved.
Keywords: Fatty acid composition; Melting behavior; GCMS; DSC; Milk fat; Surface free-fat; Inner free-fat; Encapsulated fat

1. Introduction Spray-dried dairy powders are common ingredients in many food and dairy industries. Many powder properties important in its storage, handling and nal application (e.g. wettability, dispersibility, owability and oxidative stability) are expected to be largely determined by the surface composition of the powder. If one of the milk components is preferentially present on the powder surface, these properties may be changed dramatically. Of special importance is the amount of fat on the powder surface. The presence of fat renders the powder surface hydrophobic decreasing wettability and dispersibility [1]. Fat on the surface acts as a bridge between the particles reducing owability [2,3]. It is also readily susceptible to oxidation and the development of rancidity [4,5]. Understanding the mechanism behind the formation of pow

Corresponding author. Tel.: +64 93737599x87004; fax: +64 93737463. E-mail address: d.chen@auckland.ac.nz (X.D. Chen).

der surface composition and the ability to control the surface composition will be, therefore, highly useful in product quality improvement and new product development. In order to understand the mechanism behind the formation of powder surface composition, a detailed knowledge of the surface composition of spray-dried dairy powders is rstly needed. In our previous publication [6], the surface composition of various industrial spray-dried dairy powders was investigated by means of electron spectroscopy for chemical analysis (ESCA), a method providing direct chemical analysis of the outermost surface layer (10 nm). It was found that the surface composition of powders is signicantly different from their bulk composition. Particularly pronounced is the accumulation of fat on the surface of fat-containing dairy powders, such as cream powder and whole milk powder. The top 10 nm surface layer of these powders is mainly made of milk fat (99 wt.% for cream powder, 98 wt.% for whole milk powder) [6]. This observation indicates that the surface-related properties of these powders will be greatly

0927-7765/$ see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2005.01.004

E.H. -J. Kim et al. / Colloids and Surfaces B: Biointerfaces 42 (2005) 18 Table 1 Bulk composition of the industrial spray-dried dairy powders used Product Composition (wt.%) Lactose CP WMP 12.3 36.6 Protein 11.5 27.9 Fat 71.5 26.6 Moisture 2.7 3.0 Ash 2.0 5.9

determined by the melting characteristics of the fat present on the powder surface. Milk fat is a heterogeneous mixture of various triglycerides with a melting range from 40 to 40 C [7]. In the spray-drying process, the milk fat is encapsulated by protein and lactose matrix (so called encapsulated fat), and only part of the fat can be extracted by organic solvents (so called free-fat). The free-fat is the fat that is extracted using solvent within a pre-xed time and thus originates not only from the powder surface, but also comes from the interior of the powder particle [8]. Different triglycerides can have different encapsulation efciency and the distribution of triglycerides in the powder particle may not be uniform. In order to further investigate the surface formation mechanism of fat-containing dairy powders, the melting characteristics of the fat present on both the powder surface and the bulk powder, therefore, are of great interest. The effects of different fat phases on the encapsulation efciency (level of surface fat) have been investigated using ESCA [4,911]. It was shown that the liquid oils and fully crystalline fats with high melting points were well encapsulated showing low surface fat coverage, while fats with intermediate melting points were poorly encapsulated with the highest surface fat coverage. The authors explained that the highest surface fat coverage in the latter is attributed to decreased emulsion stability of partially crystalline emulsions. A similar trend was observed when emulsions with mixtures of different fats were spray-dried, the pure fats are more efciently encapsulated than any mixtures [12]. It is not possible to distinguish between solid fat and oil by ESCA since they do not differ in the C/O ratio. Due to this methodological limitation, the previous studies assumed that the different fats uniformly distributed in various fat fractions in the powder particle and no information could not be provided regarding the melting characteristics of the fat present on the powder surface. The objective of the present study was to investigate the melting characteristics of the fat present on the surface of industrial spray-dried fat-containing dairy powders in comparison with other milk fat fractions present in the powder. To this end, the milk fat present in the powders was differentiated into three types; free-fat originated from the powder surface (dened as surface free-fat), free-fat originated from the interior of the powder particle (dened as inner free-fat) and encapsulated fat. Those various milk fat fractions were then analyzed for fatty acid composition and melting behavior. Furthermore, the melting point of fat present on the powder surface was directly estimated by testing the wettability of powders using various water temperatures.

