Unsaturated Emulsifier-Mediated Modification of ... - ACS Publications

Oct 9, 2012 - Nuclear magnetic resonance was used to measure the solid fat content .... High pressure crystallization of binary fat blend: A feasibili...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Unsaturated Emulsifier-Mediated Modification of the Mechanical Strength and Oil Binding Capacity of a Model Edible Fat Crystallized under Shear Nuria C. Acevedo,*,† Jane M. Block,‡ and Alejandro G. Marangoni§ †

Department of Food Science and Human Nutrition, Iowa State University, 2312 Food Sciences Building, Ames, Iowa 50011-1061, United States ‡ Department of Food Science and Technology, Santa Catarina Federal University, Rod. Admar Gonzaga, l346, Itacorubi, Florianópolis, Santa Catarina, Brazil, 88034-00l § Guelph-Waterloo Physics Institute, Centre for Food & Soft Materials Science, Department of Food Science, University of Guelph, 50 Stone Road East, Guelph, Ontario, Canada, N1G 2W1 S Supporting Information *

ABSTRACT: The effects of processing using a scraped surface heat exchanger (SSHE) before and after adding unsaturated monoglyceride (UM) on blends of fully hydrogenated soybean oil (FHSO) and soybean oil (SO) were studied. Mixtures of 40:60 and 45:55 FHSO:SO were melted at 80 °C for 30 min and crystallized statically or in the SSHE (shear rate of 25 s−1) at a cooling rate of 9 °C/min. Upon shearing and UM addition, polymorphic transformations toward more (β) or less (β′) stable forms were governed by the combination between system concentration, composition, and crystallization conditions, as determined by differential scanning calorimetry and powder X-ray diffraction. Nuclear magnetic resonance was used to measure the solid fat content (SFC) development which showed to increase with processing conditions due to the high nucleation rate induced. Processing conditions greatly affected the nano- and microcrystalline structures which were characterized by polarized light microscopy (PLM), cryogenic transmission electron microscopy (Cryo- TEM), and Scherrer analysis of the powder X-ray diffraction data. Crystallization under shear promoted the longitudinal growth of the nanoplatelets; nevertheless, meso structural elements showed a decrease in their dimensions under the same crystallization conditions. The relative oil loss determined gravimetrically was inversely related to the elastic modulus and yield stress of the sheared fat blends, and values were closer to the desirable usability ranges for bakery applications. Our results suggest that fully hydrogenated fats can be functionalized by crystallization in a SSHE and/or by judicious addition of an unsaturated emulsifier.

■

INTRODUCTION Most natural oils and fats have limited applications in their native state. The introduction of partial hydrogenation in the 1800s for the modification of fats and oils allowed the manufacture of margarine and shortening products with improved functional properties. It was found that certain partially hydrogenated vegetable oils had an extended stability and shelf life and produced lighter, finer-grained cakes than other types of fats.1 Oils have been hydrogenated for many decades; however, partially hydrogenated oils contain large amounts of trans fatty acids, or trans fats, which increase the risk of cardiovascular disease. As a result, many health organizations have suggested reducing the consumption of foods containing trans fatty acids.2−5 Full hydrogenation, on the other hand, involves the addition of hydrogen molecules to unsaturated double bonds within the fatty acids that make up triglycerides, obviously without leading to the accumulations of © 2012 American Chemical Society

the geometric isomers, i.e., trans fatty acids. Thus, full hydrogenation converts unsaturated fatty acids to saturated ones, and are considered zero-trans fats. Even though full hydrogenation increases the amount of saturated fat present, much of it is in the form of stearic acid that is converted by the body to oleic acid which does not raise the levels of bad cholesterol.6,7 For that reason, fully hydrogenated vegetable oils are attractive to the food industry as a replacement of partially hydrogenated fats. However, fully hydrogenated fats are very hard, have a high melting point, and have a waxy consistency, even at room temperature. This makes it very difficult for them to be used in food products as such. Received: August 20, 2012 Revised: October 4, 2012 Published: October 9, 2012 16207

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

The mechanical properties of edible fats are influenced by a series of factors, including the solid fat content (SFC), the polymorphism of the solid state, and the crystalline structure of the network.8 Fat crystallization encompasses various processes including nucleation and crystal growth but also polymorphic transformations and crystal aggregation due to van der Waals attractions which continues until a three-dimensional crystal network is created.8 Due to their high viscosity and the need to remove latent heat of crystallization efficiently, fats are usually crystallized under shear. The effect of shear on the process of fat crystallization can be significant, and detailed accounts of these effects are provided by Mazzanti et al.9−11 and Kloek et al.12 Shear has an important impact on crystallite orientation, structure, and size as well as on polymorphic transformations. Shear can also promote crystal aggregation, though at high shear rates it can induce the aggregate break-up. In addition, different shear rates can induce internal rearrangements of the aggregates, leading to more compact assemblies. Emulsifiers are added to shortenings in order to improve their functionality.13,14 The main effects of emulsifiers already present or deliberately added to a fat-based product occur in the area of fat nucleation, crystal growth, and polymorphism. These effects modify the physical properties of the fat system such as crystal size, solid fat content, and crystal arrangement.15 Whether an emulsifier promotes or retards crystallization in oils and fats is still somehow unclear, since it depends on several factors. It has been postulated that similar acyl groups promote crystallization,16 while different acyl groups retard crystal development.17 The aim of this work was to investigate the effects of unsaturated emulsifier addition on the structural and macroscopic properties of mixtures of fully hydrogenated fats in liquid oil crystallized under shear. To achieve this, we analyzed the polymorphism, solid fat content, rheological properties, and oil binding capacity of fully hydrogenated soybean oil (FHSO) in soybean oil (SO). We also addressed the analysis of the impact of both shear and emulsifier addition on the structural organization of the crystallized material.

■

Table 1. Fatty Acid Composition of Fully Hydrogenated Soybean Oil (FHCO), Soybean Oil (HOSO), and Unsaturated Monoglyceride (UM) fatty acid

SO

FHSO

UM

16:00 17:00 18:00 18:1 (total) 18:1t6-8 18:1t9 18:1t10 18:1t11 18:1t12 18:1t13 18:1c6-10 18:1c11 18:1c12 18:2 (total) 18:2c9-t12 18:2t9c12 18:2n6 20:00 20:1c11 18:3n3 22:00

