Food Chemistry 386 (2022) 132671

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Palmitoylethanolamide gels edible oils

Duncan Schwaller , Yi Sui , Alain Carvalho , Dominique Collin , Philippe J. Mésini *
Université de Strasbourg, CNRS, Institut Charles Sadron, 23 rue du Loess, F67000 Strasbourg, France

A R T I C L E I N F O A B S T R A C T

Keywords: Palmitoylethanolamide (PEA) is an endogenous compound with no adverse effect for oral intakes of a gram per
Palmitoylethanolamide day. We show that PEA gels edible oils at concentrations as low as 0.5 wt%. The elastic moduli values of the
Organogelators formed gels are 1400 Pa at 1 wt% and 9000 Pa at 2 wt%. The study of the gels by cryo-SEM, optical microscopy
Oleogelators
and WAXS show that PEA forms lamellar solid aggregates with widths of several tens of micrometers. Upon
Oil
Phase diagram
heating, the sample shows two transitions. The first one is the gel-to-sol transition, observed by rheology and
defined by the switch from a solid to a liquid behavior. During this transition, the solid particles remain but do no
longer form a network. The second transition, observed at higher temperature by DSC corresponds to the melting
of the solid particles.

1. Introduction Marangoni, 2009).

In this paper, we explore the possibility to form gels from edible oils
Saturated and trans unsaturated fats are a public health issue because and palmitoylethanolamide (PEA, Fig. 1). This simple molecule is
they represent a risk factor for cardiovascular diseases (Castelli et al., endogenous, present in the mammalian brains, liver and muscles
1986; Martin, Browner, Hulley, Kuller & Wentworth, 1986; Hu et al., (Bachur, Masek, Melmon & Udenfriend, 1965). It was shown to have
2001; Mozaffarian, Katan, Ascherio, Stampfer & Willett, 2006; Brouwer anti-inflammatory and analgesic properties (Bortolotti et al., 2012;
Wanders & Katan, 2010; Clifton & Keogh, 2017). Public health au­ Kuehl, Jacob, Ganley, Ormond & Meisinger, 1957). In oral uptake by
thorities try to reduce their intake and tend to render the labelling of rats, it has a no effect level up to 1000 mg/body wt/day (Nestmann,
trans fat compulsory. Supported by this increasingly strengthening reg­ 2017). This high level makes them interesting candidates to gel edible
ulations, much research is now devoted to find replacement products of oils, especially if the concentration required to form a gel is a few weight
these solid fats. However, these fats play a major role to texture the food percent. In the present contribution, we have explored the gelation
products: they form a network of microcrystals that endow the lipid ability of PEA was investigated in different solvents, including vegetable
phases with their mechanical properties. The sought substitution prod­ oils and found that it can form gels in the latter at concentration below 1
ucts should provide similar rheological properties. In this context, wt%. We have studied in details the gels of PEA in rapeseed oil. We have
organogelators have gained increasing interest. These compounds self- studied the transitions in this system by microDSC, turbidimetry and
associate in organic solvents to form 3D networks, which results in the rheology and we have mapped the c-T binary phase diagram. Finally, the
formation of gels at mass fraction of a few percent. Gels made of such structure of the oleogels were investigated by scanning electron micro­
organogelators and vegetal liquid oils, comprising mainly cis-unsatu­ scopy and FTIR.
rated acids and healthier, constitute good substitution products of solid
fats (Scharfe & Flöter, 2020; Wesdorp, Melnikov & Gaudier, 2014; Zetzl 2. Materials and methods
& Marangoni, 2014). A few compounds are known for their ability to
form edible oleogels: waxes (Toro-Vazquez et al., 2007; Dassanayake, 2.1. Materials
Kodali, Ueno & Sato, 2009; Hwang, Kim, Singh, Winkler-Moser & Liu,
2012), mixtures fatty alcohol and fatty acid (Gandolfo, Bot & Flöter, Rapeseed Oil (Vita D’or), Olive oil (Carrefour) and Safflower oil
2004), 12-hydroxystearic acid (12-HSA) (Elliger, Guadagni & Dunlap, (Mon-droguiste) were purchased in a local grocery shop. All the oils
1972; Rogers, 2009; Rogers & Marangoni, 2008), mixtures of γ-oryzanol were refined. Palmitoyl chloride was purchased from Alfa Aesar
and sitosterol (Bot & Agterof, 2006) and ceramides (Rogers, Wright & (Thermo Fischer Scientific) and hydroxybenzotriazole from Acros

* Corresponding author.
E-mail address: mesini@ics-cnrs.unistra.fr (P.J. Mésini).