obtained from a local dairy company. The powders were commercial products which had been freshly manufactured and packed for consumer use. The capacities of dryers where the powders were made are at least 4 t powder per h. The composition of the powders used is shown in Table 1. The hexane (95%) and iso-propanol (>99%) used in this study were purchased from Asia Pacic Specialty Chemicals Ltd. (Auckland, New Zealand). Boron triuoridemethanol was purchased from Sigma Chemical Company (St. Louis, MO, USA). Deionised water was used throughout. 2.2. Methods 2.2.1. Extraction of milk fat fractions The extraction process of different milk fat fractions in spray-dried dairy powders is schematically shown in Fig. 1. The details of each extraction procedure are as follows. The procedures were repeated until a sufcient amount of fat was obtained for analyses. In all cases we are referring to the fat that is extractable by organic solvents. 2.2.1.1. Extraction of surface free-fat. The free-fat on the surface of a powder particle is quickly dissolved by very brief exposure to the organic solvents. The free-fat from the inner part of the particle dissolves much more slowly. In order to extract only surface free-fat and minimize the extraction of free-fat from the interior of powders, only a brief wash with organic solvent was carried out. Surface free-fat was extracted by a modication of the method described by Buchheim [13]. One gram of the fresh powder was accurately weighed on a lter paper (No. 4, Whatman, Maidstone, Kent, UK), and washed with 4 5 ml of hexane. The powder residue was dried under vacuum at room temperature, and the ltrate solution containing the extracted fat was allowed to evaporate until the extracted fat residue achieved a constant weight. The extracted fat value was then recorded as g surface free-fat/g fresh powder. 2.2.1.2. Extraction of inner free-fat. The powder residue left after the extraction of surface free-fat was added to 40 ml of hexane/g powder, and shaken frequently by hand for 48 h. The powder residue and the solvent were rst separated by ltration through lter paper (No. 4, Whatman, Maidstone, Kent, UK). The powder residue was further washed with 2 2 ml hexane and then dried under vacuum at room temperature. The ltrate solution containing the extracted fat was allowed to evaporate until the extracted fat residue achieved a con-

2. Materials and methods 2.1. Materials Two industrial spray-dried fat-containing dairy powders (cream powder (CP) and whole milk powder (WMP)) were

E.H. -J. Kim et al. / Colloids and Surfaces B: Biointerfaces 42 (2005) 18

Fig. 1. Schematic diagram of extraction of different milk fat fractions from industrial spray-dried dairy powders. of inner free-fat, , extraction of encapsulated fat, , extraction of total fat.

, Extraction of surface free-fat,

, extraction

stant weight. The extracted fat value was then recorded as g inner free-fat/g fresh powder. 2.2.1.3. Extraction of encapsulated fat. The powder residue left after the extraction of inner free-fat was added to 4 ml of warm water (50 C)/g powder and vortexed for 2 min to dissolve the powder matrix and release the encapsulated fat. The resulting solution was extracted with 45 ml hexane/isopropanol (3:1, v/v)/g powder. The suspension was shaken for 15 min, and centrifuged at 1000 g for another 15 min. The clear organic phase was collected and the aqueous phase was re-extracted with the solvent mixture. The organic phase containing the extracted fat was allowed to evaporate until the extracted fat residue achieved a constant weight. The extracted fat value was then recorded as g encapsulated fat/g fresh powder. 2.2.1.4. Extraction of total fat. One gram of the fresh powder was added to 4 ml of warm water (50 C) and vortexed for 2 min. The resulting solution was extracted with 45 ml hexane/iso-propanol (3:1, v/v). The suspension was shaken for 15 min, and centrifuged at 1000 g for another 15 min. The clear organic phase was collected and the aqueous phase was re-extracted with the solvent mixture. The organic phase containing the extracted fat was allowed to evaporate until the extracted fat residue achieved a constant weight. The extracted fat value was then recorded as g total fat/g fresh powder. 2.2.2. Analysis of fatty acid composition The fatty acid composition of the milk fat fractions was determined after conversion of fatty acids into the corresponding methyl esters. An amount of 0.1 g of the milk fat sample was accurately weighed into a vial, and 10 ml of hexane was added to dissolve the sample. A 1 ml of this solution was transferred to a screw-cap reaction vessel and mixed with 2 ml of boron triuoride-methanol. The contents of the vessel were then shaken well and reacted at 70 C for 60 min. After cooling to room temperature, 1 ml of water was added