11.18 0.14 4.58

12.81 0.34 83.91

11.3

0 0 0 0 0 0 21.73 1.47 0

0.07 0.11 0.12 0.31 0.12 0.20 0.50 0.11 0.16

0.56 0.15 52.57 0.38 0.47 6.77 0

0 0 0.20 0.68 0 0 0.36

4.7 25

50

8.8

complicated and to fully analyze and understand the generated heat and fluid flow it is ultimately necessary to resort to mathematical models. However, the shear and cooling rates reported in this work even though “ill defined” attempt to provide the reader with a rough guide on the crystallization parameters established. After crystallization, the mixtures were kept at 20 °C for 48 h to allow the material to set and subsequently they were stored at 4 °C until analysis. Gas Chromatography (GC). Lipids were extracted from fat sources by the method of Folch et al.18 Briefly, 0.01 g of shortening or 10 μL of oil was added to 4 mL of chloroform (Fisher, Cat#C2984):methanol (Fisher, Cat#A452-4) solution (2:1, v/v). Samples were vortexed for 1 min, flushed with nitrogen gas (Boc gases, Guelph, ON), and incubated at 4 °C overnight. On the following day, samples were centrifuged at 1000 rpm for 10 min (21 °C) to separate phases. The lower chloroform layer was extracted and transferred to a fresh test tube and dried down with a gentle stream of nitrogen gas. The lipid was saponified in 0.5 M KOH in methanol and heated for 1 h at 100 °C. Phospholipids were converted to fatty acid methyl esters with the addition of 14% boron trifluoride (Sigma, cat#B1252)/methanol and incubation at 100 °C for 1 h. Fatty acid methyl esters were quantified on an Agilent 6890N gas chromatograph equipped with flame ionization detection and separated on an Supelco SPTM-2560 fused-silica capillary column (100 m, 0.2 μm film thickness, 0.25 mm i.d.; Sigma, cat#24056). Samples were injected in splitless mode. The injector and detector ports were set at 250 °C. Fatty acid methyl esters were eluted using a temperature program set initially at 60 °C and held for 0.2 min, increased at 13 °C/min and held at 170 °C for 4 min, increased at 6.5 °C/min to 175 °C, increased at 2.6 °C/min to 185 °C, increased at 1.3 °C/min to 190 °C, and finally increased at 13 °C/min to 240 °C and held for 13 min. The run time per sample is 37.77 min. The carrier gas was hydrogen, set to a 30 mL/min constant flow rate. Peaks were identified by retention times of fatty acid methyl ester standards (Nu-Chek-Prep, Elysian, MN) using EZchrom Elite version 3.2.1 software. Fatty acid concentrations were calculated as percent area. Solid Fat Content Determination. Crystallized samples were introduced into NMR glass tubes and stored for 24 h at 4 °C. Then, the tubes were incubated at the desired temperature for 30 min to allow a homogeneous distribution of temperature at the moment of

MATERIALS AND METHODS

Materials. Fully hydrogenated soybean oil (FHSO) and soybean oil (SO) were generously provided by Bunge Canada (Toronto, Canada). All chemicals and organic solvents were purchased from Fisher Scientific and Sigma-Aldrich (ON, Canada). Unsaturated monoglyceride (UM) provided by Danisco Canada Inc. (Scarborough, ON, Canada) was used as an emulsifier. The fatty acid composition of the materials was determined by gas chromatography and is shown in Table 1. Blend Preparation. Blends of FHSO and SO were mixed in 40:60 and 45:55 (w/w) proportions with the addition of emulsifier in concentrations of 0, 3, and 5% w/w. The blends were melted and held at 80 °C for 30 min to erase crystal memory. UM was incorporated into the melt under mild mixing conditions to ensure homogeneity. The samples were crystallized statically or using a scraped surface heat exchanger (SSHE). The SSHE used in this study is a laboratory scale batch exchanger whose vessel was cooled with ice (wall temperature of ∼0 °C). In this exchanger, the rotational speed of the rotor was 40 rpm in all cases. Considering that the scraper blades are located 2 mm away from the cooling surface, the effective shear rate can be estimated as 25 s−1. Sheared and nonsheared mixtures were crystallized for 30 min in this device. Change in temperature as a function of time, i.e., the cooling rate, was obtained using a copper-constant thermocouple (Omega Engineering, Inc., Stamford, CT), which was determined to be approximately 9 °C/min for all blends. It is well-known that the nature of the heating/cooling and mixing process in an SSHE is very 16208

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

where K is the shape factor, θ is the diffraction angle of the X-rays, fwhm is the full width at half of the maximum peak height in radians (usually from the first small angle reflection corresponding to the (001) plane), and λ is the wavelength of the X-ray. The dimensionless shape factor provides information about the “roundness” of the particle. For a spherical particle, the shape factor is 1; for all other particles, it is smaller than 1. A value of 0.9 is usually used for crystallites of unknown shape and the magnitude employed in this study. Cryogenic Transmission Electron Microscopy (Cryo-TEM). In order to remove the oil phase and better image single crystals, fat blends were treated as reported previously by Acevedo and Marangoni.21 Samples were treated at 10 °C as follows. Fat blends were suspended in cold isobutanol approximately at a ratio of 1:50 (in weight) using a glass stirring rod to obtain a uniform suspension. The fat plus isobutanol mixtures were homogenized at 30 000 rpm with a rotor-stator (Power Gen 125, Fisher Scientific) for 10 min. Then, the crystals were collected by vacuum filtration through a glass fiber filter of 1.0 μm pore size. After filtration, the recovered solid was resuspended in cold isobutanol and rehomogenized for 10 min using the rotor-stator in order to obtain a suitable dispersion of crystals. Finally, the mixtures were sonicated at 10 °C for 60 min using an ultrasonic processor (Bransonic 1210R-DTH, Branson Ultrasonic Corporation, Danburry, CT) to complete the dispersion of the fat crystals. A 5 μL portion of the obtained dispersion was placed on a copper grid with perforated carbon film (Canemco-Marivac, Quebec, Canada), and excess liquid was blotted automatically for 2 s using filter paper. A staining aqueous solution of 2% uranyl acetate was used to enhance contrast. Subsequently, the sample was transferred to a cryo holder (Gatan Inc., Pleasanton, CA) for direct observation at −176 °C in a FEI Tecnai G2 F20 Cryo-TEM operated at 200 kV in low dose mode (Eidhoven, The Netherlands). Images were taken using a Gatan 4k CCD camera. Micrographs were stored and analyzed using DigitalMicrographTM software (USA). Image J 1.42q software (USA, NIH) was employed for a semiautomatic analysis procedure. Oil Loss Determination. Oil loss studies were performed according to the technique described previously by Dibildox-Alvarado et al.22 Once crystallized, fat blends were molded into discs of 22 mm diameter and 3.2 mm thickness using polyvinyl chloride (PVC) molds and then transferred to filter papers (Whatman #5, 110 mm diameter). The amount of oil that each sample (prepared as discs) lost to filter papers was determined by the difference in weight of the filter papers before and after placing the fat disk on the paper for 24 h at 20 °C. A “blank” filter paper was included in all experiments to account for the effects of the treatments on the paper itself such as the influence of the humidity of the storage environment. Filter papers must be large enough in order to avoid the paper saturation with oil during the period of measurement. An average and standard deviation of at least five replicates (five separate disks on individual filter papers) is reported. Oil loss (%) was calculated as

