Analysis of co-crystallized free phytosterols with triacylglycerols as a functional food ingredient

Food Research International 85 (2016) 104–112

Contents lists available at ScienceDirect

Food Research International journal homepage: www.elsevier.com/locate/foodres

Analysis of co-crystallized free phytosterols with triacylglycerols as a functional food ingredient Nuria C. Acevedo ⁎, Danielle Franchetti Department of Food Science and Human Nutrition, Iowa State University, Ames, IA 50011, United States

a r t i c l e

i n f o

Article history: Received 12 January 2016 Received in revised form 2 April 2016 Accepted 9 April 2016 Available online 18 April 2016 Keywords: Free phytosterols Co-crystallization Fully hydrogenated soybean oil Oil loss Rheological properties

a b s t r a c t This research focuses on the analysis of mixtures of free phytosterols (FPSs) with fully hydrogenated soybean oil (FHSO):soybean oil (SO) mixtures as a potential zero-trans substitute for various types of shortenings. Oil binding capacity as well as the thermal, rheological and structural properties of FHSO:SO blends containing 0, 20 and 25 wt.% β-sitosterol or stigmasterol were investigated in this study. Differential interference contrast (DIC) microscopy and wide angle X-ray diffraction (WAXRD) revealed that co-crystallization of FPSs with FHSO:SO blends occurred. Polymorphic forms were characterized as a mixture of β′ and β for all samples. The addition of FPSs decreased oil loss (OL) of FHSO:SO samples. Melting profiles of the prepared FPS–TAG (triacylglycerol) blends were extended to higher temperatures compared to a commercial shortening. Rheological properties were comparable to those of commercial puff pastry shortening suggesting that FPS–TAG blends may be acceptable for bakery applications. FPSs co-crystallized with FHSO and SO may be a suitable trans-fat free substitute for a number of types of shortening, including puff pastry shortening. The manufacturing of co-crystallized /FPS-TAG matrices will possibly bring large economic benefits as their functionalization can potentially be achieved by using existing simple shear processing. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Phytosterols are found in all plant foods, with the highest concentrations occurring in vegetable oils (Ostlund, 2002). The general term ‘phytosterols’ describes plant-derived sterols and stanols with a chemical structure related to cholesterol and having a different side chain configuration (Cantrill, 2008; Spitzer & Maggini, 2013). It has been recognized that high consumption of plant sterols can reduce serum total and LDL cholesterol concentrations in humans (Ostlund, 2002; Piironen, Lindsay, Miettinen, Toivo, & Lampi, 2000); therefore reducing the risk of cardiovascular disease. Phytosterols displace cholesterol from mixed micelles decreasing cholesterol absorption in the intestine (Nissinen, Gylling, Vuoristo, & Miettinen, 2002). Recent research has suggested that plant sterols may have additional biological activities different than cholesterol lowering, such as anti-inflammatory and anticancer effects, and changes in cell membrane properties, among others (Awad, Roy, & Fink, 2003; Halling & Slotte, 2004; Navarro, De las Heras, & Villar, 2001; Ratnayake et al., 2000; von Holtz, Fink, & Awad, 1998). In previous studies, an intake of 1–3 g/day of plant sterols has been shown

Abbreviations: FHSO, fully hydrogenated soybean oil; SO, soybean oil; FPS, free phytosterol; β-Sit, β-sitosterol; Stig, stigmasterol; DSC, differential scanning calorimetry; NMR, nuclear magnetic resonance; WAXRD, wide angle X-ray diffraction; DIC, differential interference contrast; OL, oil loss. ⁎ Corresponding author. E-mail address: [email protected] (N.C. Acevedo).

http://dx.doi.org/10.1016/j.foodres.2016.04.012 0963-9969/© 2016 Elsevier Ltd. All rights reserved.

to produce a 10–15% reduction in LDL cholesterol (Kritchevsky & Chen, 2005; Kuhlmann et al., 2005; Ling & Jones, 1995). Nevertheless, the typical Western diet today contains only 150–400 mg/d (Ostlund, 2002). It is evident that a greater intake of phytosterols is necessary to compete for absorption against cholesterol and thus, supplemental doses may be required. Several studies have suggested that the LDL-cholesterollowering effect of plant sterols reaches a maximum at doses of 2–3 g/ day (Katan et al., 2003; Law, 2000; Musa-Veloso, Poon, Elliot, & Chung, 2011; Ras, Geleijnse, & Trautwein, 2014). For that reason, many health authorities have included 2 g/day PS as an optimal dose in their diet and lifestyle guidelines to manage hypercholesterolemia (American Heart Association Nutrition Committee et al., 2006; International Atherosclerosis Society (IAS) Executive Board, 2013; Gylling et al., 2014). The enrichment of foods with PSs is challenging from a technological and food quality standpoint, since PSs are insoluble in water and poorly soluble in dietary fats (Salo & Wester, 2005). Therefore, for commercial use, phytosterol and stanol powders are esterified with fatty acids in vegetable oils; a process that allows manipulation of the physical properties of these high melting powders. The characteristics of the esters are similar to edible fats and oils and can be classified as liquid or semi-liquid (Cantrill, 2008). The phytosterol fatty acid esters are incorporated into processed foods such as spreads, juices, oils, and other foods. Therefore, until recently, the majority of studies involving phytosterols or phytostanols have been strictly on the use of esterified phytosterols or phytostanols. There have been few studies addressing the