https://doi.org/10.1016/j.foodchem.2022.132671
Received 5 December 2021; Received in revised form 25 February 2022; Accepted 7 March 2022
Available online 9 March 2022
0308-8146/© 2022 Elsevier Ltd. All rights reserved.
D. Schwaller et al. Food Chemistry 386 (2022) 132671

The variable temperature experiments consisted in following, at 1 Hz,

the evolution of the complex shear modulus of the sample when tem­
perature was progressively decreased or increased between 110 and
20 ◦ C at a rate of 0.25 ◦ C/min. For each sample, the applied stress was
set to ensure the deformation belongs in the linear regime.
Prior to sample measurements, a blank experiment was conducted
without the sample in order to know the residual stress, in phase with
Fig. 1. Chemical structure of PEA. the strain, related to the experimental limit of the device. This residual
stress has been subtracted from the rheological measurements presented
Organics (Thermo Fischer Scientific). Ethanolamine was purchased from in this study.
Sigma Aldrich. EDAC was purchased from Apollo Scientific.
2.5. Differential scanning calorimetry

2.2. Palmitoylethanolamide
The thermograms were recorded with a MC-DSC (TA instrument).
The measuring cells were filled with PEA (between 1.73 and 13.89 mg).
A solution of palmitic acid (2.60 g, 10.12 mmol), hydroxybenzo­
Rapeseed oil (between 347 and 335 mg) was added in the cells to obtain
triazole (140 mg, 1.01 mmol) and ethanolamine (610 µL, 10.1 mmol) in
the desired weight concentration of PEA. The reference cell was filled
35 mL of anhydrous DMF, was stirred at 0 ◦ C. EDAC (1.57 g, 10.1 mmol)
with rapeseed oil only. The weights of the full cells had a maximal de­
was added and the reaction medium was stirred at 0 ◦ C for 1 h and at
viation of 5 mg. The presented thermograms were measured during a
25 ◦ C for 48 h. EDAC (0.78 g, 5.06 mmol) was added at 0 ◦ C and the
cycle of cooling and heating at 0.25 ◦ C/min.
mixture was stirred for 24 h at 25 ◦ C. 250 mL of 1 M aqueous HCl was
added. The product was recovered by filtration and washed several
2.6. Turbidimetry measurement
times with distilled water. The crude product was dissolved in chloro­
form, washed several times with distilled water and dried under vacuum
The turbidity of the samples at different temperatures was followed
overnight. PEA was obtained as a white powder (2.03 g, 6.78 mmol,
by the light intensity transmitted through the sample with a home-made
67% yield). 1H NMR (500 MHz, DMSO‑d6): δ (ppm) 7.75 (s, 1H, NH),
optical device. A monochromatic light beam (λ = 632.8 nm), provided
4.63 (t, 1H, OH, J = 5.4 Hz), 3.38 (q, 2H, CH2OH, J = 5.9 Hz), 3.09 (q,
by a HeNe laser (Melles Griot 05-LHP-151), successively passed through
2H, CH2NH, 6.0 Hz), 2.04 (t,2H CH2CO, J = 7.4 Hz,), 1.53–1.41 (m, 2H,
an attenuating filter (10–3), a pinhole (50 μm), then through the sample
CH2CH2CO) 1.24 (m, 24H), 0.86 (t, 3H, CH3, J = 6.8 Hz). 13C NMR (125
over a 5 mm length and through a second pinhole (150 μm). At the end,
MHz, DMSO‑d6): δ (ppm) 172.7 (NHCO), 60.5 (CH2OH), 41.8 (CH2NH),
the beam was focused onto the objective of a CCD camera (Retiga
35.8(CH2CO), 31.8 (CH2CH2CH3), 29.5, 29.5, 29.4, 29.3, 29.17 29.15,
2000R) to measure its intensity. The light intensity were measured while
25.7 (CH2CH2CO), 22.6 (CH2CH3), 14.43 (CH3). FTIR ATR (solid) cm− 1
temperature was increased from 20 to 100 ◦ C at a constant and imposed
3368 (νOH), 3286 (νNH), 2966 (νasCH3), 2954(νasCH2), 2917 (νasCH2),
rate of 0.25 ◦ C/min. Typical exposure times were comprized between
2847 (νsCH2), 1632 (amide I), 1552 (amide II), 1472 (δCH2). MS(ESI):
500 and 2000 ms. The sample temperature was controlled by an oven
322.27 (MNa+). Elemental analysis: found C% 72.29, H% 12.45, N%
driven by a PID controller (SHINKO 100).
4.79; calc. C% 72.19, H% 12.45, N% 4.68.
2.7. Optical microscopy
2.3. Gelation assay and gel formation
The solutions were observed and characterized with an optical mi­
PEA was weighed in a vial and the oil or solvent was added to reach croscope Olympus BX51, equipped with a Nikon DXM 1200 camera. The
the targeted concentration. The mixture was heated until complete studied mixtures were first heated and homogenized into the isotropic
solubilisation of PEA. The mixture was cooled to room temperature. The phase in an oven. Rectangular capillaries (CM Scientific, 0.5 mm × 2
gelation was checked by turning the vial upside down. When it forms a mm) were filled with this solution by capillarity and sealed with a
gel, the sample does not flow. blowtorch. The capillaries were then placed in a hot stage regulated at
± 0.1 ◦ C and observed by bright-field microscopy. The changes of phase
2.4. Mechanical measurements were observed upon cooling and heating at a rate of 0.3 ◦ C/min.