measurement. Solid fat content (SFC) was measured by pulse nuclear magnetic resonance (p-NMR) using a Bruker Minispec spectrometer (Bruker Optics Ltd., Milton, ON, Canada). The reported data corresponds to the average of five individual measurements. Differential Scanning Calorimetry (DSC). A differential scanning calorimeter (DSC; Q1000, TA Instruments, Mississauga, ON, Canada) was used in the thermal analysis of the different fat blends. The instrument heat capacity response was calibrated with sapphire, and the heat flow was calibrated with indium. Approximately 10 mg of the fat sample was placed in alodined pans and sealed hermetically (an empty pan served as a reference). All measurements were performed at a heating rate of 5 °C/min. Thermograms were evaluated using TA Instruments Universal Analysis Software. The peak melting temperature (Tm) and the enthalpy of melting (ΔHm) were determined. The average and standard deviation of four replicates are reported in this study. Polarized Light Microscopy and Box Counting Fractal Analysis. Polarized light microscopy (PLM) was used to observe fat microstructure. To obtain satisfactory reproducibility of slides and crystal appearance, a definite amount of the crystallized sample plus soybean oil was weighed on a slide in order to maintain a 1:1 proportion; then, the mixture was homogeneously spread in all directions and a cover glass was carefully laid over the fat to remove air and complete spreading the fat. Samples were imaged using a Leica DM RXA2 microscope with polarized light (Leica Microsystems, Richmond Hill, Canada) and equipped with a CCD camera (Q Imaging Retiga 1300, Burnaby, BC, Canada). All images were acquired using a 40× objective lens (Leica, Germany). The camera was set for autoexposure. Openlab 6.5.0 software (Improvision, Waltham, MA) was used to acquire images. Focused images were stored as uncompressed 8-bit (256 grays) grayscale TIFF files with a 1280 × 1024 spatial resolution. Five images were captured from each of the five replicates prepared. Microstructural analysis was carried out by image analysis employing the Adobe Photoshop CS 3 software (Adobe Systems Inc., San Jose, CA) and filters from the Fovea Pro 4.0 software (Reindeer Graphics, Inc., Asheville, NC). A manual threshold was applied to all the pictures to convert the grayscale images to binary images, in order to discriminate between features and background and to measure the feature sizes. The microstructural elements were analyzed using the filter tools included in the Fovea Pro software. Fat crystal network fractal dimensions (Db) were determined by the box counting method.19 Thresholded PLM images were processed using the software Benoit 1.3 (TruSoft Int’l Inc., St. Petersburg, FL) to calculate the 2D fractal dimension. A grid formed by boxes of decreasing sizes is placed over the binary images, and the number of occupied grids (N) is counted for a series of grid side length (L). Any box containing a number of particles greater than a threshold value is considered to be an occupied grid. The number of occupied boxes as a function of the size of the boxes is plotted, and the negative of the slope of the log−log plot is the box counting fractal dimension (Db). An average of 25 replicates for each sample was carried out. Powder XRD Analysis. XRD data were collected using a Rigaku Multiflex Powder X-ray Diffractometer (Rigakug, Japan). The copper lamp (λ = 1.54 Å for copper) was set to 40 kV and 44 mA. A 0.57 divergence slit, 0.57 scatter slit, and 0.3 mm receiving slit were used. For the small-angle X-ray diffraction analysis (SAXD), the samples were scanned from 0.9 to 8° at 0.05°/min. The wide-angle X-ray diffraction analysis (WAXD) was carried out scanning the samples from 16 to 35° at 0.5°/min. PeakFit software (Seasolve, Framingham, MA) and MDI’s Jade 6.5 software (Rigaku, Japan) were used to analyze the obtained SAXD and WAXD patterns, respectively. From the SAXD patterns, the crystalline domain size (ξ) can be calculated by the well-known Scherrer formula which is limited to nanoscale particles and it is not applicable to sizes larger than about 100 nm:20 ξ=

Kλ FWHM cos(ϕ)

OL (%) =

wt. paper(24 h) − wt. paper(0 h) × 100 wt. paper(0 h)

(2)

Small Deformation Rheology. After crystallization, samples were transferred to the wells of polyvinylchloride (PVC) disk molds of 3.2 mm thickness and 20 mm diameter. Rheological measurements were obtained using a TA Instrument AR2000 controlled stress dynamic rheometer (TA Instruments, Mississauga, ON, Canada). A 20 mm diameter stainless steel flat plate was selected to carry out the experiments. The temperature of the sample was held at 20 °C during analysis achieved by a Peltier system located in the base of the measurement geometry. The oscillatory stress sweep within the region of linear viscoelastic behavior (LVR) was performed from 1 to 1000 Pa (with a frequency of 1 Hz) for all samples. Compression was set to a normal force of 5 N. To prevent slippage, sandpaper (grade 60) was attached to the lower surface of the geometry and the upper surface of the Peltier base of the rheometer. The yield stress (σ*) values were determined from the stress sweep curves as the stress value (in Pa), above the LVR, at

(1) 16209

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

mixed crystal formation, which would lead to the “kinetic entrapment” of TAG molecules in a less stable but more kinetically accessible polymorphic state, namely, the β′ form. Thus, here we have learned that a very rapid crystallization process against a large temperature gradient leads to the formation of the less stable, but more functionally desirable, polymorphic form. Plastic fats in the unstable β′ form are considered to be more functional, and therefore more desirable over fats in the stable β form. This is due to its small crystal size (1 mm) and thin, needle-shaped morphology. Many authors reported that the shape and size of crystals and crystal aggregates largely affects the macroscopic elastic constant and hardness of the fat network.29,30 When the conversion from the β′ to β form occurs in a fat crystal network, large crystals are created;31 thus, sensorial properties (such as graininess) and macroscopic properties are negatively affected. On the other hand, addition of unsaturated monoglyceride to the 40:60 FHSO−SO blend at 3 and 5% levels leads to an increase in the intensity of the diffraction peak at 4.6 Å. Thus, XRD patterns clearly indicate that the presence of UM seems to favor polymorphic transformation from the β′ to β form. Unsaturated partial acylglycerols tend to increase the liquid fraction of a fat, by delaying nucleation, therefore accelerating phase transformations to more stable forms via melt-mediated polymorphic transformations.32 Additionally, these surfactants probably create local imperfections within the crystals which facilitate the mobility of triacylglycerol molecules33 and therefore a solid-state β′ to β polymorphic transformation. In fat mixtures with higher amounts of crystalline fat (45:55 FHSO:SO), the observed polymorphic behavior was different upon emulsifier addition and crystallization conditions (Figure 1B). In contrast to the 40:60 FHCO−SO samples, static crystallization from the melt leads to the formation of a mixture of β and β′ crystal forms, with higher proportions of the β′ polymorph than for the 40:60 case. Since this sample had a higher proportion of saturated fatty acids (and was thus more supersaturated), a more rapid crystallization would have taken place, with a faster and more pronounced increase in melt viscosity. This would have limited molecular mobility (diffusion), thus leading to a more pronounced kinetic entrapment of the TAG molecule in the β′ form. As expected, when this sample was crystallized in the SSHE, the formation of the β polymorph was enhanced, in agreement with former works.9,34 Contrary to what was observed for samples with 40% hardstock, addition of the unsaturated monoglyceride inhibited the β′ to β conversion. This effect was more pronounced at higher proportions of surfactant incorporated into the fat blend. These opposite results on the influence of emulsifiers at both FHSO:SO proportions imply that the effect of surfactant on the β′−β transformation is not absolute but rather varies at different FHSO:SO ratios. Interestingly, the β′ form was stable for over 6 months for these blends. The melting curves obtained by DSC for FHSO:SO blends with 0, 3, and 5% UM are shown in Figure 2. Blends with either 40 or 50% FHSO (Figure 2A and B, respectively) showed thermographs consistent with the results obtained from X-ray diffraction studies. In blends with 40% FHSO, polymorphic transformation from β′ to β was enhanced by UM. This effect is manifested by DSC as the disappearance of the “shoulder” corresponding to the melting of β′ (indicated by the arrows in Figure 2A) and the intensification of the melting peak of β as the emulsifier concentration increases. Meanwhile, an increase

which the storage modulus (G′) decreases 10% from the constant value. The reported data are the average of 6−10 individual replications. Statistical Analysis. Data were processed using GraphPad Prism 5 software (GraphPad Software, Inc., San Diego, CA). Reported values correspond to means and standard errors of the determinations. Statistical analysis was performed by one-way ANOVA (p < 0.001) using Tukey’s multiple comparisons as post-test (p < 0.005).