N.C. Acevedo, D. Franchetti / Food Research International 85 (2016) 104–112

use of free (non-esterified) phytosterols and phytostanols in commercial foods due to their limited solubility and disputes on bioavailability as compared to their esterified counterparts (Hayes, Pronczuk, & Perlman, 2004). However, researchers have reported that when free phytosterols are effectively heated and then recrystallized in fat upon cooling, they become bioavailable and therefore effective in reducing cholesterol absorption (Hayes et al., 2004; Perlman, Hayes, & Pronczuk, 2006). The utilization of non-esterified phytosterols in fat, by recrystallization, would lower the cost and increase convenience of processing to make previously insoluble non-esterified phytosterols bioavailable and soluble in dietary fats (Hayes et al., 2004). Previous strategies of phytosterol utilization include ultrafine powders, chemically modified esterified phytosterols, emulsified phytosterols, and phytosterols in water–oil microparticulate suspensions which are very susceptible to oxidation (Perlman et al., 2006). Today, there is still a need to develop a stable solid or semisolid food product incorporating free plan sterols. Hence, the justification of this research relies, in part, on the health benefits of phytosterol supplementation as a solid or semisolid ingredient, specifically chemically unmodified free phytosterols to reduce production cost in industry, as well as increase convenience in processing. On the other hand, due to current recommendations to eliminate trans fats, there has been a growing demand in the last few years to find appropriate semi-solid fat substitutes that do not increase the risk for CVD (Hunter, Zhang, & Kris-Etherton, 2010). Fully hydrogenated fats have been the focus of studies as a probable replacement of trans fatty acids containing fats (Acevedo, Block, & Marangoni, 2012; Acevedo & Marangoni, 2014; Guedes et al., 2014; Ribeiro, Grimaldi, Gioielli, dos Santos, Cardoso & Gonçalves, 2009; Ribeiro, Grimaldi, Gioielli & Gonçalvesa, 2009). Therefore, the purpose of this study was to prepare and evaluate physical blends of fully hydrogenated soybean oil (FHSO) and liquid soybean oil (SO) enriched with free phytosterols, with a view to assessing prospective applications in food products. Our novel approach encompasses the manufacture of free phytosterols (FPSs) and TAG based on their co-crystallization. We hypothesize that this strategy will effectively entrap and protect freephytosterols in fat processed foods. FPS–TAG blends with 20 wt.% and 25 wt.% β-sitosterol or stigmasterol content were prepared and analyzed in order to meet the recommended daily intake to decrease blood cholesterol with one serving of the blend produced. For example the consumption of one bakery product would be sufficient to achieve the desired effect. In the present study a commercial puff pastry shortening, which contains large amount of trans fats was specifically selected as an example of the physico-chemical properties required for use in food manufacturing.

105

the mixture up to ~ 180 °C and agitating with a Caframo Real Torque Digital overhead stirrer (Ontario, Canada) at 200 rpm. Once the mixtures were clear, they were held at 180 °C for additional 20 min to ensure that the crystal memory was erased. Subsequently, all samples were cooled at room temperature until ~ 20 °C was reached and once crystallization was complete they were stored at 4 °C until use in analysis. 2.3. Wide angle X-ray diffraction (WAXRD) Wide angle X-ray diffraction (WAXRD) patterns of crystallized FPS, FPS–TAG blends, and control fat blends were measured using a Rigaku Ultima (Rigaku, Japan) IV X-ray diffractometer. The operating conditions during experiments were 44 mA and 40 kV. The angular range using a 10 mm slit was from 1 to 30° (2θ) with steps of 1°, and the measuring time was 1 min/step. XRD patterns were analyzed with MDI Jade 9.0 software (Rigaku, Japan). In this study, three replicates of each sample were performed. 2.4. Differential scanning calorimetry (DSC) The thermal properties of the samples were measured by differential scanning calorimetry (DSC) using a Perkin Elmer Diamond DSC (Shelton, CT, USA). Heat flow calibration was made by reference to the known melting enthalpy of indium (purity N 90%). Temperature calibration was made with zinc (purity N90%). Approximately 10 mg of sample was placed in aluminum pans and sealed hermetically, an empty pan served as a reference. All measurements were performed at a heating rate of 10 °C/min. Thermograms were analyzed using Pyris Series Diamond DSC 9.0 software (Shelton, CT, USA). The peak melting temperature (Tm) and the enthalpy of melting (ΔHm) were determined. The reported data corresponds to the average and standard deviation (STD) of three replicates for each sample. 2.5. Proton-nuclear magnetic resonance (1H-NMR) Solid fat content (SFC) was analyzed by proton nuclear magnetic resonance (1H-NMR) by using a Bruker MiniSpec Bench Top NMR (Billerica, MA, USA). Crystallized samples stored at 4 °C were introduced into NMR glass tubes, then incubated for 30 min at 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100 °C to allow for a homogenous distribution of temperature at the time of measurement. The reported data correspond to the average and standard deviation (STD) of three replicates for each sample.

2. Materials and methods

2.6. Oil loss determination (OL)

2.1. Materials

Once crystallized, oil loss (OL) experiments were performed according to previously described techniques (Acevedo et al., 2012; Dibildox-Alvarado, Rodrigues, Gioielli, Toro-Vazquez, & Marangoni, 2004). Blends were molded into PVC molds of 35 mm diameter and 3.2 mm thickness to form disks that were then transferred to filter papers (Whatman #1, 125 mm diameter). The amount of oil lost over time was determined by the difference in weight of the filter papers before and after placing the disk on the paper for the designated time at 20 °C. A “blank” filter paper was included in all experiments to account for differences in possible humidity of the storage environment. Filter papers were large enough to avoid paper saturation with oil during the experiment. The reported data correspond to the average and standard deviation of at least five replicates, each separate disk on an individual filter paper. Oil loss (%) was calculated according to previously described technique as follows:

Fully hydrogenated soybean oil (FHSO) and liquid soybean oil (SO) were generously donated from ADM oils (Decatur, IL). β-Sitosterol powder (purity ≥ 70%) was obtained from Sigma-Aldrich (St. Louis, MO). Stigmasterol powder (purity N90%) was obtained from TCI America (Portland, OR). Super Bowl® puff pastry shortening consisting of partially hydrogenated soybean oil and cottonseed oils with artificial flavor and artificial colors was generously provided by Stratas Foods (Memphis, TN). 2.2. Blend preparation All samples consisted of 48 wt.% solids and 52 wt.% liquid oil. For free phytosterol-triacylglycerol (FPS–TAG) blends, the solid material consisted of either 20 wt.% or 25 wt.% FPS (β-sitosterol or stigmasterol) and 28 wt.% or 23 wt.% FHSO (totaling 48 wt.%); the 52 wt.% liquid portion consisted of soybean oil. The control sample contained 0 wt.% FPS, 48 wt.% FHSO, and 52 wt.% soybean oil. Blends were prepared by heating

