2.1. Physico-Chemical Properties
The measured physico-chemical properties of the oleogels are presented in
Table 1. The gelation time (GT) is the total time elapsed to solidify a melted oleogel under specified conditions. It was also acknowledged that GT depends on the organogelator type and concentration, oil type, melting and cooling conditions, presence of shear and ultrasound, and others [
10,
12]. Consequently, comparison of different studies seems meaningless, but GT within a study provides an insight into how fast oleogels could be formed. In this study, thyme containing oleogel (SWO-TE) was gelled at 0.75 min and cumin containing oleogel (SWO-CN) was gelled within 1.00 min at ambient temperatures. This was mostly due to the sunflower wax, which yielded gels in similar time scales in previous studies [
5,
6,
7,
13,
14,
15]. It seems that the presence of solid spice particles does not interfere with GT values. A similar finding was also observed for the oil binding capacity (OBC) values. Clearly, sunflower wax (SW) yielded a reasonably strong and able network to entrap the liquid oil, similar to the previous studies listed above. The oleogel samples were also stable at the applied centrifugal force, which indicates a strong gel nature of the SW at 10 wt% with spice present.
The solid fat index (SFI) values of the oleogel samples were 4.33% and 4.30% for the SWO-TE and SWO-CN samples. The values indicate the level of total solids at 20 °C measured by the pulsed NMR instrument. In oleogels, solid content is usually originated from the added organogelator and from the crystallized triglycerides of the oil used. In this study, 10 wt% SW and VOO yielded SFI values at 20 °C, and these values were quite similar to previous studies listed above, which utilized around 2–10 wt% of SW and oils of sunflower seed, hazelnut, and olive. Most table margarines contain around 7–20% SFI at around 20–25 °C, as indicated [
16]. Oleogels are semi-solid to solid preparations at serving temperatures with quite low SFI values, proving their health advantage, which is due to the low levels of saturated and trans-fatty acids, respectively.
The color of an oleogel is the sum result of the ingredients used. In this study, the greenish VOO and spices with yellow and red tones were used with cream-toned SW (observe graphical abstract). The VOO used had 58 L value, −11 a* value and 22 b* value as indicated in the materials section. The SWO-CN sample was brighter (66.35 L value) than the SWO-TE sample having 50.88 L value. Since thyme was a darker spice than cumin, this finding was expected. Likewise, cumin-including oleogel had more redness (7.03 a* value) than the thyme-including one (5.38 a* value). Furthermore, yellowness of the SWO-CN sample (31.27 b* value) was significantly higher than that of the SWO-TE sample (22.82 b* value). Furthermore, it was indicated in previous studies that color of olive oil oleogels were quite different than the starting liquid olive oil due to phase change, respectively [
5,
6,
13]. Overall, colors of the samples were observed as greenish-yellow tones, and these colors could be acceptable for olive oil consumers who are accustomed to consuming green-yellow colored VOOs.
To determine the stability, the free fatty acidity (FFA) and peroxide values (PV) of the samples were measured. The VOO used had 0.97% oleate free fatty acidity and 11.5 meq O
2/kg oil peroxide value. Both FFA and PV values of the SWO-TE sample were significantly higher than those of the SWO-CN sample (
Table 2). The Turkish Codex for olive oil permits maximum values of 0.8% oleate as FFA and 20 meq O
2/kg oil PV for the ‘extra virgin-EVOO’ class, and a maximum of 2.0% oleate as FFA and 20 meq O
2/kg oil PV for ‘virgin-VOO’ class oils [
17]. Similar values were also indicated for EU Council Regulation (EC) No 1234/2007 [
18]. Hence, these new spreadable oleogel products could be accepted as ‘virgin class’ olive oil products. Of course, these samples are no longer liquid oils, but spreadable samples with added spices, and could be acceptable by regulations.
2.2. Structural Properties
The X-ray diffraction pattern graphics of the samples are shown in
Figure 1. Both samples had quite similar patterns. The wide-angle (WAXS) region diffraction peaks of the samples were observed at around 3.71 and 4.09 Å for the SWO-TE sample, and at around 3.71, 4.11, and 4.59 Å for the SWO-CN sample, respectively. According to the official method Cj 2-95 [
19], a single peak at d = 4.15 Å indicates the α polymorphic form, two peaks at positions d = 3.8 Å and d = 4.2 Å indicate the β′ polymorphic form, and a peak at position d = 4.6 Å indicates the β polymorphic form.