■

RESULTS AND DISCUSSION The fatty acid composition of FHSO, SO, and unsaturated emulsifier was determined by gas liquid chromatography (Table 1). The results showed that FHSO contained 83.9% stearic acid (C18:0) and 12.8% palmitic acid (C16:0), which represent 96.7% of the total fatty acids. The predominant fatty acids in SO were linoleic acid (C18:2) with 52.6% of the total, followed by oleic (C18:1), palmitic (C16:0), and stearic acid (C18:0) with 21.7, 11.2, and 4.6%, respectively. These results are in agreement with values reported previously for FHSO23−25 and SO.1,25,26 The unsaturated emulsifier shown is comprised of 50% linoleic acid (18:2), 25% of oleic acid (18:1), 11.3% palmitic acid (16:0), and 8.8% alpha-linolenic acid (18:3, n-3). Powder X-ray diffraction patterns obtained for the FHSO:SO blends before and after the incorporation of 3 and 5% emulsifier are shown in Figure 1. The patterns corresponding to

Figure 1. X-ray diffraction patterns for 40:60 (A) and 45:55 (B) FHSO:SO samples crystallized under static (ST) and sheared conditions (SSHE) before and after the incorporation of unsaturated emulsifier (UM) to the formulation.

mixtures with 40% FHSO crystallized under static conditions (Figure 1A) exhibited three strong signals at 4.6, 3.8, and 3.7 Å, which are characteristic of the polymorphic β form.27 Surprisingly, when mixtures were crystallized in the scraped surface heat exchanger crystallizer, powder XRD patterns indicated the presence of the β′ polymorph, with wide angle reflections at 4.2 and 3.8 Å.27 However, one weak XRD reflection at 4.6 Å indicated the persistence of a small amount of the β form. These results can be understood considering the more rapid formation of a crystal network and thus viscosity increase, during SSHE processing. This increase in viscosity would decrease molecular mobility28 and would enhance more 16210

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

An increase in the SFC profile was also observed in a previous work after similar blends were subjected to laminar shear rates of 30 and 240 s−1.39 As expected, SFC values of the blends with UM were lower for all the temperatures investigated. The higher the amount of monoglyceride added, the lower the SFC value. This could logically be explained by both matrix dilution and the lower melting point of the UM. A lower melting point results in a higher solubility of the crystallized phase in the liquid phase, leading to lower SFC values.40−42 The microstructure of all the samples observed by PLM is shown in Figure S1 (Supporting Information). In general, the microstructure was shown to be similar at both FHSO proportions. As expected, in the absence of shear, the microstructure of the fat mixtures is composed of discrete spherulitic entities. These observations agree with earlier work where analogous microstructural elements were observed working with similar fat mixtures.39,43 Crystallization under shear resulted in the formation of rounded clusters of needleshaped crystals with a semiorganized spatial crystal distribution and diameters ranging between 20 and 60 μm. The rounded shape of the clusters suggests that some partial melting or softening of the crystals had occurred after their formation. Similar spherical crystal morphologies were previously observed by Acevedo and Marangoni39 when crystallizing analogous fat blends using a laminar shear processing. Furthermore, Van Aken and Visser44 reported comparable microstructural findings after crystallization of milk fat using a scraped surface heat exchanger. These changes observed at the mesoscale have been attributed to crystal aggregation and organization promoted by crystallization under low shear or mild agitation, like in this case.39,45 On the other hand, the addition of the unsaturated emulsifier to the fat blends had a significant effect on crystal morphology. The microstructure of sheared blends with unsaturated monoglyceride is characterized by a granular texture with fine crystals combined with a small number of spherical aggregates. Apparently, the presence of unsaturated emulsifier seemed to prevent the ordered organization of the mesoscale. In addition, the number of the globular aggregates was smaller in blends with 45% FHSO, indicating that the destabilizing effect of emulsifier on the mesoscale is more pronounced in samples with a higher percentage of FHSO. Crystallizations under static conditions resulted in similar meso-crystal diameters for blends with 40 and 45% FHSO, with values of 1.23 and 1.29 μm, respectively (Figure 4). Shearing caused a significant decrease in the meso-crystal sizes (Figure 4). On the other hand, incorporation of UM led to the

Figure 2. DSC melting thermograms for 40:60 (A) and 45:55 (B) FHSO:SO mixtures crystallized under static conditions (ST) or in a scraped surface heat exchanger crystallizer (SSHE) before and after the addition of unsaturated emulsifier (UM).

of the β′ form is clear in blends with 45% FHSO where UM has been incorporated in proportions of 3 and 5%. Figure 3 displays the results of solid fat content (SFC) obtained for all the blends. SFC is highly dependent on

Figure 3. Solid fat content (SFC) profiles obtained for FHSO:SO fat blends crystallized under static conditions (ST) or in a scraped surface heat exchanger crystallizer (SSHE) before and after the addition of unsaturated emulsifier (UM).

processing and crystallization conditions. Crystallization under shear induced an increase in the SFC profile at both FHSO proportions due to the high nucleation rate induced and the formation of highly developed crystals in relation to the formation of the crystal network under static conditions.35−38

Figure 4. Meso-crystal equivalent diameters of FHSO:SO blends before and after crystallization in a scraped surface heat exchanger and incorporation of unsaturated monoglyceride. Values represent means and standard deviations of at least 25 replicates. Letters represent statistically significant differences between the values (P < 0.05). 16211

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

Figure 5. Left: Cryo-TEM photomicrographs of 45:55 FHSO:SO blends crystallized statically or in a scraped surface heat exchanger crystallizer (SSHE) before and after adding unsaturated monoglyceride (UM). Right: Corresponding frequency size distributions for nanoplatelet lengths and widths obtained from the analysis of the cryo-TEM images.

Table 2. Nano-Platelet Lengths, Widths, Thickness, Aspect Ratios, and Nano-Equivalent Diameters (a) in Blends with 45% Fully Hydrogenated Soybean Oil (FHCO) and 55% Soybean Oil (HOSO) (45:55 FHSO:SO), Crystallized Statically (ST) and under External Shear Fields (SSHE) before and after the Addition of Unsaturated Monoglyceride (UM)a

a

nanoplatelet characteristic

ST, 0% UM

SSHE, 0% UM

SSHE, 3% UM

SSHE, 5% UM

length (nm) width (nm) thickness (nm) aspect ratio nanoequivalent diameter (a)

249 ± 10a 108 ± 3e 35.0 ± 2.1i ∼2.3 185 ± 6l

450 ± 25b 124 ± 4f 35.7 ± 5.3i ∼3.5 267 ± 11m

730 ± 27c 183 ± 6g 46.5 ± 3.0j ∼4 412 ± 14n

1000 ± 44d 196 ± 13h 49.1 ± 3.5k ∼5 500 ± 27o

Means and standard errors are reported. Superscript letters represent statistically significant differences between the values (P < 0.05).

Crystallization under shear fields led to the creation of nanocrystals with lengths and widths 60 and 30% larger than those observed in statically crystallized blends. In earlier works, Acevedo and Marangoni46 and Maleky et al.47 demonstrated that different cooling and shear rates markedly affected the nanoscale of fat crystalline networks. They found that both high cooling and laminar shear rates led to the formation of smaller nanocrystals. However, our results are consistent with those published by Acevedo and Marangoni39 where a predominance of larger nanoparticles was reported when working with similar blends crystallized under mild laminar shear conditions, relative to nonsheared blends. These authors argued that crystallization below a specific shear rate induces the growth of the nanocrystals, while, above this critical shear rate, progressively smaller nanocrystallites are observed. Nanoplatelets formed under shear in the presence of UM were significantly larger. Nanocrystal lengths of blends crystallized in the SSHE were 1.6 and 2.2 times larger after

formation of smaller meso-crystals with sizes of approximately 1 μm for both UM concentrations and FHSO proportions. These results indicate that the UM effect is probably related to an enhanced inhibition of meso-crystal growth, or aggregation, under shearing conditions. These effects point toward the involvement of crystal surfaces. Cryo-TEM and the sample treatment developed by Acevedo and Marangoni46 were used to systematically study the effects of shear and the unsaturated emulsifier on the nanostructure of FHSO:SO mixtures. The analysis of the nanoscale revealed opposite trends compared to those found at the mesoscale. Figure 5 shows examples of cryo-TEM micrographs and the corresponding nanoplatelet size distributions obtained for samples with 45:55 FHSO:SO before and after being subjected to shearing during crystallization and the incorporation of UM. Similar images and with analogous tendencies were observed for blends with 40% FHSO (data not shown). 16212