OLð% Þ ¼

wt:paperðX hÞ  wt:paper ð0 hÞ  100: wt:paper ð0 hÞ

106

N.C. Acevedo, D. Franchetti / Food Research International 85 (2016) 104–112

OL rate (g/h) was calculated as the average of the maximum slope of the OL vs time curves obtained. 2.7. Differential interference contrast microscopy (DIC) Differential interference contrast (DIC) microscopy was used to observe the FPS–TAG blend microstructure. Approximately 0.01 g of the melted sample was placed on a heated glass slide and a heated glass cover was carefully laid over the sample to remove air and homogenously spread the sample. After cooling at room temperature 20 °C, slides were stored at 4 °C until analysis. Samples were imaged using an Olympus BX53 System Microscope (Tokyo, Japan) with polarized light (X-Cite 120 LED) and equipped with an Olympus Q Imaging camera. All images were obtained using a 40× objective lens. Autoexposure of the camera was adjusted according to the sample. CellSens Dimension 1.9 software (Tokyo, Japan) was used to acquire focused images. At least twenty-five images in total were captured from each of the 9 replicates. 2.8. Small deformation rheology Samples were carefully spread into the wells of polyvinylchloride (PVC) molds of 3.2 mm thick and 35 mm in diameter and stored at 4 °C for 24 h prior to analysis in order to minimize changes in the matrix due to handling. A polyvinyl chloride (PVC) cylindrical plunger (35 mm diameter, 62 mm height) was used to push the disk out of the mold. Rheological measurements were obtained using a HAAKE RS 150 RheoStress rheometer (Paramus, NJ, USA). A 35 mm diameter serrated stainless steel plate (PP35Ti) was used in the analysis. Each sample was kept at a constant temperature 20 °C controlled by the base of the TC 81 Peltier temperature control system through a HAAKE F6 (Paramus, NJ, USA) water bath. Oscillatory stress sweeps were performed within the range of 50 to 2000 Pa (with a frequency of 1 Hz) inside the linear viscoelastic region (LVR). Normal force was set at 3 N for all samples. Sandpaper (grade 60) was attached to the base of the rheometer to prevent slippage of samples during analysis. The yield stress (σ⁎) values (Pa), elastic (G′) and storage (G″) modulus values (MPa) were determined from the stress sweep curves. The reported data are the average and standard deviation of eight individual replicates from each sample.

Table 1 Melting temperature (Tm) and enthalpy of melting (ΔH) of the control sample, pure phytosterols and their blends. Sample

Tm (°C) ± SD

ΔH (J/g) ± SD

48:52 FHSO:SO 20:28:52 β-Sit:FHSO:SO 25:23:52 β-Sit:FHSO:SO 20:28:52 Stig:FHSO:SO 10:10:28:52 β-Sit:Stig:FHSO:SO β-Sitosterol Stigmasterol

65.30 ± 2.50a 63.02 ± 2.00a 63.05 ± 0.17a 65.36 ± 0.72a 63.07 ± 0.37a 137.78 ± 0.79b 169.99 ± 1.54c

71.88 ± 6.90a 48.37 ± 6.43b 41.18 ± 2.74b 50.51 ± 7.37b 86.64 ± 8.14a 43.27 ± 4.34b 72.80 ± 4.64a

Different letters within columns indicate statistical difference (P b 0.05).

overlapping peaks can be observed during melting of the control blend which is indicating the presence of two coexisting polymorphic forms; probably β and β′ (Fig. 1). However, the decrease in intensity of the lower temperature peak observed with the addition of FPS suggests that FPS had an effect on the formation of the triacylglycerol crystal lattice. It seems that the disturbance of the network introduced by the presence of FPS enhanced a polymorphic transformation towards a more stable form. The occurrence of this sample's polymorphism will be confirmed by powder X-ray diffraction and discussed in the following section. FPS appears to have affected the enthalpy of melting (Δ H) of the FHSO:SO mixture (Table 1). The addition of 20 wt.% β-sitosterol or stigmasterol reduced the total ΔH by 32.71% and 29.73% respectively when compared to the control. These results may be indicating interference by FPS in the thermodynamic system as FPS served as a crystal structure modifier or as a promoting agent for polymorphism modification. Surprisingly, when both β-sitosterol and stigmasterol were conjointly added to the FHSO:SO fat blends (10:10:28:52 β-Sit:Stig:FHSO:SO), ΔH was not significantly different from the control. In FPS–TAG blends, no peaks were observed at temperatures corresponding to the Tm of FPS powders suggesting co-crystallization of the species (Fig. 1); i.e. a certain amount of FPS is integrated in a dominating triglyceride crystal lattice of FHSO:SO. Phytosterols have a chemical structure similar to that of cholesterol, though they contain

2.9. Statistical analysis GraphPad Prism 5 software (GraphPad Software, Inc., San Diego, CA) was used to process data. Reported values correspond to means and standard deviations of the measurements. Statistical analysis was carried out by one-way ANOVA (P b 0.005) with Tukey's multiple comparisons as a post-test (P b 0.05). 3. Results and discussion 3.1. Analysis of thermal properties DSC is a well-established method conducted to assess the thermal properties of fats and oils during heating and cooling. Thermograms showed a melting point Tm of 65.30 ± 2.50 °C for the 48:52 FHSO:SO (control sample) which is in agreement with previous studies on FHSO:SO blends (Acevedo et al., 2012). All FPS–TAG blend samples, i.e. 20:28:52 β-Sit:FHSO:SO, 25:23:52 β-Sit:FHSO:SO, 20:28:52 Stig:FHSO:SO and 10:10:28:52 β-Sit:Stig:FHSO:SO also showed Tm within the 63.02– 65.36 °C range probably due to the presence of similar solid–liquid ratios (Table 1); indicating that the presence of FPS did not have a significant effect on the TAG sample Tm. DSC analysis also showed a Tm for β-sitosterol powder of 137.78 ± 0.79 °C in agreement with Vaikousi, Lazaridou, Biliaderis, and Zawistowski (2007) and for stigmasterol powder of 169.99 ± 1.54 °C (Table 1). It is worth noting that two

Fig. 1. DSC thermograms of pure FPS and FHSO:SO fat blends with and without FPS.