Consequently, both samples include β′ polymorphic crystal forms. The 4.59 Å peak in the SWO-CN sample does not seem very definite, and might be due to polymorphic transformation. Since both oleogels were prepared from the same VOO and SW at the same concentrations, and the only difference was the kind of the spice added, the presence of the same β′ polymorphic form is quite expected. It has been well documented that polymorphic types of solid fat determine its crystal size, shape, and stabilities. Accordingly, among the four main types of fat polymorphs, γ polymorph (transition state) occurs upon rapid cooling of melted fat. It is transparent, quite unstable, and immediately converts to a more stable α polymorph. The α polymorph is usually characterized with fine, waxy, less stable, least dense, and lowest melting point crystals. Under different cooling regions, β′ type polymorph could form, which yields intermediate density, medium melting point, orthorhombic chain packing crystals with very fine, creamy, soft texture, and immobilizes the maximum amounts of liquid oil due to higher surface area. It was observed that under certain conditions, a fat with β′ polymorph could transform to β polymorph. The β polymorph yields the highest melting point, most dense and stable form, with a coarse and sandy texture. All polymorphs from the same oil result in exactly the same liquid oil upon melting. Generally, polymorph type is based on the oil purity and oil triglyceride configurations, cooling process parameters, storage conditions, shear, and similar factors. In the edible fat industry, certain types of oils were preferred according to their common polymorphic habits to prepare margarine, shortening, and similar products. Sensory quality and palate properties of solid fats were greatly affected by polymorph types, and β′ polymorph was usually preferred in margarine, spreads, and similar products, while β-type polymorph was preferred for sugar confectionery and bakery. It was also acknowledged that different oleogels showed different polymorph types especially according to organogelator type and processing conditions [
7,
11,
12,
13,
16].
2.3. Thermal Properties
Thermal properties of the oleogels were usually measured to determine their similarities to and differences from commercial solid fat products and to determine their melting behavior in mouth space (body temperature). The measured crystallization and melting temperatures and enthalpies are summarized in
Table 2.
The SWO-TE sample starts to melt at 48.41 °C and the melting peak occurs at 62.53 °C, while SWO-CN samples’ melting onset and peak temperatures were 53.51 and 62.83 °C, respectively. Clearly, there is no difference for the peak temperatures, but onset (starting to melt) temperatures were different. Thyme spice reduced onset temperature, possibly due to its particle properties causing lipid crystals to melt more easily or possibly due to some compounds leaching from the thyme into the oil. It has been acknowledged that melting temperature of an oleogel is mostly affected by the organogelator type and concentration, oil type (fatty acid composition and triglyceride configuration), and preparation techniques applied [
12]. It would be helpful to compare melting data of these sample with literature pertaining to SW oleogels with olive oils. In the study of Yılmaz and Öğütcü [
6], olive oil-sunflower wax oleogels were prepared at 3, 7, and 10 wt% SW, and the peak melting temperatures were measured as 58.26, 61.37, and 63.59 °C, respectively. In another study, 5 wt% SW added sunflower seed oil oleogels showed 33.41 °C peak melting temperature [
20]. Clearly, both amounts of added wax and oil type caused the differences in the measured melting temperatures. Oleogels prepared in this study had very similar thermal behavior to previously prepared similar VOO-SW oleogels. Hence, added spices have not interfered much with the melting habits. Further, it was indicated that most commercial breakfast and kitchen margarines and spreads were required to quickly melt at slightly above body temperature for cooling sensation on the palate without lingering greasiness or waxiness. Commercial shortenings are said to have somewhat higher melting temperatures based on their usage purpose [
16]. Spice-flavored oleogels intended to be used as spreadable fat-like preparations in this study showed somewhat higher melting temperatures, similar to almost all of the previous wax oleogels [
6,
7,
12,
14,
20]. Therefore, the suitability of these new oleogels as spreadable products and consumer attitudes towards them were evaluated by sensory analysis and consumer tests, and are discussed later.
2.4. Rheological Properties
At the beginning of rheological analyses, amplitude sweep tests for the samples were conducted to determine the non-destructive deformation range (the linear viscoelastic region, LVR) and upper limit of this range. Consequently, between 0.01 and 100% strain, 1 Hz frequency at 10 °C, the amplitude sweep tests were conducted and the LVR strains of the samples were determined as 0.014% and 0.015% for the SWO-TE and SWO-CN samples, respectively.