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

the addition of UM at concentrations of 3 and 5%, respectively. On the other hand, platelet widths were 1.4 and 1.6 times larger in samples with 3 and 5% UM, respectively. This effect was probably due to a delay in the nucleation process upheld by the presence of the unsaturated monoglyceride within the formulation; therefore, crystal growth at the nanoscale is enhanced and as a consequence nanocrystal sizes are considerably larger in the presence of UM. Table 2 summarizes the nanoplatelet lengths and widths obtained from the analysis of the cryo-TEM images of 45:55 FHSO:SO blends. In addition, the thickness values acquired from the Scherrer analysis of the powder X-ray data in the small angle region (SAXD), the nanocrystal aspect ratios (length to width ratio), and the calculated equivalent diameters are also included in order to provide a more comprehensive analysis of the effects of shear forces and emulsifier addition on the fat blends nanoscale. It can be clearly seen that crystallization using a scraped surface heat exchanger led to the formation of nanoplatelets with significantly larger lengths, widths, and thicknesses. As can be inferred from the nanoplatelets’ features, there is an enhancement of the longitudinal growth of the nanocrystals when using the mechanical crystallizer, since the nanoplatelet aspect ratios were higher after crystallization under these conditions. There is a good agreement between this work and previous studies where a large increase in the nanoparticles’ aspect ratios was reported in similar fat mixtures crystallized under the influence of external fields.39,46 In these works, laminar shear rates of 30 s−1 yielded platelet lengths and widths up to 2.6 and 1.5 times larger, respectively, than those observed in nonsheared samples.39 On the other hand, shearing at 300 s−1 led to the formation of smaller nanoplatelets in fat systems.45 Nevertheless, both studies revealed an enhancement to a lesser or greater extent of the longitudinal growth of the nanocrystals, and, therefore, an increase in the aspect ratio of the nanoparticles. It is also worth noting that the addition of unsaturated emulsifier induced an increase in the size of the crystal nanoparticles; furthermore, the higher the amount of UM introduced in the formulation, the larger the dimensions and the aspect ratio of the extracted nanocrystals. Another interesting finding is that UM seems to have a more pronounced effect on the longitudinal growth of the nanoplatelets, in particular at high concentrations, since it is possible to observe nanoplatelet lengths 5 times larger than the widths at emulsifier concentrations of 5%. Meanwhile, at 0 and 3% UM, the nanocrystal lengths were approximately 3.5 and 4 times larger than the widths, respectively. It is interesting that in our study there is an opposite behavior between both the nano- and mesoscale. Even though there is an increase in the nanoplatelet dimensions induced by shear, the opposite effect is observed at the mesoscale, as smaller mesocrystals were observed after crystallization in the SSHE with and without emulsifier (Figure 4). We studied the rheological properties of all the blends, since they are strongly correlated with the macroscopic functionality of the fat system. Figure 6 shows the shear storage moduli (G′) and yield stress (σ*) values obtained for all the studied fat blends. No significant differences were found between the elastic moduli and yield stress values of blends with different relative amounts of FHSO, independent of the crystallization conditions and emulsifier concentration. One exception was found for G′ of fat blends crystallized in the absence of

Figure 6. (A) Elastic moduli (G′) and (B) yield stress (σ*) values of blends with 40% FHSO (black bars) and with 45% FHSO (gray bars) before and after crystallization under shear rate and the incorporation of unsaturated monoglyceride (UM). Values represent means and standard deviations of at least six replicates. Letters represent statistically significant differences between the values (P < 0.05).

emulsifier where the storage moduli were slightly lower in blends with lesser amounts of FHSO. Both G′ and σ* decreased with crystallization under shearing conditions. In particular, this decrease was more significant for σ*, implying that this parameter is more sensitive than G′ to changes in crystallization conditions. For instance, samples with 40% FHSO showed an ∼90% decrease in yield stress when sheared in the SSHE relative to static crystallization, while the reduction in G′ was of ∼80% for the same blends. Unexpectedly, from Figure 3, it is possible to notice an increase in the SFC profile of fat blend after crystallization in a SSHE. Therefore, since generally a higher SFC translates into a more solid-like character of the fat, it was expected to observe higher values of G′ and σ*.48−50 Our results are opposite to that, which can be explained by the fact that the mechanical properties of fats are not only governed by the amount of solid fat in the network but also by the structure of this crystal network. These observations can be explained in part on the basis of the work by Marangoni and Rogers51 who proposed a structural definition for the yield stress (σ*) of a particle network: σ* ≈

6δ 1/(d − D) ϕ a

(3)

where the yield stress is a function of the crystal−melt interfacial tension (δ), the primary particle diameter (a), the Euclidean dimension of the embedding space (d), and the fractal dimension (D). Our findings agree with this model. When shear is applied during crystallization and/or unsaturated monoglyceride is added to the blend, σ* is significantly lower, which can be due to an increase in the primary crystal size (nanoplatelet size) induced by these conditions (Table 2). As expected, G′ presented an analogous behavior to that observed for σ*. Furthermore, the results show that unsaturated emulsifier had no significant effect on the rheological properties of the 16213

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

We also investigated the oil binding capacity of the blends through analysis of the oil loss (OL) which results are shown in Figure 7. Samples with higher amounts of FHCO had lower OL

system. There were in general no significant changes in rheological parameters of the sheared samples independent of the amount of UM incorporated into the mixture (Figure 6). Moreover, the results presented in Figure 3 show that UM addition lowered the SFC profile of the monoacylglycerolcontaining blends. These findings could be justified as a consequence of sample dilution, since unsaturated monoacylglycerols melt at low temperatures.52 However, the decrease in SFC is definitely more than either 3 or 5%, which implies that perhaps UM somehow causes a reduction in the amount of fat that crystallizes in the system, even though these changes are not manifested on the rheological properties. Of great interest is how the physical properties of fats (such as mechanical strength) can be influenced by the microstructure. Much work has demonstrated that the rheological properties of fat crystal networks are the outcome of a collective effect of SFC and microstructure which includes the shape, size, and spatial distribution of the fat crystals.53−55 The microstructure of fat crystal networks can be quantified through the determination of the microscopic fractal dimension using different methods such as box counting, particle counting, and Fourier transform of polarized light microscopy images. In this work, we determined the fractal dimension (Db) of all the fat blends by box counting in order to relate structure to the physical properties of the systems. Only fat mixtures with 40% FHSO seemed to show a slight decrease in Db upon processing in the SSHE independent of the presence and concentration of unsaturated emulsifier. Instead, Db in 45:55 FHSO:SO blends did not show a significant change in their values (Table 3).

