Food Biophysics DOI 10.1007/s11483-016-9467-y

ORIGINAL ARTICLE

Crystallization and Gelation Behavior of Low- and High Melting Waxes in Rice Bran Oil: a Case-Study on Berry Wax and Sunflower Wax Chi Diem Doan 1,2 & Iris Tavernier 1 & Mohd Dona Bin Sintang 1 & Sabine Danthine 3 & Davy Van de Walle 1 & Tom Rimaux 4 & Koen Dewettinck 1

Received: 12 November 2016 / Accepted: 20 December 2016 # Springer Science+Business Media New York 2016

Abstract Low-melting berry wax (BEW) has proven to be a good oil gelator with a positive contribution to the consistency and flexibility of the structured oil. Nevertheless, the properties of BEW and the corresponding oleogel have not yet been investigated in-depth. In this research, the difference in crystallization and gelling behavior between sunflower wax (SW), a high melting wax, and BEW, a low-melting wax, in rice bran oil (RBO) was investigated. The difference in melting and Highlights • High-melting sunflower wax (mainly long-chain wax esters) and lowmelting berry wax (mainly short-chain fatty acids) were compared. • Wax esters contribute to platelet crystals in sunflower wax oleogel. • Short-chain fatty acids give rise to tiny microplatelet crystals in berry wax oleogel. • Polymorphic transition appears in both sunflower wax and berry wax oleogels during storage at 5 °C. • Re-organization in molecular structure of berry wax oleogel results in a stronger gelation over 4 weeks at 5 °C. Electronic supplementary material The online version of this article (doi:10.1007/s11483-016-9467-y) contains supplementary material, which is available to authorized users.

crystallization temperatures can be explained by the different chemical composition (long-chain wax esters in SW and shortchain fatty acids in BEW). The heterogeneity in crystal habits (unidirectional platelets versus microcrystalline particles) and polymorphism (orthorhombic versus hexagonal) are responsible for the varying gel strength and hardness of the respective SWand BEW-oleogels. The microcrystalline BEW particles aligned and reorganized during 1-month storage at 5 °C, which leaded to an increase in the gel strength and hardness of BEW-oleogel. The gelling property of SW-oleogel however did not significantly differ after 4 weeks at 5 °C, despite of the appearance of spherulitic crystalline clusters. The changes in the physical properties of wax-based oleogels during storage time were further explored using differential scanning calorimetry, polarized light microscope, powder X-ray diffraction and rheology. Keywords Rice bran oil . Sunflower wax . Berry wax . Crystallization . Gelation

Introduction * Chi Diem Doan [email protected]; [email protected] * Tom Rimaux [email protected]

1

Laboratory of Food Technology and Engineering, Faculty of Bioscience Engineering, Ghent University, 653 Coupure Links, 9000 Ghent, Belgium

2

Department of Food Technology, College of Agriculture and Applied Biology, Cantho University, Cantho, Vietnam

3

Department of Food Science, University of Liège, Gembloux, Belgium

4

Vandemoortele R&D Centre, Prins Albertlaan 79, 8870 Izegem, Belgium

In the past few decades, there has been a clear demand of consumers and policy makers for healthier food products. Several studies have related the excessive consumption of trans fatty acids to a higher incidence of cardiovascular diseases, and recently the positive health impact of (partially) replacing saturated fat in the daily diet with unsaturated fats was confirmed [1, 2]. Therefore, alternative and healthier ways to structure liquid oil are intensively researched. Oleogels are promising oil-structuring systems and could be an alternative to traditional trans- and saturated fat structuring. Oleogels are semi-solid systems containing building blocks that entrap liquid oil in a three-dimensional network [3]. Those building blocks are self-assembled structures formed

Food Biophysics

by crystalline materials (monoglycerides, diglycerides, triglycerides, fatty acids, fatty alcohols, hydrocarbons, wax esters or natural waxes), by low-molecular compounds (fibers, strands or tubules…), or by polymeric substances [4]. Among these various structurants, natural waxes appear to be the most promising ones because of their high oil binding capacity even at very low concentrations (as low as 0.5%wt) and their thermo-reversibility [5]. Furthermore, wax crystals can stabilize water-in-oil emulsions and they can be applied in food products as an indirect additive [6]. Therefore, extensive research has been conducted to study the use of waxes in oil gelation [5, 7–11] and the influence of external factors on the gelling behavior of waxes in liquid oils, such as solvent type and cooling rate [12, 13], gelator concentration and oil type [14], shear rate [15], or high intensity ultrasound [16]. Waxbased oleogels have even already successfully been applied as a replacer for trans- and saturated fats in food products such as ice cream [17], cookies [18] or confectionery fillings [19]. Waxes are chemically comprised of hydrocarbons, fatty acids, fatty alcohols, wax esters, ketones and/or aldehydes [20, 21]. Natural waxes with different chemical composition will expose different melting points and therefore also different crystallization and gelation behavior in liquid oils [22]. During cooling, the building blocks formed by the crystalline aggregates can entrap the liquid oil into a solid-like structure. The intramolecular (weak bonding such as H-bonding and polar-polar interaction) and intermolecular interactions (dipole-dipole interactions) between the wax constituents stabilize the structure of the crystalline network, resulting in the formation of an oleogel [11]. At temperatures below the crystallization temperature, the crystalline particles of waxes thermodynamically undergo further reorganization of the structural network [3]. The strength, consistency, brittleness and stability of the wax-based oleogels can be governed by changing the wax concentration, the cooling rate, shear rate or the solvent [13, 23, 24]. The modification of these factors will alter the gelator-gelator interaction and the solvent-gelator interaction, leading to a change in gelling properties. Therefore, it is of utmost importance to study the crystallization and gelation behavior of wax-based oleogels, while taking the melting temperature, wax concentration, cooling rate and shear rate into account. Most of the studied waxes are high-melting materials (more than 50 °C) like rice bran wax, sunflower wax, carnauba wax, candelilla wax or bees wax [10]. However, to the author’s knowledge, the gelation ability of low-melting waxes, such as berry wax, has not been in-depth studied. Also, these waxes have the potential to reduce the amount of trans- or saturated fats in low-temperature food systems such as ice-cream or whipped cream. To make a comprehensive case, high-melting sunflower wax and low-melting berry wax were applied at varied concentrations (1.0 to 5.0%wt) to structure rice bran oil into wax-based oleogels. The influence of the storage time was examined by observing the