Applying the frequency range of 0.1–100 Hz at LVR strains and 10 °C constant temperature, the frequency sweep tests were completed. The frequency sweep graphics are given in
Figure 2. Frequency sweep tests usually provide the time-dependent behavior of a sample in the non-destructive deformation (LVR) range. Generally, this test yields information about behavior and inner structure of a gel, as well as its storage stability. High frequencies simulate fast motion on a short time scale, while low frequencies simulate slow motion in long time scales. In this study, we preferred low-to-medium frequencies, which most food products would face during processing and distribution [
21]. The changes in the storage (G′) and loss (G″) modulus of the samples can be observed from the graphics. The SWO-TE sample had 13,000–100,000 Pa storage modulus and 12,000–14,000 Pa loss modulus values within the applied frequency range. Similarly, the SWO-CN sample had 16,000–101,000 Pa storage and 12,000–14,000 Pa loss modulus values. In both oleogels, storage modulus values were higher than those of the loss modulus values (G′ > G″), indicating that the materials were mostly similar to solid nature, in other words, they were true-gels.
Within the applied frequency range, the G′ > G″ condition prevails, and this condition proved that the samples were in a gel state. Descriptions of complex viscosity (η*) are useless in practice since for low frequencies, the η* curve approaches infinity. In rheology science, it was stated that the G′ of a sample resembles solid-like properties and describes the elastic portion, while the G″ resembles liquid-like or viscous properties. Consequently, if a sample had higher G′ values than G″ values, it would be more like a solid. In the samples of this study, this means that the prepared oleogels were in real gel-state, and must have enough storage stability since through all applied frequency ranges, the condition was maintained. Furthermore, during the measurement range, no crossover point (G′ = G″) was observed, indicating gel stability. Further, the loss factor (G″/G′ ratio) values ranged between 0.14 and 0.92 for SWO-TE, and between 0.14 and 0.75 for SWO-CN samples. This ratio indicates that these oleogels were also resistant to syneresis. Very similar findings were found previously for the SW and other plant wax oleogels [
6,
7,
10,
12,
14,
20,
21].
Time-dependent oscillatory viscoelastic behaviour of the prepared oleogel samples were evaluated by time sweep tests (
Figure 3). In these types of tests, both amplitude and frequency (1 Hz) were kept constant, and three different shear values were applied at three different time-domains to simulate the resting, destruction, and recovery-regions, as explained in the methods section.
Both oleogel samples were in gel state at resting condition under the LVR strains at the first region, respectively. In the second-domain, high strain was applied purposefully to destroy the structure, and both oleogels were deformed and became free flowing liquids, as evidenced from the lowering G′ and G″ values. Clearly the storage modulus values of both samples were lowered below their loss modulus values (G′ << G″), indicating that the gel network was broken down by mechanical effects. Most importantly, at the third time-domain, the high strain was removed from the samples, and samples re-gelled again, as evidenced from the enhancing G′ values, which again became higher than their loss modulus values to yield the G′ > G″ condition again. Consequently, this test proved that both samples were able to recover their lost structure due to high strain as soon as the strain was lowered. This behavior was typical in most wax oleogels studied before [
6,
8,
9,
13]. Overall, the mechanical stability of the prepared oleogels indicated that these new spreadable VOO preparations could be stable during food processing and distribution operations, where mixing, agitation, and some mechanical stress would be unavoidable. If the gel breaks down, then, after the cease of energy input, it can re-gel to provide solid-like structure to thus be spreadable again. This could be accepted as an advantage.
The temperature ramp test of the samples was presented in
Figure 4. These graphics provide the data to observe changes in the storage and loss moduli during the sample heating process under constant strain and frequencies. Consequently, these changes provide information about how the oleogels respond to temperature enhancements. Clearly, in both samples, the G′ and G″ values were almost constant and linear as temperature increases from 0 to around 52 °C. In this range the samples were well solidified and gelled. Softening of both gels starts at around 52–53 °C, and complete melting occurs at around 60 °C as the crossover point (G′ = G″) is reached. After the crossover point, the oleogels became free flowing liquids. These data concur with DSC-determined melting peak points, which indicate around 62 °C melting temperature. While DSC data provides only melting onset and peak temperatures and enthalpies, these rheological temperature ramp graphics provide the opportunity to follow gel structure during the heating process. Accordingly, these spreadable VOO oleogels would remain solid until around 52 °C, and hence would remain solid in summer season at ambient temperature. This condition has always been mentioned as an advantage of oleogels not requiring refrigeration during handling [
12,
14,
20]. Further, these new oleogels would remain a little longer in mouth space but eventually would melt to yield a good palate.
