Structuration, elastic properties scaling, and mechanical reversibility of candelilla wax oleogels with and without emulsifiers

Structuration, elastic properties scaling, and mechanical reversibility of candelilla wax oleogels with and without emulsifiers

Food Research International 122 (2019) 471–478 Contents lists available at ScienceDirect Food Research International journal homepage: www.elsevier...

2MB Sizes 0 Downloads 5 Views

Food Research International 122 (2019) 471–478

Contents lists available at ScienceDirect

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

Structuration, elastic properties scaling, and mechanical reversibility of candelilla wax oleogels with and without emulsifiers

T



Pérez-Martínez J.D.a, , Sánchez-Becerril M.a, Marangoni A.G.b, Toro-Vazquez J.F.a, Ornelas-Paz J.J.c, Ibarra-Junquera V.d a

Facultad de Ciencias Químicas, Centro de Investigación y Estudios de Posgrado, Universidad Autónoma de San Luis Potosí, San Luis Potosí 78210, Mexico Department of Food Science, University of Guelph, Guelph, ON N1G 2W1, Canada c Centro de Investigación en Alimentación y Desarrollo, A.C. Unidad Cuauhtémoc, Av. Río Conchos S/N, Parque Industrial, 31570, Mexico d Universidad de Colima, Bioengineering Laboratory, Km. 9 carretera Coquimatlán-Colima, C.P. 28400 Coquimatlán, Colima, Mexico b

A R T I C LE I N FO

A B S T R A C T

Keywords: Oleogel Candelilla wax Fractal dimension Monoglycerides PGPR Rheology Elastic modulus Mechanical reversibility

The crystal network development, elastic properties scaling behavior, and mechanical reversibility of candelilla wax (CW) oleogels with and without emulsifiers were studied. Saturated monoglycerides (MG) and polyglycerol polyricinoleate (PGPR) were added at 1 or 2 times the critical micelle concentration. Although the micelles of both emulsifiers act as nucleation sites for the mixture of aliphatic acids and alcohols of CW, they did not affect the oleogel's thermodynamic stability. It was established that the crystal network of CW consists of at least two types of crystals, one rich in n-hentriacontane and other rich in aliphatic acids. Both crystals species contributed significantly to the oleogel elasticity. The elastic properties scaling behavior of CW oleogels fitted the fractal model within the weak-link regime. The setting temperature and added emulsifier modified the crystal network fractal dimension. During shearing, oleogels had massive breaking of junction zones, causing the loss of fractality in the crystal network, which in turn decreased the system's elasticity.

1. Introduction Eliminating trans fats from foods requires the development of innocuous matrices delivering the functionality of partially hydrogenated vegetable oils (PHVO) at equal or lower productions costs than the products currently marketed. In this regard, the first action of food producers has been the replacement of PHVOs with mixtures of high melting fats (saturated) and vegetable oils (unsaturated) (Kodali, 2014). Using these mixtures allows the elimination of trans fats, but overlooks another major challenge, the reduction of saturated fat in foods (ToroVazquez et al., 2013). Within this context, vegetable oil gelation has remained as a promising strategy for trans fats elimination, and saturated fats reduction for a myriad of confectionery and baking products (Marangoni & Garti, 2011). Wax oleogels have been thought as a healthy replacement of saturated fats in products such as margarines, confectionery toppings and spreads (Dassanayake, Kodali, Ueno, & Sato, 2009; Patel & Dewettinck, 2016; Toro-Vazquez et al., 2013). Crystallization of ≥1% candelilla wax (CW) in vegetable oil solutions produces thermodynamically stable oleogels with elevated oil binding capacity, but much softer than baking margarine (Blake, Co, & Marangoni, 2014; Toro-Vazquez et al., 2007). Improving the



rheological functionality requires among others, to fully understand the structuration process of CW as well as the elastic properties scaling as has been done for plastic fats structured with TAGs (Marangoni & Rousseau, 1996). As CW can be used to produce baking margarine and other complex systems, it is relevant to know the effect of a variety of lipophilic molecules, such as emulsifiers, on the gelation process and the thermo-mechanical properties of gels. Model mixtures of CW, comprised by n-hentriacontane and tritiacontanoic acid (C30), have shown that the particular microstructure of CW is the result of two concomitant processes occurring during crystallization: i) the incorporation of small amounts of C30 and, probably, other aliphatic acids and/or alcohols in the n-hentriacontane crystalline lattice, and ii) the heterogeneous nucleation of n-hentriacontane over high melting polar lipids (i.e., C30) (Serrato-Palacios et al., 2015). Nonetheless, it is not clear whether the presence of low molecular weight emulsifiers affects the structuring process of CW and their impact on the oleogels' rheology (Toro-Vazquez et al., 2013). These phenomena are not trivial since emulsifiers are typically added to margarine, not only to stabilize the emulsion (e.g., protect it against tearing), but also to improve the aeration and the crumb uniformity in baked goods (O'Brien, 2009). In this research, we analyze the effect of two food-grade lipophilic

Corresponding author at: Facultad de Ciencias Químicas-CIEP, Zona Universitaria, Av. Dr. Manuel Nava 6, Zona Universitaria, San Luis Potosí SLP 78210, México. E-mail address: [email protected] (J.D. Pérez-Martínez).

https://doi.org/10.1016/j.foodres.2019.05.020 Received 12 January 2019; Received in revised form 7 May 2019; Accepted 12 May 2019 Available online 15 May 2019 0963-9969/ © 2019 Elsevier Ltd. All rights reserved.

