Document [original]

RESEARCH ARTICLE

www.ejlst.com

Inﬂuence of Minor Oil Components on Sunﬂower, Rice

Bran, Candelilla, and Beeswax Oleogels

Maria Scharfe,* Jonas Niksch, and Eckhard Flöter

The impact of the solvent composition on wax oleogels is addressed by (1)

increasing polar components (PC) in sunﬂower and canola oil through

thermal treatment and (2) removing minor components from untreated oils by

column chromatography. Subsequently, oleogels are produced at 0.05 and

10 °Cmin

−1using 4% or 10% w/w of either sunﬂower, rice bran, candelilla, or

beeswax. Oleogels ﬁrmness, break-up behavior during amplitude sweeps, and

gelation and dissolution are studied using penetration tests, rheology, and

diﬀerential scanning calorimetry (DSC), respectively. Moreover, the crystal

morphology of 4% w/w samples, gelled at 10 and 0.05 °Cmin

−1, is studied

using bright ﬁeld microscopy. Distinct eﬀects caused by the presence or

absence of PCs on the characteristics mentioned above are observed,

depending on the wax type. The formation of highly ordered wax crystal

structures is favored in oils without PCs and low cooling rates. Data on gel

formation and dissolution reveal a decrease in wax solubility in the absence of

PCs. In contrast, the critical gelation concentration (CGC) decreases when

PCs are present, independent of their concentration, indicating that PCs aid

network formation. Moreover, the break-up behavior during oscillatory stress

is signiﬁcantly diﬀerent, leading to more network fragments and higher

energy dissipation with increasing strain.

Practical applications: It is found that the oil composition, in particular, the

fatty acid composition of TAGs and dissolved minor polar oil components,

profoundly aﬀect wax oleogel properties. Although not all mechanisms

leading to these changes can be unraveled within this study, a fundamental

understanding of solvent composition’s role on oleogel formation,

dissolution, and network properties is vital in the light of product applications.

Moreover, trustworthy and comparable oleogel research can only be achieved

if the impact of solvent composition is considered in experiments. That way,

the capability of oleogels for industrial applications might be maximized. For

that, a detailed characterization of oil quality, particularly the fatty acid

composition and presence of minor polar components, is required to conduct

reliable scientiﬁc work in oleogel research.

The ORCID identiﬁcation number(s) for the author(s) of this article

can be found under https://doi.org/10.1002/ejlt.202100068

published by Wiley-VCH GmbH. This is an open access article under the

terms of the Creative Commons Attribution-NonCommercial-NoDerivs

License, which permits use and distribution in any medium, provided the

original work is properly cited, the use is non-commercial and no

modiﬁcations or adaptations are made.

DOI: 10.1002/ejlt.202100068

1. Introduction

Waxes are organic, lipophilic mixtures of

long alkyl chains having several functional

groups such as carboxyl, hydroxyl, ketones,

aldehydes, and esters. Additionally, the

chains may comprise unsaturated bonds.

Hence, plant waxes contain a heteroge-

neous mixture of wax esters (WEs), free

fatty acids (FFAs), fatty alcohols (FAOHs),

hydrocarbons (HCs), and up to 10% other

minor components.[1] The chemical com-

position, e.g., alkyl chain length and num-

ber of unsaturated bonds, and the ratio

of these components deﬁnes the waxes’

physicochemical characteristics such as dis-

solution temperature (TD) and solubility in

a particular solvent.

The composition considerably depends

on the wax source, growing conditions,

and extraction and puriﬁcation methods in-

dicated by the compositional diﬀerences

found in the literature.[1–11] Besides the

main constituents, minor wax components

such as sterol esters, esters of pentacyclic

triterpenoids, and alcohols and their con-

centration may inﬂuence, i.a., transition

temperatures.[12] However, in their native

state, waxes are solid at ambient temper-

ature and usually melt between 50 and

80 °C.[13]

When dissolved in low polar, organic sol-

vents such as edible oils at suﬃcient con-

centrations, they may form a 3D crystalline

network that can immobilize the liquid (see

Table 1). The resulting oleogels are promis-

ing candidates to replace traditionally struc-

tured lipid phases based on semi-solid

networks with high saturated fatty acids

contents.

Waxes of industrial signiﬁcance are beeswax (BWX), carnauba

(CRX), candelilla (CLX), rice bran (RBX), and sunﬂower wax

(SFX). The ability of waxes to gel, e.g., plant oils, relates to their

low solubility and crystal structure, which is a function of their

M. Scharfe, J. Niksch, E. Flöter

Department of Food Processing

Technical University Berlin

Berlin 13353, Germany

E-mail: [email protected]

Technology published by Wiley-VCH GmbH

2100068 (1 of 19)

www.advancedsciencenews.com www.ejlst.com

Table 1. Selection of literature, including the CGC determination of SFX, RBX, CLX, and BWX in diﬀerent oils. The list does not aim to cover all publications

measuring the CGC of waxes in edible oils. Values were determined at ambient temperature (20 or 25 °C), except.[50]

Wax CGC [% w/w] Δ% w/w Oil Method Ref.

SFX 0.5 0.5 Soybean Flow test [7]

1.0 1.0 Canola Flow test [3]

0.5 0.5 High oleic sunﬂower Flow test +Rheology [50]

1.0 0.5 Rice bran Flow test [26]

0.3, 0.5 &1.0 0.2, 0.5, and 1.0 Soybean, almond, canola, corn, grape seed, saﬄower &

sunﬂower (all 0.3)

olive, peanut, pumpkin, sesame, and walnut (all 0.5),

ﬂaxseed (1.0)

Flow test [51]

RBX 1.0 0.5 and 1.0 Olive oil Flow test [5]

1.0 1.0 Canola Flow test [3]

5.0 2.0 Rice bran Flow test [52]

5.0 0.5 Rice bran Flow test [26]

CLX 2.0 0.5 and 1.0 Reﬁned soybean Flow test [53]

2.0 0.5 and 1.0 Olive oil Flow test [5]

1.0-2.0 0.5 Soybean Flow test [7]

2.0 1.0 Canola Flow test [3]

0.75 0.25 High oleic sunﬂower Flow test [50]

1.0 0.5 Rice bran Flow test [26]

BWX 2.0–3.0 0.5 Soybean Flow test [7]

1.0 0.25 High oleic sunﬂower Flow test+Rheology [50]

4.0, 4.0 2 MCT, LCT Flow test [52]

1.5 0.5 Rice bran Flow test [26]

chemical composition and the interactions between individual

components and the gels’ production conditions, such as cooling

rate and shear. Moreover, the solvent composition and dissolved

PCs could impact the gelling ability of waxes by interacting with

the main wax components or modifying the network-deﬁning

interactions between the crystals. Similar eﬀects have been re-

ported for interactions in fat crystal networks in the presence of

molecules with functional groups.[14–17]

Hence, the main wax components and potential interaction

points will be discussed brieﬂy in the following. Hydrocarbons

with odd and even-numbered alkyl chains (27–36) self-assemble

into lamellar structures, forming crystalline microplatelets in hy-

drophobic solvents. These aggregate and form a 3D network sta-

bilized by mainly van der Waals forces (vdW).[18] The melting

temperature increases with increasing carbon chain length, si-

multaneously decreasing the solubility.[19] That applies to all wax

constituents.

Combinations of long-chain FFAs and FAOHs reportedly gel

vegetable oils at concentrations as low as 2% w/w and show syn-

ergistic eﬀects at speciﬁc composition ratios when having the

same chain length.[20] In addition, due to their carboxylic group,

hydrogen bonds form during the self-assembling process result-

ing in stronger network interactions than HCs. These bonds re-

portedly boost interactions in other oleogels systems as well.[21]

Consequently, their melting temperatures are higher than HCs

with the same number of carbon atoms.[20] Natural waxes con-

tain FAOHs with even-numbered carbon chains between 24–34

carbon atoms.[22] Their weight fraction in waxes ranges from less

than1%inRBXandSFX,2and6%inBWXandCLX,tomore

than 30% in CRX. Similar to FFAs strong hydrogen bonds form

between adjacent FAOHs through their hydroxyl groups.

Finally, wax esters are the main components in SFX, RBX,

BWX, and CRX, possibly impacting oleogel formation and

properties the most. They contain FA and FAOH moieties with

carbon chain lengths between 16–24 and 18–31, respectively.[22]

The proportion of the two alkyl chains in WE’s inﬂuences the

phase transition, with symmetric WE showing higher melt-

ing temperatures.[23,24] The self-assembly of WE is due to the

interlocking of alkyl chains which is more favored if adjacent

chains have the same number of carbon atoms.[25] These struc-

tural units form large plates of stacked crystals (symmetrical),

while asymmetric chains result in smaller, more needle-like

shapes.[23]

Besides processing conditions and wax composition, weak,

noncovalent bonds between crystals such as hydrogen bonds,

vdW forces, and 𝜋–𝜋-interactions additionally impact network

strength.[6,26,27] These may form by bridging between adjacent

crystals. Moreover, minor oil and wax components alter crys-

tallization kinetics and crystal morphologies. Indeed, a study

showed that minor wax components with functional groups such

as secondary alcohols or ketones do not develop the orthorhom-

bic structure typical for aliphatic compounds. That results from

reduced crystallinity which favors more loosely mixed molecular

packing.[3] Therefore, minor components may profoundly aﬀect

crystal morphology, changing macroscopic properties such as gel

hardness. Possible eﬀects due to minor oil and wax components’

are diﬃcult to assess. The more so because of the chemical diver-

sity and varying concentrations.

Technology published by Wiley-VCH GmbH

2100068 (2 of 19)

www.advancedsciencenews.com www.ejlst.com

Nevertheless, several publications addressed the role of oil type

on wax oleogels. The rheological and thermal behavior of SFX,

BWX, and paraﬃn wax (PFX, only HCs: C20+C40) was studied

in olive, corn, soybean, sunﬂower, saﬄower, and canola oil.[28]

Dissolution temperatures (TD) and transition enthalpies were af-

fected by the oil type, whereas the degree of change diﬀered for

all waxes. Similarly, the storage modulus, which represents the

number of interaction points within a network,[31] showed con-

siderable diﬀerences in oil and wax types. Another study assessed

the crystallization of RBX in a canola/soybean oil mix, camellia

oil, and olive oil.[29] While the gel formation temperature (TF)was

not signiﬁcantly aﬀected, TDvaried slightly with the oil type, in-

dicating minor changes in wax solubility. In line with that, pen-

etration hardness was aﬀected by the oil type. However, X-ray

diﬀraction (XRD) patterns revealed two RBX crystal types and

anisotropic crystal growth, resulting in needle-like crystals.

In addition to varying the oil type, adding small molecules

with functional groups is a popular method to tailor also wax

oleogel properties.[30,31] However, without standardization of the

solvent, the eﬀects of minor oil components and admixed molec-

ular species are challenging to unravel due to superimposition.

The admixing of adipic acid (2 carboxyl groups, Tm≈151 °C)

to CRX/soybean oil oleogels resulted in hydrogen bond forma-

tion visible in FTIR spectra.[32] These may form between the car-

boxyl groups of the acid and FFAs (84–85%) of CRX. Further-

more, XRD revealed modiﬁcations in the crystallinity in oleogels

with adipic acid manifested in ﬁbrillar instead of needle-like crys-

tals.