2.5. Volatile Aromatics Composition of the Oleogels
The headspace SPME-collected and GC-MS-quantified volatile aromatics compositions of the prepared oleogel samples are presented in
Table 3. The table shows the retention times, aroma descriptions, and mean peak % values of the determined volatiles. There were 22 compounds quantified in the SWO-TE and 20 compounds in the SWO-CN samples. Among them, 12 compounds (ethanol, 1-propen-2-ol-acetate, acetic acid, ethyl acetate, 1-penten-3-ol, hexanal, E-2-hexenal, cymol, limonene, gamma-terpinene, nonanal, farnesene) were common in both samples.
These compounds were usually described with ethereal, fruity, pungent, green, fresh, terpenic, citrus, and woody aroma descriptors. Since both oleogel samples were formulated with the same VOO and SW, most of these common volatiles must have originated from these ingredients. However, some would come from the spices (thyme and cumin), and would be common in both spices as well. Since the volatiles must come from the ingredients used to prepare the oleogels, it would be helpful to compare volatiles of the oleogels with the volatiles of VOO, SW, and the spices published in the literature.
As a virgin and natural oil, olive oils contain more than 100 listed volatile aromatic compounds [
1]. A very comprehensive recent review about volatile aromatic compounds, their analysis, and sensory perceptions of various olive oils were published [
22]. In this review, a pair of tables listed approximately 700 volatiles identified from olive oil samples in different studies. Most of these volatiles were characterized by molecular weight < 300 Da, with high and variable volatility, variable solubility, and capability to bind proteins or sensory receptors. When literature listed VOO volatiles were compared with the volatiles in
Table 3, it was observed that most of the identified volatiles in oleogel samples were listed in VOO samples in the literature. Only acetol, diethyl ketone, geranyl butyrate, cymol, 2-caren-10-al, carvacrol, and bergamotene were absent among the volatiles listed for VOO in the current literature, but were found in the oleogel samples prepared in this study. Consequently, these compounds must come from the spices or SW used. The VOO volatiles identified in the oleogel samples were defined with green, grassy, fruity, musty, fatty, leafy, tomato, and sour aroma terms.
To determine the aromatic volatiles originating from thyme spice in the SWO-TE sample,
Table 3 and literature about thyme volatiles were compared [
23,
24]. The aromatic volatiles of beta-pinene, beta-phellandrene, beta-myrcene, carene, limonene, beta-ocimene, gamma-terpinene, 2-carene, carvacrol, and bergamotene found in the oleogel sample were listed among the volatile constituents of thyme spice in the literature. These volatiles were associated with terpentine, herbal, spicy, minty, pine, woody, and citrus aroma notes.
Similarly, volatiles coming from the added cumin in the SWO-CN oleogel sample were identified by comparing with the literature indicating the volatiles found in cumin samples [
25,
26]. After screening
Table 3 with these references, the volatiles originating from cumin spice in the SWO-CN oleogel were determined as beta-pinene, beta-myrcene, gamma-terpinene, 2-caren-10-al, and carvacrol. These volatiles were identified with herbal, spicy, citrus, and woody aroma descriptions.
Generally, volatile aromatic compositions of the oleogel samples were in agreement with literature regarding the volatiles of the VOO and the spices. The volatile aromatic compositions of the spices alone as solid samples were not measured in this study, since it is well known that aromatic release of spices is strongly dependent on the food matrix, and the chemical nature (polar or non-polar) of the matrix [
27]. Consequently, we only referred to literature to identify the volatile aromatics originating from the spices used. Unfortunately, there is no data available regarding the volatile composition of sunflower wax (SW). In fact, the SW used in this study was quite pure, refined, faint, and odorless, as described by the producer. We assume that compounds yielding some fatty, aldehydic, and possibly waxy aroma notes must have originated from the SW. Since the literature about volatile aromatics composition, sensory descriptions, and consumer tests for various oleogels are scarce, these data would contribute significantly to the oleogel literature. The volatile aromatics data could provide information about the kinds and amounts of the chemical compounds responsible for the perceived aroma but cannot provide a human perception of the attributes and whether the consumer would accept the samples or not. Therefore, descriptive sensory analyses and consumer tests were also completed in this study.