changes in crystallization and gelation behavior of these wax-based oleogels using a variety of complimentary techniques (differential scanning calorimetry, polarized light microscope, cryo-SEM, powder X-ray diffraction and rheology).

Materials and Methods Materials Rice bran oil (RBO, onset and peak melting temperatures: Tm, onset = −73.95 °C, Tm, peak = −11.14 °C) (Suriny brand, Surin Bran Oil Company, Thailand) was purchased from a local supermarket in Ghent (Belgium). The fatty acid composition of RBO includes oleic acid (44.2%wt), linoleic acid (30.0%wt), palmitic acid (19.9%wt), stearic acid (2.1%wt) and arachidic acid (1.1%wt). Sunflower wax – a by-product from the sunflower-oil dewaxing (SW, 96% of long chain wax esters – WEs, 3% of free long chain fatty acids – FFAs, 0.3% of long chain free fatty alcohols – FALs and 0.2% of hydrocarbons - HCs) and berry wax – being extracted from the fruits of the Rhus verniciflua tree (BEW, 0.7% of WEs, 17% shortchain FFAs of C16, C18 and C18:1, 12% diglycerides of C32, C34 and C36, 63.7% triglycerides of C48, C50, C52, C54 and 4.24% FALs) were kindly donated by Kalhwax GmbH & Co. KG (Trittau, Germany). The compositional analysis of natural waxes, including SW and BEW has been reported [22]. Preparation of Samples To prepare the samples, SW and BEW were dispersed in RBO at different concentrations (1.0–5.0%wt). The samples were heated at 90 °C under mild agitation (200 rpm) using a magnetic stirrer (Model EM300T, Labotech Inc., Germany) until clear solutions were obtained (10 to 30 min). The clear oily solutions were subsequently cooled down to 5 °C at a cooling rate of 5 °C/min and stored at 5 °C overnight in a thermal cabinet for further experiments. Thermal Behavior in DSC The thermal profiles of waxes in bulk and in RBO were examined with a Q1000 DSC (TA Instruments, New Castle, Delaware, USA) equipped with a refrigerated cooling system. Nitrogen was used as purge gas. The cell constant and temperature were set with indium (TA Instruments). An additional temperature calibration was done using azobenzene (SigmaAldrich, Bornem, Belgium) and undecane (Acros Organics, Geel, Belgium). The sample (weighing from 6.0 to 8.0 mg/ cup) was placed inside an aluminum pan and sealed with an aluminum lid (TA Instruments). The DSC cup was placed inside the heating chamber and followed a specific thermal procedure. To observe the influence of various wax

Food Biophysics

concentrations on the thermal profiles of corresponding oleogels, the samples were cooled from 90 °C to 0 °C at a rate of 5 °C/min, kept isothermally at 0 °C for 10 min before being heated to 90 °C at a rate of 5 °C/min. Characteristic parameters of the thermal curves, including onset temperature (Tc, onset and Tm, onset), peak maximum temperature (Tc, peak and Tm, peak) and melting enthalpy (ΔHc and ΔHm) were obtained using the software TA Universal Analysis provided by the instrument supplier. Polarized Light Microscopy The crystal morphology of the wax-based oleogels was compared at a constant wax concentration of 5.0%wt. After placing the sample on a glass microscope slide and coverage by a slit, the samples were equilibrated at 90 °C and cooled to 5 °C at a cooling rate of 5 °C/min on a hot stage connected to a Linkam T95 System Controller (Linkam Scientific Instrument Ltd., Surrey, UK). For the storage experiment, after each time interval (1, 2, 3 and 4 weeks), the slide containing the waxbased oleogel was equilibrated at 5 °C on this stage to ensure the crystallizing state of the wax particles. The microstructure of wax crystals was observed under polarized light using a Leica DM2500 microscope (Wetzlar, Germany) equipped with a color camera Leica MC170 HD. Powder X-ray Diffraction Spectroscopy Polymorphism of the neat wax crystals and the waxes in RBO was investigated by XRD using a Bruker D8-Advanced Diffractometer (Bruker, Germany) (λ Cu = 1.54178A°, 40 kV, and 30 mA), equipped with an Anton Paar temperature control system composed of a TTK 450 low-temperature chamber connected to a waterbath (Lauda) and heating device (TCU 110 Temperature Control Unit) (Anton Paar, Graz, Austria). The samples were heated at 90 °C for 10 min and subsequently cooled to 5 °C at a cooling rate of 5 °C/min. To observe the polymorphic transition in 5.0%wt wax-based oleogels, the samples were measured after each storage-time interval at 5 °C (1, 2, 3 and 4 weeks). The short-spacing runs were recorded during cooling and at 5 °C using a Vantec-1 detector (Bruker, Germany). D-values were directly calculated by Diffract. Suite Eva software. Cryo-Scanning Electron Microscopy (cryo-SEM) In order to visualize the wax crystals under cryo-SEM, wax-based oleogels were partly de-oiled using isopropanol and ethanol to remove the surface oil and the oil bound within the crystal network. A known quantity of oleogel was respectively extracted with isopropanol for 2 days and ethanol for 1 day in a glass vial without stirring. The mixing ratio was 1 oleogel:50 solvent. The vial was placed inside a thermocabinet