N.C. Acevedo, D. Franchetti / Food Research International 85 (2016) 104–112

107

an additional hydrophobic carbon chain (Rozner & Garti, 2006). As a result, it is possible that the formation of FFS–TAG mixed crystals due to partial structural similitudes between FFS and fatty acid molecules that allow their molecular packing. It has been previously stated in the literature that minor lipids, such as diacylglycerols, monoacylglycerols, free fatty acids, phospholipids, glycolipids and sterols can affect every stage of the crystallization process, from the nucleation to the post-crystallization steps (ToroVazquez, Vargas, Dibildox-Alvarado, & Charó-Alonso, 2005). Furthermore, the nature of the interactions between minor lipids, and the growing triacylglycerol crystals can result in a structural competitive effect or the permanent incorporation in the crystalline structure, which inhibits or promotes growth (Metin & Hartel, 2005; Ribeiro et al., 2015). However, whether minor lipids exert a promoting or inhibitory effect on the nucleation or crystal growth stages is not yet fully understood (Foubert, Vanhoutte, & Dewettinck, 2004; Metin & Hartel, 2005; Wright & Marangoni, 2002). The absence of a defined melting enthalpy at temperature within the range of pure FPS indicates incorporation of FPS, at least in the concentrations analyzed, in the crystalline structure. A few studies have highlighted the fact that that some types of fat crystallization modifiers suffer co-crystallization with TAGs because of the similarity between their chemical structures, i.e. they associate with the TAG molecules by their hydrophobic groups. These additives would exert their effects mainly during the stages of nucleation, polymorphic transition and crystal growth leading to modifications of the physical properties of fats including crystal size, solid fat content and microstructure (Garti, 2002). 3.2. Polymorphic form characterization WAXRD patterns of the control consisting of 48:52 FHSO:SO and all FPS–TAG blend samples showed the presence of both polymorphs, β′ and β (Fig. 2) (Sato, 2001). The results of this study agree with those of Acevedo et al. (2012), stating that a mixture of β′ and β could be detected in a binary mixture of FHSO and SO by this method (Fig. 2a). Additionally, it can be confirmed that both endothermic peaks observed in the DSC thermograms of the control sample correspond to β′ and β polymorphs (Fig. 1). On the other hand, XRD patterns from the FPS– TAG blends showed greater characteristics of β polymorphism than of β′ (Fig. 2b, c). This result is in agreement with that from the DSC determination, where it can be seen that the addition of β-sitosterol or stigmasterol caused a significant decrease or even disappearance of the first endothermic peak corresponding to β′ polymorph (Fig. 1). The addition of 20 wt.% FPS to the molten FHSO:SO mixture, prior to the crystallization process, caused a significant change in the stability of the β′ polymorph and only traces of β′ from can be detected by X-ray diffraction. As stated before, FFS served as crystal structure modifiers or as accelerating agents for the β′ to β modification. Polymorphism is predominately determined by the rate of nucleation, being governed by thermodynamic and kinetic influences (Sato, 2001). When nucleation is induced under sizeable kinetic factors, e.g., supercooling or supersaturation the metastable form nucleates first prior to the most stable form. This is the Ostwald step rule on the law of successive reactions (Mutaftschiev, 1993). However, when some other external influences (pressure, temperature fluctuation, ultrasonic stimulation, template, seeding, etc.) are applied this “law” may be broken (Rousseau, Hodge, Nickerson, & Paulson, 2005; Sato, 2001). It appears that the addition of FPS as an external influence into the FHSO:SO system is prompting the polymorphism transition from the metastable form β′ to most stable β. The addition of FPS may be kinetically allowing the molecules to rearrange in a crystalline stable form. It is most likely that FPS molecules gave rise to a reduced steric hindrance favorable to the accommodation or of triacylglycerols resulting in an optimized spatial rearrangement of the molecules; thus, X-ray diffraction pattern shows an evident predominance of the β polymorphic form upon addition of FPS to the fat matrices (Fig. 2b, c).

Fig. 2. Wide angle X-ray diffraction (WAXRD) patterns of control (a), β-sitosterol and its mixture with FHSO:SO (b), and stigmasterol and its mixture with FHSO:SO (c). d-Values of possible new structures formed in free phytosterol fat blends as compared to free phytosterols are indicated with arrows (b and c).

β-Sitosterol d-spacing values (lattice spacing values) were in agreement with previous research (Christiansen, Rantanen, von Bonsdorff, Karjalainen, & Yliruusi, 2002; von Bonsdorff-Nikander, Karjalainen, Rantanen, Christiansen, & Yliruusi, 2003). However, there is a lack of previous studies performed to assess the nanostructure of stigmasterol using X ray diffraction which makes it challenging to compare and discuss results within the frame of previous works performed. Nevertheless, it is important to mention that the XRD patterns of FPS–TAG blends presented additional d-values absent in the control and pure FPS powder XRD patterns (Fig. 2b and c). These additional d-values, indicated with arrows, could be attributed to the presence of new structures formed by co-crystallization of FPS with triacylglycerols, also suggested by DSC thermograms (Fig. 1). However, further studies need to be carried out to fully elucidate the structure of these matrices.

108

N.C. Acevedo, D. Franchetti / Food Research International 85 (2016) 104–112

3.3. Microstructure analysis Differential interference contrast (DIC) microscopy was used for microstructural analysis due to the polarized light microscope's ability to exploit the high contrast between the solid and liquid fraction in fats and oils (Murphy et al., 1998, 2003). Fig. 3 shows digitalized images of samples thermostatically crystallized at 20 °C. The microstructure of the control sample consisted of aggregated spherulites, similar to previous research where comparable fat blends were analyzed (Acevedo et al., 2012; Bouzidi, Omonov, Garti, & Narine, 2013). The morphology of β-sitosterol and stigmasterol was represented by shiny needle-like crystals, analogous to previous research on phytosterols (AlHasawi & Rogers, 2013). From the FPS–TAG samples, a microstructural network characterized by mixed crystals, that differs from the crystalline structure of either β-sitosterol or stigmasterol can be observed (Fig. 3). These results agree with those obtained by DSC (Fig. 1) and WAXRD (Fig. 3) which indicated that phytosterols induced a crystal lattice disturbance in the growing TAG network. For instance, the addition of stigmasterol into the FHSO:SO matrix caused morphological differences in the aggregated spherulite crystalline structure; long needle-like crystals arose as indicated in the powder stigmasterol sample. In the case of βsitosterol-FHSO:SO, the samples presented well developed shiny flower-like crystals, a different morphology than that observed in the pure components. Only the addition of β-sitosterol at the highest proportion (25 wt.%) seems to exhibit a trivial presence of separated needle-like crystals resembling those observed in the individual FPS component. In contrast, Rodrigues, Torres, Mancini-Filho, and Gioielli (2007) reported that the addition of phytosterol esters (PE) into milk fat (MF) did not seem to drastically modify the structure of MF crystals, nonetheless the PE still showed a small presence within the MF crystal network The possible reason for some of the differences found is the added fatty acid chain to the phytosterol molecules via esterification. Thus, phytosterol esters exhibit lower melting points and higher oil solubility than the non-esterified phytosterols (FPSs) used in this work. Additionally, it is worth mentioning that the inhabitance of both FPS, β-sitosterol and stigmasterol, in the 10:10:28:52 β-Sit:Stig:FHSO:SO sample displayed a branch-like crystal structure differing from both the β-sitosterol and stigmasterol structures as well as from the TAG mixture. These observations are in excellent agreement with the DSC (Fig. 1) and WAXRD determination (Fig. 2) pointing out cocrystallization of FPS with triglycerides from melt at the studied weight ratio. 3.4. Oil binding capacity Oil binding capacity in this study was analyzed through the measurement of the amount of oil loss (OL) over time and the rate of oil loss, which is a quantified representation of the kinetics of oil loss