Food Research International 122 (2019) 471–478

J.D. Pérez-Martínez, et al.

2. Materials and methods

emulsifiers commonly used to tailor W/O emulsions: polyglycerol polyricinoleate and saturated monoglycerides (Rafanan & Rousseau, 2017; Toro-Vazquez et al., 2013). Toro-Vazquez et al. (2013) reported that monoglycerides cause a slight softening of candelilla wax organogels. On the other hand, polyglycerol polyricinoleate is largely used to reduce the yield value of chocolate suspensions as it causes steric hindrance among cocoa butter crystals and sugar solids (Middendorf, Juadjur, Bindrich, & Mischnick, 2015). At microscale the solid phase of wax oleogels is a continuous threedimensional structure consisting of branched platelets forming microchannels were the liquid phase is arrested by capillary forces (Imai, Nakamura, & Shibata, 2001; Miyazaki & Marangoni, 2014; SánchezBecerril et al., 2018). These microplatelets are structured by nanoplatelets (4–40 nm length) which can be regarded as the primary particles of a colloidal gel (Sánchez-Becerril et al., 2018). Thus, it is reasonable to assume that the oleogel's solid-network is a collection of fractal flocs, analogous to a fat crystal network (Marangoni & Rousseau, 1996). Determining the fractal organization in these oleogels it is possible by using a scaling theory relating the microstructure to the elastic properties. The model developed by Shih, Shih, Kim, Liu, and Aksay (1990), which has been proven to be suitable for plastic fats and colloidal gels well above the gelation threshold (δ < < 45°), establishes that the scaling of the elastic modulus (G') and the yield strain (γ*) with the solid phase fraction (ϕ) is dictated by the fractal dimension of the individual flocs (Marangoni & Rousseau, 1996; Narine & Marangoni, 1999a, 1999b; Shih et al., 1990). Depending on the strength of the intra and inter flocs links, this theory considers two regimes. The strong-link regime predominates at low particle concentrations when flocs are very large, and the inter-floc is greater than the intra-floc strength. Shih et al. (1990) predicted that within this regime γ* decreases with the increasing of ϕ. In the weak-link regime, flocs are smaller, the system elasticity is determined by inter-floc interactions, and γ* increases with ϕ (Shih et al., 1990). Plastic fats show a power-law increment of G' as a function of ϕ within the weak-link regime according to Eq. (1): 1

G′ = γϕ (D − 3)

2.1. Materials CW and high oleic safflower oil (HOSFO) were obtained from Multiceras (Monterrey, Mexico) and Coral Internacional (San Luis Potosí, Mexico), respectively. Polyglycerol polyricinoleate 4180 (PGPR) and a mixture of monoglycerides MG93 (MG) were provided by Palsgaard Industri de México (San Luis Potosí, Mexico). 2.2. Methods 2.2.1. Critical micelle concentration The critical micellar concentration (CMC) of emulsifiers in HOSFO was determined at room temperature (25 °C) using interfacial tension measurements obtained with a DuNouy ring tensiometer 70,545 (CSC Scientific Company, EUA). Solutions in mass percentage of the emulsifier in HOSFO were prepared by dissolving the corresponding amount of stock solutions prepared by dissolution of the emulsifier in hot HOSFO (80 °C) under constant stirring for 20 min. The corresponding amount of stock solution was mixed with HOSFO at room temperature in a volumetric flask (25 mL), slightly shaken to avoid bubbles formation, and filled with HOSFO until the liquid reached the calibration mark. The flask was capped and gently inverted until the contents were thoroughly mixed. The water-oil interfacial tension calculation was performed according to the equipment manual. In short, 20 mL of miliQ water was pipetted in the bottom of cylindrical glass containers (Kymax® 2300; 50 mm Length × 35 mm Diameter), the ring attached to the measuring system was placed ~1 mm below the water surface and the emulsifier solution (20 mL) was carefully pipetted to develop a uniform interface. After that, the ring was taken to the interface adjusting the container position, and the torsion in the measuring system was increased until the film at the interface broke. The torsion reading at this point was used to calculate the interfacial tension. The CMC was obtained from the break in the interfacial tension vs. the logarithm of the emulsifier concentration. Both emulsifiers reduced the interfacial tension to slightly below 10 mN/m. For PGPR, the break associated with the CMC was detected at ~0.25% (dotted arrow in Fig. S1). For the MG, the break associated with the CMC could not be established (Fig. S1) since interfacial tension measurements for solutions with > 0.35% MG were inconsistent, and solutions with ≥0.4% MG showed partial crystallization at room temperature. As crystallization occurs only at concentrations higher than the CMC, we established the CMC for MG at 0.35%. Thus, the emulsifier concentration used to produce the oleogels were set as 0% emulsifier, 0.25% or 0.5% for PGPR, and 0.35% or 0.7% for MG.