at 5 °C to remove the oil and to ensure that the crystals do not melt. After decanting the solvent, the crystalline sample was collected and dried at 5 °C inside the thermocabinet overnight. This method will minimize the percentage of collapsed structure and remain mostly the crystalline network of the oleogel. The extracted sample was then placed on the sample holder, which was plunge-frozen in liquid nitrogen and transferred into the cryo-preparation chamber (PP3010T Cryo-SEM Preparation System, Quorum Technologies, UK). The frozen sample was subsequently sublimated, sputter-coated with Pt and examined in a JEOL JSM 7100F SEM (JEOL Ltd., Tokyo, Japan). Rheological Behavior All the rheological measurements were carried out using an advanced rheometer AR2000ex (TA Instruments, New Castle, USA) equipped with a Peltier system and water bath (Julabo, Seelbach, Germany) for temperature control. Parallel plate (cross–hatched; diameter, ϕ = 40.0 mm; gap =1000 μm) geometry was used to observe the linear visco-elastic region (LVR) of the wax-based oleogels by logarithmically increasing the oscillation stress from 0.01 to 1000 Pa at a frequency of 1.0 Hz. The starch pasting cell (shear rate factor = 4.500 s−1, shear stress factor = 48,600 1/m3, gap =5500 μm) was utilized to record the gelling points during cooling from 90 to 5 °C at a cooling rate of 5 °C/min. The oleogels were then kept at 5 °C for 3 h to investigate the isothermal change in gelling properties of wax-based oleogels. To study the influence of shearing on the gelling behavior of oleogels, the measurement was done by applying different shear rates (50, 75 and 100 s−1) during cooling the wax-based solution from 90 to 5 °C at a constant cooling rate of 5 °C/min. Frequency sweeps (0.01–100 Hz) were subsequently performed at 5 °C and a stress value within LVR to investigate the time-dependent deformation behavior of oleogels after each individual cooling-shear step.

Results and Discussion Characterization of Oleogels Prepared with SW and BEW Thermal Behavior The thermal properties of the neat waxes and wax-based oleogels were investigated using differential scanning calorimetry (DSC). The oleogels were prepared at concentrations varying between their minimum gelling concentration in RBO and 5.0%wt. The minimum gelling concentration is the threshold concentration at which the wax-based system rheologically behaves as a gel (storage modulus G’ > loss modulus G^ ≥ 50 Pa) and does not flow within a certain experimental timeframe (24 h) at the setting temperature (5 °C) [25]. Below

Food Biophysics

this concentration, crystalline particles are present but in an insufficient amount to create a space-filling network. The minimum gelling concentration for both SW and BEW was determined to be 1.0%wt [25]. Figure 1 depicts the crystallization profile (A – B) and melting profile (C – D) of the neat waxes (straight lines) and their wax-based oleogels (shown as dotted lines and zoomed in the inset). Neat SW is considered to be a mono-component wax, containing a high amount of long-chain WEs (96%, mainly formed by arachidic acid C24, behenic acic C22, lignoceric acid C24 bound with 1-tetracosanol C24, 1-hexacosanol C26 and 1octacosanol C28) (Table S1) which explains the single, narrow and intense crystallization peak at a high temperature (75.86 ± 0.03 °C). This prominent peak also suggests the dominant activity of WEs in the crystallization behavior of SW in liquid oils. The small shoulder following this main peak indicates the co-crystallization of other molecular components, most likely the FFAs - the second predominant compound in SW (3%). However, this shoulder disappears on the melting

curve (Fig. 1c). It is possible that the co-crystallized component (causing the shoulder on the cooling curve) has an overlapping melting point with the predominant WE fraction, resulting in a broader melting peak as compared to the crystallization peak. Neat BEW displays three low-crystallization peaks, suggesting that the chemical components in BEW fractionate into three separated crystalline portions without co-crystallizing: one at a higher temperature (36.40 ± 0.16 °C) of FFAs, almost overlapping with the most prominent peak at 35.64 ± 0.19 °C of DAGs and the least intense peak at a lower temperature of 14.25 ± 0.07 °C of TAGs. After remaining the sample at 5 °C in 10 min, the distinctive melting peaks of FFAs and DAGs were clearer identified than those observed in cooling curves. The low melting and crystallization temperatures of BEW can be explained by the high amount of short-chain FAs, mostly palmitic acid C16, stearic acid C18 and oleic acid C18:1 (Table S1). Due to the dilution effect, only the predominant chemical components will govern the crystallization of waxes in RBO, which is revealed by the main crystallization peaks in DSC

Fig. 1 Thermal profiles of the neat waxes (main figure) and their corresponding 5.0%wt wax-based oleogels (inset figure): a crystallization curve SW; b crystallization curve BEW; c melting curve SW and d melting curve BEW (Exo up)