Similarly, combining CLX with diﬀerent concentrations of ei-

ther monoglyceride (MAG, C18, C22) and fully hydrogenated

crambe oil (FHCO, C22, C18) in high oleic sunﬂower oil (7% sat-

urated FAs) and soybean oil (16% saturated FAs) altered oleogel

properties.[33] Although in DSC thermograms, an increase of

transition enthalpy and TDis visible at high MAG and FHCO con-

centrations, this does not involve an equal increase in gel hard-

ness. Instead, gel ﬁrmness dropped to about 20% of the initial

value of CLX oleogels. The gel hardness increased by increasing

the additive concentration further due to the formation of indi-

vidual networks by FHCO or MAGs.

In contrast, tripalmitate (PPP) addition leads to a synergistic

eﬀect in CLX oleogels.[34] Here, systems with 1% CLX and up

to 1% PPP showed higher G’ than 3% CLX oleogels. Reportedly,

that is due to the cocrystallization of PPP and CLX.

However, all studies considered utilized oils from supermar-

kets without detailed characterization of, e.g., unknown levels

of minor polar components, FA proﬁle, or viscosity. The stan-

dardization of oils is recommended to avoid the superimpo-

sition of several eﬀects contributing to the modiﬁcations of

oleogels.[21,35–37] To the best of our knowledge, only one study in-

cluded an oil puriﬁcation procedure to remove minor oil compo-

nents before preparing SFX oleogels.[38] Diﬀerent soybean oils

with varying iodine values (112, 136, and 159) and PC levels

(7.5%, 6.2%, 2.8%, and stripped) were used, exemplifying diﬀer-

ent oil polarities. The absence of PC increased the transition en-

thalpies, indicating more crystalline material. Since the enthalpy

was highest for the most polar oil (highest IV), one could assume

that SFX is less soluble in more polar oils. However, the puriﬁca-

tion did not aﬀect gel ﬁrmness, but G′decreased in all stripped

samples compared to natural oils.

Secondary wax and oil components containing functional

groups may cause interactions with wax constituents. These

might interfere with molecular stacking during gel formation,

modify wax solubility, and promote mixed molecular packing

formation.[3,39] In addition to wax and oil composition, the pro-

cessing conditions such as cooling rate, application of mechani-

cal forces, and storage time essentially aﬀect wax oleogels.[2,12,22]

Nevertheless, these factors have been discussed in recent liter-

ature and will be kept constant in this study. However, distinct

eﬀects inﬂuencing wax oleogel properties were reported in the lit-

erature. Since these superimpose in non-standardized oils com-

prising PCs, it is impossible to disentangle individual contribu-

tions. Thus, the utilization of stripped oils oﬀers a more holistic

approach.

This study aims to provide a more coherent and relevant ap-

proach to describe the impact of solvent composition (PC and

TAG composition) on wax oleogels. To this end, canola and sun-

ﬂower oil with varying concentrations of PC (0–19.4%) are used

to produce wax oleogels utilizing SFX, RBX, CLX, and BWX. Sub-

sequently, the gel hardness, crystal morphology, rheological and

thermal properties, and CGC are analyzed. The latter was deter-

mined at 20 °C using a rheological approach with minimal con-

centration steps (0.05% w/w). Finally, an attempt is made to link

the data on solvent composition to the microscopic and macro-

scopic properties of diﬀerent wax oleogels, considering the indi-

vidual wax compositions. That hopefully enables a better under-

standing of the composition-functionality relation of wax oleogels

and provides new insights into the impact of PC on the formation

and properties of wax crystal networks in edible oils.

2. Experimental Section

2.1. Material

BWX, CLX, SFX, and RBX were kindly provided by KAHL

G.m.b.H. & Co. K.G., Trittau, Germany. The waxes were stored in

plastic bags at 6 °C. Canola oil (Canolin 10 770,) and sunﬂower oil

(Sonnin 70 020) were kindly provided by Walter Rau AG, Neuss,

Germany. All oils were stored in opaque containers at 3 °Cim-

mediately after delivery to prevent deteriorating reactions.

2.2. Oil Puriﬁcation and Increment of Polar Oil Components

Untreated canola, sunﬂower, and ﬂaxseed oil were stripped by

combining the oﬃcial method for the determination of polar

compounds and two methods developed in previous studies

[DGF C-III 3b (13)].[40,41] The details can be found elsewhere.[35]

Sunﬂower and canola oil (2 l each) were heated to 180 °Cand

vigorously stirred to induce oil deterioration and increase the con-

tent of polar components. Samples were taken from the batch,

cooled to ambient temperature, and stored at 5 °C in opaque

containers. Samples were taken when the initial PC value deter-

mined using Testo 270 cooking oil tester (Titisee-Neustadt, Ger-

many) increased by 0.5%, 1.0%, 2.5%, and 15.0% from the initial

Testo value, which was 5.0% for canola and 9.0% for sunﬂower

oil. That way, several stages of oil deterioration were reached. It

should be mentioned that natural sunﬂower oil did not actually

Technology published by Wiley-VCH GmbH

2100068 (3 of 19)

www.advancedsciencenews.com www.ejlst.com

contain 9.0% of total polar components. The Testo cooking oil

tester measures the permittivity of a liquid and correlates the

value to a reference oil. This reference oil has a lower iodine value

(IV) than sunﬂower oil. Therefore, more unsaturated bonds re-

sult in a higher permittivity. The measured values thus overshoot

the actual PC value. The value of 3.7 ±0.2 veriﬁed this according

to the oﬃcial method (DGF, C-III 3b (13)).

It should be mentioned that the same oils (natural, stripped,

and with increased levels of polar components) were used in pre-

vious studies. Data on the eﬀectiveness of the oil puriﬁcation pro-

cedure will hence not be shown in this study. For further infor-

mation, refer to refs. [35, 42].

Except for oil analysis and oleogel ﬁrmness, only stripped and

untreated oils and oils with the highest level of deterioration

products were used for further analysis. They will be referred to

as –P, –N, and –D, respectively.

2.3. Oil Analysis

Dielectric Constant: A parallel plate type electrode was built with

two stainless steel plates. The plates were ﬁxed at a constant

distance of 0.5 mm by 4 PTFE washers (Ø 4 mm) and suitable

PTFE screws (12 mm in length). Each plate was equipped with

a 150 mm stranded wire (cross-sectional area 1.5 mm2) and con-

nected to a precision LCR meter (4280A, Hewlett Packard). The

capacitor was placed in a sealed PTFE housing ﬁlled with the

respective oil. The setup and the oil samples were tempered at

20 °C before analysis to ensure a constant and evenly distributed

temperature, and the temperature varied less than 0.5 °C during

measurements. The dielectric constant (𝜖) was determined by di-

viding the capacitance of the oil (Cx) by the capacitance of the air

(C0)

𝜀=Cx

(1)

All measurements were performed at 1 MHz and carried out

in triplicates.

Total Polar Components: In addition to the Testo cooking oil

tester, the content of polar components was determined using the

oﬃcial method employing column chromatography.[43] Brieﬂy,

25 g silica (0.063-0.2 mm, VWR International, Pennsylvania,

USA) with a water content if 5 g/100g is added to a mixture of

diethyl ether (13% v/v) and petrol ether (87% v/v) (both HPLC

grade and purchased from VWR International, Pennsylvania,

USA) and ﬁlled into a chromatography column (21 mm inner di-

ameter, length 45 cm, Lenz Laborglas GmbH, Berlin, Germany).

Subsequently, 2.5 g of oil was weighed and dissolved in 50 mL of

the solvent mixture. 20 mL of the oil solvent mixture were trans-

ferred into the column and eluted using another 150 mL of the

solvent mixture. The elution time was 60–70 min, and the sam-

ples were collected in a dried ﬂask. Once elution was completed,

the solvent was evaporated at 50 °C using a vacuum rotary evapo-

rator (600 mbar). The content of polar components can be calcu-

lated from the initial mass of the oil sample and the eluted mass

collected in the ﬂask. Measurements were carried out in tripli-

cates.

Peroxide Value, and Free Fatty Acids: Free fatty acids (FFA) and

peroxide value (PV) of oil samples were determined by titration

(Excellence T5, Mettler Toledo, Columbus, USA). The PV was de-

termined according to DGF method C-VI 6a Part 2(02) (Wheeler

method). 3–5 g oil was diluted in a mixture of chloroform and

acetic acid (3:2 v/v, AppliChem GmbH, Darmstadt, Germany).

Subsequently, 1 mL of saturated potassium iodate (GPR REC-

TAPUR, VWR International, Pennsylvania, USA) solution was

added. After stirring at 300 rpm for 180 s, 50 mL of deionized

water were added. The titration was performed with 10.0 mol−1

sodium thiosulfate (Alfa Aesar, Haverhill, USA) and a redox elec-

trode (DMi140-SC, Mettler Toledo, Columbus, USA).

The content of FFAs, expressed as oleic acid content in

mg/100 g, was determined according to DGF method C-III 4

(06) with a pH electrode (DG113-SC, Mettler Toledo, Colum-

bus, USA). Brieﬂy, 2–4 g of oil was dissolved in 60 mL of an

ethanol/diethyl ether solution (1:1 v/v, both HPLC-grade, VWR

International, Pennsylvania, USA). The titration was performed

using potassium hydroxide (GPR RECTAPUR, VWR Interna-

tional Pennsylvania, USA) in ethanol (0.02 mol/l). A blank value

was determined for each new batch of solvent. The titer was de-

termined by dissolving 25 mg benzoic acid (GPR RECTAPUR,

VWR International Pennsylvania, USA) in the solvent and was

measured in triplicates every day. All measurements were carried

out in triplicates.

2.4. Oleogels

Gel Preparation: 200 mL stock solutions of 4, and 10% w/w wax in

oil were prepared by adding the solids to the oil and heating the

mixture in 400 mL glass beakers to 80 °C on a heating plate (MR

Hei-Tech, Heidolph Instruments GmbH & Co.KG, Schwabach)

agitated using a magnetic stirrer at 200 rpm. Before further pro-

cessing, the solution was kept at 80 °C for 20 min to ensure com-

plete dissolution.

For rheology and microscopy, hot wax solutions were directly

transferred into the respective measurement environment to

avoid any changes in the gels due to sample preparation and

transfer. Three individual stock solutions were prepared for each

oil, and measurements were repeated as described in the follow-

ing sections.

Gel Firmness: Freshly prepared 10% w/w stock solutions were

poured into preheated (60 °C) glass Petri dishes (Ø 110 mm)

up to a height of 13 mm (50 g ±1 g). Samples were trans-

ferred into a programmable climate chamber (HPP260, Mem-

mert GmbH, Schwabach, Deutschland) and cooled to 20 °Cata

constant cooling rate of 10 or 0.05 °Cmin

−1. Subsequently, the

samples were stored at the same temperature for 24 h before

ﬁrmness was measured using a static material testing machine

(Zwick GmbH & Co. KG, Germany) equipped with a 0.5-inch

cylindrical probe. When the preset force of 0.02 N was detected,

the cylinder penetrated the sample to a depth of 3 mm (<30% of

sample height), and the associated program testExpertII recorded

the force–displacement motion curves. Each Petri dish was pen-

etrated ﬁve times, and the distance between penetration points

and the wall of the petri dish was about 10 mm.