(1)

where D is the fractal dimension of the flocs' network and the pre-exponential factor (γ) a parameter dependent on the Hamaker's constant, the floc size, the number of particles per floc, the size of the primary particles and the distance between them (Narine & Marangoni, 1999a). Another topic of interest for wax oleogels is their low level (< 10%) of elasticity recovery (i.e., partial thixotropy) after shearing (RamírezGómez et al., 2016). Under stress, particle gels undergo an elastic deformation of the network structure, but also a gradual breakdown at the junction zones (Miyazaki & Marangoni, 2014; Van den Tempel, 1961). Crystal-crystal junction zones can be reversible or irreversible. Reversible junction zones occur between neighboring crystals attracted by van der Waals forces, in contrast, irreversible junction zones exist where crystals are mechanically interpenetrated. A reversible junction zone broken under shear will reform, allowing a macroscopic recovery of elasticity. On the other hand, the breakdown of irreversible junction zones will produce permanent softening. To overcome the permanent loss of elasticity in wax-based oleogels, we developed a composite crystal network with CW and TAGs crystals to avoid the CW crystals intertwining and increase the number of reversible junction zones (Ramírez-Gómez et al., 2016). On the other hand, Stortz and Marangoni (2014) used a large amount of emulsifier to change the ethylcellulose solubility in vegetable oil to produce fully thixotropic oleogels. Within this framework, the objective of the present research was to determine the effect of two liposoluble emulsifiers on the gelling process, elastic properties scaling and mechanical reversibility after shear for candelilla wax oleogels. Emulsifiers in oleogels were at 1 or 2 times the critical micelle concentration (CMC), as the CMC is the limit concentration of emulsifier dissolved in the bulk phase above which emulsifier micelles can form and stabilize the dispersed phase.

2.2.2. Oleogel preparation Oleogels were produced from HOSFO gel-forming solutions with a mass percentage of 1–8% CW and 0% emulsifier, 0.35% MG, 0.7% MG, 0.25% PGPR or 0.5% PGPR. These solutions were prepared by mixing the corresponding amount of materials and heating to 90 °C for 20 min. After that, solutions were aliquoted on the corresponding sample holders and processed according to Sections 2.2.3, 2.2.4, 2.2.5 and 2.2.6. 2.2.3. Differential scanning calorimetry Crystallization and melting profiles of gel-forming solutions were determined in a differential scanning calorimeter (Q1000; TA Instruments, New Castle, DE, USA) equipped with a refrigerated cooling unit. The DSC equipment was calibrated with Indium. Gel-forming solutions (5 to 7 mg) were weighed and sealed in aluminum pans and placed in the measuring cell, heated at 90 °C for 20 min and then cooled (5 °C/min) until −15 °C. After 1 min at this temperature the system was heated (5 °C/min) up to 90 °C. The temperature at the onset (TO) of the exotherm was calculated using the software Universal Analysis 2000 472

Food Research International 122 (2019) 471–478

J.D. Pérez-Martínez, et al.

Fig. 1. Complex modulus (G*) rheogram (filled circles) overlaid on the corresponding crystallization thermogram (solid line). Exotherms are in down direction in thermograms. A) 1% CW, B) 3% CW, C) 8% CW, D) 1% CW-0.5% PGPR, E) 3% CW-0.5% PGPR, F) 8% CW-0.5% PGPR, G) 1% CW-0.7% MG, H) 3% CW-0.7% MG, I) 8% CW-0.7% MG.

Ver. 4.2E (TA Instruments—Waters LLC). Similarly, we calculated the temperature at the maximum (TM) of the melting endotherm, and the corresponding melting enthalpy (ΔHM).

frequency. With this information, we programmed the rheometer to apply a variable strain to the sample, within the LVR, at the corresponding time and temperature conditions.

2.2.4. Rheology All rheological measurements were obtained using a rheometer MCR 302 (Anton Paar, Germany) fitted to a 25 mm parallel plate geometry, equipped with the TruGap™ system (PP25/TG, Anton-Paar, Germany) and controlled by the rheometer software (RheoPlus, AntonPaar, Germany). Temperature control was achieved by a Peltier system in the base of the lower plate (P-PTD200/62/TG, Anton-Paar, Germany) and a Peltier hood covering the measuring probe (H-PTD200, Anton-Paar, Germany). Gel-forming solutions were placed on the lower plate, and the superior plate was positioned on the sample leaving a gap of 1 mm between plates. Samples were held at 90 °C for 20 min, and then cooled at 5 °C/min to the setting temperature (Tset; 25 or 5 °C). During the cooling, the samples viscoelastic properties (i.e., G*, G′) were measured. After that, the samples were maintained at Tset until G′ remain constant for 60 min, this value was regarded as the elastic modulus under static conditions (G'S). Then, the oleogel was sheared (10 s−1) for 60 s under zero normal force, and let recover under quiescent conditions. Elastic properties of sheared oleogels were measured during recovery by oscillatory measurements; the G′ value after 60 min was regarded as the recovered elastic modulus (G'R). G*, G′ and δ were measured within the linear viscoelastic region (LVR) using a deformation frequency of 1 Hz. For each mixture, the LVR was assessed through amplitude sweeps applying an increasing strain from 0.0001 up to 100% at constant temperature (90, 60, 40, 25 or 5 °C) and

2.2.5. Solid phase content The oleogels solid phase content (SPC) was determined by pNMR using a Bruker Minispec spectrometer (Bruker, Milton, ON, Canada) at a resonance frequency of 20 MHz. Gel-forming solutions (4 mL) were aliquoted in NMR tubes, heated at 90 °C for 20 min, and stored at Tset (25 or 5 °C). The SPC was determined after 24 h of storage. The solid fraction (ϕ) was calculated as SPC/100%. 2.2.6. X-ray diffraction X-ray diffraction patterns of 8% CW oleogels were collected using a diffractometer Miniflex 600 (Rigaku, Japan) equipped with an X-ray tube with Cu Kα radiation (λ = 1.54 Å) set at 40 kV and 15 mA, and a Dtex ultra detector (Rigaku, Japan). In the incident beam, a soller slit was collimated with a Ni filter and a slit of 0.5 mm. The diffraction scan was performed in symmetrical geometry (2θ) from 3 to 60° with a step of 0.01° at 3°/min. Gel-forming solutions were placed in the sample holder and processed as described in Section 2.2.5. Diffractograms were analyzed using PeakFit software (Seasolve, Framingham, MA, USA). 2.3. Statistical analysis Plots and regression analysis were done with Prism 5.0 software (GraphPad Software Inc. San Diego, CA, USA). Statistical analysis for the effects of CW and emulsifier content was performed by two-way 473

Food Research International 122 (2019) 471–478

J.D. Pérez-Martínez, et al.