(Acevedo et al., 2012; Bouzidi et al., 2013; Dibildox-Alvarado et al., 2004). Furthermore, the initial amount of oil receptive to be lost, usually referred to as propensity for oil loss (POL) was also considered. POL is dependent upon how much oil is experimentally added to the system (Bouzidi et al., 2013). To eliminate this variable, the control and all FPS samples were prepared with equal amounts of POL at 52% total weight of SO. Fig. 4 reports OL values as a function of time while Fig. 5 reports OL rates in %/h. All curves in Fig. 4 showed similar behavior of liquid oil migrating out of the gel network over time. At the initial stage, liquid oil was quickly lost from the gel network. Then, OL decreased gradually until finally reaching a plateau. The 48:52 FHSO:SO control sample presented 42 ± 5% OL after reaching the plateau at about 1000 h (~ 42 days) with an OL rate of 0.19 ± 0.03%/h. While the FPS–TAG blends had OL values ranging from 28 to 46% and OL rates of 0.09–0.14 g/h. Low oil migration values indicated samples with high stability and structural integrity, as samples are able to entrap and retain more oil inside the network for a long period of time (Dibildox-Alvarado et al., 2004). Thus, these results suggest that the addition of FPS has a positive effect on the system's ability to lock oil within the network. The addition of β-sitosterol into the FHSO:SO matrix at 20 wt.% and 25 wt.% decreased the amount of total OL compared to the control by 13.5% and 16.5%, respectively. Surprisingly, for the case of addition of stigmasterol (20:28:52 wt Stig:FHSO:SO) or a mixture of both FPS (10:10:28:52 wt β-Sit:Stig:FHSO:SO), fat blends did not show a significant decrease in the amount of oil lost (P b 0.05) compared to the control. These results were interesting considering that all the samples contained the same total amount of crystalline network and that oil binding capacity in plastic fats is dependent in grand part of this crystalline mass. It is evident that molecular interactions and crystalline structure affected the ability of the matrix to retain oil. The significantly lower OL values of FHSO:SO in the presence of βsitosterol probably reflect a structural improvement induced into the network upon crystallization. However, there is a lack of literature discussing the effects of FPS powders on OL% in FHSO:SO blends and further studies should be conducted to elucidate their role during fat network crystallization. OL% and OL rate analysis was also conducted on commercial puff pastry shortening to ascertain the differences between FPS:FHSO:SO blends and assess the mixtures functionality from the perspective of oil binding capacity. It is possible to note in Fig. 4a and c that the control sample, without FPS, showed significantly higher OL values (~40%) than the puff pastry shortening after long storage at 20 °C (~30%) showing a lower long term oil binding capacity. On the other hand, similar OL% resulted between 20:28:52 wt β-Sit:FHSO:SO (28 ± 2%), 25:23:52 wt βSit:FHSO:SO (34 ± 3%), and puff pastry shortening over time. However, blends containing stigmasterol or its mixture with β-sitosterol lost significantly larger amounts of oil (P b 0.05) than the commercial puff

Fig. 3. DIC microscopy microstructure of FPS powders, FPS–TAG blends and FHSO:SO control.

N.C. Acevedo, D. Franchetti / Food Research International 85 (2016) 104–112

109

Fig. 5. Oil loss (OL) rate (%/h) from slope of OL as a function of time. Letters represent statically significant differences between the values (P b 0.05).

Fig. 4. Oil loss of FHSO:SO control (a), FPS–TAG blends (b) and puff pastry shortening (c) as a function of time.

pastry shortening. Although final OL% values in Stigmasterol-containing blends were higher than the limit set by the commercial shortening, it is worth to bring up that all FPS–TAG blends presented significantly lower (P b 0.05) OL rates (0.09–0.14%/h) than puff pastry shortening (0.52 ± 0.05%/h), suggesting the FPS–TAG blends may release oil at a slower rate, which would be desirable for shelf life (Fig. 5). It is also worth noting that OL% values for all FPS–TAG mixtures were significantly lower than those observed for the commercial shortening up to 336 h (14 days) of storage, after which they continued increasing while OL% in puff pastry reached a plateau (Fig. 4b and c). 3.5. Solid fat content (SFC) analysis by NMR Solid fat content (SFC) analysis was conducted to determine the SFC profiles of all samples and to compare them to a commercial puff pastry shortening (Fig. 6a and b). SFC results from puff pastry shortening were in agreement with previous research where a similar product was studied; for instance ~24% SFC was observed at 20 °C in line with those findings reported by Ghotra, Dyal, and Narine (2002) (Fig. 6b). The control

sample (48:52 FHSO:SO) reached 0% SFC at 65–70 °C, consistent with DSC results previously discussed; however FPS blends did not achieve 0% SFC even at temperatures higher than 65–70 °C (Fig. 6a). This phenomenon could be due the existence of a threshold in the cocrystallization of FPS with FHSO:SO; beyond a certain FPS to FHSO:SO ratio not all of the FPS present may be co-crystallizing. There may be a phase separation of the components where some FPS did not integrate into the FSHO:SO matrix. XRD patterns showing new d-values suggested co-crystallization occurred (Fig. 2). However, NMR indicated that probably a portion of FPS in the matrix is not co-crystallized. A sample in homogeneities as a result of partial phase separation may explain the absence of a high temperature melting peak in the DSC thermograms (Fig. 1). Previous research has been conducted on the addition of phytosterols into a matrix on SFC. However the work encompassed phytosterol esters (PE) blended with milk fat (MF), instead of FPS (Rodrigues et al., 2007). The MF PE blends presented a decrease in SFC when a higher proportion of PE was present, even though PE independently presented higher SFC than MF at all temperatures (Rodrigues et al., 2007). A phenomenon was hypothesized regarding the interaction between MF and PE in that the PE interfered with the formation of a MF crystalline structure, resulting in lower SFC values (Rodrigues et al., 2007). Results obtained in this work are conflicting with those reported by Rodrigues et al. (2007) as samples with FPS showed a statistically significant (P b 0.05) increment in the SFC profile relative to the observed in the control. It is worth mentioning that at 20 °C no differences were observed in SFC values between the FPS blends, even though different types of FPS and different percentages (20 wt.% versus 25 wt.% FPS) were utilized (Fig. 6b). Therefore, it is proposed that the presence of FPS in the fat matrix affected the SFC at room temperature while the type or percentage of FPS studied did not have an effect as shown in Fig. 6b. Again, this may be due to high melting FPS not fully incorporated into the fat crystalline matrix. Similar trends were observed for SFC values within the whole range of temperatures analyzed (Fig. 6a). SFC has been recognized as one of the major factors determining the texture of plastic fats (deMan, deMan, & Blackman, 1989); thus with the aim to evaluate FPS blends functionality in terms of consistency and texture, SFC profiles were compared with a commercial puff pastry shortening used as a reference. As observed in Fig. 6a, the SFC profile of the commercial shortening shows a steady decrease with increasing temperature up to ~60 °C where it reaches 0%. On the other hand, fat blends with and without FPS showed SFC profiles steeper than that of the