All the rheological studies were performed with a commercial stress- 2.8. FTIR
controlled rheometer (Haake, Mars III), in the oscillatory mode, with a
double Couette cell (DG41, Haake, outer gap of ~400 μm). For one Mid-IR spectra (600–4000 cm− 1) of the samples were measured as a
experiment, a plate-plate cell was used in order to apply different function of temperature between 28 and 120 ◦ C with a Bruker Vertex 70
compression stresses to the sample and to measure for each of them the spectrometer using a mercury cadmium telluride detector, a KBr beam
complex shear modulus. splitter, and a blackbody source. The spectral resolution was 2 cm− 1, and
The temperature was regulated by a heated bath (Haake F3) and 64 scans were coadded for each spectrum. The gel was inserted in a
controlled between 20 and 110 ◦ C, within ±0.05 ◦ C. The gel was formed homemade liquid cell equipped with two NaCl windows spaced by an
in the rheometer by cooling the sol from 110 to 20 ◦ C at a rate of indium flat O-ring. The cell was inserted in a thermoregulated Linkam
0.25 ◦ C/min. The formation of the gel was monitored by measuring the heating stage. The cooling or heating rate was the same as for DSC ex­
complex shear modulus (e.g. for c = 4 wt%, freq. of 1 Hz, stress of 0.2 Pa, periments (0.25 ◦ C/min).
the formation occurred at ~76 ◦ C). The gel was let at rest 1 h at 20 ◦ C
before the measurements upon heating. The range of linear regime was 2.9. Cryo-SEM
tested by measuring the shear modulus at a given frequency (1 Hz) as a
function of the applied stress σ. For a sample of concentration 4 wt%, G′ A piece of gel (2 wt%) of 5 × 5 × 2 mm3 was immersed in n-hexane
is constant below 20 Pa (Figure S1). The samples at low temperature during 24 h to remove all the rapeseed oil. Every 6 h, the n-hexane was
were subjected to a stress within this linear regime with frequencies renewed and the exchange continued. The gel placed onto a cryo-holder,
between 10-3 to 1 Hz: G′ plateaus over the whole range and is always quickly plunged into a nitrogen slush, and subsequently transferred
greater than G′′ (Example for c = 4 wt%, 20 ◦ C, σ = 0.2 Pa, Figure S2). under vacuum into the Quorum PT 3010 chamber attached to the

2
D. Schwaller et al. Food Chemistry 386 (2022) 132671

microscope. Then, the frozen sample was fractured with a razor blade
and etched at –70 ◦ C to reveal the details of the morphology. The sample
was eventually transferred in the FEG-cryo-SEM (Hitachi SU8010) and
observed at 1 kV at –150 ◦ C.

3. Results and discussion

3.1. Solubility and gelation

The structure of the studied compound is depicted in Fig. 1. PEA is

able to form gels in alkanes (hexane, cyclohexane, trans-decalin and in
several edible oils for wt. fraction of 0.5%. or higher: rapeseed, olive and
safflower oil. At room temperature, PEA is insoluble in the oils; it is fully
solubilized when the mixture is heated. When the solution is cooled back
at room temperature, the gel forms. These gels are turbid and are stable
for months. In this work, rapeseed oil was chosen for the physico­
chemical studies.