Food Biophysics

profiles of wax-based oleogels. Similar to neat SW, only one intense peak occurred on cooling and melting curve of 5.0%wt SW-oleogel, suggesting the dominance of WE fraction in crystallization activity of SW in RBO. Interestingly, only one peak appeared on thermal curves of 5.0%wt BEW-oleogel. It is believed that this peak originates from FFA fraction because TAGs and DAGs have a slow crystallization in liquid oils [26]. According to Martini, Tan and Jana [27] the crystallization of waxes in a triglyceride system (liquid oil) is most likely initiated by the supersaturation of waxes in this liquid oil, affected by the wax concentration. The dependence of the thermal properties on the wax concentration is illustrated in Table 1. The crystallization and melting temperatures drop with a decreasing concentration of wax in RBO and are lower compared to the neat waxes due to the dilution effect in RBO. At a concentration of 1.0%wt in RBO, BEW starts to crystallize at a very low temperature (Tc, onset = 3.35 ± 0.01 °C), which is lower than the setting experimental temperature (5 °C). Only after overnight storage at 5 °C, a gel was formed (G’ > G^ > 50 Pa). This can be explained by the insufficient undercooling at 5 °C and the slow crystallization of TAGs in BEW, which results in a longer gelling time at that temperature. The crystallization of BEW in RBO still continues after 10 min at 5 °C, demonstrating by a higher onset melting temperature than onset crystallization temperature. Due to a high amount of long chain WEs in SW, even at a concentration as low as 1.0%wt, SW started to crystallize at higher temperature (50.34 °C) and this value increases with the increasing SW concentration.

Table 1

Thermal parameters of SW and BEW in bulk and in rice bran oil (cooling rate 5 °C/min and heating rate 5 °C/min) Neat wax

ΔHc

ΔHm

Tc, onset Tc Tm, onset Tm

The energy required to crystallize and to melt the waxbased oleogels was much smaller than that for the neat wax, which can be seen from enthalpy values (Table 1). The melting enthalpy is a measure for the amount of crystalline mass in the oleogels, which is in its turn influenced by the initial wax concentration [28]. As the wax concentration increases, a higher supersaturation (more driving force for the wax crystallization) is obtained which induces the formation of more crystals, explaining the higher melting enthalpy at higher wax concentrations. Table 1 shows the enthalpy values obtained from the DSC measurements and those obtained by dividing the enthalpy of 100% neat wax for the percentage of wax in RBO. At any concentration of wax, the crystallization and melting enthalpy of SW-oleogel was higher than that of BEW-oleogel which indicates the higher resistance to temperature changes of SW-oleogel as compared to BEW-oleogel. Higher melting enthalpy and melting temperature values of SW-oleogel compared to BEW-oleogel is also related to the formation of more stable polymorphs in SW-oleogel, which is confirmed in Fig. 3. Except for 5.0%wt, 3.0%wt and 4.0%wt SW-oleogel that the actual enthalpy values are higher than the calculated values, the measured enthalpies of other SWand BEW-oleogels are smaller than the calculated enthalpy. The appearance of hysteresis (crystallization enthalpy ΔHc > melting enthalpy ΔHm) in BEW-oleogel was related to the heat of dissolution (the enthalpy change associated with the dissolution of the crystals) during melting [29]. As confirmed by Rocha et al. [30] the temperature increase caused a network rupture and the exothermic dissolution of the wax

5.0%wt

4.0%wt

3.0%wt

2.0%wt

1.0%wt

SW 216.73 ± 1.86 Calculated value* BEW 105.50 ± 1.99

11.91 ± 0.41 10.83 2.90 ± 0.01

9.76 ± 0.34 8.67 1.45 ± 0.02

6.87 ± 0.26 6.50 0.97 ± 0.03

2.84 ± 0.03 4.33 0.59 ± 0.01

2.04 ± 0.17 2.16 0.25 ± 0.02

Calculated value* SW 213.10 ± 4.95 Calculated value* BEW 102.17 ± 4.67 Calculated value* SW 75.86 ± 0.03 BEW 37.12 ± 0.27 SW 75.69 ± 0.09 BEW 35.64 ± 0.19 SW 71.69 ± 0.06 BEW 16.20 ± 0.05 SW 76.50 ± 0.06 BEW 44.39 ± 0.18

5.28 11.13 ± 0.47 10.66 2.95 ± 0.16 5.11 63.61 ± 0.50 12.69 ± 0.13 60.42 ± 0.78 9.80 ± 0.13 48.88 ± 0.08 11.07 ± 0.06 62.76 ± 0.23 18.32 ± 0.23

4.22 9.42 ± 1.10 8.52 1.02 ± 0.06 4.08 62.97 ± 0.05 10.85 ± 0.06 59.32 ± 0.05 6.86 ± 0.05 46.09 ± 0.39 10.36 ± 0.03 61.36 ± 0.09 16.86 ± 0.30

3.17 7.26 ± 0.18 6.39 0.86 ± 0.02 3.07 57.41 ± 0.50 8.34 ± 0.25 55.69 ± 0.84 3.88 ± 0.12 36.31 ± 0.59 7.80 ± 0.21 59.81 ± 0.56 14.96 ± 0.18

2.11 2.85 ± 0.28 4.26 0.43 ± 0.01 2.04 54.46 ± 0.45 6.28 ± 0.06 49.57 ± 0.07 1.47 ± 0.06 34.52 ± 0.17 6.26 ± 0.06 56.96 ± 0.46 10.38 ± 0.12

1.05 2.61 ± 0.06 2.13 0.21 ± 0.01 1.02 50.34 ± 0.11 3.35 ± 0.01 46.90 ± 0.64 0.32 ± 0.02 33.40 ± 0.34 5.03 ± 0.11 54.55 ± 0.11 8.83 ± 0.48

The experimental data are expressed as means ± standard deviation of three repetitions *Calculated values of enthalpy from relatively concentrations (enthalpy value of 100% wax/percentage of wax in oleogel)