110

N.C. Acevedo, D. Franchetti / Food Research International 85 (2016) 104–112

Fig. 6. Solid fat content (SFC) profile (a) and SFC values at 20 °C (b). Letters represent statically significant differences between the values (P b 0.05).

reference shortening. Even though FPS blends had higher profiles than the commercial shortening, FPS samples exhibited a large SFC reduction within the 40–60 °C range. This characteristic would make it a suitable choice to use as a filler fat because these type of products have a SFC profile significantly steeper than that of an all-purpose bakery shortening (Ghotra et al., 2002; Metzroth, 2005). Furthermore, future studies analyzing different FPS to TAG proportions as well applying different processing conditions will be conducted to address this issue. 3.6. Analysis of rheological properties by small deformation oscillatory rheology Small deformation rheology with an oscillatory stress sweep was conducted in order to maintain a linear viscoelastic region (LVR) where measurements of G′ and G″ were obtained. G′ is the storage modulus which represents the elastic behavior of the sample (Marangoni & Wesdorp, 2013). G′ has been related to the strength or hardness and has also been a strong indicator of the SFC of the fat crystal network (Acevedo et al., 2012; Marangoni & Wesdorp, 2013). No significant differences were observed in the G′ values of the blends except for commercial puff pastry shortening and 25:23:52 wt β-Sit:FHSO:SO which G′ values were significantly (P b 0.05) lower (Fig. 7a). These results demonstrated that FPS addition to the FHSO:SO matrix did not affect mechanical properties up to concentrations of 20 wt.%. However, the sample containing 25 wt.% FPS had lower G′ compared to samples containing 20 wt.% FPS. This could be explained by incompatibilities between components of the FHSO:SO versus FPS at concentrations higher than 25 wt.%. For instance, incompatibility between cocoa butter and milk fat has previously been studied in chocolate and it has been correlated to the cause of undesirable chocolate loss of hardness and blooming (Marangoni & Wesdorp, 2013). Similar to the incompatibilities within fat systems studied in chocolate, FPS at proportions close to 25 wt.% may not be compatible with the FHSO:SO matrix. It has been widely demonstrated that the elastic modulus is a strong function of the matrix solid fat content (Liang, Shi, & Hartel, 2008; Narine & Marangoni, 1999; Rousseau, Forestiere, Hill, & Marangoni, 1996). However, a significant reduction in G′ upon addition of 25 wt.% β-sitosterol was accompanied by no significant changes in the SFC value which can be explained by the fact that mechanical properties of fats are influenced not only by SFC but also by the structure of the fat crystalline network (Acevedo & Marangoni, 2014; Narine & Marangoni, 1999). Comparisons with the commercial shortening revealed that the sample with a higher percentage of phytosterols (25 wt.% vs. 20 wt.%)

had similar properties, in regard to G′ and G″, as compared to commercial puff pastry shortening. G″ represents the viscosity of the fat crystal network also known as the loss modulus (Marangoni & Wesdorp, 2013). Similar to the aforementioned G′ results, no significant effect was observed on G″ upon addition of 20 wt.% FPS and there were no significant differences in G″ between the 25:23:52 wt β-Sit:FHSO:SO sample and the commercial puff pastry shortening suggesting parallel viscoelastic properties (Fig. 7b). The measurement of yield stress (σ*) is one of the most imperative macroscopic properties in regard to assessing fats; it is strongly correlated to the sensory perception, material stability and spreadability (Marangoni & Wesdorp, 2013). The apparent σ* of a plastic solid is usually defined as the point at which, when the stress is increased, the deforming solid first begins to show liquid-like behavior. In this work, we considered the stress at the limit of linearity (after a change in G′ of 10%) as the yield stress. Previous research states that if the yield stress for shortenings falls within the 200–800 Pa range, the product may be acceptable for bakery applications (Haighton, 1959). Not only did the puff pastry commercial shortening fall within this range, but so did all of the FPS:FHSO:SO samples (Fig. 7c). Additionally, the 20:28:52 wt βSit:FHSO:SO sample and the commercial puff pastry shortening were not significantly different from each other in respect to σ* values, suggesting analogous sensory, stability and spreadability characteristics. In puff pastry shortening applications, the shortening is layered between sheets of dough (Ghotra et al., 2002). In order to create separate flaky layers during the baking process it is imperative that replacements for puff pastry shortenings have similar spreadability (Ghotra et al., 2002). Thus, the yield stress values found for the studied fat blends put forward for consideration their capability for use in bakery applications. 4. Conclusions In this work, a series of experiments were performed to produce and analyze structure and functional properties of FPS–TAG blends. In DSC thermograms, the absence of fusion peaks at high temperatures in the FPS blends suggested co-crystallization of FPS and the FHSO:SO fat crystal network. XRD patterns revealed the presence of new d-values not present in patterns corresponding to the control sample or the FPS powders. These new d-values insinuated new structures formed in the presence of FPS. Overall, the addition of FPS decreased the OL% of the matrix, which was comparable to commercial puff pastry shortening. In regard to SFC, the addition of FPS into the FHSO:SO matrix increased the SFC profile compared to the control. Furthermore, it is possible that the

N.C. Acevedo, D. Franchetti / Food Research International 85 (2016) 104–112

111

oils, is of great interests on both, research and industrial fields, in particular those focused on the development of healthier functional fats. Furthermore, the prospect of being able to engineer food ingredients that can best deliver free-phytosterols can significantly advance this field of study, and lead to better food and food ingredient choices for an aging, obese population prone to chronic diseases. References

Fig. 7. G′ (a), G″ (b) and yield stress (σ*) (c) at 20 °C. Letters represent statically significant differences between the values (P b 0.05).

existence of a threshold in co-crystallization of the FPS–TAG samples did not achieve 0% SFC at the Tm of ~65 °C as reported in DSC thermograms. Despite differences in SFC melting profiles, the rheological properties of the FPS enriched shortenings were comparable to those of commercial puff pastry shortening, notably all FPS yield stress values fell within the range suitable for bakery applications and with potential as a replacement for puff pastry shortenings containing trans fats. Furthermore, the amount of phytosterol present in one serving (one tablespoon or 12 g) of the trans fat-free shortening produced in this study, contains 2.4–3 g of FPS per serving. This amount meets the recommended phytosterol intake in order to reduce cholesterol levels established at 2 g/ day. However, further studies are necessary to assess the relative bioavailability of these different mixtures. In vitro testing, and/or LDL-C lowering efficacy assessment using animal or human models, requires exploration to clarify the clinical benefit that the produced fat blends can provide. The potential applications of fully hydrogenated fats enriched with phytosterols as a replacement of trans-fat rich partially hydrogenated