to stop the reaction, followed by a further 1 ml of hexane. A 2 l of the organic phase was injected into a fused silica capillary column (DB-WAX; 30 m 0.25 mm ID; lm thickness 0.25 m; J&W Scientic Co., Folsom, CA, USA) which was placed into a Shimadzu 17A gas chromatograph equipped with a ame ionization detector connected to a QP5000 quadrupole mass spectrometer (Shimadzu Corp., Kyoto, Japan). The oven temperature was programmed to start at 50 C (2 min) and increase to 240 C at 2 C/min. The injector temperature and the detector temperature were maintained at 230 C. The carrier gas was helium and its ow rate was 1.5 ml/min. The fatty acid methyl esters were identied by comparison of retention times with those of a standard mixture from Sigma (St. Louis, MO, USA). Usually, 35 fatty acids could be detected. For clarity, only the main fatty acids and groups of fatty acids (e.g. C18:1 as the sum of all positions of the double bond) were presented. There can be some differences between analyses, mainly with the most volatile compound. This can have an inuence at the proportion of fatty acids. To avoid this problem, the most volatile fatty acid C4:0 was excluded from the results. 2.2.3. Differential scanning calorimetry (DSC) A differential scanning calorimeter (DSC-SP, Rheometric Scientic, NJ, USA) was calibrated with an Indium standard (mp 156.66 C, Hf 28.41 J/g,). About 10 mg portions of the fat samples were accurately weighed in aluminium crucibles, which were sealed and placed into the calorimeter. An empty sealed crucible was used for reference. DSC melting thermograms were determined by heating the samples at 70 C for 10 min to completely melt the fat and eliminate all crystal nuclei, after which the samples were cooled to 65 C at a cooling rate of 10 C/min. The samples were kept at 65 C for 10 min before heating to 60 C at a heating rate of 5 C/min. Solid fat content (SFC) at various temperatures was calculated from the melting thermograms with the RSI Orchestrator partial area program (Rheometric Scientic, NJ, USA).

E.H. -J. Kim et al. / Colloids and Surfaces B: Biointerfaces 42 (2005) 18 Table 3 Individual fatty acid composition of the milk fat fractions extracted from industrial spray-dried cream powder (CP) Fatty acids Surface free-fat 6:0 8:0 10:0 12:0 14:0 15:0 16:0 16:1 18:0 18:1 18:2 18:3 Caproic Caprylic Capric Lauric Myristic Pentadecylic Palmitic Palmitoleic Stearic Oleic Linoleic Linolenic 2.28 1.40 3.54 4.45 13.99 1.77 32.11 2.02 12.72 22.33 1.91 1.49 Composition (%) Inner free-fat 2.03 1.28 3.33 4.23 13.53 1.79 31.67 2.10 12.99 23.41 2.04 1.61 Encapsulated fat 1.83 1.00 2.54 3.44 11.83 1.60 30.39 1.97 13.53 24.69 5.10 2.09 Total fat 1.87 1.17 3.14 4.05 13.10 1.77 31.34 2.10 13.33 23.71 2.63 1.79

2.2.4. Wetting (wettability) test The wettability of powders was determined by static wetting test as described by Freudig et al. [14] with some modication. A constant amount of powder (1 g) was transferred into a glass cylinder placed on a glass slide covering a water reservoir (diameter = 50 mm). The glass slide was then withdrawn to bring the powder into contact with 25 C water. The wetting time, which is necessary for the submersion of the last powder particle, was measured. The measurements were performed using water at various temperatures (1090 C).