Food Biophysics

crystallites might win against the visualization of the endothermic melting event. Hysteresis is observed for the all BEW-oleogels, even at the lowest gelling concentration (1.0%wt). For SW-oleogel, the hysteresis only occurred at a SW concentration higher than 4.0%wt. The thermal analysis confirmed that the chemical composition of the waxes is a determining factor for the crystallization behavior of the waxes in liquid oil. Crystal Morphology and Polymorphism Under cooling, the wax crystallites start to appear and connect into a network which is strongly dependent on the morphology and polymorphism of crystalline particles. The appearance of those wax crystals can be visualized with different techniques, among which polarized light microscopy (PLM) and cryo-SEM are the most interesting ones. Figure 2a shows the crystals of SW under PLM as long needles which are in reality the edges of platelet crystals [31]. The internal structure of the gel network can be exposed with cryo-SEM after doing sublimation and sputtering. Figure 2c displays the crystals of SW as thin and wide platelets imbrating together. Those wide-surface platelets are composed of different thinner platelets which are piled upon each other. The occurrence of platelet crystals are attributed to the presence of long-chain WEs, which was confirmed by Chambers, Ritchie and Booth [32], Koch and Ensikat [33] and Hwang, Kim, Singh, Winkler Moser and Liu [7]. The appearance of 5.0%wt BEW-oleogel is much more transparent as compared to SW-oleogel. This is because the BEW microcrystallites are much smaller than the SW crystals Fig. 2 Crystal morphology of 5.0%wt SW-oleogel (left side) and 5.0%wt BEW-oleogel (right side) under PLM (a and b) and cryo-SEM (c and d)

(Fig. 2b vs a). Cryo-SEM visualization of the BEW oleogel shows a porous structure composing of thin tiny platelets with irregular margins, orientating perpendicularly from the surface and imbricating like tiles (Fig. 2d). During the isothermal stage at the setting temperature (5 °C), different platelet crystals could grow simultaneously from one nucleation site or a new crystal could emerge from a pre-existing crystal. Those platelet crystals tend to pile upon each other, forming crystal clusters and aggregates (Fig. 2c and d). As observed under PLM, the single crystals in waxbased oleogels reached their final size right after the time that the oleogels were cooled to below their melting temperatures. During the rest of the cooling step, the crystals aggregates restructure the crystalline network (Fig. S1). Figure 3 depicts the wide angle X-ray diffractograms of neat SW, neat BEW and their corresponding wax-based oleogels during cooling from 90 to 5 °C at a cooling rate 5 °C/min. Figure 3a shows two distinct short-spacing peaks (WAXD) at d-value of 0.415 nm and 0.373 nm (d = n x wavelength/2 x sin(theta)), which is characteristic for an orthorhombic sub-cell structure (β’-morphology) [34]. The large amount of long-chain WEs present in SW induced crystallization at very high temperatures, which coincides with the thermal behavior observed in DSC. Figure 3b reveals a similar lateral packing of 5.0%wt SW-oleogel although the shortspacing peaks are less intense due to the dilution effect. Figure 3c shows the appearance of one small peak at 0.415 nm in the wide angle region of neat BEW, which is evidence for a hexagonal symmetry (α-morphology). Similarly, not influenced by the dilution effect, BEW-oleogel exhibits a small short-spacing peak at d-value of 0.415 nm, indicating its hexagonal lateral packing (Fig. 3d). According

Food Biophysics

Fig. 3 Wide angle X-ray diffractograms of SW (a and b) and BEW (c and d) in bulk (left side) and in RBO (right side)

to Larsson and Larsson [35], when orthorhombic waxes show a phase transition at temperatures close to their melting points, the orientation order of the Bzig-zag^ planes was lost and the molecules could rotate around their length axis to form a hexagonal symmetry, which could be the case for BEW-oleogel. Rheological Gelling Behavior of SW- and BEW-Oleogels Under an applied oscillation stress, the deformation of a viscoelastic oleogel will show a Bductile^ or Bbrittle^ type, depending on the uniformity of the linkage strength among the crystalline particles [11, 36]. In the stress sweep tests, it can be observed that both SW- and BEW-oleogels have a higher G’ than G^, evidencing its gel state at a concentration as low as 1.0%wt (Fig. 4a). Wang, Liu, Xiong and Li [37] reported that the higher amount of junctions formed by the platelet crystals are responsible for the construction of a strong molecular network. Despite a higher average storage modulus (G’LVR), SW-

oleogel presented a narrower linear visco-elastic region (LVR), indicating that SW-oleogel was more brittle than BEW-oleogel. However, the oscillation yield stress (the point at which the structural linkage is broken down, G’ = G^) of BEW-oleogel was a little bit lower than SW-oleogel, confirming the better stability of SW-oleogel under the applied stress. A weaker gel strength of BEW-oleogel is attributed to the less stable alpha morphology as compared to the more stable β’ morphology in SW-oleogel (Fig. 3). The timedependent deformation behavior of the two oleogels is illustrated in Fig. 4b. The G’ and G^ curves were mostly linear in the range of angular frequency from 0.1 to 500 rad/s (G’ > G^), indicating that the gels had good tolerance to the rate of deformation and the bonds forming the network were almost permanent within the time frame of the performed test. A later dropping point at a higher angular frequency again demonstrates a higher stability in the structure of SW-oleogel than BEW-oleogel.