2.6. Descriptive Sensory Analysis
A trained sensory panel decided to describe the oleogel samples with 13 sensory descriptor terms, and the collected data are summarized in
Table 4. Sensory ‘hardness’ was described as the force needed to push a knife into the sample, and the maximum and minimum points (10 and 0 scores) were referenced with tallow fat and yoghurt. Both samples had around 8.15 and 8.25 scores, indicating that the oleogels were not as hard as tallow, but much harder than yoghurt. In fact, they were similar to breakfast margarines. These sensory scores were also in good agreement with the rheological data given in
Figure 2. Both oleogel samples were found as fully ‘spreadable’, in comparison with the cream cheese reference for a 10-point score. This finding was observed as quite fulfilling since the goal was to develop spreadable VOO oleogels with added spices. Apparently, spice particles did not impact spreadability. The ‘liquefaction’ was defined as the amount of fat melting during spreading onto the bread surface. It measures the amount of melting by mechanical energy input. Both samples had fairly low (0.50) liquefaction scores. ‘Sandiness’ was defined as the perceived gritty texture on the tongue, and was found as 1.00 for both samples, in comparison with 10 score for semolina as the reference. The fresh olive oil itself was the 10-score reference for the ‘olive fruit’ descriptor, and the oleogels had around a little less than half (4.00–4.20) of the full-score. Since oleogels include SW and most importantly the aromatic spices, this decrease of ‘olive fruit’ description was expected. The ‘grassy’ score of the SWO-CN sample (7.00) was significantly higher than that of the SWO-TE sample (5.15). Usually, fresh VOO was characterized with a grassy or green aroma, and this difference could be attributed to the differences of the aroma potencies of the spices added. Clearly, some grassy notes were masked more by the thyme spice.
The ‘waxy’ scores of both samples were low (0.5) and not different. In previously prepared olive oil-SW oleogels the waxy scores were a little higher than this study [
6,
14]. The aromatic spices have masked the waxy odor attribute coming from the SW used. In fact, this situation could be well accepted, since the higher waxy odor was not preferred in oleogels [
12,
13,
14,
20]. In all fat-containing and long-term stored foods, ‘rancid’ was defined as aromas associated with oxidized oil, and referenced with used frying oil. The oleogels luckily had very low (0.50–0.52) rancid scores. Since fresh VOO was used, and the spices added were strong anti-oxidants, not much oil oxidation occurred during oleogel preparation. ‘Thyme’ and ‘cumin’ were defined for the oleogels with the spices themselves as the references. Of course, each of these aromas was only detected in the corresponding samples. The panel perceived more thyme aroma (5.23) in the SWO-TE sample than cumin aroma (3.00) in the SWO-CN sample (
Table 4). Of course, aroma intensity and volatility in oil media differ for the different spices, although both were added at 1 overall wt% level. The ‘hay’ attribute had around 1.00 score in the samples, and it was low but still perceivable. It was associated with dry straw and most probably originated from the added spices. As a mouth feeling attribute, ‘cooling’ was defined as the cold feeling inside the mouth during oleogel consumption. Compared to 10 score of menthol candy, the oleogels had around 2.05 and 2.55 scores. In fact, solid fats and chocolate provide some cooling sensation during mastication in mouth due to heat absorbed by the melting fat crystals [
16]. Since oleogels had some crystallized components, this sensation was perceived. ‘Mouth coating’ was defined as the perceived fatty coating on the palate and referenced with butter. The oleogels had a lower mouth coating than butter, but it was quite sensible. Overall, descriptive sensory analysis data is indispensable to provide real human sense descriptions of a food sample to the readers and has been very important in product formulations, comparison, and modifications to end up with successful samples. In fact, results of consumer tests were evaluated with sensory descriptive analysis and volatiles’ data to determine the defects of the sample to take corrective actions or to improve the attributes of the sample to achieve higher consumer acceptance. Of course, the other analytical data (physical, rheological, thermal, etc.) can serve the same purpose.