3. Results and discussion 3.1. Masses of milk fat fractions The amounts of various milk fat fractions sequentially extracted from industrial spray-dried dairy powders are shown in Table 2. In cream powder, only about 30% of total fat present were encapsulated and a substantial part of total fat was present as free-fat. Particularly, about 40% of total fat was located the near surface region in the form of a thick layer. In whole milk powder, almost 90% of total fat present were encapsulated and only a small part of total fat was present as free-fat. It is expected the sum of the masses of fat fractions (surface free-fat, inner free-fat and encapsulated fat) to be equal to the mass of total fat. However, there was a slight difference in the masses. The sum of the masses of fat fractions (surface free-fat, inner free-fat and encapsulated fat) was low by approx. 0.01 g/g powder, in both cream powder and whole milk powder. This difference might be due to the extracted fat remained behind on the lter paper and the glassware. 3.2. Fatty acid composition of milk fat fractions Milk fat is a complex mixture of triglycerides, composed of a large number of different fatty acids. This great variety of fatty acids leads to a heterogeneous composition of triglycerides and a very broad melting range, which varies between approximately 40 to +40 C [7]. The fatty acid composition is, therefore, an important and a most basic parameter for determining melting characteristics of milk fat. The fatty acid compositions of the milk fat fractions extracted from industrial spray-dried cream powder are presented by individual fatty acids in Table 3 and, for simplicity of presentation, by selected fatty acid groupings (according
Table 2 Amount of the milk fat fractions extracted from industrial spray-dried dairy powders Product Amount of fat extracted (g fat/g fresh powder) Surface free-fat CP WMP 0.266 0.007 Inner free-fat 0.211 0.015 Encapsulated fat 0.228 0.235 Total fat 0.715 0.266

to similarities in melting points) in Fig. 2. Firstly, the results show that there is a compositional difference between the free-fat and the encapsulated fat. The free-fat fractions (surface free-fat and inner free-fat) had slightly higher concentrations of C6C18 saturated fatty acids and slightly lower concentrations of C16C18 unsaturated fatty acids compared to the total fat, whereas the encapsulated fat showed the reverse tendency with more marked concentration differences. These observations on the free-fat fractions concur with the trend obtained by Lindquist and Brunner [15], although their results were based on spray-dried whole milk powder and only minor dissimilarities were observed. Secondly, the results show that there is a compositional difference even between the free-fat fractions. The surface free-fat had slightly higher concentrations of C6C18 saturated fatty acids and slightly lower concentrations of C16C18 unsaturated fatty acids than the inner free-fat.

Fig. 2. Fatty acid composition (selected fatty acid groupings) of the milk fat fractions extracted from industrial spray-dried cream powder (CP). satd.: saturated fatty acids, unsatd.: unsaturated fatty acids.

E.H. -J. Kim et al. / Colloids and Surfaces B: Biointerfaces 42 (2005) 18 Table 4 Individual fatty acid composition of the milk fat fractions extracted from industrial spray-dried whole milk powder (WMP) Fatty acids Surface free-fat 6:0 8:0 10:0 12:0 14:0 15:0 16:0 16:1 18:0 18:1 18:2 18:3 Caproic Caprylic Capric Lauric Myristic Pentadecylic Palmitic Palmitoleic Stearic Oleic Linoleic Linolenic 2.31 1.98 5.42 6.47 15.81 1.46 30.76 1.58 12.10 19.57 1.35 1.19 Composition (%) Inner free-fat 2.08 1.87 5.24 6.51 16.15 1.48 30.80 1.61 12.17 19.58 1.29 1.20 Encapsulated fat 2.03 1.82 5.03 6.25 15.55 1.47 30.17 1.67 12.58 20.55 1.55 1.33 Total fat 2.08 1.86 4.90 6.22 15.38 1.44 30.40 1.69 12.57 20.65 1.52 1.28