Table 1 Thermal parameters obtained from cooling and heating thermograms of candelilla wax (CW) solutions with polyglycerol poliricionoleate (PGPR) or monoglycerides (MG). TO and TM are the temperatures at the crystallization onset and the maximum of the melting endotherm, respectively. The melting enthalpy ΔHm is the global melting enthalpy. CW (%)

Emulsifier

TO (°C)

TM (°C)

1 1 1 1 1 2 2 2 2 2 3 3 3 3 3 6 6 6 6 6 8 8 8 8 8

0.00% 0.25% 0.50% 0.35% 0.70% 0.00% 0.25% 0.50% 0.35% 0.70% 0.00% 0.25% 0.50% 0.35% 0.70% 0.00% 0.25% 0.50% 0.35% 0.70% 0.00% 0.25% 0.50% 0.35% 0.70%

31.71 (0.01)a,a 32.23 (0.29)b,a 32.51 (0.08)c,a 31.71 (0.01)a,a 31.82 (0.07)a,a 35.57 (0.08)a,b 36.14(0.10)b,b 36.72 (0.01)c,b 35.95 (0.07)d,b 35.79 (0.11)d,b 38.46 (0.01)a,c 38.96 (0.13)b,c 39.18 (0.17)b.c 39.04 (0.02)b,c 38.93 (0.20)b,c 43.49 (0.06)a,d 43.90 (0.11)b,d 44.16 (0.16)b,d 43.65 (0.01)a,d 43.69 (0.02)a,d 45.76 (0.04)a,e 46.02 (0.06)b,e 46.09 (0.06)b,e 45.98 (0.06)a,e 45.81 (0.01)a,e

30.89 30.24 31.28 31.98 23.20 35.57 35.49 36.41 36.05 36.19 38.33 39.22 38.99 39.15 39.60 44.50 44.41 44.41 44.04 44.63 46.41 46.21 46.40 46.15 46.10

PGPR PGPR MG MG PGPR PGPR MG MG PGPR PGPR MG MG PGPR PGPR MG MG PGPR PGPR MG MG

(1.57)a,a (0.18)a,a (1.58)a,a (0.49 a,a (0.41)b,a (1.03)a,b (0.64)a,b (0.02)a,b (0.19)a,b (0.36)a,b (0.13)a,c (0.09)a,c (0.24)a,c (0.15)a,c (0.63)b,c (0.19)a,d (0.06)a,d (0.01)a,d (0.01)a,d (0.04)a,d (0.02)a,e (0.18)a,e (0.05)a,e (0.01)a,e (0.04)a,e

ΔHM (J/g) 0.73 (0.13)a,a 1.00 (0.06)a,a 1.02 (0.07)a,a 0.86 (0.41)a,a 1.55 (0.33)a,a 1.94 (0.01)a,b 1.99 (0.48)a,b 2.49 (0.10)a,b 2.49 (0.14)a,b 2.93 (0.69)a,b 4.77 (0.13)a,c 5.25 (0.04)a,c 4.22 (0.18)a,c 4.29 (0.54)a,c 4.47 (0.62)a,c 8.98 (1.29)a,d 9.18 (0.76)a,d 9.53 (0.58)a,d 8.06 (0.31)a,d 7.30 (0.74)a,d 11.12 (0.59)a,e 12.04 (1.60)a,e 11.55 (0.42)a,e 11.55 (0.07)a,e 10.96 (0.28)a,e

ANOVA (P < .05) and orthogonal contrasts using STATISTICA V 7.1 (StatSoft, Inc. Tulsa, OK). 3. Results 3.1. CW gelling under dynamic supercooling MG and PGPR in HOSFO showed a differential behavior during the cooling process. Neither 0.25% nor 0.5% PGPR crystallized within the analyzed temperature range, while 0.7% MG solutions showed two exotherms corresponding to the α-lamellar organization, and aliphatic chains crystallization (i.e., crystalline sub-α structure) at 16.76 ± 0.45 °C and 9.73 ± 0.50 °C, respectively (Fig. 1). In 0.35% MG solutions, MG crystallized directly to the sub-α structure at ~1.47 °C. In any case, CW crystallization in solutions started while emulsifiers were dissolved or aggregated as micelles in the HOSFO (Fig. 1). Cooling thermograms of 1–8% CW solutions with and without emulsifier showed two exotherms associated to the CW crystallization (Fig. 1); the higher crystallization peak corresponded to the mixture of aliphatic acids and alcohols, the lower to the of n-hentriacontane crystallization as reported by Serrato-Palacios et al. (2015). Independently on the CW concentration, PGPR raised the TO for CW (< 1.2 °C), a similar effect was produced by MG only in solutions with 2–3% CW (Table 1). The seeding effect of low concentrations (0.5–1.0%) of partial glycerides on triglycerides mixtures have been reported elsewhere (Basso et al., 2010; Foubert, Vanhoutte, & Dewettinck, 2004). In those cases, micellar structures of partial glycerols facilitated the nucleation process, reduced the induction time under isothermal conditions, reduced the crystallization temperature on dynamic crystallization, and change the microstructure of triacylglycerol crystals. In this study, the seeding effect of MG on CW was limited, while PGPR micelles effectively perform as nucleation sites for the mixture of aliphatic acids and alcohols of CW. Our group reported polarized light microscopy images showing the increment of the CW crystals amount resulting from adding PGPR in oleogels with the compositions here studied (Sánchez-Becerril et al., 2018).