3.2. Study of the gel-to-sol transitions

Samples at different concentrations were heated from 25 ◦ C to 100 ◦ C

at a controlled rate. Their behaviors were followed by different tech­
niques: differential scanning calorimetry, mechanical measurements,
optical microscopy and turbidimetry. The heating rate was fixed of
0.25 ◦ C/min for all techniques except for optical microscopy at 0.3 ◦ C/
min. The different experiments have been carried out for concentrations
between 0.5 and 4 wt%. Fig. 2A shows the measurements on a sample of
concentration 4 wt%.
The mechanical measurements consist in measuring the loss (G′′ ) and
elastic (G′ ) moduli while the temperature is increased. At low temper­
ature, the sample shows a solid behavior with the tangent of loss angle
tanδ ≪ 1 and G′ > G′′ . The elastic modulus is 1400 Pa for the 1 wt%
sample and 9000 Pa for 2 wt%. When T increases, tanδ, G′ , and G′′ in­
crease. At a given temperature tanδ = 1 and the curves G′ and G′′ cross
over; when T further increases, tanδ ≫ 1 and G′ becomes smaller than
G′′ , which is characteristic of a liquid behavior. The temperature of the
gel-to-sol transition is taken as the temperature at which tanδ = 1 and G′
= G′′ . In the case of a gel at 4 wt% with an applied stress of 0.2 Pa, the
transition temperature is TGS = 81.3 ◦ C.
Turbidimetry experiments consist in measuring the light transmitted
through the sample. At room temperature, part of the light is scattered,
which shows that large particles are present in the medium. When the
temperature increases, the intensity keeps constant and shows a steep
increase starting around 85 ◦ C and reaches a plateau at 87.5 ◦ C. This
increase shows the disappearance of aggregates large enough to scatter
the light as the sol forms.
The thermogram measured by DSC shows an endotherm with a
maximum at 86.5 ◦ C (Fig. 2A). This endotherm represents the enthalpic
Fig. 2. Variations of different quantities of the PEA/rapeseed oil mixture; A and
changes accompanying the dissociation of the compound upon heating. B: at 4 wt% (applied stress 0.2 Pa); B enlargement of the variations of G′ and
It shows a shouldering peak around 90 ◦ C, which suggests it comprises viscosity (η ~ G′′ /ω) close to TDSC; variations of the kinematic viscosity ν,
two events. We have recently shown (Schwaller, Christ, Legros, Collin & coherent with an Arrhenian behavior (dashed line). C: gel at 0.5% (applied
Mésini, 2021) that the temperature of the full dissolution of the gelator is stress 0.1 Pa).
best estimated at the inflection point (TDSC = 90.6 ◦ C). The difference
between TGS and TDSC is 8 ◦ C, although all the experiments have been suppl. info). To summarize, the gap between TGS and TDSC diminishes
carefully thermoregulated. after the extrapolation but still has a significant value of 21 ◦ C.
For a concentration of 0.5 wt%, the difference between TGS and TDSC In the interval between TGS and TDSC, the mixture is a sol and has a
is even larger. The Fig. 2C shows the TGS value for 0.1 Pa stress, 41 ◦ C. liquid behavior, but a large part of the gelator is not melted and forms a
The thermogram measures at the inflection point a transition tempera­ solid fraction. It can be also detected by the slow decrease of G′ in the
ture TDSC = 79 ◦ C, which represents a difference of 38 ◦ C between TDSC same intervals. For c = 0.5 wt%, it reaches the floor value only around
and TGS. Part of this difference is due to a too large applied strain. As a 62 ◦ C, 21 ◦ C above TGS. For c = 4 wt%, the decrease of G’ is visible as a
matter of fact, the measured value of TGS increases from 41 to 56 ◦ C shoulder starting at 82.5 and ending at 90 ◦ C, close to TDSC (Fig. 2A and
when the applied stress decreases from 0.1 to 0.02 Pa (Fig. S3 suppl. 2B). These values of G′ , higher between TGS and TDSC than above TDSC,
info). When the gel is heated, part of the solid network progressively are probably the signature of a suspension of particles in a liquid. Only
dissociates, leading to a weaker gel. The applied stress may disrupt the after the end of the shoulder at TDSC, the shear viscosity, deduced from
gel before the actual transition. For this reason, the values of TGS were the G′′ values in the flow regime (η = G′′ /ω) follows a linear behavior
extrapolated at null stress to find a value of 59.2 ◦ C for 0.5 wt% (Fig. S4,