Food Biophysics

Fig. 4 LVR (a) and frequency sweep (b) of 1.0%wt wax-based oleogels; Gelling behavior (G’ and G^) of 5.0%wt wax-based oleogels during cooling from 90 °C to 5 °C at a rate of 5 °C/min (c), and during isothermal cooling at 5 °C in 3 h (d); critical stress (e) and oscillation

yield stress (f) of wax-based oleogels at different wax concentrations; and complex modulus (g) and delta (h) of 5.0%wt wax-based oleogels after being sheared at different rates

Food Biophysics

At a cooling rate of 5 °C/min, the gelling points (the temperature at which G’ = G^) of 5.0%wt SW and BEW oleogels are 47.1 ± 0.9 °C and 7.1 ± 0.3 °C, respectively. The difference between Tgel and Tc, peak of SW-oleogel is much larger than that of BEW-oleogel, demonstrating a prominent delay in gelation of the high-melting wax (SW) as compared to the lowmelting wax (BEW), for which the crystallization and gelation occurs almost simultaneously at very low temperatures. At the end of the cooling, the oleogel formed by SW crystals already achieved a strong gelling behavior (G’ - G^ ≥ 1 decade), while BEW-oleogel behaved like a weak gel (G’ - G^ < 1 decade) due to the undercooling of low-melting BEW in a short time at 5 °C (Fig. 4c). Nevertheless, during 3 h at 5 °C, the gelation of the low-melting wax (BEW) continued undergoing to form a stronger crystalline network (Fig. 4d). This is attributed to the further crystallization of slow crystallizing components in BEW (DAGs and TAGs) at 5 °C. By contrast, the gelation of the high-melting wax (SW) already finished during the cooling step. There was no more increase in the gel strength even after 3 h at 5 °C. Despite the development of the crystalline network, BEW-oleogel still exhibited a weaker gelling behavior than SW-oleogel at the end of the aging period. Also the wax concentration strongly influences the gelling behavior of wax-based oleogels (Fig. 4e and f). It is shown that SW-oleogel had a higher oscillation yield stress and a lower critical stress than BEW-oleogel, at every wax concentration. At a small concentration (≤ 2.0%wt), the difference in critical stress between SW and BEW oleogels is more prominent than at higher concentrations (> 3.0%wt). The critical stress and oscillation yield stress linearly increased with the increasing wax concentration. This results coincided with the data obtained by Dassanayake, Kodali, Ueno and Sato [8] and Hwang, Kim, Singh, Winkler Moser and Liu [7]. To further understand the impact of shearing on the structure of 5.0%wt SW- and BEW-oleogels, mixed flowoscillatory measurements were carried out by first subjecting the samples to a certain shear rate (3 shear rates were chosen 50, 75 and 100 s−1) during cooling at a constant cooling rate (5 °C/min), followed by doing frequency sweeps (0.01– 1000 rad/s). The graphs of after-shear moduli plotted as a function of angular frequency are shown in Fig. 4g and h. There is a clear and big difference in the time-deformation trend between SW- and BEW-oleogel after being sheared. BEW-oleogel still possessed a strong gelling network, as can be deduced from the higher complex modulus \G*\ (G’ > G^) and a lower delta degree as compared to SW-oleogel. The \G*\ curves of BEW-oleogel had a small increase at the initial small frequency, and remained a structural network until 10 rad/s before the structure breakdown occurred. However, even at the higher frequencies, delta degrees were lower than 45, indicating that BEW-oleogel still behaved as a good gel. On the contrary, after being sheared, the molecular structure of SWoleogel was almost ruptured, indicated by a dramatically

increase in \G*\ (G^ > G’) and delta degree (Fig. 4h). When the samples were sheared to the metastable conditions (10 °C above the crystallization temperature), the permanent junction zones would develop and prevent crystal breakage. However, shearing during the nucleation stage might impart some damage to a developing crystalline network [15, 24]. It is interesting that when the frequency surged to 500 rad/s, delta degree started to drop, demonstrating recovery in the gelling property of BEW-oleogel. This result coincided with the strongest thixotropic behavior of BEW-oleogel reported in the previous paper [25], while SW-oleogel exhibited the lowest thixotropic property among the strong oleogels (the other strong oleogels prepared with candelilla wax and bees wax). Mazzanti, Guthrie, Sirota, Marangoni and Idziak [38] also confirmed that a thixotropic system will be able to recover from the shearing damage by reforming the network connections. In this work, the influence of 3 different shear rates (50, 75 and 100 s−1) was not significantly different. However, a lower shear rate was less likely to rupture the network connections as compared to a higher shear rate, mitigating the network destruction. Change in Crystallization and Gelation Properties of Wax-Based Oleogels during Storage Wax-based oleogels are lipid-based semi-materials which will be utilized as food-grade additives in food products. The changes in chemico-physical properties of wax-based oleogels can directly influence the properties of food products during marketing time (transportation, distribution, marketing, household use…). In this research, SW- and BEW-oleogel have been investigated during 4 weeks at 5 °C by examining their changes in crystal morphology, melting temperature, frequency sweep and polymorphism. Visually, the oleogels prepared with SW and BEW in RBO did not show any phase separation (syneresis). During storage, the microplatelet crystalline particles at the junction zones and within the network have re-organized their molecular packing and achieve a higher level of structural organization [9]. Aggregation and clustering among the microplatelet particles took place and re-enforced the structure of the oleogels over the storage time. After 3 weeks, also big spherulite crystals appeared in the microstructure of SW-oleogel apart from the aggregates of initial platelets (Fig. 5). The transformation in the microstructural network is clearly reflected by the change in the melting temperatures of wax-based oleogels. A new small peak appeared at 20.7 ± 0.05 °C on the melting curve of 5.0%wt SW-oleogel after 4 weeks at 5 °C, could originate from the new spherulite clusters (Fig. 6a). There was almost no change in melting temperature of the main peak around 62.66 ± 0.21 °C after the instant cooling to after 4 weeks at 5 °C.