Gel–Sol Transition: DSC was performed using a Netzsch

204 Polyma (Netzsch-Gerätebau GmbH, Selb, Germany). Pure

waxes and oleogels comprising 4% and 10% w/w wax were

gelled at ambient temperature and stored for 24 h. Subsequently,

Technology published by Wiley-VCH GmbH

2100068 (4 of 19)

www.advancedsciencenews.com www.ejlst.com

Figure 1. Exemplary plot of storage and loss modulus of wax oleogels dur-

ing amplitude sweep, explanation in the text.

10–20 mg were cut from the gel samples’ center, weighed into

aluminum pans, and hermetically sealed. Samples were kept

at 95 °C for 30 min in the calorimeter to erase crystal memory.

Samples were then cooled at constant rates of 1, 10, and 40 °C

min−1. After an isothermal period of 20 min at 20 °C, samples

were heated at 10 °Cmin

−1and the gel–sol transition temper-

atures and enthalpies were determined using Proteus software

(Netzsch-Gerätebau GmbH, Selb, Germany). The measurements

were carried out in duplicates.

Sol–Gel Transition and Viscoelastic Behavior: Sol–gel transition

temperatures were determined via dynamic mechanical thermal

analysis (DMTA) using an MCR 302 Rheometer (Anton Paar

GmbH, Austria) with a plate-plate geometry (gap 0.2 mm). The

upper plate was sandblasted to avoid slipping of the sample. Hot

wax solutions (10% w/w) were pipetted (preheated pipette tips,

0.8 mL) onto the preheated plate (90 °C). Subsequently, the solu-

tion was cooled from 90 to 10 °C at a ﬁxed cooling rate of 10 °C

min−1. The measurements were performed at a constant strain

within the linear viscoelastic region (LVR) of both the liquid and

the solid (0.05%) and an angular frequency of 10 rad s−1.The

sol–gel transition temperature was calculated using the associ-

ated program (Rheoplus, Anton Paar, Austria). It is deﬁned as

the crossover of the loss (G′′) and storage modulus (G′) upon

cooling. All measurements were carried out in triplicates.

After gelation occurred, the samples were left to rest for 20 min

at 20 °C. Then, a strain sweep from 0.01–100% was performed

at 10 rad s−1and 20 °C. The data was used to determine G′′max

within the LVR and the strain at which the sample starts to

be irreversibly damaged (𝛾max). The threshold value for deter-

mining 𝛾max was set to 5% of the G′′max in agreement with the

literature.[44] Moreover, ΔG′′ was determined by deducting G′′

within the LVR from G′′max (see Figure 1).

CGC: The CGC in this study was determined for all waxes in

combination with stripped, untreated, and deteriorated oils (PC

+10%). It refers to the concentration at which G′exceeds G′′ at

20 °C. To this end, the CGC was approximated in a ﬁrst step by

measuring the sol–gel transition temperature of wax solutions

with varying concentrations. Subsequently, the results were plot-

ted and extrapolated to estimate the CGC at 20 °C. The approxi-

mation of the precise CGC at 20 °C was carried out by lowering

the wax concentration stepwise (0.05% w/w) from a concentra-

tion roughly above the estimated CGC. The respective solution

was cooled from 90 to 20 °C at a ﬁxed cooling rate of 10 °Cmin

−1,

a constant strain of 0.05%, and an angular frequency of 10 rad

s−1. Subsequently, samples were kept at 20 °C for 20 min, and

a strain sweep was performed for each concentration step. The

lowest concentration yielding G′>G′′ was taken as the CGC for

the respective wax/oil combination. It needs to be mentioned that

samples might not be in a macroscopic solid state at this point,

but the deﬁnition is in line with the rheological deﬁnition of a

gel.[45]

Microscopy: Brightﬁeld light microscopy (BFM) images were

taken using the Axio Scope.A1 KMAT (Zeiss, Jena Germany)

equipped with an AxioCam ICm1 Rev.1 camera. Samples were

prepared similarly to those in Petri dishes. Hot solutions of 4%

w/w wax in oil were pipetted on preheated microscope slides. Af-

ter the cover glass was placed, samples were kept at 80 °Cfor

10 min. Then cooling rates of 10 and 0.05 °Cmin

−1were applied,

analogous to the Petri dishes used for gel ﬁrmness. Images were

taken after a storage time of 24 h at 20 °C without further pro-

cessing (20×magniﬁcation).

In addition, preheated microscope slides were cooled from 80

to 10 °Cat10°Cmin

−1using a temperature-controlled micro-

scope unit (Linkam Scientiﬁc Instruments Ltd., Tadworth United

Kingdom), and pictures were taken every 2.5 s. The image with

the ﬁrst visible crystals was taken as crystallization onset and

compared to DSC onset.

All BFM-images were processed using ImageJ 1.52a software

to derive quantitative information from BFM images. After scal-

ing and thresholding, the number average crystal size [𝜇m2], the

fractal dimension D, and the samples’ 2-D porosity were calcu-

lated. The fractal dimensions were determined by using the box-

counting method provided by ImageJ. All calculations were based

on the evaluation of at least two images. For detailed information,

please refer to ref. [46].

2.5. Statistical Analysis

All experiments were replicated according to the descriptions in

the sections above. In tables, relevant ﬁndings were expressed

as mean ±standard deviation. Statistical diﬀerences were de-

termined using SPSS software, and running paired T-test or

ANOVA and posthoc test (Bonferroni) analysis. The p-value was

set to 0.05. In the results and discussion section, the oils stripped

from minor components served as the reference. Posthoc tests

were used to establish whether samples with minor oil compo-

nents are signiﬁcantly diﬀerent from stripped samples. More-

over, it was assessed whether a further increase of PCs in heated

oils aﬀected oil and gel properties.

3. Results and Discussion

3.1. Oil Analysis

It has been recently reported that minor oil components have

a considerable impact on the macroscopic and microscopic

Technology published by Wiley-VCH GmbH

2100068 (5 of 19)

www.advancedsciencenews.com www.ejlst.com

Table 2. Oil permittivity, free fatty acids, peroxide value, viscosity, water content, and polar components (from left to right) determined for stripped (P),

natural (N), and various stages of deteriorated (0.5, 1.0, 2.5, 5.0, and 10.0) canola (C-) and sunﬂower (S-) oil. n.a. – not applicable.

Sample 𝜖[−]FFA[mg/100g]PV[meqkg

−1]𝜂[mPa s] H2O [ppm] PC ISO [g/100 g]

C-P 3.0027 ±0.002 n.a. 0.1 ±0.1 25.5 ±0.02 13.4 ±1.1 n.a.

C-N 3.0576 ±0.002 4.0 ±0.5 0.5 ±0.1 33.2 ±0.02 46.8 ±4.4 3.8 ±0.5

C-0.5 3.0678 ±0.006a) 12.3 ±31.9a) 5.4 ±0.2a) 32.9 ±0.10 192.7 ±2.3a) 3.9 ±0.4

C-1.0 3.0891 ±0.006a) 3.8 ±1.0 10.3 ±0.4a) 32.6 ±0.15a) 147.8 ±1.6a) 4.9 ±0.6a)

C-2.5 3.1078 ±0.006a) 2.2 ±0.4a) 17.9 ±5.5a) 33.2 ±0.20 197.1 ±10.6a) 5.8 ±0.7a)

C-5.0 3.1273 ±0.006a) 2.3 ±0.2a) 13.3 ±1.2a) 34.4 ±0.23a) 172.3 ±1.4a) 8.6 ±1.4a)

C-10.0 3.1673 ±0.006a) 2.3 ±0.5a) 17.1 ±1.4a) 38.2 ±0.10a) 267.1 ±1.6a) 17.6 ±1.4a)

S-P 3.0102 ±0.003 n.a. 0.9 ±0.3 21.4 ±0.09 19.9 ±1.2 n.a.

S-N 3.1225 ±0.001 2.2 ±3.1 2.6 ±0.4 30.4 ±0.05 224.0 ±8.7a) 3.7 ±0.2

S-0.5 3.1943 ±0.006a) 2.8 ±6.3 36.2 ±1.1a) 30.2 ±0.00 232.2 ±19.8 4.7 ±0.4a)

S-1.0 3.2143 ±0.007a) 3.9 ±1.4a) 43.5 ±0.8a) 30.7 ±0.34 195.7 ±0.4a) 6.0 ±0.2a)

S-2.5 3.2413 ±0.006a) 1.9 ±0.3a) 41.8 ±3.5a) 31.9 ±0.10a) 336.7 ±28.2a) 8.7 ±1.7a)

S-5.0 3.2683 ±0.007a) 2.6 ±1.3a) 63.3 ±1.4a) 33.8 ±0.00a) 529.6 ±25.3a) 11.7 ±1.8a)

S-10.0 3.3248 ±0.007a) 4.8 ±1.0a) 72.5 ±0.1a) 38.0 ±0.05a) 505.0 ±4.8a) 19.4 ±0.8a)

a) indicates signiﬁcant diﬀerence between oils containing minor components. p=0.05.

properties of sterol/sterol ester oleogels.[21,35] For studying their

eﬀect on wax oleogels, it appears necessary to determine basic

oil quality parameters. Table 2 shows the analytical data for

stripped, untreated, and deteriorated canola (C-) and sunﬂower

(S-) oil. It needs to be mentioned that natural and oils at all

stages of deterioration—expressed as the increase in PCs—

diﬀer signiﬁcantly from the stripped oils. Therefore, Table 2

only includes statistical signiﬁcance (a) for oils containing minor

components.

The puriﬁcation treatments reduces the oils’ permittivity, vis-

cosity, water content, and eliminates FFAs and peroxides. More-

over, no polar components could be detected using the oﬃcial

method. On the other hand, most parameters presented in Ta-

ble 2 increased signiﬁcantly with each stage of oil deterioration.

However, the increase was not continuous for FFAs, PV, and wa-

ter content since these form and decompose during deteriora-

tion. Interestingly, the increase of oil permittivity from untreated

oil (S-N) to the highest deterioration stage (S-10.0) was greater for

sunﬂower oil (Δ𝜖≈0.21) than canola oil (Δ𝜖≈0.11). That is likely

due to the formation of diﬀerent reaction products indicated by

the variations in PV, water content, and FFA development in both

oils.

Additionally, sunﬂower oil contains more polar components at

the end of the heating procedure, even though both oils contain

similar amounts in their natural state. That is linked to the high

polyunsaturated FAs (mainly linoleic acid) content in sunﬂower

oil, which is generally more reactive than oleic acid, the main FA

found in canola oil. However, samples with the highest oxidation

level are only included in this study to highlight the changes in-

troduced by nontriglyceride oil components and have no practical

relevance.

One could argue that the increase in solvent viscosity impedes

the formation of wax crystals during oleogel preparation by re-

ducing the solutes’ diﬀusion rate. However, temperature sweeps

have shown that the viscosity diﬀerence around the relevant tem-

peratures for gelation (50–60 °C) is almost negligible. In contrast,

the ﬁrmness measurements were executed at ambient tempera-

ture. In that case, solvent viscosity might aﬀect the results since it

relates to the ﬂow of the solvent through the crystalline network

upon deformation.

3.2. Firmness

This section discusses the impact of oil type (degree of unsatura-

tion, stripped oils) and content of PCs on wax oleogel ﬁrmness.

The results are more descriptive at this point but will be put into

context with, e.g., thermal behavior and microstructure later on.

Eﬀect of Oil Type: The ﬁrmness of oleogels utilizing stripped

canola or sunﬂower oil and cooling rates of 10 and 0.05 °Cmin

−1

are depicted in Figure 2. The columns without a pattern (black:

canola; white: sunﬂower) are related to the high cooling rate,

while columns with a pattern represent the ﬁrmness of samples

prepared at 0.05 °Cmin

−1(dotted: canola; striped: sunﬂower).