3
D. Schwaller et al. Food Chemistry 386 (2022) 132671

similar to that observed for the kinematic viscosity ν (Fig. 2B). This
result shows that the solution behaves like an Arrhenian simple fluid.
This observation, along with the DSC, suggests that after the sol-to-gel
transition, there are still solid aggregates, but are no longer connected
to form a network.
This is also confirmed by turbidimetry. For c = 0.5 wt%, the trans­
mitted intensity shows increases and reaches a plateau at 67 ◦ C and
corresponds to the transition to a transparent solution. Around 42 ◦ C the
intensity shows some fluctuations. It occurs around the sol-to-gel tran­
sition observed for an applied stress of 0.1 Pa as shown in Fig. 2C. This
stress is the order of magnitude of the stress corresponding to the weight
of the sample in the cell σ = ρgh with ρ the density of the sample, h the
height of the sample above the beam imprint and g the gravity accel­
eration. With the characteristics of our sample, σ = 0.09 Pa. The fluc­
tuations of intensity around 42 ◦ C correspond to movements of large
particles in the beam. The particles are not fully melted but are no longer
connected into a network and can move in a liquid.
In conclusion, this particular example of organogels illustrates that
rheology and DSC measure different quantities. DSC measures TDSC, the
temperature above which all the gelators is dissolved and represents the
liquidus. Rheology measures the temperatures TGS representing the limit
between the gel and the sol. In the range of temperatures between these
both limits the sample is a sol: it has the mechanical behavior of a liquid,
although a large part of the gelator forms solid particle. These particles
are suspended in the continuous liquid phase without forming a 3D
network. Hwang et al. have observed such gaps upon cooling between
the onset of crystallization (observed by cloud point) and gelation
(Hwang et al., 2012).

3.3. Morphological studies

In order to confirm the behavior of the gel in this intermediate

temperature range, the morphology of the gel was studied by cryo-SEM
and optical microscopy. Its direct observation by cryo-SEM showed
almost no structure because all the structures were embedded in the oil. Fig. 3. A: Cryo-SEM image of gel of PEA in rapeseed oil (2 wt%) after 24 h of
It was not possible to sublimate the oil to etch the gel and uncover the solvent exchange with n-hexane, showing large aggregates with lamellar
aggregates. Therefore, the oil was exchanged with a more volatile sol­ structures. B,C, D: optical micrographs of the same mixture at 70, 80 and 90 ◦ C.
vent. We found out that n-hexane was the most suitable solvent. Its
sublimation within the cryo-SEM, etched the gel and enabled the visu­ been observed that some of them are jammed suspensions (Collin et al.,
alization of the solid fraction of the gel (Fig. 3). This fraction consists of 2013). We have investigated that question by compressions assays: a
large crystals, several tens of micrometers wide, and rather flat. The sample at 4 wt% was formed as described in the experimental part in a
surface of some particles shows steps, which proves that the crystallites plate-plate rheometer, and the gap between the plates was gradually
have a lamellar structure. This lamellar structure is demonstrated by the reduced, thus applying a static compression stress to the sample. For
WAXS pattern of the gels, which display a series of Bragg peaks at 0.132 each compression rate, the storage modulus G′ was measured for fre­
0.385, 0.513 & 0.767 Å− 1) which shows a lamellar structure with a quencies varying between 0.1 and 30 Hz. For each compression rate, the
repeat distance of 47.6 Å (Fig S5 suppl. info). In order to follow the sample showed the characteristic response of an elastic gel at low
change in morphology with temperature, we have observed the gels by
optical microscopy while they were heated (rate 0.3 ◦ C/min). As an
example, the Fig. 3 shows the optical micrographs of a sample at 2 wt%
at 70, 80 and 90 ◦ C. At 70 ◦ C (Fig. 3B), the sample is in the gel state and
shows crystallites of several tens of µm wide, corresponding to those
observed by cryo-SEM. When the temperature increases, the size and the
number of crystallites decrease. Above the gel-to-sol transition tem­
perature, the crystallites are still present but lose their connectivity
(Fig. 3C, 80 ◦ C). Eventually, above the melting point, (Fig. 3D, 90 ◦ C)
those structures are no longer visible and the samples are homogeneous
solutions. The OM experiments thus confirm the existence of an inter­
mediate regime between the gel and the homogenous solution. This
regime behaves mechanically like a liquid and is actually a suspension of
solid particles not connected with each other.

3.4. Mechanical nature of the gels

Fig. 4. Mean values of the elastic modulus plateaus of a PEA/rapeseed oil gel at
The present oleogels are not formed by fibrillar particles, but by large 4 wt% submitted to different static compression ratios and loss angles. The
lamellar aggregates. Therefore, the question arises whether these gels errors bars represent the dispersion of the values around the mean value of G′
are genuine physical gels. This is the case of many organogels but it has on the plateau.

4
D. Schwaller et al. Food Chemistry 386 (2022) 132671

frequency, with constant values of G′ as frequency varies. The Fig. 4

reports the mean values of G′ ; the error bar is the dispersion of the values
around this mean values. The storage modulus G′ increases with
compression rates up to ~60%. The elastic modulus increases because
the volume density of nodes increases, which is characteristic of an
elastic behavior. If the gel was a jammed suspension, the sample should
show a thixotropic behavior and G′ should be constant during the static
compression.