As already mentioned above, the saturated and unsaturated fatty acid content of milk fat affects its melting behavior. In general, increases in short-chain fatty acids (C4C8) and long-chain unsaturated fatty acids (C16:1 and C18:1C18:3) with concurrent decreases in long-chain saturated fatty acids (C16:0 and C18:0) result in milk fat with lowered melting points and a greater proportion of low-melting triglyceride species. Conversely, increases in long-chain saturated fatty acids with concurrent decreases in short-chain fatty acids and long-chain unsaturated fatty acids result in milk fat with higher melting points and a greater proportion of high-melting triglyceride species [1621]. Taking this into account, the results here indicate that high-melting triglyceride species are more concentrated in the free-fat whereas more low-melting triglyceride species in the encapsulated fat. Furthermore, the high-melting triglyceride species present in free-fat are accumulated at the surface of powder. Table 4 and Fig. 3 show the fatty acid composition of the milk fat fractions extracted from industrial spray-dried whole milk powder by individual fatty acids and by selected fatty acid groupings, respectively. The results show similar trends to those observed in the cream powder, i.e. higher concentrations of C6C18 saturated fatty acids and lower concentrations of C16C18 unsaturated fatty acids in the free-fat fractions (surface free-fat and inner free-fat) compared to the encapsulated fat. However, little compositional differences between the surface free-fat and the inner free-fat were observed. Here, data for the encapsulated fat and total fat are virtually identical. This would be expected in that the encapsulated fat accounts for about 90% of the total fat, as already mentioned in Section 3.1. This close agreement of data also conrms the precision of the analytical method. 3.3. Melting prole of milk fat fractions The melting behavior of the milk fat fractions extracted from the two powders was studied by means of DSC. DSC provides a generalized view of the melting of milk fat over

Fig. 3. Fatty acid composition (selected fatty acid groupings) of the milk fat fractions extracted from industrial spray-dried whole milk powder (WMP). satd.: saturated fatty acids, unsatd.: unsaturated fatty acids.

its melting range (melting prole) and the actual percentage of solid fat at various temperature (SFC prole) [18]. Figs. 4 and 5 show the melting proles of the milk fat fractions extracted from industrial spray-dried cream powder and whole milk powder, respectively. In both gures, the melting thermograms of the surface free-fat, the inner free-fat and the encapsulated fat all were very similar to that of the corresponding total fat. All exhibited a wide melting range from 40 to +40 C and four distinct peaks, i.e. a broad shoulder representing high-melting triglyceride species, one peak representing middle-melting triglycerides species and two peaks representing low-melting triglyceride species. These observations indicate that the surface free-fat, the inner free-fat and the encapsulated fat all are complex mixtures of vari-

Fig. 4. DSC melting thermograms of the milk fat fractions extracted from industrial spray-dried cream powder (CP).

E.H. -J. Kim et al. / Colloids and Surfaces B: Biointerfaces 42 (2005) 18

Fig. 7. Solid fat content prole of the milk fat fractions extracted from industrial spray-dried whole milk powder (WMP).

Fig. 5. DSC melting thermograms of the milk fat fractions extracted from industrial spray-dried whole milk powder (WMP).

ous triglycerides with melting points ranging from 40 to +40 C. The same as the total fat. There were general similarities observed, but the size of particular peaks differed between fat fractions. The consequence of these peak size differences is discussed in Section 3.4. 3.4. Solid fat content prole of milk fat fractions SFC prole of the milk fat fractions extracted from industrial spray-dried cream powder is shown in Fig. 6. SFC obtained by DSC is a relative measure, and we assume 100% crystallinity at 40 C. As already observed in the melting thermograms, all milk fat fractions showed gradual melting throughout their melting range in a very similar manner. However, the free-fat fractions (surface free-fat and inner free-fat) had higher a solid fat content at all temperatures than the encapsulated fat. The solid fat content of total fat fell between the encapsulated and surface free-fat fractions. It was nearly identical to the prole of the inner free-fat fraction. The total fat prole was also very similar to the surface free-fat prole. This is not surprising since the free-fat fractions account for