Food Biophysics

Fig. 5 Change in crystal morphology of SW-oleogels (top) and BEW-oleogel (down) during aging storage

For BEW-oleogel, larger spherulitic crystals started to appear and aggregated to form denser network already after one week (Fig. 5). Most of the FAs in neat BEW are not free, but esterified in a glycerol backbone as TAGs and DAGs, which display a slow crystallization and lateral packing after the

isothermal storage at a certain temperature [26]. The growth and clustering of BEW-microcrystallites occurred simultaneously with the development of the rosette spherulites, showing the occurrence of polymorphic transition and Ostwald ripening in the system during the storage (Fig. 4d). The

Fig. 6 Change in melting profiles (a and b) and rheological frequency sweeps (c and d) of wax-based oleogels at different time intervals: SW-oleogel (a and c) and BEW-oleogel (b and d)

Food Biophysics

difference in melting profile of 5.0%wt BEW-oleogel is clearly observed after one week at 5 °C (one major peak at 32.54 ± 0.12 °C and one minor peak at 13.10 ± 1.05 °C – Fig. 6b), as compared to its melting profile after 5 min at 5 °C (one peak at 18.32 ± 0.23 °C – Fig. 1d) and after 90 min at 5 °C (one major peak at 19.8 ± 0.15 °C and one small shoulder at 32.00 ± 0.25 °C). The minor peak, which appeared at week 1, grew bigger from week 2 and shifted to a higher temperature (21.2 ± 0.16 °C) and the area of the major melting peak became larger. There is no significant difference in melting profile of 5.0%wt BEW-oleogel after 2 and 3 weeks. However, after 4 weeks at 5 °C, all the melting peaks came together and formed only one large peak around 33–34 °C. A frequency sweep measurement was subsequently conducted to test the difference in time-dependent deformation behavior of wax-based oleogels after each time interval. Despite the occurrence of big spherulite crystals in the microstructure of SW-oleogel, the G’ value in week 4 was not significant higher than that in week 3, 2 and 1 (Fig. 6c). The G’ of BEW-oleogel had a dramatically increase from week 1 to week 2 and week 3; however, there was no significant difference in G’ between week 3 and week 4 (Fig. 6d). The increase in the gel strength is related to the sintering between crystals via non-covalent bonds, which is undesirable for food products [26, 39]. In summary, the strength of SW-oleogel was more stable than BEW-oleogel, from which SW-oleogel exhibits a higher potential in food application. The WAXD short-spacing runs were carried out to confirm the results obtained in PLM and DSC and to examine the polymorphic transition of wax crystals during storage. Upon comparing the lateral packing in Figs. 7a and 3b, SW-oleogel still exhibited the β’-morphology after 2 weeks at 5 °C. In

week 3, one more short-spacing peak occurred at d-value of 0.473 nm, indicating a transition in polymorphism. This phenomenon even more clear after 4 weeks as 2 more peaks appeared at 0.453 and 0.387 nm, demonstrating the existence of both β and β’-morphology (Fig. 7a). Considering BEWoleogel, only one small WAXD peak appeared at 0.415 nm after the instant cooling from 90 to 5 °C, representing the αmorphology (Fig. 3d). However, after one week, two more peaks appeared at 0.473 nm and 0.387 nm, indicating a transition from α to β morphology (Fig. 7b). After 4 weeks, more β crystalline particles appeared and strengthened the microstructural network for BEW-oleogel.

Conclusions In summary, the difference in chemical composition between SW and BEW resulted in different thermal profiles, which dominate the crystallization habits in RBO. The high amount of long-chain WEs in SW resulted in a very high crystallization temperature compared to BEW, which mainly consisted of short-chain FAs (present both in a free form and bindingform on the glycerol backbones). The gelation of the lowmelting wax took place concurrently with its crystallization, while there was a prominent delay in the gelation of the highmelting wax. The high-melting WEs were also responsible for the formation of platelet crystals and a strong gelling property of SW-oleogel. The tiny microcrystallites in BEW-oleogels were attributed to the predominant short-chain FAs, which showed a slow crystallization and lateral packing, leading to the appearance and clustering of larger spherulites during the aging period. As a result, the molecular structure of the BEW-

Fig. 7 Polymorphic transition of SW (a) and BEW (b) crystals in RBO at different time intervals

Food Biophysics

oleogel underwent the re-organization and exhibited a stronger texture over 4 weeks at 5 °C. A better understanding on the changes of wax-based oleogels is helpful in controlling the quality of food products, in which these oleogels can act as structuring agents. This study would be an important basis to select the right wax for each unique food application.

15. 16. 17. 18. 19.

Acknowledgements This research has been funded with support from the European Commission. This publication reflects the views only of the author, and the Commission cannot be held responsible for any use which may be made of the information contained therein. Authors want to thank Dr. Ashok R. Patel for his scientific inputs. Vandemoortele is recognized for its financial help in the acquisition of the Leica polarized light microscope and the scientific input. Hercules foundation is recognized for its financial support in the acquisition of the scanning electron microscope JEOL JSM-7100F equipped with cryo-transfer system Quorum PP3000T (grant number AUGE-09-029).

20. 21. 22. 23.

24.

References 1. 2. 3. 4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14.