In line with the literature, SFX oleogels are the hardest, irre-

spective of oil type and cooling rate, while BWX samples are the

softest.[1,47]

There is no signiﬁcant diﬀerence connected to the type of oil

in SFX, BWX, and CLX. However, in RBX samples, more unsat-

urated oil produces harder gels.

In contrast, the cooling rate signiﬁcantly changes the oleogel

ﬁrmness of SFX and RBX gels, while there was no considerable

diﬀerence in BWX and CLX samples.

Surprisingly, cooling at 0.05 °Cmin

−1resulted in signiﬁcantly

softer gels for SFX, while RBX-based gels became harder at the

low cooling rate. In general, a lower cooling rate enables highly

ordered crystals which are usually larger (lower supersaturation)

due to a shift in the balance of nucleation and growth rates.

Consequently, the formation of separate instead of mixed crys-

tals is likely. In line with that, RBX samples cooled at 0.05 °C

min−1show high deviations in gel ﬁrmness compared to the

gels produced with 10 °Cmin

−1. That indicates the formation

Technology published by Wiley-VCH GmbH

2100068 (6 of 19)

www.advancedsciencenews.com www.ejlst.com

Figure 2. Firmness of oleogels prepared using puriﬁed sunﬂower or

canola oil and 10% w/w sunﬂower (SFX), bees- (BWX), rice bran (RBX),

or candelilla wax (CLX), cooled at 10 (black and white columns) or 0.05 °C

min−1(pattern-ﬁlled columns).

of inhomogeneous areas within the Petri dish potentially caused

by separate crystallization. On the other hand, in BWX and

CLX samples, oleogel hardness seems to be independent of the

cooling rate and the type of oil used.

Eﬀect of PCs: The hardness data depicted in Figure 2 was fur-

ther used as the reference to relate the data of oleogels from un-

treated and deteriorated oils. These gels were subjected to the

same cooling rates. Figure 3 depicts the relative oleogel ﬁrmness

of canola (A, B) and sunﬂower oil (C, D) oleogels over PC content

(Table 2). The ﬁrst group of columns relates to gels made with

untreated canola oil (PC 3.8 and 3.7). Column groups 2, 3, 4, 5,

and 6 relate to the oil deterioration stages (see Table 2). It should

be mentioned that the statistical correlation of gel hardness and

several oil quality parameters such as PV, FFA and water content

was, to the greatest possible extent, inconclusive.

However, distinct trends can be seen for the type of wax, the

cooling rate, and the oil used. For example, at 0.05 °Cmin

−1the

hardness of SFX gels is signiﬁcantly higher at the lowest two PC

levels (Figure 3A,C). In contrast, there is no signiﬁcant diﬀerence

to the reference at higher PC content. For the high cooling rate,

changes in gel hardness are insigniﬁcant. Hence, it is fair to as-

sume that SFX’s network is not primarily aﬀected by the type and

concentration of PC but the cooling rate.

In contrast, RBX oleogels are signiﬁcantly ﬁrmer with in-

creased PC levels in canola oil, and the eﬀect is more pronounced

for samples cooled at 0.05 °Cmin

−1. Similar to the results re-

ported for hardness in stripped oils, the deviations are relatively

high in gels subjected to 0.05 °Cmin

−1. In general, BWX and

RBX samples are the softest. Therefore, slight changes in gel

ﬁrmness result in signiﬁcant diﬀerences in relative hardness.

Hence, the trends seen in Figure 3 could be disproportionate.

Interestingly, for sunﬂower oil (Figure 3C,D), RBX gel hardness

appears not signiﬁcantly diﬀerent from the reference at 0.05 °C

min−1. However, at 10 °Cmin

−1, a signiﬁcant reduction in gel

hardness occurred on PC increase. Still, it is essential to note that

the initial hardness values of RBX in stripped sunﬂower oil were

considerably higher than in canola oil (Figure 2). However, that

does not fully explain the signiﬁcant drop (up to 60%) in gel hard-

ness at high cooling rates. Thus, the RBX crystal network’s ﬁrm-

ness seems to be aﬀected by the type of oil and its deterioration

products.

The hardness of BWX oleogels appears to be more sensitive to

the cooling rate. All samples containing PCs were signiﬁcantly

softer than the reference at lower cooling rates (Figure 3A,C)

without a clear relation to the PC level. However, the opposite

trend on increasing PC levels was observed at high cooling rates,

whereas the eﬀect is much more pronounced in sunﬂower oil

(Figure 3B,D). Like RBX, the ﬁrmness of all gels was very low.

Again, the trends seen in Figure 3 could be disproportionate.

At 0.05 °Cmin

−1, BWX in sunﬂower oil decreases gel hard-

ness’ for all TPC levels. That is in line with the observation in

gelled canola oil. However, the gels become signiﬁcantly ﬁrmer

with increasing PC content at 10 °Cmin

−1. That was observed in

canola oil to a much lesser extent. Since BWX contains compo-

nents diﬀerent from WE (Table 2), this eﬀect likely relates to the

synergistic eﬀects of oil deterioration products. However, BWX

oleogel hardness appears to be sensitive to both cooling rate and

the content of polar components, while the type of oil is negligi-

ble.

Finally, CLX oleogels showed lower gel hardness than the ref-

erence in both oils when cooled at 0.05 °Cmin

−1, whereat the

diﬀerence was not always signiﬁcant. However, in oils with high

PC content (Figure 3A at 17.6 and Figure 3C at 11.7 and 19.4), a

signiﬁcant increase in gel hardness was observed, indicating ei-

ther a modiﬁcation of CLX solubility or a substantially diﬀerent

crystal structure or network interactions. When using the high

cooling rate, CLX ﬁrmness appeared independent of the PC level

(except canola oil 17.6). That indicates that CLX network ﬁrmness

is aﬀected by the cooling rate and PC content and only marginally

by the type of oil used.

In summary, PCs seem to have distinct eﬀects on the oleogels

for the wax types studied, while the unsaturation of the oil does

not seem to have a signiﬁcant impact. At this point, it remains

unclear whether these observations are related to changes in wax

solubility, crystal morphology, network interactions, or a combi-

nation of those. Consequently, the thermal behavior of the gels

was studied to address the ﬁrst parameter.

3.3. Thermal Behavior

Since the eﬀect of PC increment on oleogel ﬁrmness is most pro-

nounced for the highest content of PC except a few exceptions, in

the following, only the results obtained using stripped (–P), un-

treated (–N), and most deteriorated oil with the highest PC con-

tent (–D) will be compared.

DSC measurements were performed to validate whether the

eﬀects on oleogel hardness are related to wax solubility modiﬁ-

cations. Moreover, studying the thermograms allows identifying

distinct crystallization or melting events induced in the presence

or absence of PC.

Figure 4 shows the transition enthalpies during gel dissolution

(endothermic >0) and gel formation (exothermic <0), plotted

on the left vertical axis. The right vertical axis displays the respec-

tive transition temperatures of wax oleogels from stripped canola

Technology published by Wiley-VCH GmbH

2100068 (7 of 19)

www.advancedsciencenews.com www.ejlst.com

Figure 3. Relative hardness of 10% w/w A,B) canola and C,D) sunﬂower oil oleogels over PC content determined using the oﬃcial method. A,C: Slow

cooling (0.05 K min−1), B,D: Fast cooling (10 K min−1). Red line indicates the reference value.

Figure 4. Enthalpy (endothermic >0, exothermic <0) of gel dissolution and gel formation (columns, left vertical axis) and dissolution and gelation

temperatures (symbols, right vertical axis) of wax oleogels of puriﬁed canola (C-P, circles) and sunﬂower oil (S-P, triangles). Left ﬁgure: 10% w/w wax;

right ﬁgure: 4% w/w wax. Samples were cooled and heated at 10 K min−1.

(circles) and sunﬂower oil (triangles). The higher temperature al-

ways relates to gel dissolution’s peak temperature and the lower

temperature to the peak during gel formation. The left ﬁgure de-

picts the data of 10% w/w oleogels, while 4% w/w wax were used

on the right. Samples with 4% w/w wax were produced to inten-

sify the impact of polar components introduced when untreated

and deteriorated oils were used (discussed further down).

The results in Figure 4 serve two purposes. On the one hand,

it validates the applied method and serves as a reference for oil

quality eﬀects. Indeed, comparing the left and right graphic of

Figure 4, one can assume that the behavior of a speciﬁc wax type

is independent of the wax concentration and oil type. There were

no signiﬁcant diﬀerences in the transition enthalpies between

oleogels comprising either stripped canola or sunﬂower oil.

Overall, SFX and RBX samples showed the highest transition

enthalpies. That is due to their high WE content (see Table 3). The

minute enthalpy diﬀerence between dissolution and formation—

formation in line with expectation slightly lower—suggests that

neither additional crystallization nor polymorphic transitions oc-

curred during stabilization (20 min). Preliminary stabilization

Technology published by Wiley-VCH GmbH

2100068 (8 of 19)

www.advancedsciencenews.com www.ejlst.com

Table 3. Chemical composition of waxes in % adapted.[22]

SFX RBX BWX CLX

HC 0.2 0.3 26.8 72.9

WE 96.2 93.5 58.0 15.8

FFA 3.3 6.0 8.8 9.4

FA-OH 0.3 0.2 6.4 2.2

tests over 72 h at 20 °C showed no postcrystallization events in

oleogel samples (data not shown). In line with the higher WE

chain length, the dissolution temperature of RBX is higher than

SFX.

Interestingly, the gap between gel formation and dissolution is

more signiﬁcant in RBX samples, implying that the WEs compo-

sition of SFX enables better nucleation (Figure 4). That is likely

related to the smaller chain length span within the WE-fraction

of SFX compared to RBX.[22] Indeed, network formation was

found to be promoted in WE compositions with narrow chain

length distribution.[23] However, there seems to be a small eﬀect

of the oil type on the solubility of SFX and RBX since the dis-

solution, and the gel formation temperatures are slightly higher

in stripped sunﬂower oil. However, this observation could not be

reconﬁrmed in 4% w/w oleogels and is possibly related to minor

deviations caused during sample preparation.

In BWX and CLX gels, gel formation and dissolution tem-

peratures appear independent of the oil type. For BWX-based

oleogels, a larger undercooling than in CLX-based gels was ob-

served, which could be caused by the diﬀerences in the composi-

tion’s heterogeneity (Table 3). CLX is characterized by a dominant

component comprising about 60% w/w of C31 HC.[22] Unlike

the other waxes studied, there is a considerable discrepancy be-

tween the transition enthalpies in CLX, indicating the crystalliza-

tion process of the secondary components (19% WE and 9% FFA)

was not completed during the cooling and stabilization process.

However, the data in Figure 4 establish a sound base as a refer-

ence point for the quantiﬁcation to evaluate changes introduced

by the presence of minor polar components.

Figure 5 shows the relative dissolution enthalpies of 4% w/w

oleogels utilizing untreated (–N) and deteriorated (–D) canola

(left) and sunﬂower oil (right). Again the data has been related to

that of oleogels from stripped oils. For all samples, the enthalpy

is signiﬁcantly lower than the reference. However, the magnitude

appears somewhat diﬀerent for canola and sunﬂower oil.