Acevedo, N. C., Block, J. M., & Marangoni, A. G. (2012). 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 Discussions, 158, 171–194. Acevedo, N. C., & Marangoni, A. G. (2014). Functionalization of non-interesterified mixtures of fully hydrogenated fats using shear processing. Food and Bioprocess Technology, 7, 575–587. AlHasawi, F. M., & Rogers, M. A. (2013). Ternary phase diagram of beta-sitosterol-gammaoryzanol-canola oil. Journal of the American Oil Chemists' Society, 90, 1533–1540. American Heart Association Nutrition Committee, Lichtenstein, A. H., Appel, L. J., Brands, M., Carnethon, M., Daniels, S., ... Wylie-Rosett, J. (2006). Diet and lifestyle recommendations revision 2006: A scientific statement from the American Heart Association Nutrition Committee. Circulation, 114, 82–96. Awad, A. B., Roy, R., & Fink, C. S. (2003). Beta-sitosterol, a plant sterol, induces apoptosis and activates key caspases in MDA-MB-231 human breast cancer cells. Oncology Reports, 10(2), 497–500. von Bonsdorff-Nikander, A., Karjalainen, M., Rantanen, J., Christiansen, L., & Yliruusi, J. (2003). Physical stability of a microcrystalline β-sitosterol suspension in oil. European Journal of Pharmaceutical Sciences, 19(4), 173–179. Bouzidi, L., Omonov, T. S., Garti, N., & Narine, S. S. (2013). Relationships between molecular structure and kinetic and thermodynamic controls in lipid systems. Part I: Propensity for oil loss of saturated triacylglycerols. Food & Function, 4(1), 130–143. Cantrill, R. (2008). Phytosterols, phytostanols and their esters: Chemical and technical assessment. Paper presented at the The 69th Joint FAO/WHO Expert Committee on Food Additives (JECFA) (pp. 1–13) (http://www.fao.org/fileadmin/templates/agns/pdf/ jecfa/cta/69/Phytosterols.pdf). Christiansen, L. I., Rantanen, J. T., von Bonsdorff, A. K., Karjalainen, M. A., & Yliruusi, J. K. (2002). A novel method of producing a microcrystalline β-sitosterol suspension in oil. European Journal of Pharmaceutical Sciences, 15(3), 261–269. Dibildox-Alvarado, E., Rodrigues, J. N., Gioielli, L. A., Toro-Vazquez, J. F., & Marangoni, A. G. (2004). Effects of crystalline microstructure on oil migration in a semisolid fat matrix. Crystal Growth & Design, 4(4), 731–736. Foubert, I., Vanhoutte, B., & Dewettinck, K. (2004). Temperature and concentration dependent effect of partial glycerides on milk fat crystallization. European Journal of Lipid Science and Technology, 106, 531–539. Garti, N. (2002). In A. G. Marangoni, & S. S. Narine (Eds.), Physical properties of lipids. CRC Press: Boca Raton. Ghotra, B. S., Dyal, S. D., & Narine, S. S. (2002). Lipid shortenings: A review. Food Research International, 35(10), 1015–1048. Guedes, A. M. M., Ming, C. C., Ribeiro, A. P. B., da Silva, R. C., Gioielli, L. A., & Gonçalves, L. A. G. (2014). Physicochemical properties of interesterified blends of fully hydrogenated Crambe abyssinica oil and soybean oil. Journal of the American Oil Chemists' Society, 91(1), 111–123. Gylling, H., Plat, J., Turley, S., Ginsberg, H. N., Ellegård, L., Jessup, W., ... European Atherosclerosis Society Consensus Panel on Phytosterols (2014). Plant sterols and plant stanols in the management of dyslipidaemia and prevention of cardiovascular disease. Atherosclerosis, 232, 346–360. Haighton, A. J. (1959). The measurement of the hardness of margarine and fats with cone penetrometers. Journal of the American Oil Chemists' Society, 36(8), 345–348. Halling, K. K., & Slotte, J. P. (2004). Membrane properties of plant sterols in phospholipid bilayers as determined by differential scanning calorimetry, resonance energy transfer and detergent-induced solubilization. Biochimica Biophysica Acta, 1664(2), 161–171. Hayes, K. C., Pronczuk, A., & Perlman, D. (2004). Nonesterified phytosterols dissolved and recrystallized in oil reduce plasma cholesterol in gerbils and humans. The Journal of Nutrition, 134(6), 1395–1399. von Holtz, R. L., Fink, C. S., & Awad, A. B. (1998). Beta-sitosterol activates the sphingomyelin cycle and induces apoptosis in LNCaP human prostate cancer cells. Nutrition and Cancer, 32(1), 8–12. Hunter, J. E., Zhang, J., & Kris-Etherton, P. M. (2010). Cardiovascular disease risk of dietary stearic acid compared with trans, other saturated, and unsaturated fatty acids: A systematic review. The American Journal of Clinical Nutrition, 91(1), 46–63. International Atherosclerosis Society (IAS) Executive Board (2013d). An International Atherosclerosis Society position paper: Global recommendations for the management of dyslipidemia. Full report (http://www.athero.org/download/IASPPGuidelines_ FullReport_2.pdf). Katan, M. B., Grundy, S. M., Jones, P., Law, M., Miettinen, T., Paoletti, R., & Stresa Workshop Participants (2003). Efficacy and safety of plant stanols and sterols in the management of blood cholesterol levels. Mayo Clinic Procedures, 78, 965–978. Kritchevsky, D., & Chen, S. C. (2005). Phytosterols—Health benefits and potential concerns: A review. Nutrition Research, 25(5), 413–428. Kuhlmann, K., Lindtner, O., Bauch, A., Ritter, G., Woerner, B., & Niemann, B. (2005). Simulation of prospective phytosterol intake in Germany by novel functional foods. British Journal of Nutrition, 93(03), 377–385. http://dx.doi.org/10.1079/BJN20041364. Law, M. (2000). Plant sterol and stanol margarines and health. British Medical Journal, 320, 861–864.