3.5. Phase diagram upon heating

The phase diagram of PEA in rapeseed oil was mapped upon heating
(Fig. 5). As discussed above, the TGS values, measured by the cross-over
of G′ and G′′ vary with the applied stress. For this reason, they have been
measured for several applied stresses and extrapolated to null stress and
Fig. 5. c-T phase diagram of PEA/rapeseed oil measured at a heating rate of
the represented diagram has been plotted with these extrapolated values 0.25 ◦ C/min. Sol 1 and gel are both biphasic, comprising a continuous liquid
(Fig. 5). Even after extrapolation, it shows a gap between the transitions phase and a solid phase made of solid particles of PEA. In the gel, these particles
measured by rheology and the ones measured by microDSC. The first are connected and form a 3D network whereas in sol 1, they are too few or too
transition temperatures TGS mark the boundary between two mechani­ small to form a 3D network. The gel has the mechanical behavior of a solid, sol
cal states in the diagram. They refer to a state diagram. The second 1 that of a liquid. Sol 2 is a monophasic liquid, with PEA fully solubilized in oil.
transition temperatures TDSC define the disappearance of the solid phase.
Therefore, it defines a change in the number of phases and refers to a
genuine phase diagram. The discrepancy between both boundaries can
be explained by the shape and the size of the particles. When they have
partially melted their sizes diminishes, and because of their low aspect
ratio, they are no longer connected. For most gelators, both diagrams are
superimposed. But the present example shows that both diagrams may
be distinct and reminds the different natures of the mechanical and
thermodynamical transitions. The temperatures measured by turbi­
dimetry lie between TGS and TDSC, because at the measured temperatures
the particles have become smaller than those forming the network, no
longer detectable but not fully melted.

3.6. Molecular interactions

A 2 wt% sample of PEA/rapeseed oil was studied by infrared upon

heating from 28 ◦ C to 120 ◦ C at a rate of 0.25 ◦ C/min (Fig. 6). At low Fig. 6. Infrared of a 2 wt%. PEA/rapeseed oil from 28 ◦ C (blue curve) to 120 ◦ C
temperature (upper blue curve), the OH and NH stretching are found (red curve). (For interpretation of the references to colour in this figure legend,
respectively as single bands at 3200 and 3297 cm− 1 (Falk & Whalley, the reader is referred to the web version of this article.)
1961; Prystupa, Anderson & Torrie, 1994). It shows that all these groups
are H-bonded. When T increases, the intensities of these two bands interest for food or cosmetics applications. The phase diagram shows a
decrease, and three bands appear at 3413, 3450 and 2548 cm− 1. The gap between the sol-to-gel transition and the liquidus. This gap is
shouldering band around 3450 cm− 1 is attributed to the NH stretching explained by the large size and low aspect ratio of the particles. The
band of the free amide, the band at 3548 and 3413 cm− 1 to the OH domain between both transitions corresponds to a biphasic system
stretching of the free and H-bonded hydroxyl groups. The presence of comprising a liquid continuous phase and solid particles, too few or too
the second band shows that even in the melt, part of the hydroxyl groups small to form a 3D network.
are H-bonded, probably with other gelator molecules, but this associa­
tion takes places in the liquid state and with weaker H-bonds. Similar CRediT authorship contribution statement
facts are observed between 1700 and 1500 cm− 1, at low temperature
(blue curve). The band at 1640 cm− 1 corresponds to the amide I band Duncan Schwaller: Investigation, Formal analysis, Writing – review
and the two bands around 1560 cm− 1 to the amide II band and these & editing, Writing – original draft. Yi Sui: Investigation, Formal anal­
frequencies show the amide groups are H-bonded. At 80 ◦ C, the amide I ysis. Alain Carvalho: Investigation. Dominique Collin: Investigation,
band shifts to higher wavenumbers, around 1680 cm− 1 and the amide II Visualization, Formal analysis, Writing – review & editing. Philippe J.
shifts to lower wavenumber 1510 cm− 1, corresponding to non-bonded Mésini: Conceptualization, Supervision, Funding acquisition, Writing –
amide groups. In conclusion, in the solid, both the amide groups and review & editing, Writing – original draft.
the hydroxyl are self-associated through strong H-bonds which disap­
pear or become weaker in the melt. Declaration of Competing Interest