Fig. 2. X-ray diffraction patterns of CW oleogels with and without emulsifiers as produced at different setting temperature (Tset). A) 8% CW, B) 8% CW-0.5% PGPR, C) CW-0.7% MG. Scattering associated with the liquid phase was subtracted using the software PeakFit.

0.35% MG and 0.25–0.5% PGPR did not affect the melting temperature (TM) of oleogels. But, 0.7% MG had a significant effect on the melting temperature of systems with 1 and 3% CW (P > .05). Particularly, for oleogels structured with 0.70% MG and 1% or 3% CW the highest endothermic peak resulted from the concomitant melting of MG and CW. Melting enthalpy for oleogels produced under dynamic 474

Food Research International 122 (2019) 471–478

J.D. Pérez-Martínez, et al.

Table 2 Fractal dimension (D) and pre-exponential term (Log γ) considering the weaklink regime for candelilla wax oleogels with and without emulsifier. Pearson correlation coefficient (R2) for the linear fitting between Log G' vs Log ϕ for each condition is reported. Emulsifier

0% 0.25% PGPR 0.5% PGPR 0.35% MG 0.7% MG

5 °C

25 °C 2

D

Log (γ) (Pa)

R

2.63 2.66 2.69 2.62 2.57

10.27 10.68 11.03 10.21 9.81

0.963 0.953 0.932 0.988 0.994

D

Log (γ) (Pa)

R2

2.71 2.72 2.73 2.74 2.79

11.33 11.46 11.74 11.93 12.95

0.959 0.961 0.948 0.958 0.930

oleogels with 0.35–0.7% MG showed an additional peak at d = 47 Å, which is characteristic of the lamellar ordering of monoacylglycerols in the inverse lamellar and sub-α crystalline phases (Chen & Terenjev, 2011). These results confirmed that the MG and CW crystallized independently under dynamic cooling conditions. To elucidate the structuration process of CW with and without emulsifiers, we perform comparisons of each G* rheogram and the corresponding thermogram (Fig. 1). G* is formally defined as the ratio of the stress to strain for a material subjected to a sinusoidal load applied within the viscoelastic linear region (Mezger, 2014). That is, G* is a measure of the sample's resistance to deformation. The complementary information of both signals revealed several major events occurring during the cooling process. Close to TO, the sharp increment for G* was associated with the development of a spanning crystal network formation (i.e., the percolation threshold). In systems with 1% CW and 0% emulsifier or 0.7% MG the increment occurred at temperatures below the first CW exothermic peak, which is associated with the crystallization of a mixture of aliphatic acids and alcohols in CW. Similar behavior was observed in the systems with 1% CW and 0.5% PGPR, but the increase of G* was closer to the corresponding TO. Below TO, all mixtures showed a sharp increment of G* concomitantly with the exothermic peak corresponding to the n-hentriacontane crystallization (Fig. 1). Both G* increments could be seen clearly during the cooling of solutions with 6–8% CW (Fig. 1C, F & I, data not shown for 6% CW). G* did not increase during the MG crystallization of gels with ≥2% CW (Figs. 1B-I). On the other hand, 0.7% MG produced one order magnitude increment in the system with 1% CW (Fig. 1G). From this differential behavior we conclude that to produce a significant increase in the system strength (i.e., G*), a secondary crystal species must develop a spanning network intertwined with the primary crystal network. This event might occur only when the primary crystal network is not too dense (i.e., 1% CW), allowing the secondary crystal species (i.e., MG crystal) to growth through the primary network interstices to develop a secondary network. If primary network density is high enough to block the growth of a secondary network, the contribution of the secondary crystal to the global system strength would be not significant.

3.2. CW oleogels elastic properties scaling Fig. 3. Relationship between the elastic modulus for oleogels structured under static conditions (G'S) and their solid fraction (ϕ). CW oleogels with and without emulsifiers were set at either 5 or 25 °C. A) CW without emulsifier, B) CW with 0.5% PGPR, C) CW with 0.7% MG.

Independently on the type and amount of emulsifier, CW oleogels had a good linear correlation between Log G'S and Log ϕ at 5 °C and 25 °C (Fig. 3). The effect of temperature on the Log G'S vs. Log ϕ relationship was more evident in oleogels with 0.7% MG (Fig. 3C). Although CW oleogels had lower ϕ (~0.01–0.06) than TAGs crystal networks (TCNs) (i.e., ϕ ~ 0.25–0.75), increments of G' as a function of the solid phase for oleogels were within the range reported for TCNs (Marangoni & Rousseau, 1996; Narine & Marangoni, 1999a). Assuming these systems develop a fractal organization within the weak-link regime, which is observed for plastic fats, oleogels with 1–8% CW reach D values (2.57–2.79) within the range observed for milkfat and palm oil, produced through reaction-limited aggregation (Table 2; Walstra, 2002;

supercooling conditions increased with the CW concentration but, was not significantly modified by emulsifiers despite both SPC and XRD patterns at 5 and 25 °C evidenced the presence of MG crystals (Table 1, Fig. 2; Table S1). Regardless of Tset and the emulsifier type, CW oleogels showed the distinctive peaks for the n-hentriacontane stable polymorph at room temperature, the orthorhombic subcell packing (4.14 Å and 3.74 Å; Fig. 2) (Serrato-Palacios et al., 2015). In the small angle region, 475

Food Research International 122 (2019) 471–478

J.D. Pérez-Martínez, et al.