Figure 7. Oil loss values (OL) of blends with 40% FHSO (black bars) and with 45% FHSO (gray bars) before and after crystallization under shear rate and the incorporation of unsaturated monoglyceride (UM). Values represent means and standard deviations of five replicates. Letters represent statistically significant differences between the values (P < 0.05).

values, compared to blends with less FHSO. As anticipated, there is a significant increase in OL upon shearing the blends with or without the addition of unsaturated monoglyceride. From the rheological results, it is possible to notice that OL is inversely correlated to both rheological parameters, since blends with lower elastic moduli or yield stresses displayed the greatest OL values, which is in agreement with previous works.22,39 The significantly higher OL values after crystallization in the SSHE probably reflect a structural damage inflicted to the network upon shearing. It is also worthwhile to note that UM yielded a significant change in oil loss compared to sheared blends without emulsifier. The UM effect does not follow a trend with its final concentration in the blend. In mixtures with 40% FHSO crystallized in the SSHE, the incorporation of either 3 or 5% UM led to a reduction of 27% in the oil loss value. Instead, the addition of 3% UM caused a reduction of 50% in OL for blends with 50% FHSO, while a rise in UM concentration to 5% led to an increase in OL of 12%. These results are in close agreement with earlier analyses carried out by Acevedo and Marangoni39 where similar fat blends processed under laminar shear exhibited significantly higher OL values, even at a shear rate comparable to the one used in this work. To further investigate the relationship between structure and oil transport, we calculated the permeability coefficient (B) of the oil through the crystal network via Darcy’s law, as used by Bremer et al.:56

Table 3. Box Counting Fractal Dimension (Db) and Permeability Coefficients (B) of Blends of Fully Hydrogenated Soybean Oil (FHSO) and Soybean Oil (SO) before and after Shearing and Addition of Emulsifiera Db

FHSO:SO 40:60-ST-0%UM 40:60-SSHE-0%UM 40:60-SSHE-3%UM 40:60-SSHE-5%UM 45:55-ST-0%UM 45:55-SSHE-0%UM 45:55-SSHE-3%UM 45:55-SSHE-5%UM

1.10 0.96 1.01 1.03 1.04 1.07 1.01 1.05

± ± ± ± ± ± ± ±

0.14a 0.08b 0.03b 0.03b 0.19b 0.06a,c 0.04b 0.043a

B (m2) 12.8 7.80 7.48 7.56 9.69 7.33 6.37 7.67

× × × × × × × ×

10−12 10−12 10−12 10−12 10−12 10−12 10−12 10−12

a

Means and standard deviations of at least 25 replicates are reported. Superscript letters represent statistically significant differences between the values (P < 0.05).

a 2 2/(D b − 3) Φ (5) τ This expression predicts that an increase in fractal dimension (Db) and meso-particle size (a) and a reduction in solids’ volume fraction (Φ) lead to a higher permeability coefficient. τ is the tortuosity of the migration path which is considered equal to 1 in this work due to the high volume fraction of solids in the system. The permeability coefficient represents the flow of water (in this case oil) through a unit cross-sectional area under a unit hydraulic gradient at a determined temperature.57 The obtained results surprisingly show that there is a decrease in the permeability induced by the SSHE processing which is further decreased by the addition of UM. These results are opposite to the OL values obtained for all the samples where the amount of oil lost from blends crystallized statically is always lower than B=

According to the models reported by Narine and Marangoni50 and Marangoni and Rogers,51 the shear modulus (G′, eq 4) and yield stress (σ*, eq 3) are related to the volume fraction of solids in a power law manner as a function of the network fractal dimension. (4) G′ = λ Φ1/(3 − D) According to these expressions, a lower fractal dimension should result in a higher elastic modulus and yield stress. However, we observed a decrease in the mechanical properties induced by processing and UM which was accompanied with a decrease or no variation in the values of Db. Probably, the increase in the solid fat content (or volume fraction of solids) and the substantial increase in the nanoparticle sizes that make up the network counteracted the changes in Db. 16214

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

that of mixtures processed in the SSHE. Nevertheless, B values correlate well with the changes observed in crystal sizes at the mesoscale (Figure 4); moreover, it is important to mention that UM yielded indeed lower B values, in particular at a concentration of 3%. These contradictory results could be explained by the possibility that in our case the mechanism of oil diffusion is affected not only by the architecture of the microstructural elements but also by parameters such as tortuosity and porosity whose changes are not taken into account in this work. Better data on diffusivities of the individual samples should be considered in the study of the determining factors of the oil migration through the crystalline matrix and to ultimately explain the behavior of the physical properties of fats. In an effort to evaluate the effect of shear and emulsifier on the functionality of the produced fats, their mechanical properties and oil loss values are compared in Table 4 to those obtained from a commercial roll-in shortening.

influence crystallization rates and polymorphic transformations of fats.60 In crystal growth, the emulsifiers are adsorbed at the kinks or steps of fat crystals at the crystal/liquid interfaces and disturb crystal growth. According to Garti,60 crystal morphology can be altered by the presence of surfactants in three ways: (1) when the emulsifier has little miscibility in the fat, it thus acts as an impurity, inducing imperfect fat crystals and promoting or delaying polymorphic crystal growth; (2) when the amphiphilic molecule is highly miscible in the fat, forming molecular compounds; and (3) when there is total immiscibility of the emulsifier and the fat in the solid state; hence, the surfactant can behave as seeds for nucleation. In an attempt to clarify the effects of adding UM on the systems studied, we calculated the entropy of melting changes (ΔSm) in all fat mixtures (Table 5). In general, ΔSm values were Table 5. Entropy Change Values (ΔSm) Obtained for Blends of Fully Hydrogenated Soybean Oil (FHSO) and Soybean Oil (SO) before and after Shearing and Addition of Unsaturated Emulsifiera

Table 4. Elastic Moduli (G′), Yield Stress (σ*), and Oil Loss (OL) Values Obtained for a Commercial Roll-in Shortening and Blends of Fully Hydrogenated Soybean Oil (FHSO) and Soybean Oil (SO) before and after Shearing and Addition of Emulsifiera FHSO:SO 40:60-ST-0%UM 40:60-SSHE-0%UM 40:60-SSHE-3%UM 40:60-SSHE-5%UM 45:55-ST-0%UM 45:55-SSHE-0%UM 45:55-SSHE-3%UM 45:55-SSHE-5%UM commercial roll-in shortening a

G′ (×106, Pa) 8.89 1.40 1.51 1.23 11.56 2.99 1.87 1.84 1.08

± ± ± ± ± ± ± ± ±

0.05 0.20 0.31 0.30 0.09 0.37 0.36 0.34 0.20

σ* (Pa) 2750.0 270.6 368.7 369.8 3053.7 555.1 430.7 389.1 835

± ± ± ± ± ± ± ± ±

31.5 37.5 1.08 40.0 28.9 31.2 33.0 35.3 22.7

FHSO:SO 40:60-SSHE-0%UM 40:60-SSHE-3%UM 40:60-SSHE-5%UM 45:55-SSHE-0%UM 45:55-SSHE-3%UM 45:55-SSHE-5%UM

OL (%) 0.60 9.56 6.96 7.06 0.30 4.36 2.19 4.75 2.80

± ± ± ± ± ± ± ± ±

0.07 0.50 1.08 1.24 0.06 1.09 0.50 1.09 0.25

ΔSm (J/mol) 206.9 203.4 195.1 244.8 233.2 239.6

± ± ± ± ± ±

1.3 6.0 4.9 14.1 4.7 7.2

a

Means and standard deviations of at least three replicates are reported.

lower in fat systems after the addition of UM and polymorphic transformations to the more stable β form were delayed or prevented. A lower ΔSm is indicative of a more disordered solid state structure (resembles the liquid more) and is probably related to the effect of cocrystallization, creating crystalline imperfections within the crystal networks. It is worth noting that such changes were associated, in turn, with the increase in nanocrystal size (possibly induced by the molecular disorder in the presence of emulsifier), as well as the observed decrease in mechanical strength of the materials (lower G′ and σ*).

Means and standard deviations of at least five replicates are reported.