For SFX samples, varying the oil type and level of PCs (un-

treated vs deteriorated) does not have a considerable eﬀect. That

is in line with the results on gel hardness (Figure 3). In con-

trast, the drop in dissolution enthalpy is more signiﬁcant in RBX

samples (up to 38%) and seems to vary with the PC content

and oil type (larger eﬀect in canola oil). Remarkably, this evo-

lution of values does not seem to correlate to the gel hardness

(Figure 3).

Similarly, the increase in BWX oleogel hardness does not cor-

relate with the data obtained for dissolution enthalpies. Moreover,

there is no clear correlation with the oil type. In both oils, the en-

thalpies are reduced, but the eﬀect is again more pronounced in

canola oil samples.

In CLX oleogels, the dissolution enthalpy is reduced by about

10% in untreated oil and about 16% in high-PC oils. Here, both

oils show comparable results, and an increase in dissolution en-

thalpy with a higher PC content can be observed, in line with the

results on gel hardness.

Overall, the dissolution enthalpy results remain inconclusive

and cannot be put into context with other parameters. Neverthe-

less, the enthalpy is signiﬁcantly reduced in all samples contain-

ing PCs, relating to the assumption that the build-up of highly

ordered crystalline structures is less disturbed in stripped oils

than in systems containing PCs. However, these diﬀerences do

not automatically increase oleogel hardness since other factors

such as crystal network interactions, crystal size, and surface. A

lower dissolution enthalpy in oils containing polar components

suggests a modiﬁed solubility of the waxes, which might arise

from solute-solute (PC-wax) interactions. However, that implies

that more material remains dissolved after equilibrium has been

reached, resulting in lower dissolution temperatures. Interest-

ingly, the oleogel dissolution and formation temperatures were

practically invariant for all waxes and independent of the oil type

(Figure 6A,B).

Figure 6 (top) shows that the dissolution temperature does not

vary signiﬁcantly with the PC content for oleogels produced at

10 °Cmin

−1. Hence, there is no considerable eﬀect on wax sol-

ubility within the range studied. However, the peak formation

temperature appears to be slightly modiﬁed in the presence of

PC, shown in Figure 6C,D.

Although the formation is subjected to somewhat greater devi-

ations, distinct eﬀects can be seen for the diﬀerent wax types. The

formation of SFX oleogels is not considerably aﬀected by PCs.

In contrast, there is a slight decrease in gel formation temper-

ature for RBX, BWX, and CLX samples. However, this appears

Figure 5. Relative dissolution enthalpy of wax oleogels (4 % w/w). Left: canola and right: sunﬂower oil. –N indicates untreated oils and –D oils with the

highest PC content. Samples were cooled and heated at 10 K min−1.

Technology published by Wiley-VCH GmbH

2100068 (9 of 19)

www.advancedsciencenews.com www.ejlst.com

Figure 6. A,B) Relative gel–sol temperature and C,D) relative sol–gel temperature of wax olgeogels. A,C: canola oil; B,D: sunﬂower oil. All gels prepared

with 4 % w/w wax. –N indicates untreated oils and –D oils with the highest PC content. Samples were cooled and heated at 10 °Cmin

−1.

Figure 7. Dissolution enthalpy (columns, left vertical axis) and dissolution peak temperature (symbols, right vertical axis) of 4% w/w wax oleogels from

puriﬁed (C-P, squares), untreated (C-N, circles) and deteriorated (C-D, triangles) canola oil. Left graphic: gelation at 1 °Cmin

−1, right: gelation at 40 °C

min−1.

to be independent of the oil type and could result from weak

interactions, retarding the formation of wax crystals. Substan-

tial eﬀects on the gel formation temperature, similar to those re-

ported for the sterol/sterol ester systems, cannot be seen.[35] The

self-assembly and structuring of the sterol/sterol ester structur-

ing system widely depend on hydrogen bonds. Hence, they are

prone to the presence of polar components able to form similar

bonds.[48,49]

In contrast, the formation of wax crystals is much less depen-

dent on hydrogen bonds. Only FFA and FAL may form hydrogen

bonds with polar molecules through carboxyl or hydroxyl groups.

However, their content varies in the waxes studied (3.5% SFX,

6.2% RBX, 15.1% BWX, and 11.6% CLX[22]), and their arrange-

ment in the nonpolar triglyceride oil is unknown. Potentially, in

a dissolved state, they form clusters with the polar head groups

facing each other.

However, the data on oleogels ﬁrmness (Figure 3) suggests that

diﬀerent structures might form depending on the cooling rate. To

this end, samples of stripped, untreated, and deteriorated (high-

est PC) canola were subjected to a fast (40 °Cmin

−1)andslow

(1 °Cmin

−1) gelling procedure, followed by an isothermal period

of 20 min at 10 °C. Subsequently, the gel dissolution was recorded

at 10 °Cmin

−1.

Figure 7 shows the dissolution enthalpies (columns) and gel-

sol transition peak temperatures (symbols) of these oleogels. The

dissolution temperatures are not signiﬁcantly diﬀerent in sam-

ples containing PCs in fast-gelled samples (Figure 7 left). Even

though the dissolution temperatures vary a bit more with PC at

0.05 °Cmin

−1, the diﬀerences are insigniﬁcant. Hence, the low

cooling rate likely induces this variation (Figure 7, right). The fact

that the slowly crystallized sample always shows higher gel–sol

transition temperatures conﬁrms the above statement. It is also

conﬁrmed that this kinetic depression eﬀect is least pronounced

in more homogenous waxes, SFX and CLX. That is in line with

the observation that the enthalpy of dissolution is lower for fast

cooled samples based on stripped oils.

Technology published by Wiley-VCH GmbH

2100068 (10 of 19)

www.advancedsciencenews.com www.ejlst.com

These might show a suboptimal crystal packing due to kinetic

eﬀects or be based on energetically less favorable kinetically in-

duced mixed crystals. Considering the eﬀect of PC on the en-

thalpy of transition, it is fair to conclude that no diﬀerence be-

tween the two PC levels could be identiﬁed throughout all sam-

ples. The reduction encountered on PC presence is more pro-

nounced in samples after cooled ad 0.05 than at 10 °Cmin

−1.

However, this statement is misleading because, in the presence

of polar components, the enthalpy of dissolution in SFX and RBX

appears to be independent of the cooling rate. In samples based

on CLX and BWX, the enthalpy of gel–sol transition is higher for

PC-containing samples when the cooling rate is high. A possible

explanation of this ﬁnding based on the contribution of crystal-

lized FFA’s, which are present in both waxes and deteriorated

oils, falls short because of the strong dependency of the observa-

tion on cooling rate.

3.4. Rheology – CGC

Regarding oleogels, the solvent composition (PC and TAG com-

position) could also aﬀect some commonly measured parameters

such as gel hardness and critical gelling concentration (CGC).

The latter is usually postulated as the minimum concentration

of a gelator to produce a non-ﬂowing state. However, it is often

determined arbitrarily by increasing the wax concentration grad-

ually, using large increments and diﬀerent procedures. Hence,

the relevant literature reports very diverse values of CGC for

the same type of wax (Table 1). Table 1 shows the CGC found

in the literature of SFX, RBX, CLX, and RBX, the concentra-

tion increase used (Δ% w/w), the oil type, and the determina-

tion method. Considerable variations of CGC can be seen for all

wax types. These are related to the individual wax constituents,

the increment of concentration, and the oil type. It is crucial to

consider the impact of oil composition since the FA proﬁle of

triglycerides (TAGs) might aﬀect wax solubility (solvent-solute in-

teractions). Besides, dissolved minor polar oil components could

interfere with the self-assembly of the structurants by forming

similar interactions to those between wax components explained

earlier (solute–solute interactions). Only one publication consid-

ers a combined approach, where a range of concentrations was

prepared, and samples were ﬂipped over to determine the non-

ﬂowing state at 5 °C (Table 1).

Subsequently, dynamic-mechanical-thermal analysis (DMTA)

(temperature sweep at low oscillation and frequency) and am-

plitude sweeps were used to detect the lowest concentration at

which G′>G′′. Unfortunately, the increments are large (0.5%

w/w), and it remains undisclosed if the samples used for the ﬂow

test were transferred onto the rheometer or if hot wax solutions

were used for DTMA. Transferring viscous samples partly breaks

up structures and generates falsiﬁed results.

However, the results render the ﬂow-test determination of

CGC very unprecise, subjective, and unsuitable for scientiﬁc pub-

lications since the CGC is regularly used to characterize the

oleogel system further. Besides, if a sample’s ﬂowing is solely

considered, the CGC also appears unpractical for product appli-

cations since it provides no useful information about consistency

and sensitivity.

A precise method to determine CGC was developed within this

study to study the diﬀerences introduced by PCs. Figure 8 (right)

shows the data for all wax oleogels as a function of solvent per-

mittivity for either stripped (lowest 𝜖), untreated (moderate 𝜖), or

deteriorated (highest 𝜖) canola (ﬁlled symbols) or sunﬂower oil

(open symbols). In line with the ﬂow test results reported in the

literature (Table 1), the rheological CGC increases in the order

SFX <CLX <RBX <BWX. Again this underlines that the gener-

ation of a solid-like structure in wax oleogels is promoted by (1)

high WEs content and (2) a limited diﬀerence in WE chain length

(SFX), and (3) more homogenous composition with high concen-

trations of components that co-crystallized (HC, predominantly

C31 in CLX).

For all waxes studied, the data gathered here indicate a lower

CGC than given in Table 1. In general, the presence of PCs signif-

icantly reduces the CGC of all waxes but SFX. This eﬀect is more

pronounced in BWX. None of the systems appeared to be sensi-

tive to the variation in the level of PC. Nevertheless, in stripped

oils, the CGC appears independent of the oil type (Figure 8).

Simultaneously, there are minor variations between canola and

sunﬂower oils containing PCs, likely caused by the diﬀerences of

the non-TAG molecular species.

However, the reduction of CGC with PCs leaves two possible

explanations: the polar components could contribute to establish-

ing a space-ﬁlling scaﬀolding so that gelation can be achieved at

lower levels of solid material. That is in analogy to the contribu-

tion of small amounts of emulsiﬁers to fat crystal networks.[14,17]

Alternatively, PCs might reduce the solubility of the waxes in the

oil. Therefore, the necessary amount of solid material to form a

gel is lower. However, the second interpretation is in conﬂict with

the increase in gel–sol transition enthalpy in gel from stripped

oils reported in the previous section.

Moreover, gel dissolution temperatures were invariant, indi-

cating no substantial changes in wax solubility. Consequently, a

synergistic eﬀect of PCs on network formation appears likely.

At this point, it has to be discussed what is necessary to create

a structure where G′′ >G′′. Initially, there has to be supersatura-

tion which leads to nucleation and growth of wax crystals. There-

fore, another approach to process the DSC data is presented in

Table 4. The gel–sol transition enthalpy data for two wax concen-

trations (4%, 12% w/w) allows estimating the enthalpy of melting

of 100% wax (ΔHpure,calc). Dividing the diﬀerence in the two sam-

ples gel–sol enthalpies by their concentration yields the melting

enthalpy when dissolution eﬀects and the temperature at which

the transition from solid to liquid occurs are ignored. The data for

gels with stripped oils are in good agreement with pure waxes’ ac-

tual values (ΔHpure). The values obtained gels from untreated and

deteriorated oils are, except RBX gels, lower. These reduced val-

ues of the normalized slid to liquid transition enthalpy indicate

that PCs support dissolution. That is in line with the fact that

the lowest solubility was recorded for stripped oils. Furthermore,

the data enable determining the amount of dissolved material

in stripped oils (xdiss% w/w) by plotting the dissolution enthalpy

over structurant concentration and extrapolating ΔH=0. The

values of xdiss have per deﬁnition to be lower than the CGC. For

the stripped oil oleogels, this condition is satisﬁed and allows de-

termining the amount of solid material present at CGC (Table 4).