Fig. 4. Phase angle (δ) rheogram for oleogels recovery after shear at either 5 or 25 °C. A) 1% CW, B) 2% CW, C) 3% CW, D) 6% CW, E) 8% CW.

25 °C, respectively. As the CW crystallization onset was minimally affected by PGPR (Section 3.1), and the pre-exponential factor (γ) had only slight changes as a function of the PGPR content, we consider the primary particle size was not greatly changed. This means CW oleogels with PGPR produced larger flocs than those without an emulsifier. This effect was associated to the steric force barrier produced by PGPR molecules adsorbed in the crystal's surface, which has been reported to be effective over relatively large distances (120 Å) (Dedinaite & Campbell, 2000). The effect of MG on the ξ/σ ratio could not be

Narine & Marangoni, 1999a). For low-concentration colloidal gels the floc size (ξ)/particle size (σ) ratio, changes as a function of the ϕ and D according to Eq. (2) (Uriev & Ladyzhinsky, 1996). 1

ϕ (D − 3) =

ξ σ

(2)

The ξ/σ ratios for 8% CW oleogels with 0% emulsifier were 1096 and 13,816, while those with 0.5% PGPR were 5515 and 28,615 at 5 °C and 476

Food Research International 122 (2019) 471–478

J.D. Pérez-Martínez, et al.

Table 3 Intercept with y axis (Υ) and slope (k) for the linear fitting between Log G'R vs ϕ for candelilla wax oleogels without emulsifier, 0.25% PGPR, 0.5% PGPR, 0.35% MG or 0.7% MG. Pearson correlation coefficient (R2) for each condition is reported. Emulsifier

0% 0.25% PGPR 0.5% PGPR 0.35% MG 0.7% MG

5 °C

25 °C

Υ (Pa)

k

R

2.814 3.148 3.080 2.562 2.811

48.17 52.41 53.83 58.50 55.56

0.975 0.997 0.955 0.985 0.988

2

Υ (Pa)

k

R2

3.286 3.202 3.367 3.173 3.238

34.98 37.45 35.49 36.09 31.41

0.970 0.955 0.944 0.988 0.96

engineering the crystal network topology, and so the associated physical properties. Considering the primary particles in 8% CW oleogels are irregularly shaped nanoplatelets with a particle size between 4 and 40 nm (Sánchez-Becerril et al., 2018), as well as, this system has a ϕ = 0.0734 (Table S1), and a fractal dimension of D = 2.63, the floc size estimated through Eq. (1) is ~5–40 μm. Such estimation is within the CW crystals size imaged through SEM by Sánchez-Becerril et al. (2018). This preliminary result implies that CW microcrystals in these oleogels are fractal agglomerates structured by a large number of nanocrystals.

3.3. CW oleogels elastic recovery after shear Regardless of the type and amount of emulsifier, CW olegels at 25 °C recover the solid-like behavior, δ < 45°, in less than ~9 s after stopping the shearing, while oleogels at 5 °C required up to ~24 s. Recovering the solid-like behavior after shear implies the re-formation of the solid network by the development of pre-existing and new reversible junction zones. The quicker recovery observed to the higher temperature was ascribed to an increased diffusivity of colloidal particles. According to the Stokes-Einstein's equation (Walstra, 2002), diffusivity is increased by reducing particle size and lowering viscosity. The particle size of crystals and crystal aggregates is unknown, but diffusivity in vegetable oil is approximately 3 times higher at 25 °C than at 5 °C as follows from the viscosity decrement in the continuous media. Aside from the faster recovery, 1–2% CW olegels at 25 °C also recover a more solid-like character (i.e., lower δ) than gels with the same composition at 5 °C (Fig. 4). Oleogels produced with either emulsifier showed a similar recovery. Emulsifiers had a differential effect on the G′ measured after 60 min of recovery (G'R). Regardless of the CW amount, the effect of either emulsifier was more evident at 5 °C. Particularly, oleogels with 6–8% CW and 0.25–0.5% PGPR or 0.35–0.7% MG had a G'R higher than those produced without emulsifier and the same CW content (P < .05, Table S1). On the other hand, oleogels with 1% CW had a similar effect, but only with the higher emulsifier content (i.e., 1% CW-0.5% PGPR and 1%CW-0.7% MG). At 25 °C the positive effect of the emulsifier on G'R was shown only in oleogels with 6% CW and 0.35% MG, 0.25% or 0.5% PGPR, as well as those with 1% CW and 0.5% PGPR (P < .05, Table S1). A slightly negative effect of the emulsifier was detected only in two oleogels, 3% CW-0.35% MG at 5 °C, and 8% CW-0.7% MG at 25 °C (P < .05, Table S1). These results were not conclusive regarding the mechanism or general effect of MG or PGPR on the CW oleogels recovery after shear, but the general trend showed that these emulsifiers could be used to increase the G′ recovery of oleogels after shearing if they were added to a higher concentration than those used in this research. The scaling of G'R after shearing had a semilogarithmic correlation with ϕ (i.e., Log G'R = kϕ + Υ) instead of the Log G'S -Log ϕ correlation derivate from a crystal network with fractal structure, this was shown regardless Tset and emulsifier concentration (Fig. 5). A semilogarithmic