Despite their low oil loss, statically crystallized samples show G′ and σ* values considerably higher than those of the commercial shortening which point out a limited functionality due to their extremely brittle, not spreadable, and hard consistency. According to Haighton,58 crystallized fats with yield stresses above 1000 Pa are too hard and in the limit of spreadability. σ* values of sheared blends before and after the addition of UM have always yield stresses below 1000 Pa and showed no significant change in their plastic properties during storage (data not shown). Furthermore, in consonant to the “usability range” for some fat products reported by Haighton,58 every mixture crystallized in the SSHE can function as normal or hard bakery shortening, since their yield values ranged from 200 to 700 Pa. Some interesting findings resulted from comparison of the oil binding capacities observed in the mixtures. The addition of UM decreased the amount of oil lost from the samples in particular at the lowest concentration studied (3%). This would seem to be related to the additional decrease in the mesocrystal sizes in blends with emulsifier. A fat matrix consisting of large crystals cannot incorporate as much liquid oil as others constituted by small crystals; hence, the product becomes oily and unable to efficiently retain liquid oil.59 The effect of emulsifiers on triacylglycerol systems has been of scientific interest over many years, since they are known to

■

CONCLUSIONS This work describes the effects of processing using a surface heat exchanger crystallizer and an unsaturated emulsifier based on the triacylglycerol structure, mechanical properties, and oil binding capacity of the network. Our results show that, although it has been previously demonstrated that shear enhances polymorphic transformation from a less stable form to a more thermodynamically stable polymorph, it seems to be a mechanism dependent on a combination of system concentration and crystallization conditions. Furthermore, unsaturated monoglyceride addition can influence the polymorphism of the mixture but will be dependent on fat content and emulsifier concentration. We found opposite effects at the nano- and microscale structures of the fat mixtures depending on processing and emulsifier addition. In general, there was a decrease in the size of the microstructural elements at the microscale with SSHE processing and/or unsaturated emulsifier, while at the nanoscale larger crystalline elements could be observed. 16215

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

(7) Mensink, R. P.; Zock, P. L.; Kester, A. D.; Katan, M. B. Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a metaanalysis of 60 controlled trials. Am. J. Clin. Nutr. 2003, 77, 1146−1155. (8) Marangoni, A. G.; Acevedo, N. C.; Maleky, F.; Co, E.; Peyronel, F.; Mazzanti, G.; Quinn, B.; Pink, D. Structure and functionality of edible fats. Soft Matter 2012, 8, 1275−1300. (9) Mazzanti, G.; Guthrie, S. E.; Sirota, E. B.; Marangoni, A. G.; Idziak, S. H. J. Orientation and phase transitions of fat crystals under shear. Cryst. Growth Des. 2003, 3, 721−725. (10) Mazzanti, G.; Guthrie, S. E.; Sirota, E. B.; Marangoni, A. G.; Idziak, S. H. J. Novel shear-induced phases in cocoa butter. Cryst. Growth Des. 2004, 4, 409−411. (11) Mazzanti, G.; Marangoni, A. G.; Idziak, S. H. J. Modeling phase transitions during the crystallization of a multicomponent fat under shear. Phys. Rev. E 2005, 71, 041607. (12) Kloek, W.; van Vliet, T.; Walstra, P. Mechanical properties of fat dispersions prepared in a mechanical crystallizer. J. Texture Stud. 2005, 36, 544−568. (13) Stauffer, C. E. Properties of emulsifiers. In Fats and Oils; Stauffer, C.E., Ed.; Eagen Press: St. Paul, MN, 1996; pp 29−33. (14) Chrysam, M. M. Table spreads and shortening. In Bailey’s Industrial Oil and Fat Products; Applewhite, T., Ed.; John Wiley and Sons, Inc.: New York, 1985; Vol. 3, pp 42−83. (15) Garbolino, C.; Bartoccini, M.; Floter, E. The influence of emulsifiers on the crystallisation behaviour of a palm oil-based blend. Eur. J. Lipid Sci. Technol. 2005, 107, 616−626. (16) Katsuragi, T. Interactions between surfactants and fats. In Physical Properties of Fats, Oils, and Emulsifiers; Widlak, N., Ed.; AOCS Press: Champaign, IL, 1999; pp 211−219. (17) Litwinenko, J. W.; Singh, A. P.; Marangoni, A. G. Effects of glycerol and Tween 60 on the crystallization behavior, mechanical properties, and microstructure of a plastic fat. Cryst. Growth Des. 2004, 4, 161−168. (18) Folch, J.; Lees, M.; Sloane Stanley, G. H. A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem. 1957, 226, 497−509. (19) Tang, D. M.; Marangoni, A. G. Quantitative study on the microstructure of colloidal fat crystal networks and fractal dimensions. J. Am. Oil Chem. Soc. 2006, 83, 377−388. (20) West, A. R. Solid State Chemistry and Its Applications; John Wiley & Sons: Chichester, West Sussex, England, 1984. (21) Acevedo, N. C.; Marangoni, A. G. Characterization of the nanoscale in triacylglycerol crystal networks. Cryst. Growth Des. 2010, 10, 3327−3333. (22) Dibildox-Alvarado, E.; Neves Rodrigues, J.; Gioielli, L. A.; ToroVazquez, J. F.; Marangoni, A. G. Effects of Crystalline Microstructure on Oil Migration in a Semisolid Fat Matrix. Cryst. Growth Des. 2004, 4, 731−736. (23) List, R.; Mounts, T. L.; Orthoefer, F.; Neff, W. E. Margarine and shortening oils by interesterification of liquid and trisaturated triglycerides. J. Am. Oil Chem. Soc. 1995, 72, 379−382. (24) Lee, J. H.; Akoh, C. C.; Lee, K. T. Physical properties of trans free bakery shortening produced by lipase-catalyzed interesterification. J. Am. Oil Chem. Soc. 2008, 85, 1−11. (25) Ribeiro, A. P. B.; Grimaldi, R.; Gioielli, L. A.; Gonçalves, L. A. G. Thermal behavior, microstructure, polymorphism, and crystallization properties of zero trans fats from soybean oil and fully hydrogenated soybean oil. Food Res. Int. 2009, 42, 401−410. (26) O’Brien, R. D. Fats and oils − formulating and processing for applications; CRC Press: New York, 2004. (27) Sato, K. Crystallization behavior of fats and lipids: a review. Chem. Eng. Sci. 2001, 56, 2255−2265. (28) Andersen, A. J. C.; Williams, P. N. Margarine, 2nd revised ed.; Pergamon Press: Oxford, U.K., 1965; pp 1−7. (29) Marangoni, A. G.; Narine, S. S. Elasticity of fractal aggregate networks: Mechanical arguments. In Crystallization and solidification properties of lipids; Widlak, N., Hartel, R., Narine, S., Eds.; AOCS Press: Champaign, IL, 2001; pp 153−159.

The contrasting trends observed between the nano- and mesoscale is proof that fat crystals grow by aggregation and there is no guarantee that the different length scales will behave in a similar fashion. Processing conditions can affect and determine the size of the agglomerates at the microscale and the primarily crystal size at the nanoscale. However, based on the opposing tendencies observed, it is not possible to say at the present time which length scale is more closely related to the functional properties of the system. Therefore, it is very important to take into account both structural levels when designing a process. In this work, we also demonstrated that crystallization under laminar shear and unsaturated emulsifier addition improved the mechanical properties and oil binding capacity of the fat blends and therefore can help the enhancement of the functional properties of the final product. At this stage of study, we are unable to specify the most appropriate processing conditions for FHSO in SO blends. However, we have demonstrated that SSHE processing and unsaturated monoglyceride addition is a good choice for the functionalization of a shortening.

■

ASSOCIATED CONTENT

S Supporting Information *

Polarized light microscopy images of FHSO:SO blends before and after the incorporation of UM in concentrations of 3 and 5% and crystallized under static conditions and with a scraped surface heat exchanger crystallizer (SSHE). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +1-515-294-3011. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors acknowledge The Natural Sciences and Engineering Research Council of Canada and Advanced Foods and Materials network for the financial support. The Brazilian National Council of Technological and Scientific Development ́ (CNPq − Conselho Nacional de Desenvolvimento Cientifico e Tecnológico) provided a postdoctoral fellowship for Dr. Jane M. Block. We also wish to thank Professor David Ma (Department of Human Health and Nutritional Sciences, University of Guelph) for his analysis of the fatty acid composition of the oil mixtures.