The values generated are very low, 0.075–0.722% w/w. However,

Technology published by Wiley-VCH GmbH

2100068 (11 of 19)

www.advancedsciencenews.com www.ejlst.com

Figure 8. Left: CGC measurements for various wax concentrations (c [% w/w]) in natural canola oil to estimate the CGC at 20 °C, x-axis logarithmic

scale. Right: CGC of waxes in puriﬁed (lowest 𝜖), natural and deteriorated (highest 𝜖) oil, ﬁlled symbols: canola oil; open symbols: sunﬂower oil.

Table 4. Validation of CGC determination, xdiss -amount of wax dissolved in stripped oil, CGC%- CGC in stripped oil, ΔHpure -dissolution enthalpy pure

wax, ΔHpure,calc dissolution enthalpy pure wax extrapolated from ΔHof 4% and 12% w/w wax oleogels.

xdiss % w/w CGC % w/w CGC-xdiss [% w/w] ΔHpure [J g−1] C-P C-N C-D

ΔHpure,calc [J g−1]ΔHpure,calc [J g−1]ΔHpure,calc [J g−1]

SFX 0.075 0.15 0.075 195 194 173 167

RBX 0.153 0.65 0.497 187 191 194 197

BWX 0.228 0.95 0.722 160 161 159 147

CLX 0.228 0.35 0.122 146 151 139 139

the values of xdiss have to be lower than the CGC, which is valid for

all waxes using stripped oil. In oils comprising PC, the calculated

dissolution enthalpy did not agree with that of pure wax. Hence,

the calculated dissolved material is always higher than the CGC,

independent of whether the actual dissolution enthalpy of pure

wax or the calculated values were chosen. That results from the

overall lower dissolution enthalpy in oils comprising PC (see sec-

tion DSC). However, if the calculated ΔHovershoots the actual

value, interactions between the network particles can be assumed

and vice versa. Hence, in oils with PC, only RBX shows increased

crystalline interactions in the presence of PCs. That conﬂicts with

the lower CGC in these oils, indicating that the establishment of

inter-crystalline bonds might be promoted with PC. However, the

number of these interaction points highly depends on the num-

ber of crystals and their surface area (size-related).

3.5. Rheology – Strain Sweep

Strain sweeps of 10% w/w oleogels were used to determine the

maximum G′within the LVR as well as the ΔG′′ and 𝛾max (see

Figure 1 for details).

Figure 9 shows the results of the relative values of G′′max,ΔG′′

and 𝛾max. The values were always related to those of the respec-

tive stripped oil. Unfortunately, at 10% w/w wax concentration,

the results of G′′max are less expressive due to the high wax con-

centration. That leads to relatively large deviations between the

repetitions since the networks are very dense.[3,23] Repeating the

experiments at, e.g., 4% w/w wax concentration would hence be

beneﬁcial. However, the break-up behavior shows a much better

response at high wax concentrations.

Nevertheless, Figure 9A illustrates the relative G′′max values for

all waxes in untreated and deteriorated oils (reference is stripped

oil). Various studies reported the absolute values of G′′ for dif-

ferent waxes, which increase in the following order RBX <CLX

<BWX <SFX.[22,23,50] In SFX samples, there is no considerable

eﬀect considering the type of oil and PC concentration. How-

ever, the results indicate a higher G′′max for gels from oils con-

taining PCs, which relates to more interaction points within the

network.[45] Considering the ﬁrmness of SFX oleogels, this does

not necessarily translate into a higher penetration hardness (Fig-

ure 3). The G′′max of RBX oleogels seems to be largely indepen-

dent of PCs but is also subjected to irregularities, especially at

the highest PC concentration for both oils. In contrast, in BWX

and CLX samples, there appears to be an increase of G′′max when

utilizing oils with the highest PC. However, when utilizing un-

treated oils, CLX samples show a lower G′′max than the reference,

while in BWX gels, it is slightly higher.

It remains unresolved whether the increase observed in some

samples results from the formation of additional connection

points in the network due to PCs or smaller crystals, which pro-

vide a larger structuring surface. Vice versa, a decrease could be

linked to the formation of larger crystals or the prevention of

crystal–crystal interactions due to high amounts of PCs block-

ing the interaction points, similar to the results reported for fat

crystal networks.[17]

Figure 9B shows ΔG′′ determined for gels based on PC-

containing oils according to Figure 1. Again, all values are

considerably higher than in the reference. Moreover, the eﬀect

is most extensive in RBX oleogels and lowest in SFX’s highly

ordered structure. In SFX, BWX, and CLX samples, a further

Technology published by Wiley-VCH GmbH

2100068 (12 of 19)

www.advancedsciencenews.com www.ejlst.com

Figure 9. Relative values of A) G′max within the LVR, B) ΔG′′ during ampli-

tude sweep and C) 𝛾max representing the end of the LVR. Samples gelled

under low oscillation conditions at 10 K min−1. All samples with 10%

w/w wax. C-X canola oil; S-X: sunﬂower oil. X: N-untreated, D-deteriorated

(highest PC).

increase of ΔG′′ can be seen comparing oils utilizing untreated

and deteriorated oil. Generally, G′′′′ describes the portion of de-

formation energy lost by internal friction. As the strain increases

during amplitude sweeps (before the ﬂow point), individual

network bonds rupture. The break-up relates to their individual

strength. However, these movable network fragments show

internal friction increasing G′′ until the sample starts to ﬂow.

Hence, network interactions appear to be modiﬁed in gels with

PCs so that more fragments form during the sweep, causing

higher internal friction. A similar eﬀect has been reported in the

presence of PCs for sterol/sterol ester oleogels.[21,36]

Finally, Figure 9C depicts the strain at which the sample starts

to be irreversibly damaged, which relates to the systems’ ability

to store deformation energy during amplitude sweeps. The value

was determined in agreement with the oﬃcial methods deﬁn-

ing the end of the LVR as the strain at which G′deviates by 5%

from its maximum value.[44] In SFX and BWX oleogels, there

is no apparent diﬀerence to the reference when untreated oils

are used. In contrast, in RBX and CLX samples, 𝛾max is consider-

ably higher than in the reference sample. Clearly, 𝛾max increases

tremendously in oleogels utilizing oils with the highest PC levels,

and the eﬀect is highest in BWX and CLX samples. That might re-

late to the crystalline network formed in these samples, whereas

smaller network structures are known to store deformation en-

ergy better.[45]

In summary, wax oleogels appear to store deformation energy

better if PCs are present in the continuous phase, and the ef-

fect depends on the PC level. However, there is also more in-

ternal friction due to loose aggregates once the sample is irre-

versibly damaged but macroscopically intact. That shows that

the network interactions are signiﬁcantly modiﬁed in the pres-

ence of PC, but that does not necessarily translate into equal

changes in macroscopic properties such as hardness. Neverthe-

less, these modiﬁcations are likely linked to the speciﬁc network

features present in each wax’s oleogel. Their appearance and po-

tential alterations in the presence of PC will be discussed in the

following.

3.6. Microscopy

The crystal size and number, their interactions (e.g., sintering),

and the general arrangement of the crystal network are vital for

the macroscopic properties of wax oleogels, such as hardness.[54]

Therefore, this section presents and discusses the microstructure

of 4% w/w wax oleogels cooled at 10 and 0.05 °Cmin

−1in the

context of the results presented in the previous sections.

After gelation, tile images were taken of all samples to identify

representative areas. Subsequently, two images of the typical

structures were captured and processed using ImageJ software.

The box-counting fractal dimension (D), average crystal size

[μm2], and the porosity of each sample are shown in Table 5.

The latter is a measure of the magnitude of crystal-free space

in the sample. In contrast, the fractal dimension relates to the

spatial distribution of the crystals, indicating the homogeneity

of their distribution. Hence, higher D-values connect to more

evenly ﬁlled samples. In general, it is expected that at lower

cooling rates, larger crystals form due to lower supersaturations.

Thus, the porosity likely increases while the fractal dimension

decreases. The image analysis was performed similarly to pre-

vious studies, and the magnitude of values obtained is in good

agreement with the literature for a similar sample preparation

procedure.[55] Although the crystal size results were often incon-

clusive due to either very small crystals indistinguishable from

one another or poor contrast between crystal and background,

they were still included for completeness.

Figures 10–13 show brightﬁeld images of SFX, RBX, BWX,

and CLX, respectively. The top 3 images (A–C) show samples

Technology published by Wiley-VCH GmbH

2100068 (13 of 19)

www.advancedsciencenews.com www.ejlst.com

Table 5. Results of image analysis using Image J software for 4% w/w canola oil wax oleogels.

Oil Cooling rate [°Cmin

−1] Ø size [μm2] Fractal dimension D[−] Porosity [−]

SFX C-P 5.58±1.53 1.85±0.03 0.73±0.035

C-N 10 9.38±0.51a) 1.90±0.01a) 0.71±0.011

C-D 4.92±0.12a) 1.77±0.00a) 0.78±0.002a)

C-P 2.29±0.22 1.59±0.06 0.90±0.032

C-N 0.05 4.58±0.19a) 1.74±0.06a) 0.80±0.010a)

C-D 3.38±0.20a) 1.58±0.06 0.88±0.023

RBX C-P 5.48 ±1.06 1.62±0.03 0.84±0.044

C-N 10 10.02±0.90a) 1.56±0.05a) 0.88±0.040a)

C-D 6.96±0.59a) 1.50±0.01a) 0.90±0.017a)

C-P 3.55±1.28 1.58±0.26 0.82±0.007

C-N 0.05 2.41±0.19a) 1.43±0.04a) 0.93±0.020a)

C-D 3.55±0.46 1.54±0.01 0.87±0.035a)

BWX C-P 8.59±1.43 1.92±0.01 0.63±0.021

C-N 10 4.24±0.04a) 1.84±0.03a) 0.73±0.004a)

C-D 6.31±1.30a) 1.88±0.02a) 0.68±0.019a)

C-P 3.91±0.22 1.83±0.01 0.67±0.023

C-N 0.05 5.15±0.72a) 1.92±0.00a) 0.74±0.006a)

C-D 9.58±0.09a) 1.84±0.01 0.85±0.019a)

CLX C-P 4.45±2.82 1.90±0.03 0.74±0.007

C-N 10 4.79±1.99 1.91±0.02 0.73±0.101

C-D 3.65±1.93 1.91±0.03 0.70±0.045

C-P 8.10±0.23 1.94±0.05 0.63±0.002

C-N 0.05 7.67±0.09 1.83±0.01a) 0.74±0.031a)

C-D 7.67±0.26 1.88±0.03a) 0.71±0.026a)

a) indicates signiﬁcant diﬀerence to stripped oil. p=0.

Figure 10. Images of 4 % w/w SFX oleogels. A–C) cooled at 10 K min−1, from left to right: stripped (A), untreated (B), and deteriorated (C), highest PC)

canola oil. Bottom: cooled at 0.05 K min−1from left to right: D) stripped, E) untreated, and F) deteriorated, highest PC) canola oil. Scale bar: 100 μm.