Fig. 5. Relationship between the logarithm of the recovered elastic modulus (G'R) and their solid fraction (ϕ). CW oleogels with and without emulsifiers were set at either 5 or 25 °C. A) CW without emulsifier, B) CW with 0.5% PGPR, C) CW with 0.7% MG.

assessed as different amounts of MG were in the solid phase at both Tset (Table S1). Increasing Tset from 5 to 25 °C also had a preponderant effect on the ξ/σ ratio, producing decrements over one magnitude order. Possibly, the ongoing crystallization below 25 °C produced primary particles inside the crystal network to form more filled flocs. From these results, we conclude that Tset and emulsifiers are important variables for 477

Food Research International 122 (2019) 471–478

J.D. Pérez-Martínez, et al.

correlation has also been established for the elastic properties and solid fraction of consolidated powders (Holman & Leuenberger, 1988, 1991). Holman and Leuenberger (1991) showed that the slope for the semilogarithmic relationship between the elasticity and solid fraction is inversely proportional to the material ductility when a continuous network of pores occurs in the compact (i.e., just above the percolation threshold). This trend was consistent with the differential effect of temperature on the Log G'R 2 ϕ slopes (Table 3); where regardless the type and amount of emulsifier were steeper at 5 °C when CW crystals are less ductile.

200400979. Holman, L. E., & Leuenberger, H. (1988). The relationship between solid fraction and mechanical properties of compacts - the percolation theory model approach. International Journal of Pharmaceutics, 46(1–2), 35–44. https://doi.org/10.1016/ 0378-5173(88)90007-5. Holman, L. E., & Leuenberger, H. (1991). The significance of slopes of the semilogarithmic relationship between hardness and solid fraction of porous compacts. Powder Technology, 64(3), 233–247. https://doi.org/10.1016/0032-5910(91)80138-9. Imai, T., Nakamura, K., & Shibata, M. (2001). Relationship between the hardness of an oil-wax gel and the surface structure of the wax crystals. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 194(1–3), 233–237. https://doi.org/10. 1016/S0927-7757(01)00799-3. Kodali, D. R. (2014). Trans fats: Health, chemistry, functionality, and potential replacement solutions. Trans fats replacement solutions (pp. 1–39). . Marangoni, A. G., & Garti, N. (2011). An overview of the past, present, and future of organogels. In A. G. Marangoni, & N. Garti (Eds.). Edible oleogels: Structure and health implications(1st ed.). Champaing: AOCS Press. Marangoni, A. G., & Rousseau, D. (1996). Is plastic fat rheology governed by the fractal nature of the fat crystal network? JAOCS. Journal of the American Oil Chemists' Society, 73(8), 991–994. https://doi.org/10.1007/BF02523406. Mezger, T. G. (2014). The rheology handbook: For users of rotational and oscilatory rheometers (4th ed.). Hanover: Vincentz Network. Middendorf, D., Juadjur, A., Bindrich, U., & Mischnick, P. (2015). AFM approach to study the function of PGPR's emulsifying properties in cocoa butter based suspensions. Food Structure, 4, 16–26. https://doi.org/10.1016/j.foostr.2014.11.003. Miyazaki, Y., & Marangoni, A. G. (2014). Structural-mechanical model of wax crystal networks-a mesoscale cellular solid approach. Materials Research Express, 1(2), 0–12. https://doi.org/10.1088/2053-1591/1/2/025101. Narine, S. S., & Marangoni, A. G. (1999a). Fractal nature of fat crystal networks. Physical Review E - Statistical Physics, Plasmas, Fluids, and Related Interdisciplinary Topics, 59(2), 1908–1920. https://doi.org/10.1103/PhysRevE.59.1908. Narine, S. S., & Marangoni, A. G. (1999b). Mechanical and structural model of fractal networks of fat crystals at low deformations. Physical Review E - Statistical Physics, Plasmas, Fluids, and Related Interdisciplinary Topics, 60(6), 6991–7000. https://doi. org/10.1103/PhysRevE.60.6991. O'Brien, Richard D. (2009). Fats and Oils: Formulating and Processing for Applications. In Boca Raton (Fla.) (Ed.). CRC. Patel, A. R., & Dewettinck, K. (2016). Edible oil structuring: An overview and recent updates. Food and Function, 7(1), 20–29. https://doi.org/10.1039/c5fo01006c. Rafanan, R., & Rousseau, D. (2017). Dispersed droplets as active fillers in fat-crystal network-stabilized water-in-oil emulsions. Food Research International, 99(April), 355–362. https://doi.org/10.1016/j.foodres.2017.04.008. Ramírez-Gómez, N. O., Acevedo, N. C., Toro-Vázquez, J. F., Ornelas-Paz, J. J., DibildoxAlvarado, E., & Pérez-Martínez, J. D. (2016). Phase behavior, structure and rheology of candelilla wax/fully hydrogenated soybean oil mixtures with and without vegetable oil. Food Research International, 89, 828–837. https://doi.org/10.1016/j. foodres.2016.10.025. Sánchez-Becerril, M., Marangoni, A. G., Perea-Flores, M. J., Cayetano-Castro, N., Martínez-Gutiérrez, H., Andraca-Adame, J. A., & Pérez-Martínez, J. D. (2018). Characterization of the micro and nanostructure of the candelilla wax organogels crystal networks. Food Structure, 16, 1–7. https://doi.org/10.1016/j.foostr.2018.02. 001. Serrato-Palacios, L. L., Toro-Vazquez, J. F., Dibildox-Alvarado, E., Aragón-Piña, A., Morales-Armenta, M. D. R., Ibarra-Junquera, V., & Pérez-Martínez, J. D. (2015). Phase behavior and structure of systems based on mixtures of n-hentriacontane and melissic acid. JAOCS, Journal of the American Oil Chemists' Society, 92(4), 533–540. https://doi.org/10.1007/s11746-015-2623-6. Shih, W. H., Shih, W. Y., Kim, S. I., Liu, J., & Aksay, I. A. (1990). Scaling behavior of the elastic properties of colloidal gels. Physical Review A, 42(8), 4772–4779. Stortz, T. A., & Marangoni, A. G. (2014). The replacement for petrolatum: Thixotropic ethylcellulose oleogels in triglyceride oils. Green Chemistry, 16(6), 3064–3070. https://doi.org/10.1039/c4gc00052h. Toro-Vazquez, J. F., Mauricio-Pérez, R., González-Chávez, M. M., Sánchez-Becerril, M., Ornelas-Paz, J. J., & Pérez-Martínez, J. D. (2013). Physical properties of organogels and water in oil emulsions structured by mixtures of candelilla wax and monoglycerides. Food Research International, 54, 1360–1368. https://doi.org/10.1016/j. foodres.2013.09.046. Toro-Vazquez, J. F., Morales-Rueda, J. A., Dibildox-Alvarado, E., Charó-Alonso, M., Alonzo-Macias, M., & González-Chávez, M. M. (2007). Thermal and textural properties of organogels developed by candelilla wax in safflower oil. JAOCS, Journal of the American Oil Chemists' Society, 84(11), 189–1000. https://doi.org/10.1007/ s11746-007-1139-0. Uriev, N. B., & Ladyzhinsky, I. Y. (1996). Fractal models in rheology of colloidal gels. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 108(1), 1–11. https:// doi.org/10.1016/0927-7757(95)03305-X. Van den Tempel, M. (1961). Mechanical properties of plastic-disperse systems at very small deformations. Journal of Colloid Science, 16, 284–296. Walstra, P. (2002). Physical chemistry of foods (1st ed.). New York: CRC Press.