4. Conclusion The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
We have showed that PEA, a natural molecule well studied for its the work reported in this paper.
biological properties, is able to form stable gels in a variety of solvents,
especially in rapeseed oil where, its binary phase diagram was estab­ Acknowledgements
lished for a decade of concentration upon heating. The gel is made of
large lamellar crystallites forming a network in the mixture and The facility of polymer characterization is acknowledged for the use
endowing it with its mechanical properties. These new organogels are of of the UV and FTIR spectrometers. Anaïs de Maria is acknowledged for

5
D. Schwaller et al. Food Chemistry 386 (2022) 132671

the FTIR measurements. Mélanie Legros is acknowledged for her help Falk, M., & Whalley, E. (1961). Infrared Spectra of Methanol and Deuterated Methanols
in Gas, Liquid, and Solid Phases. The Journal of Chemical Physics, 34(5), 1554–1568.
with microDSC. The electron microscopy facility and Marc Schmutz are
https://doi.org/10.1063/1.1701044
acknowledged for their help. Thanks to Jean-Philippe Lamps for his help Gandolfo, F. G., Bot, A., & Flöter, E. (2004). Structuring of edible oils by long-chain FA,
with syntheses. Guillaume Fleith is acknowledged for WAXS measure­ fatty alcohols, and their mixtures. Journal of the American Oil Chemists’ Society, 81(1),
ments. Funding This work was funded by the Institut Carnot MICA 1–6. https://doi.org/10.1007/s11746-004-0851-5
Hu, F. B., Manson, J. E., & Willett, W. C. (2001). Types of Dietary Fat and Risk of
(Project Oleogel). D. S. is supported by fellowship of the French Coronary Heart Disease: A Critical Review. Journal of the American College of
Department of High Education and Research (MESR, ED 182). As part of Nutrition, 20(1), 5–19. https://doi.org/10.1080/07315724.2001.10719008
the Interdisciplinary Institute HiFunMat, the work was supported by Hwang, H.-S., Kim, S., Singh, M., Winkler-Moser, J. K., & Liu, S. X. (2012). Organogel
Formation of Soybean Oil with Waxes. Journal of the American Oil Chemists’ Society,
IdEx Unistra (ANR-10-IDEX-0002) and SFRI (STRAT’US project, ANR- 89(4), 639–647. https://doi.org/10.1007/s11746-011-1953-2
20-SFRI-0012) under the framework of the French Investments for the Kuehl, F. A., Jacob, T. A., Ganley, O. H., Ormond, R. E., & Meisinger, M. A. P. (1957). The
Future Program and the International Center for Frontier Research in Identification of N-(2-Hydroxyethyl)-palmitamide as a Naturally Occurring Anti-
Inflammatory Agent. Journal of the American Chemical Society, 79(20), 5577–5578.
Chemistry, Strasbourg (Synergy project). https://doi.org/10.1021/ja01577a066
Martin, M. J., Browner, W. S., Hulley, S. B., Kuller, L. H., & Wentworth, D. (1986). Serum
Appendix A. Supplementary data Cholesterol, Blood Pressure, and Mortality: Implications from a Cohort of 361 662
Men. The Lancet, 328(8513), 933–936. https://doi.org/10.1016/S0140-6736(86)
90597-0
Supplementary data to this article can be found online at https://doi. Mozaffarian, D., Katan, M. B., Ascherio, A., Stampfer, M. J., & Willett, W. C. (2006).
org/10.1016/j.foodchem.2022.132671. Trans Fatty Acids and Cardiovascular Disease. New England Journal of Medicine, 354
(15), 1601–1613. https://doi.org/10.1056/NEJMra054035
Nestmann, E. R. (2017). Safety of micronized palmitoylethanolamide (microPEA): lack of
Bibliography toxicity and genotoxic potential. Food Science & Nutrition, 5(2), 292–309. https://
doi.org/10.1002/fsn3.392
Bachur, N. R., Masek, K., Melmon, K. L., & Udenfriend, S. (1965). Fatty Acid Amides of Prystupa, D. A., Anderson, A., & Torrie, B. H. (1994). Raman and infrared study of solid
Ethanolamine in Mammalian Tissues. Journal of Biological Chemistry, 240(3), benzyl alcohol. Journal of Raman Spectroscopy, 25(2), 175–182. https://doi.org/
1019–1024. https://doi.org/10.1016/S0021-9258(18)97531-9 10.1002/jrs.1250250206