Technology published by Wiley-VCH GmbH

2100068 (14 of 19)

www.advancedsciencenews.com www.ejlst.com

Figure 11. Images of 4% w/w RBX oleogels. A–C) Cooled at 10 K min−1, from left to right: stripped (A), untreated (B), and deteriorated (C), highest PC)

canola oil. Bottom: cooled at 0.05 K min−1from left to right: D) stripped, E) untreated, and F) deteriorated, highest PC) canola oil. Scale bar: 100 μm.

Figure 12. Images of 4% w/w BWX oleogels. A–C) Cooled at 10 °Cmin

−1, from left to right: stripped (A), untreated (B), and deteriorated (C), highest

PC) canola oil. Bottom: cooled at 0.05 °Cmin

−1from left to right: D) stripped, E) untreated, and F) deteriorated, highest PC) canola oil. Scale bar: 100

μm.

cooled at 10 °Cmin

−1while the cooling rate was 0.05 °Cmin

−1for

the images depicted at the bottom (D–F). The PC concentration

increases from left to right from stripped to untreated to deteri-

orated (highest PC concentration). The ﬁgures mentioned above

only show oleogels produced with canola oil since equivalent im-

ages were obtained utilizing sunﬂower oil.

In SFX oleogels cooled at 10 °Cmin

−1, the surface seems

packed with the needle-like structures reported in numerous

publications (Figure 10).[3,28,50,56,57] It should be mentioned that

this shape is due to the 2D perspective since it was shown using

scanning electron microscopy (SEM) that they are an intercon-

nected network of platelets.[23] However, even though samples

Technology published by Wiley-VCH GmbH

2100068 (15 of 19)

www.advancedsciencenews.com www.ejlst.com

Figure 13. Images of 4% w/w CLX oleogels. A–C) Cooled at 10 K min−1, from left to right: stripped (A), untreated (B), and deteriorated (C), highest PC)

canola oil. Bottom: cooled at 0.05 K min−1from left to right: D) stripped, E) untreated, and F) deteriorated, highest PC) canola oil. Scale bar: 100 μm.

were cooled rapidly, diﬀerences due to the presence and absence

of PC can be seen. These modiﬁcations could be quantiﬁed well

due to the good contrast between crystals and background. The

crystal size was largest in samples with untreated (Figure 10B)

canola oil and decreased in stripped (A) and deteriorated oil (C).

Moreover, the fractal dimension was highest and the porosity low-

est in samples comprising untreated oil. That indicates that a

moderate level of minor oil components somewhat promotes the

formation of a denser network that is relatively homogenous. In-

terestingly, the more ordered structure does not translate into an

increased gel hardness (Figure 2). A similar eﬀect has been re-

ported for sterol/sterol ester oleogels.[35] However, this does not

automatically result in ﬁrmer gels (see Section 3.3).

The eﬀect of PCs is even more pronounced in samples cooled

at 0.05 °Cmin

−1. In stripped oils (Figure 10D), needle-like struc-

tures turn into highly ordered crystal shards, which appear to

grow 2D and look astonishingly similar to images of pure WE.[23]

Hence, the low cooling rate promotes the crystallization into a

highly ordered WE crystal that is largely absent in oils containing

PC, especially at the highest PC level. Hence it is fair to assume

that PC acts as a crystal modiﬁer in SFX oleogels. Unfortunately,

the shards were not detected suﬃciently during image analysis

due to poor contrast, resulting in the underevaluation of the crys-

tal sizes. However, that indicates that the size of the crystalline

building blocks must be very small.

RBX (Figure 11) produces signiﬁcantly diﬀerent crystal shapes

than SFX, although their compositions appear quite similar at

ﬁrst sight (Table 3). In the literature, diﬀerent shapes of wax crys-

tals were reported in oleogels utilizing diﬀerent oils. For exam-

ple, in olive, canola, and peanut oil, RBX formed needle-shaped

crystals similar to those observed in SFX oleogels.[3,29,57] How-

ever, more irregular-shaped crystals were observed in rice bran

oil.[26,58] The eﬀects appear to be related to minor wax, and po-

tentially oil components since the bleaching of RBX resulted in

diﬀerent crystal shapes than those found in crude RBX.[52]

Additionally, the diﬀerences induced by variations of the cool-

ing rate and polar components level appear even more distinct

than in SFX samples (Figure 11). Knob-like structures can be

seen in stripped oils at high cooling rates, while they appear elon-

gated and cluster more in oils comprising PC. Moreover, other,

more loosely organized crystalline structures can be seen in gels

comprising untreated and deteriorated canola oil. The presence

of these smaller structures might be connected to the tremen-

dous increase in ΔG′′ observed during amplitude sweeps. Vice

versa, the monocrystalline knobs observed in stripped oils do not

form many movable fragments when gels are exposed to stress.

However, this is very speculative.

Unfortunately, the clustered crystals were detected as single

crystals during the crystal size determination, resulting in the

largest crystal size and high deviations. However, in samples

from stripped oils, crystals are more evenly distributed with less

space between them, indicated by a higher fractional dimension

and lower porosity. From visual observations, it is further sug-

gested that the crystal size distribution is more homogenous.

However, samples made with untreated and deteriorated oil at

10 °Cmin

−1were signiﬁcantly harder, suggesting that the for-

mation of distinct crystal types is beneﬁcial in RBX oleogels. In

line with that, gels cooled at 0.05 °Cmin

−1were considerably

harder than those cooled at 10 °Cmin

−1. From Figure 11, bot-

tom, it can be seen that RBX does not crystallize uniformly un-

der these conditions (low supersaturation). The presence of large

spherulitic crystal arrangements next to a ﬁnely distributed mesh

of small crystals is diﬃcult to interpret. The large crystals could

be due to limited nucleation events followed by slow growth. If

these nuclei result from homogeneous nucleation due to compo-

nents with high melting points or emerge from heterogeneous

Technology published by Wiley-VCH GmbH

2100068 (16 of 19)

www.advancedsciencenews.com www.ejlst.com

nucleation remains unresolved here. The size and density of the

ﬁne crystals appear to be modiﬁed in the presence of PC. The

crystal sizes for 0.05 °Cmin

−1were signiﬁcantly smaller than for

samples gelled at 10 °Cmin

−1. That contradicts the assumptions

made at the beginning of this chapter. However, it needs to be

mentioned that the large crystals observed in slowly cooled sam-

ples were not taken into account for image analysis.

Figure 12 shows images of 4% w/w BWX oleogels. At ﬁrst

glance, a ﬁnely distributed network of very small crystals can

be seen independently of the cooling rate and PC concentra-

tion. However, their distribution and porosity appear diﬀerent. At

high cooling rates, there is a very dense mesh of crystals. In line

with that, the porosity is lowest and the fractal dimension high-

est, indicating a dense network with homogeneously distributed

crystals. These structures appear looser and less dense in sam-

ples produced from untreated oils, with more free space between

them. The size distribution in the sample with the highest PC did

not conﬁrm the visual observations, possibly due to a lack of con-

trast between crystals and background and low magniﬁcation. In

samples cooled at 0.05 °Cmin

−1, the samples from stripped oil

seem to be even ﬁner than at 10 °Cmin

−1, with some larger, more

ordered structures in between. In gels produced with untreated

oil, a dense network of slightly bigger crystals can be seen, which

is very homogenous, having a fractal dimension of 1.92. How-

ever, there due to larger crystals, there is more free space and thus

higher porosity. At high PC levels, crystals are even larger (9.58

μm2) and appear more ordered, leaving more free space between

them (porosity 0.85). Moreover, the looser structures observed for

slowly cooled samples comprising PC seem to relate to a lower

gel hardness.

Finally, Figure 13 shows the samples produced with CLX at

diﬀerent cooling rates and concentrations of PC. Like BWX

oleogels, a dense network of small crystals is formed by CLX

when samples are cooled at 10 °Cmin

−1(Figure 13A-C). All three

samples showed similar porosity and fractal dimension. How-

ever, in contrast to BWX, the crystals appear denser in gels pro-

duced with stripped oils, which is likely due to these samples’

controlled crystallization of hentriacontane (C31). Nevertheless,

the more delicate structures observed in oleogels comprising de-

teriorated (C) oil result in a considerable increase in hardness

(see Figure 3).

A diﬀerent picture emerges for gels produced using 0.05 °C

min−1(Figure 13D–F). Large and highly ordered structures can

be seen, especially in stripped (D) and untreated oil (E), indicat-

ing a crystallization into separate crystalline structures similar to

RBX. Fine crystals accompany them, homogeneously distributed

in stripped oils (fractal dimension 1.94), leaving less free space

between (porosity 0.63). In contrast, in untreated oils, these are

partly attached to the crystals associated with hentriacontane. The

gaps are ﬁlled with very small crystals, which cannot be seen in

samples comprising stripped oils. In gels from deteriorated oil

(F), these crystals make up most of the structure. Moreover, no

highly ordered structures can be seen, indicating that mixed crys-

tals’ formation is promoted with PC. In line with SFX results,

the gels with mixed crystals have a higher gel hardness than the

highly ordered structures (see Figure 3).

Interestingly, in sunﬂower oil with the highest PC concentra-

tion, RBX crystals were modiﬁed and formed needle-like struc-

tures similar to those typical for SFX samples (Figure 10 and Fig-

Figure 14. Image of a 4% w/w SFX oleogel produced with deteriorated

sunﬂower oil, cooled at 10 K min−1. Red circle indicates formation of struc-

ture similar to SFX oleogels.

ure 14). Thus, the data gathered once more illustrate the impor-

tance of minor oil components in oleogel technology. Their pres-

ence signiﬁcantly aﬀects the wax crystal appearance in edible oils,

resulting in considerable modiﬁcations of the gels’ macroscopic

properties such as hardness and break-up behavior under stress.

Furthermore, it was found that the eﬀect of minor components is

modiﬁed by the cooling rates applied to induced the sol-gel tran-

sition in wax oleogels.

4. Conclusion

The stripped oils serve as a reference for investigating the ef-

fects of polar components on wax oleogels. Signiﬁcant eﬀects of

the cooling rate on several gel characteristics were shown. These

modiﬁcations were particular for each wax type; hence no general

conclusion could be formulated.

The gel characteristics of samples with PCs revealed signiﬁ-

cant diﬀerences in gel formation, enthalpy, rheological proper-

ties, and CGC compared to stripped oils. These were generally

more pronounced at the slow cooling rate. Furthermore, the data

on macroscopic hardness are complemented by rheological data,

which in essence conﬁrm that it is diﬃcult to formulate any clear

overarching relations of cooling rate and gel characteristics in the

presence of PCs. Even more so, while macroscopic hardness is as-

sessed to be reduced due to minor components, the rheological

data indicate that the presence of PCs increases the resistance of

the gel to deformation.

The thermal analysis (DSC) of oleogels based on stripped oils

with diﬀerent levels of structurants appeared to be very consis-

tent and indicates that the applied cooling rate (10 °Cmin

−1)and

stabilization procedures widely relate to solid-liquid equilibrium

states. Building on this, the evolution of the kinetics of the crys-

tallization and formation of kinetically induced mixed crystals at

Technology published by Wiley-VCH GmbH

2100068 (17 of 19)

www.advancedsciencenews.com www.ejlst.com

diﬀerent cooling rates and PC levels could be related to variations

in cooling rate and serves as an explanation for speciﬁc phenom-

ena encountered. The interpretations of the observations made

on parameter variation are further supported by the light micro-

scopical analysis of the samples. These indicate clearly that the

eﬀects of PCs and cooling rate are signiﬁcant and speciﬁc for ev-

ery wax.