about 70% of the total fat in cream powder (Table 2). The surface free-fat had greatest solid fat content at all temperatures. For example, at 20 C, the surface free-fat contained 35% solids, the inner free-fat 30%, the encapsulated fat 28%. These results indicate that there is some fractionation with the slight accumulation of high-melting triglyceride species at the surface of the powders. The results here are in excellent agreement with the results obtained in Section 3.2 and the solid fat content prole clearly revealed the differences found in their fatty acid composition. Fig. 7 shows the SFC prole of the milk fat fractions extracted from industrial spray-dried whole milk powder. The results showed the similar trend as observed in the spray-dried cream powder, i.e. greatest solid fat content in the surface free-fat and lowest solid fat content in encapsulated fat. For example, at 20 C, the surface free-fat contained 31% solids, the inner free-fat 28%, the encapsulated fat 26%. Buchheim [22] reported that the surface of whole milk powder was locally covered with thin fat layers and the surface fat showed the crystalline state (few monomolecular layers of approx. This previous microscopic observation was found to 50 A). be in good agreement with the results obtained here. 3.5. Melting point of surface free-fat In the observations above, signicant differences in fatty acid composition and melting behavior between the different milk fat fractions were not found. However, the small differences, consistently detected, seem to be important. Because the extraction procedure of surface free-fat extracts not only the real surface free-fat but also near surface free-fat, the fat extracted is a blend of free-fat localized near surface of powder and the small differences are seen as a net effect in the substantial fraction. Much larger differences are, therefore, expected to occur in real surface fat. In order to directly measure the melting characteristics of real surface fat, the melting point of surface fat can be estimated by measuring the wettability of powders using various water temperatures. In general, it has been known that fatcontaining powders are not easily wetted in cold water and can achieve a satisfactory wettability when the water temper-

Fig. 6. Solid fat content prole of the milk fat fractions extracted from industrial spray-dried cream powder (CP).

E.H. -J. Kim et al. / Colloids and Surfaces B: Biointerfaces 42 (2005) 18

fractionation with the accumulation of high-melting triglyceride species in the free-fat and even more at the surface of the powders. The results were further conrmed by wettability tests.

Acknowledgements The rst two authors thank Fonterra Research Centre for a PhD scholarship which facilitates a research program in the area.

Fig. 8. Wettability of industrial spray-dried fat-containing dairy powders at various water temperatures.

References
[1] P. F aldt, B. Bergenst ahl, Spray-dried whey protein/lactose/soybean oil emulsions. 2. Redispersability, wettability and particle structure, Food Hydrocoll. 10 (4) (1996) 431439. [2] C.I. Onwulata, R.P. Konstance, V.H. Holsinger, Flow properties of encapsulated milkfat powders as affected by ow agent, J. Food Sci. 61 (6) (1996) 12111215. [3] E.H.-J. Kim, X.D. Chen, D. Pearce, Effect of surface composition on the owability of industrial spray-dried dairy powders, in preparation. [4] K. Granelli, et al., Inuence of surface structure on cholesterol oxidation in model food powders, J. Sci. Food Agric. 71 (1996) 75 82. [5] N. Hardas, et al., Accelerated stability studies of microencapsulated anhydrous milk fat, Lebensmittel-Wissenschaft und Technol. 33 (2000) 506513. [6] E.H.-J. Kim, X.D. Chen, D. Pearce, Surface characterization of four industrial spray-dried dairy powders in relation to chemical composition, structure and wetting property, Coll. Surf. B: Biointerf. 26 (3) (2002) 197212. [7] P. Walstra, R. Jenness, Dairy Chemistry and Physics, Wiley, New York, 1984. [8] T.J. Buma, Free fat in spray-dried whole milk. 10. A nal report with a physical model for free fat in spray-dried milk, Netherlands Milk Dairy J. 25 (1971) 159174. [9] P. F aldt, B. Bergenst ahl, G. Carlsson, The surface coverage of fat on food powders analyzed by ESCA (Electron Spectroscopy for Chemical Analysis), Food Struct. 12 (1993) 225234. [10] P. F aldt, B. Bergenst ahl, Fat encapsulation in spray-dried food powders, J. Am. Oil Chem. Soc. 72 (2) (1995) 171176. [11] G.P. Pedersen, et al., Solid characterisation of a dry emulsion: a potential drug delivery system, Int. J. Pharm. 171 (1998) 257270. [12] A. Millqvist-Fureby, Characterisation of spray-dried emulsions with mixed fat phases, Coll. Surf. B: Biointerf. 31 (2003) 6579. [13] W. Buchheim, Extraction method for the determination of surface fat of milk powders, Milchwissenschaft 31 (7) (1976) 393396. [14] B. Freudig, S. Hogekamp, H. Schubert, Dispersion of powders in liquids in a stirred vessel, Chem. Eng. Process. 38 (1999) 525 532. [15] K. Lindquist, J.R. Brunner, Composition of the free fat of spray-dried whole milk, J. Dairy Sci. 45 (1962) 661. [16] P.S. Dimick, S.y. Reddy, G.R. Ziegler, Chemical and thermal characteristics of milk-fat fractions isolated by a melt crystallization, J. Am. Oil Chem. Soc. 73 (12) (1996) 16471652. [17] F.M. Fouad, et al., A critical evaluation of thermal fractionation of butter oil, J. Am. Oil Chem. Soc. 67 (12) (1990) 981988. [18] K.E. Kaylegian, R.C. Lindsay, Handbook of Milkfat Fractionation Technology and Applications, AOCS Press, Champaign, IL, 1995. [19] G.A.v. Aken, et al., Composition and crystallization of milk fat fractions, J. Am. Oil Chem. Soc. 76 (11) (1999) 13231331.