4. Conclusions At one and two times the CMC, MG and PGPR in HOSFO can act as nucleation sites for the mixture of aliphatic acids and alcohols of CW but did not affect the thermodynamic stability of CW oleogels. The CW oleogels' crystal network structure consists of at least two types of crystals species, one rich in n-hentriacontane and other rich in aliphatic acids and alcohols. Both crystal species contributed significantly to the oleogel elasticity. The MG developed a crystal network at low CW contents, when the CW crystal network was open enough to allow MG crystals intertwining. Such MG crystal network contributed significantly to the system elasticity. The scaling elastic properties of CW oleogels developed under quiescent conditions fitted the fractal model within the weak link regime. Nonetheless, after shearing the oleogels, the massive breaking of junction zones caused the loss of fractality in the oleogel's crystal network, such structural organization was not recovered within a 60 min period, this caused a significant elasticity decrease in most systems. Adding one or two times the critical micelle concentration of MG or PGPR to CW oleogels produced only moderated changes in the G′ recovery after shear, but the general trend points out that to a higher concentration these emulsifiers could increase the recovery. Supplementary data to this article can be found online at https:// doi.org/10.1016/j.foodres.2019.05.020. Acknowledgments This research was funded by the National Council for Science and Technology of Mexico (CONACYT) through the Fondo Sectorial de Investigación para la Educación (Investigación Básica SEP-CONACYT; grant number: 256100). References Basso, R. C., Ribeiro, A. P. B., Masuchi, M. H., Gioielli, L. A., Gonçalves, L. A. G., Santos, A. O.d., & Grimaldi, R. (2010). Tripalmitin and monoacylglycerols as modifiers in the crystallisation of palm oil. Food Chemistry, 122(4), 1185–1192. https://doi.org/10. 1016/j.foodchem.2010.03.113. Blake, A. I., Co, E. D., & Marangoni, A. G. (2014). Structure and physical properties of plant wax crystal networks and their relationship to oil binding capacity. JAOCS, Journal of the American Oil Chemists' Society, 91(6), 885–903. https://doi.org/10. 1007/s11746-014-2435-0. Chen, C. H., & Terenjev, E. E. (2011). Monoglycerides in oils. In A. G. Marangoni, & N. Garti (Eds.). Edible oleogels: Structure and health implications (pp. 173–201). (1st ed.). Urbana: AOCS Press. Dassanayake, L. S. K., Kodali, D. R., Ueno, S., & Sato, K. (2009). Physical properties of rice bran wax in bulk and organogels. JAOCS, Journal of the American Oil Chemists' Society, 86, 1163–1173. https://doi.org/10.1007/s11746-009-1464-6. Dedinaite, A., & Campbell, B. (2000). Interactions between mica surfaces across triglyceride solution containing phospholipid and polyglycerol polyricinoleate. https://doi.org/10. 1021/la991018u. 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(8), 531–539. https://doi.org/10.1002/ejlt.

478