The CGC increases with wax inhomogeneity and WE chain

length diﬀerence (BWX >RBX >CLX >SFX). The wax concen-

tration necessary to form a gel (CGC) is lower in the presence

of polar components. That illustrates that the polar components

contribute to the 3D network and support the solid matter’s eﬀec-

tiveness in establishing a gel. However, these eﬀects are less pro-

nounced than in, for example, the sterol/sterol ester gels since,

in this system, they strongly depend on hydrogen bonds.[49]

In conclusion, the data gathered illustrates that dissolved PCs

signiﬁcantly aﬀect the gels’ characteristics. The eﬀects appear to

be wax-type speciﬁc. However, it is consistently found that polar

components increase the wax’s solubility and support the forma-

tion of the gel itself. These two counteracting eﬀects complicate

the overall assessment and hinder any direct functional relation

formulation.

To further address the topic, the authors suggest relating

waxes’ physical and chemical properties and their respective

oleogels in standardized oils (depleted from minor components).

Subsequently, the addition of relevant minor components could

provide information about the gels’ modiﬁcations regarding their

functional groups.[21] To this end, XRD, the determination of

FTIR spectra, and molecular modelling have proven to be power-

ful tools.[48,59,60]

Acknowledgements

Open access funding enabled and organized by Projekt DEAL.

Conﬂict of Interest

The authors declare no conﬂict of interest.

Data Availability Statement

The data that support the ﬁndings of this study are available from the cor-

responding author upon reasonable request.

Keywords

microscopy, minor oil components, oil structuring, oleogels, wax oleogels

Received: March 30, 2021

Revised: March 1, 2022

Published online: May 9, 2022

[1] C. D. Doan, Ph.D. Thesis, Ghent University, Ghent, Belgium 2017.

[2] A.I.Blake,J.F.Toro-Vazquez,H.-S.Hwang,inEdible Oleogels, (Eds:

A. Marangoni, N. Garti) Elseview, London, United Kingdom 2018,p.

133.

[3] A. I. Blake, E. D. Co, A. G. Marangoni, J. Am. Oil Chem. Soc. 2014,91,

885.

[4] A. A. Carelli, L. M. Frizzera, P. R. Forbito, G. H. Crapiste, J. Am. Oil

Chem. Soc. 2002,79, 763.

[5] L.S.K.Dassanayake,D.R.Kodali,S.Ueno,K.Sato,J. Am. Oil Chem.

Soc. 2009,86, 1163.

[6] J. Gómez-Estaca, A. M. Herrero, B. Herranz, M. D. Álvarez, F.

Jiménez-Colmenero, S. Cofrades, Food Hydrocolloids 2019,87, 960.

[7] H. - S. Hwang, S. Kim, M. Singh, J. K. Winkler-Moser, S. X. Liu, J. Am.

Oil Chem. Soc. 2012,89, 639.

[8] A.R.Patel,inAlternative Routes to Oil Structuring (Ed: A. R. Patel),

Springer, Cham 2015, pp. 15–27.

[9] A. F. Harron, M. J. Powell, A. Nunez, R. A. Moreau, Ind. Crops Prod.

2017,98, 116.

[10] G. V. Buitimea-Cantúa, S. O. Serna-Saldívar, E. Pérez-Carrillo, T. J.

Silva, D. Barrera-Arellano, N. E. Buitimea-Cantúa, Grasas y Aceites

2022,73, 110267.

[11] S. R. Vali, Y. - H. Ju, T. N. B. Kaimal, Y. - T. Chern, J. Am. Oil Chem. Soc.

2005,82, 57.

[12] C. D. Doan, I. Tavernier, P. K. Okuro, K. Dewettinck, Innovative Food

Sci. Emerging Technol. 2018,45, 42.

[13] A. Papadaki, N. Kopsahelis, D. M. G. Freire, I. Mandala, A. A. Kouti-

nas, Biomolecules 2020,10, 10.

[14] D. Johansson, B. Bergenståhl, J. Am. Oil Chem. Soc. 1992,69, 705.

[15] D. Johansson, B. Bergenståhl, J. Am. Oil Chem. Soc. 1995,72, 205.

[16] D. Johansson, B. Bergenståhl, J. Am. Oil Chem. Soc. 1992,69, 728.

[17] D. Johansson, B. Bergenståhl, J. Am. Oil Chem. Soc. 1995,72, 911.

[18] H. J. Ensikat, M. Boese, W. Mader, W. Barthlott, K. Koch, Chem. Phys.

Lipids 2006,144, 45.

[19] Structure Function of Edible Fats, (Ed: A. G. Marangoni), AOCS Press,

Urbana, IL 2012.

[20] F. G. Gandolfo, A. Bot, E. Flöter, J. Am. Oil Chem. Soc. 2004,81,1.

[21] M. Scharfe, D. Prange, E. Flöter, J. Am. Oil Chem. Soc. 2022,99,57.

[22] C. D. Doan, C. M. To, M. de Vrieze, F. Lynen, S. Danthine, A. Brown,

K. Dewettinck, A. R. Patel, Food Chem. 2017,214, 717.

[23] H. Brykczynski Wax-based oleogels- characterization and re—

engineering of waxes. Master Thesis, Technical University Berlin,

Berlin 2021.

[24] B. T. Iyengar, H. Schlenk, Lipids 1969,4, 28.

[25] G. Avendaño-Vásquez, A. De la Peña-Gil, M. E. Charó-Alvarado, M.

A. Charó-Alonso, J. F. Toro-Vazquez, Front. Sustain. Food Syst. 2020,

4,1.

[26] C. D. Doan, D. van de Walle, K. Dewettinck, A. R. Patel, J. Am. Oil

Chem. Soc. 2015,92, 801.

[27] Z. Meng, K. Qi, Y. Guo, Y. Wang, Y. Liu, Food Chem. 2018,246, 137.

[28] S. Martini, C. Y. Tan, S. Jana, J. Food Sci. 2015,80, C989 .

[29] L. S. K. Dassanayake, D. R. Kodali, S. Ueno, K. Sato, J. Oleo Sci. 2012,

61,1.

[30] A. Lopez-Martínez, M. A. Charó-Alonso, A. G. Marangoni, J. F. Toro-

Vazquez, Food Res. Int. 2015,72, 37.

[31] A. J. Gravelle, M. Davidovich-Pinhas, A. K. Zetzl, S. Barbut, A. G.

Marangoni, Carbohydr. Polym. 2016,135, 169.

[32] A. Aliasl khiabani, M. Tabibiazar, L. Roufegarinejad, H. Hamishehkar,

A. Alizadeh, Food Chem. 2020,333, 127446.

[33] A. Aliasl Khiabani, M. Tabibiazar, L. Roufegarinejad, H. Hamishehkar,

A. Alizadeh, J. Am. Oil Chem. Soc. 2018,95, 797.

[34] J. F. Toro-Vazquez, M. Alonzo-Macias, E. Dibildox-Alvarado, M. A.

Charó-Alonso, Food Biophys. 2009,4, 199.

[35] M. Scharfe, Y. Ahmane, J. Seilert, J. Keim, E. Flöter, Eur. J. Lipid Sci.

Technol. 2019,121, 1800487.

[36] M. Scharfe, On the Importance of Minor Components and Oil Properties

for Sterol/Sterol Ester Strength, Sevilla, Spain 2019.

[37] V. Giacintucci, C. D. Di Mattia, G. Sacchetti, F. Flamminii, A. J. Grav-

elle, B. Baylis, J. R. Dutcher, A. G. Marangoni, P. Pittia, Food Hydro-

colloids 2018,84, 508.

Technology published by Wiley-VCH GmbH

2100068 (18 of 19)

www.advancedsciencenews.com www.ejlst.com

[38] H. - S. Hwang, J. D. Gillman, J. K. Winkler-Moser, S. Kim, M. Singh,

J. A. Byars, R. L. Evangelista, J. Am. Oil Chem. Soc. 2018,95, 557.

[39] J. A. Morales-Rueda, E. Dibildox-Alvarado, M. A. Charó-Alonso, R. G.

Weiss, J. F. Toro-Vazquez, Eur. J. Lipid Sci. Technol. 2009,111, 207.

[40] A. - M. Lampi, A. Kamal-Eldin, J. Am. Oil Chem. Soc. 1998,75, 1699.

[41] A. Mariod, B. Matthäus, I. H. Hussein, J. Am. Oil Chem. Soc. 2011,

88, 603.

[42] M. Scharfe, D. Prange, E. Flöter, J. Am. Oil Chem. Soc. 2022,99, 43.

[43] International Organization for Standardization, Animal and veg-

etable fats and oils- determination of content of polar com-

pounds: Revised Version 2018 2002, International Organization for

Standardization.

[44] Deutsches Institu for Normung e.V., DIN 53019-4:2016-

10, Rheometrie_- Messung von Fließeigenschaften mit

Rotationsrheometern_- Teil_4: Oszillationsrheologie, Beuth Verlag

GmbH, Berlin. doi: 10.31030/2560192 .

[45] T. G. Mezger, Das Rheologie Handbuch: Für Anwender von Rotations-

und Oszillations-Rheometern, Vincentz Network, Hannover 2016.

[46] T. Wettlaufer, B. Hetzer, E. Flöter, Eur. J. Lipid Sci. Technol. 2021,123,

2000345.

[47] I. Tavernier, Ph.D. Thesis, Ghent University, Ghent, Belgium 2018.

[48] A. B. Matheson, V. Koutsos, G. Dalkas, S. Euston, P. Clegg, Langmuir

2017,33, 4537.

[49] G. Dalkas, A. B. Matheson, H. Vass, A. Gromov, G. O. Lloyd, V. Kout-

sos, P. S. Clegg, S. R. Euston, Langmuir 2018,34, 8629.

[50] A. R. Patel, M. Babaahmadi, A. Lesaﬀer, K. Dewettinck, J. Agric. Food

Chem. 2015,63, 4862.

[51] H. - S. Hwang, M. Singh, J. K. Winkler-Moser, E. L. Bakota, S. X. Liu,

J. Food Sci. 2014,79, C1926 .

[52] S. Pandolsook, S. Kupongsak, J. Food Eng. 2017,214, 182.

[53] 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. 2013,50, 318.

[54] A. G. Marangoni, Edible Oleogels: Structure and Health Implications,

Elsevier, Cambridge MA 2018.

[55] A. I. Blake, The Microstructure and Physical Properties of Plant-Based

Waxes and their Relationship to the Oil Binding Capacity of Wax

Oleogels, Dissertation, Guelph 2015.

[56] H. - S. Hwang, S. Kim, K. O. Evans, C. Koga, Y. Lee, Food Struct. 2015,

5, 10.

[57] A. I. Blake, A. G. Marangoni, Food Biophys. 2015,10, 403.

[58] I. Tavernier, C. D. Doan, D. van de Walle, S. Danthine, T. Rimaux, K.

Dewettinck, RSC Adv. 2017,7, 12113.

[59] A. Matheson, G. Dalkas, R. Mears, S. R. Euston, P. S. Clegg, Soft Mat-

ter 2018,14, 2044.

[60] E. Yılmaz, M. Ö˘

gütcü, J. Am. Oil Chem. Soc. 2014,91, 1007.

Technology published by Wiley-VCH GmbH

2100068 (19 of 19)