ature is higher than the melting point of the fat [23,24]. To improve the wettability of these powders in cold water, a natural surfactant soya lecithin is usually coated on the powder surface. However, an earlier study [25] has shown that the wettability of fat-containing powders, which are not easily wetted in cold water, can be signicantly improved by coating at least a part of the powder surface with mono-and/or di-glyceride of a medium chain fatty acid. This suggests that, in the absence of a surfactant, the wettability of powders is determined by the melting point of the fat present on the outermost powder surface, not the melting point of the fat. Fig. 8 shows the wettability of industrial spray-dried cream powder and whole milk powder at various water temperatures (1090 C). The values are an average of three measurements. As shown, industrial spray-dried cream powder and whole milk powders could not be completely wetted in cold water (10 to 37 C) within reasonable periods (> 15 min). However, the wetting time of the powders was sharply decreased at 42 C for cream powder and 38 C for whole milk powder. This indicates that the fats present on the surface of cream powder and whole milk powder melt at 42 and 38 C, respectively. Though the composition analysis shows a small tendency that the cream powder had harder fractions on surface than that of the whole milk powder, the wettability tests show marked differences.

4. Conclusions Many surface-related powder properties will be inuenced by the quantity of fat on the powder surfaces, but the melting characteristics of surface fat is also of great importance. In this study, the melting characteristics of fat present on the surface of industrial spray-dried cream powder and whole milk powder were investigated. To this end, various milk fat fractions (surface free-fat, inner free-fat and encapsulated fat) present in the powders were sequentially extracted and analyzed their fatty acid composition and melting behavior. The study showed that, in both powders, all milk fat fractions are complex mixtures of various triglycerides, but there is some

E.H. -J. Kim et al. / Colloids and Surfaces B: Biointerfaces 42 (2005) 18 [23] N. King, Dispersibility and reconstitutability of dried milk, Dairy Sci. Abstr. 28 (3) (1966) 105118. [24] Z. Zbikowski, A. Zbikowska, S. Ziajka, Studies on reconstitution of milk from milk powder depending on water temperature, Acta Alimentaria Polonica 15 (3) (1989) 243251. [25] I. Asano, M. Kozaburo, Fat-containing powder product quickly dispersible in cold water and process for preparing the same, United States Patent 4,737,369 (1988).

[20] A.R. Bhaskar, S.S.H. Rizvi, J.W. Sherbon, Anhydrous milk fat fractionation with continuous countercurrent supercritical carbon dioxide, J. Food Sci. 58 (4) (1993) 748752. [21] M.A. Amer, D.B. Kupranycz, B.E. Baker, Physical and chemical characteristics of butterfat fractions obtained by crystallization from molten fat, J. Am. Oil Chem. Soc. 62 (11) (1985) 15511557. [22] W. Buchheim, Electron microscopic localization of solventextractable fat in agglomerated spray-dried whole milk powder particles, Food Microstruct. 1 (1982) 233238.