112

N.C. Acevedo, D. Franchetti / Food Research International 85 (2016) 104–112

Liang, B., Shi, Y., & Hartel, R. W. (2008). Correlation of rheological and microstructural properties in a model lipid system. Journal of the American Oil Chemists' Society, 85, 397–404. Ling, W. H., & Jones, P. J. H. (1995). Dietary phytosterols: A review of metabolism, benefits and side effects. Life Sciences, 57(3), 195–206. deMan, L., deMan, J. M., & Blackman, B. (1989). Physical and textural evaluation of some shortenings and margarines. Journal of the American Oil Chemists Society, 66(1), 128–132. Marangoni, A. G., & Wesdorp, L. H. (2013). Structure and properties of fat crystal networks (2nd ed.). CRC Press. Metin, S., & Hartel, R. W. (2005). Crystallization of fats and oils. In F. Shahidi (Ed.), Bailey's industrial oil and fat products. New York: Wiley Interscience. Metzroth, D. (2005). Shortenings: Science and technology (6th Ed.). In F. Shahidi (Ed.), Bailey's industrial oil & fat products. Vol. 4. (pp. 83–123). Hoboken, NJ: John Wiley & Sons. Murphy, D. H., Hinsch, J., Salmon, E., Spring, K., Brandmaier, C., Brenner, M., Schwartz, S., Keller, H. E., Abramowitz, M., Pluta, M., Parry-Hill, M., Sutter, R. T., Fellers, T. J., & Davidson, M. W. (1998, 2003). Molecular expressions optical microscopy primer specialized techniques: Differential interference contrast. Retrieved March 12th 2015, from http://micro.magnet.fsu.edu/primer/ techniques/dic/dichome.html Musa-Veloso, K., Poon, T. H., Elliot, J. A., & Chung, C. (2011). A comparison of the LDLcholesterol lowering efficacy of plant stanols and plant sterols over a continuous dose range: Results of a meta-analysis of randomized, placebo-controlled trials. Prostaglandins Leukotriens and Essential Fatty Acids, 85, 9–28. Mutaftschiev, B. (1993). In D. T. J. Hurle (Ed.), Handbook on crystal growth (pp. 187) (Amsterdam: North-Holland). Narine, S. S., & Marangoni, A. G. (1999). Relating structure of fat crystal networks to mechanical properties: A review. Food Research International, 32, 227–248. Navarro, A., De las Heras, B., & Villar, A. (2001). Anti-inflammatory and immunomodulating properties of a sterol fraction from Sideritis foetens. Clem. Biological & Pharmaceutical Bulletin, 24(5), 470–473. Nissinen, M., Gylling, H., Vuoristo, M., & Miettinen, T. A. (2002). Micellar distribution of cholesterol and phytosterols after duodenal plant stanol ester infusion. American Journal of Physiology. Gastrointestinal and Liver Physiology, 282(6), G1009–G1015. Ostlund, R. (2002). Phytosterols in human nutrition. Annual Review of Nutrition, 22(1), 533–549. Perlman, D., Hayes, K., & Pronczuk, A. (2006). Prepared foods containing triglyceriderecrystallized non-esterified phytosterols: Patent US 7144595 B2. Piironen, V., Lindsay, D. G., Miettinen, T. A., Toivo, J., & Lampi, A. M. (2000). Plant sterols: Biosynthesis, biological function and their importance to human nutrition. Journal of the Science of Food and Agriculture, 80, 939–966. Ras, R. T., Geleijnse, J. M., & Trautwein, E. A. (2014). LDL-cholesterol-lowering effect of plant sterols and stanols across different dose ranges: A meta-analysis of randomised controlled studies. The British Journal of Nutrition, 112(2), 214–219.

Ratnayake, W. M., L'Abbe, M. R., Mueller, R., Hayward, S., Plouffe, L., Hollywood, R., & Trick, K. (2000). Vegetable oils high in phytosterols make erythrocytes less deformable and shorten the life span of stroke-prone spontaneously hypertensive rats. Journal of Nutrition, 130(5), 1166–1178. Ribeiro, A. P., Grimaldi, R., Gioielli, L. A., & Gonçalvesa, L. A. G. (2009a). Zero trans fats from soybean oil and fully hydrogenated soybean oil: Physico-chemical properties and food applications. Food Research International, 42(3), 401–410. Ribeiro, A. P., Grimaldi, R., Gioielli, L. A., dos Santos, A. O., Cardoso, L. P., & Gonçalves, L. A. G. (2009b). Thermal behavior, microstructure, polymorphism, and crystallization properties of zero trans fats from soybean oil and fully hydrogenated soybean oil. Food Biophysics, 4(2), 106–118. Ribeiro, A. P., Masuchi, M. H., Miyasaki, E. K., Domingues, M. A. F., Stroppa, V. L. L. Z., de Oliveira, G. M., & Kieckbusch, T. H. (2015). Crystallization modifiers in lipid systems. Journal of Food Science and Technology, 52(7), 3925–3946. Rodrigues, J. N., Torres, R. P., Mancini-Filho, J., & Gioielli, L. A. (2007). Physical and chemical properties of milkfat and phytosterol esters blends. Food Research International, 40(6), 748–755. Rousseau, D., Forestiere, K., Hill, A. R., & Marangoni, A. G. (1996). Restructuring butterfat through blending and chemical interesterification. 1. Melting behavior and triacylglycerol modifications. Journal of the American Oil Chemists' Society, 73, 963–972. Rousseau, D., Hodge, S. M., Nickerson, M. T., & Paulson, A. T. (2005). Regulating the β′ → β polymorphic transition in food fats. Journal of the American Oil Chemists' Society, 82(1), 7–12. Rozner, S., & Garti, N. (2006). The activity and absorption relationship of cholesterol and phytosterols. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 435–456. Salo, P., & Wester, I. (2005). Low-fat formulations of plant stanols and sterols. American Journal of Cardiology, 96, 51–54. Sato, K. (2001). Crystallization behaviour of fats and lipids — A review. Chemical Engineering Science, 56(7), 2255–2265. Spitzer, V., & Maggini, S. (2013). Chapter 26 — Phytosterols and micronutrients for heart health. In R. R. Watson, & V. R. Preedy (Eds.), Bioactive food as dietary interventions for cardiovascular disease (pp. 393–413). San Diego: Academic Press. Toro-Vazquez, J. F., Vargas, E. R., Dibildox-Alvarado, E., & Charó-Alonso, M. A. (2005). Crystallization of cocoa butter with and without polar lipids evaluated by rheometry, calorimetry and polarized light microscopy. European Journal of Lipid Science and Technology, 107(9), 641–655. Vaikousi, H., Lazaridou, A., Biliaderis, C. G., & Zawistowski, J. (2007). Phase transitions, solubility, and crystallization kinetics of phytosterols and phytosterol–oil blends. Journal of Agricultural and Food Chemistry, 55(5), 1790–1798. Wright, A. J., & Marangoni, A. G. (2002). Effect of DAG on milk fat TAG crystallization. Journal of the American Oil Chemists' Society, 79, 395–402.