C.M. Oomen, M.C. Ocké, E.J. Feskens, M.-A.J. van Erp-Baart, F.J. Kok, D. Kromhout, Lancet 357(9258), 746–751 (2001) D. Mozaffarian, M.B. Katan, A. Ascherio, M.J. Stampfer, W.C. Willett, N. Engl. J. Med. 354(15), 1601–1613 (2006) A.G. Marangoni, N. Garti, Edible oleogels: structure and health implications (AOCS Press, Urbana, 2011) A.R. Patel, N. Cludts, M.D.B. Sintang, A. Lesaffer, K. Dewettinck, Food Funct. 5(11), 2833–2841 (2014) A.R. Patel, D. Schatteman, W.H. De Vos, K. Dewettinck, RSC Adv. 3(16), 5324–5327 (2013) A.R. Patel, D. Schatteman, W.H. De Vos, A. Lesaffer, K. Dewettinck, J. Colloid Interface Sci. 411, 114–121 (2013) H.S. Hwang, S. Kim, M. Singh, J.K. Winkler Moser, S.X. Liu, J. Am. Oil Chem. Soc. 89(4), 639–647 (2012) L.S.K. Dassanayake, D.R. Kodali, S. Ueno, K. Sato, J. Am. Oil Chem. Soc. 86(12), 1163–1173 (2009) J.F. Toro-Vazquez, J.A. Morales-Rueda, E. Dibildox-Alvarado, M. Charó-Alonso, M. Alonzo-Macias, M. González-Chávez, J. Am. Oil Chem. Soc. 84(11), 989–1000 (2007) J. Daniel, R. Rajasekharan, J. Am. Oil Chem. Soc. 80(5), 417–421 (2003) A.G. Marangoni, J. Am. Oil Chem. Soc. 89(5), 749–780 (2012) L.S.K. Dassanayake, D.R. Kodali, S. Ueno, K. Sato, J. Oleo Sci. 61(1), 1–9 (2012) A.I. Blake, A.G. Marangoni, Food Biophysics. 10(4), 456– 465 (2015) A.J. Martins, M.A. Cerqueira, L.H. Fasolin, R.L. Cunha, A.A. Vicente, Food Res. Int. 84, 170–179 (2016)

25. 26. 27. 28. 29. 30.

31. 32. 33. 34. 35. 36. 37. 38. 39.

A.I. Blake, A.G. Marangoni, Food Biophys. 1–13 (2015) S. Jana, S. Martini, J. Agric. Food Chem. 62(41), 10192– 10202 (2014) D.C. Zulim Botega, A.G. Marangoni, A.K. Smith, H.D. Goff, J. Food Sci. 78(9), C1334–C1339 (2013) A. Jang, W. Bae, H.-S. Hwang, H.G. Lee, S. Lee, Food Chem. 187, 525–529 (2015) C.D. Doan, A.R. Patel, I. Tavernier, N. De Clercq, K. Van Raemdonck, D. Van de Walle, C. Delbaere, K. Dewettinck, Eur. J. Lipid Sci. Technol. 118(12), 1903–1914 (2016) P.E. Kolattukudy, Chemistry and biochemistry of natural waxes (Elsevier Scientific Pub. Co., Amsterdam, 1976) M. Regert, Organic mass spectrometry in art and archaeology, (2009), p. 97–129 C.D. Doan, C.M. To, M. De Vrieze, et al., Food Chem. 214, 717– 725 (2017) M. Chopin-Doroteo, J.A. Morales-Rueda, E. Dibildox-Alvarado, M.A. Charó-Alonso, A. de la Peña-Gil, J.F. Toro-Vazquez, Food Biophysics. 6(3), 359–376 (2011) F. Alvarez-Mitre, J. Morales-Rueda, E. Dibildox-Alvarado, M. CharóAlonso, J. Toro-Vazquez, Food Res. Int. 49(1), 580–587 (2012) C.D. Doan, D. Van de Walle, K. Dewettinck, A.R. Patel, J. Am. Oil Chem. Soc. 92(6), 801–811 (2015) D. Johansson, B. Bergenståhl, J. Am. Oil Chem. Soc. 72(8), 911– 920 (1995) S. Martini, C.Y. Tan, S. Jana, J. Food Sci. 80(5), C989–C997 (2015) A.I. Blake, A.G. Marangoni, J. Am. Oil Chem. Soc. 91(6), 885–903 (2014) D.J. Abdallah, L. Lu, R.G. Weiss, Chem. Mater. 11(10), 2907–2911 (1999) J.C.B. Rocha, J.D. Lopes, M.C.N. Mascarenhas, D.B. Arellano, L.M.R. Guerreiro, R.L. da Cunha, Food Res. Int. 50(1), 318–323 (2013) A.I. Blake, A.G. Marangoni, Food Struct. 3, 30–34 (2015) T. Chambers, I. Ritchie, M.A. Booth, New Phytol. 77(1), 43–49 (1976) K. Koch, H.-J. Ensikat, Micron 39(7), 759–772 (2008) H. Ensikat, M. Boese, W. Mader, W. Barthlott, K. Koch, Chem. Phys. Lipids 144(1), 45–59 (2006) K. Larsson, K. Larsson, Lipids: molecular organization, physical functions and technical applications (Oily Press, Dundee, 1994) A. Patel, M. Babaahmadifooladi, A. Lesaffer, K. Dewettinck, J. Agric. Food Chem. 63(19), 4862–4869 (2015) R. Wang, X.-Y. Liu, J. Xiong, J. Li, J. Phys. Chem. B 110(14), 7275–7280 (2006) G. Mazzanti, S.E. Guthrie, E.B. Sirota, A.G. Marangoni, S.H. Idziak, Cryst. Growth Des. 3(5), 721–725 (2003) A.P.B. Ribeiro, M.H. Masuchi, E.K. Miyasaki, et al., J. Food Sci. Technol. 52(7), 3925–3946 (2015)