Bortolotti, F., Russo, M., Bartolucci, M. L., Alessandri Bonetti, G., Gatto, M. R., & Rogers, M. A. (2009). Novel structuring strategies for unsaturated fats – Meeting the
Marini, I. (2012). Palmitoylethanolamide Versus a Nonsteroidal Anti-Inflammatory zero-trans, zero-saturated fat challenge: A review. Food Research International, 42(7),
Drug in the Treatment of Temporomandibular Joint Inflammatory Pain. Journal of 747–753. https://doi.org/10.1016/j.foodres.2009.02.024
Orofacial Pain, 26(2), 99–104. Rogers, M. A., & Marangoni, A. G. (2008). Non-Isothermal Nucleation and Crystallization
Bot, A., & Agterof, W. G. M. (2006). Structuring of edible oils by mixtures of γ-oryzanol of 12-Hydroxystearic Acid in Vegetable Oils. Crystal Growth & Design, 8(12),
with β-sitosterol or related phytosterols. Journal of the American Oil Chemists’ Society, 4596–4601. https://doi.org/10.1021/cg8008927
83(6), 513–521. https://doi.org/10.1007/s11746-006-1234-7 Rogers, M. A., Wright, A. J., & Marangoni, A. G. (2009). Oil organogels: The fat of the
Brouwer, I. A., Wanders, A. J., & Katan, M. B. (2010). Effect of Animal and Industrial future? Soft Matter, 5(8), 1594–1596. https://doi.org/10.1039/B822008P
Trans Fatty Acids on HDL and LDL Cholesterol Levels in Humans – A Quantitative Scharfe, M., & Flöter, E. (2020). Oleogelation: From Scientific Feasibility to Applicability
Review. PLOS ONE, 5(3), Article e9434. https://doi.org/10.1371/journal. in Food Products. European Journal of Lipid Science and Technology, 2000213. https://
pone.0009434 doi.org/10.1002/ejlt.202000213
Castelli, W. P., Garrison, R. J., Wilson, P. W. F., Abbott, R. D., Kalousdian, S., & Schwaller, D., Christ, E., Legros, M., Collin, D., & Mésini, P. J. (2021). Investigation of an
Kannel, W. B. (1986). Incidence of Coronary Heart Disease and Lipoprotein Organogel by Micro-Differential Scanning Calorimetry: Quantitative Relationship
Cholesterol Levels: The Framingham Study. JAMA, 256(20), 2835–2838. https://doi. between the Shapes of the Thermograms and the Phase Diagram. Gels, 7(3), 93.
org/10.1001/jama.1986.03380200073024 https://doi.org/10.3390/gels7030093
Clifton, P. M., & Keogh, J. B. (2017). A systematic review of the effect of dietary Toro-Vazquez, J. F., Morales-Rueda, J. A., Dibildox-Alvarado, E., Charó-Alonso, M.,
saturated and polyunsaturated fat on heart disease. Nutrition, Metabolism and Alonzo-Macias, M., & González-Chávez, M. M. (2007). Thermal and Textural
Cardiovascular Diseases, 27(12), 1060–1080. https://doi.org/10.1016/j. Properties of Organogels Developed by Candelilla Wax in Safflower Oil. Journal of
numecd.2017.10.010 the American Oil Chemists’ Society, 84(11), 989–1000. https://doi.org/10.1007/
Collin, D., Covis, R., Allix, F., Jamart-Grégoire, B., & Martinoty, P. (2013). Jamming s11746-007-1139-0
transition in solutions containing organogelator molecules of amino-acid type: Wesdorp, L. H., Melnikov, S. M., & Gaudier, E. A. (2014). 13—Trans Fats Replacement
Rheological and calorimetry experiments. Soft Matter, 9(10), 2947–2958. https:// Solutions in Europe. In D. R. Kodali (Ed.), Trans Fats Replacement Solutions (pp.
doi.org/10.1039/C2SM26091C 287–312). AOCS Press. https://doi.org/10.1016/B978-0-9830791-5-6.50018-6.
Dassanayake, L. S. K., Kodali, D. R., Ueno, S., & Sato, K. (2009). Physical Properties of Zetzl, A. K., & Marangoni, A. G. (2014). 10—Structured Emulsions and Edible Oleogels as
Rice Bran Wax in Bulk and Organogels. Journal of the American Oil Chemists’ Society, Solutions to Trans Fat. In D. R. Kodali (Ed.), Trans Fats Replacement Solutions (pp.
86(12), 1163. https://doi.org/10.1007/s11746-009-1464-6 215–243). AOCS Press. https://doi.org/10.1016/B978-0-9830791-5-6.50015-0.
Elliger, C. A., Guadagni, D. G., & Dunlap, C. E. (1972). Thickening action of
hydroxystearates in peanut butter. Journal of the American Oil Chemists Society, 49(9),
536–537. https://doi.org/10.1007/BF02628900