■

REFERENCES

(1) Erickson, D. R. Practical handbook of soybean processing and utilization; American Oil Chemists’ Society: Champaign, IL, 1995. (2) Trans Fat Task Force. TRANSforming the Food Supply − Report of the Trans Fat Task Force. Minister of Health 2006. www. healthcanada.ca/transfat. (3) Institute of Medicine. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein and Amino Acids; National Academies Press: Washington, DC, 2002. (4) Who. Diet, Nutrition and the Prevention of Chronic Diseases. WHO Technical Series Report 916, Geneva, 2003. (5) Health Canada. Trans Fat Monitoring Program−Data Tables for the First Set of Monitoring Data. Minister of Health, 2007. www.hc-sc. gc.ca/fn-an/nutrition/gras-trans-fats/tfa-age_e.html. (6) Bonanome, A.; Grundy, S. M. Effect of dietary stearic acid on plasma cholesterol and lipoprotein levels. N. Engl. J. Med. 1988, 318, 1244−1248. 16216

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217

Langmuir

Article

(52) Hagemann, J. W. Thermal behavior and polymorphism of acylglycerides. In Crystallization and Polymorphism of Fats and Fatty Acids; Garti, N., Sato, K., Eds.; Marcel Dekker Inc.: New York, 1988; pp 9−95. (53) Rousseau, D.; Marangoni, A. G. The effects of interesterification on physical and sensory attributes on butterfat and butterfat-canola oil spreads. Food Res. Int. 1999, 31, 381−388. (54) Nederveen, C. J. Dynamic mechanical behavior of suspensions of fat particles in oil. J. Colloid Sci. 1963, 18, 276−291. (55) Wright, A. J.; Scanlon, M. G.; Hartel, R. W.; Marangoni, A. G. Reological properties of milk fat and butter. J. Food Sci. 2001, 66, 1056−1071. (56) Bremer, L. G. B.; vanVliet, T.; Walstra, P. Theoretical and experimental study of the fractal nature of the structure of acid casein gels. J. Chem. Soc., Faraday Trans. 1 1989, 85, 3359−3372. (57) ASCE (American Society of Civil Engineers), Manual 40Ground Water Management, 1985. Netlink: http://or.water.usgs.gov/ projs_dir/willgw/glossary.html#p (accessed June 2005). (58) Haighton, A. J. The measurement of the hardness of margarine and fats with cone penetrometers. J. Am. Oil Chem. Soc. 1959, 36, 345−348. (59) Chrysam, M. In Bailey’s Industrial oil and fat products; Applewhite, T. H., Ed.; John Wiley & Sons: New York, 1985; Vol. 3, pp 41−126. (60) Garti, N. Food emulsifiers: Structure-reactivity relationships, design, and applications. In Physical Properties of Lipids; Marangoni, A. G., Narine., S. S., Eds.; Marcel and Dekker, Inc.: New York, 2002; pp 265−340.

(30) Narine, S. S.; Marangoni, A. G. Structure and mechanical properties of fat crystal networks. Adv. Food Nutr. Res. 2002, 44, 33− 145. (31) deMan, J. M.; deMan, L. Polymorphism and texture of fats. In Crystallization and solidificaton properties of lipids; Widlak, N., Hartel, R., Narine, S. S., Eds.; AOCS Press: Champaign, IL, 2001; pp 225− 235. (32) Smith, K. W.; Bhaggan, K.; Talbot, G.; van Malssen, K. F. Crystallization of fats: influence of minor components and additives. J. Am. Oil Chem. Soc. 2011, 88, 1085−1101. (33) Schlichter-Aronhime, J.; Garti, N. Mechanistic considerations of polymorphic transformations of tristearin in the presence of emulsifiers. J. Am. Oil Chem. Soc. 1987, 64, 529−533. (34) Maleky, F.; Marangoni, A. G. Process development for continuous crystallization of fat under laminar shear. J. Food Eng. 2008, 89, 399−407. (35) Manley, D. J. Technology of Biscuits, Crackers and Cookies; Ellis Horwood: Chichester, England, 1983; pp 61−79. (36) Litwinenko, J. W.; Rojas, A. M.; Gerschenson, L. N.; Marangoni, A. G. Relationship between crystallization behavior, microstructure, and mechanical properties in a palm oil-based shortening. J. Am. Oil Chem. Soc. 2002, 79, 647−654. (37) Martini, S.; Herrera, M. L.; Hartel, R. W. Effect of processing conditions on microstructure of milk fat fraction/sunflower oil blends. J. Am. Oil Chem. Soc. 2002, 79, 1055−1062. (38) Herrera, M. L.; Hartel, R. W. Effect of processing conditions on crystalization kinetics of a milk fat model system. J. Am. Oil Chem. Soc. 2000, 77, 1177−1187. (39) Acevedo, N. C.; Block, J. M.; Marangoni, A. G. Critical laminar shear-temperature effects on the nano- and mesoscale structure of a model fat and its relationship to oil binding and rheological properties. Faraday Discuss. 2012, 158, 171−194. (40) Himawan, C.; Starov, V. M.; Stapley, A. G. F. Thermodynamic and kinetic aspects of fat crystallization. Adv. Colloid Interface Sci. 2006, 122, 3−33. (41) Zhou, Y.; Hartel, R. W. Phase behavior of model lipid systems: Solubility of high-melting fats in low-melting fats. J. Am. Oil Chem. Soc. 2006, 83, 505−511. (42) Vereecken, J.; Foubert, I.; Meeussen, W.; Lesaffer, A.; Dewettinck, K. Fat structuring with partial acylglycerols: Effect on solid fat profiles. Eur. J. Lipid Sci. Technol. 2009, 111, 259−272. (43) Omonov, T. S.; Bouzidi, L.; Narine, S. S. Quantification of oil binding capacity of structuring fats: A novel method and its application. Chem. Phys. Lipids 2010, 163, 728−740. (44) Van Aken, G. A.; Visser, K. A. Firmness and crystallization of milk fat in relation to processing conditions. J. Dairy Sci. 2000, 83, 1919−1932. (45) Chaiseri, S.; Dimick, P. S. Dynamic crystallization of cocoa butter. II. Morphological, thermal, and chemical characteristics during crystal growth. J. Am. Oil Chem. Soc. 1995b, 72, 1497−1504. (46) Acevedo, N. C.; Marangoni, A. G. Towards nanoscale engineering of triglyceride crystal networks. Cryst. Growth Des. 2010, 10, 3334−3339. (47) Maleky, F.; Smith, A.; Marangoni, A. G. Laminar shear effects on crystalline alignments and nanostructure of a triacylglycerol crystal network. Cryst. Growth Des. 2011, 11, 2335−2345. (48) Rousseau, D.; Forestiere, K.; Hill, A. R.; Maragoni, A. G. Restructuring butter fat through blending and chemical interesterification. I. Melting behavior and triacylglycerol modifications. J. Am. Oil Chem. Soc. 1996, 73, 963−972. (49) Shi, Y.; Liang, B; Hartel, R. W. Correlation of rheological and microstructural properties in a model lipid system. J. Am. Oil Chem. Soc. 2008, 85, 397−404. (50) Narine, S. S.; Marangoni, A. G. Relating structure of fat crystal networks to mechanical properties: a review. Food Res. Int. 1999, 32, 227−248. (51) Maragoni, A. G.; Rogers, M. A. Structural basis for the yield stress in plastic disperse systems. Appl. Phys. Lett. 2003, 82, 3239− 3241. 16217

dx.doi.org/10.1021/la303365d | Langmuir 2012, 28, 16207−16217