Journal Pre-proofs

The presence of crystalline sugar limits the influence of emulsifiers on cocoa

butter crystallization

Selvyn Simoes, Enea Lelaj, Dérick Rousseau

PII: S0308-8146(20)32710-2
DOI: https://doi.org/10.1016/j.foodchem.2020.128848
Reference: FOCH 128848

To appear in: Food Chemistry

Received Date: 21 July 2020

Revised Date: 5 December 2020
Accepted Date: 6 December 2020

Please cite this article as: Simoes, S., Lelaj, E., Rousseau, D., The presence of crystalline sugar limits the
influence of emulsifiers on cocoa butter crystallization, Food Chemistry (2020), doi: https://doi.org/10.1016/
j.foodchem.2020.128848

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2020 Published by Elsevier Ltd.

 The presence of crystalline sugar limits the influence of emulsifiers on cocoa butter

crystallization

Selvyn SIMOES, Enea LELAJ, and Dérick ROUSSEAU*

Department of Chemistry and Biology, Ryerson University, Toronto, Canada

*Corresponding author

Address:

Department of Chemistry and Biology

Ryerson University

350 Victoria St

Toronto, Ontario, Canada

Phone: +1-416-979-5000 x552155

Fax: +1-416-979-5044

Email: rousseau@ryerson.ca

1
ABSTRACT

The effects of 2 wt% emulsifier and crystalline sugar on the isothermal solidification and

polymorphic behavior of cocoa butter were evaluated. The emulsifiers investigated were

commercial soy lecithin, polyglycerol polyricinoleate (PGPR), citric acid esters of mono- and

diacylglycerols (CITREM) and ammonium phosphatides (AP). All emulsifiers accelerated

cocoa butter nucleation and growth from the melt, with PGPR showing the smallest

enhancement. Lecithin and AP minimally affected the polymorphic form IV-to-V transition

contrary to PGPR and CITREM, which both promoted the formation of form V crystals. The

presence of sugar alone accelerated cocoa butter solidification while limiting the ability of the

emulsifiers to do so. Sugar alone, and in the presence of emulsifier, hindered the polymorphic

form IV-to-V transition. This study shows that the effects of emulsifiers on the isothermal

crystallization of cocoa butter can be muted in the presence of crystalline sugar, suggesting a

complex interplay dependent on emulsifier molecular structure.

Keywords: Cocoa butter, crystallization, polymorphism, emulsifier, sugar crystals.

2
1. INTRODUCTION

Emulsifiers are a key component in confectionery manufacturing as they help to control the

viscosity and yield stress of molten chocolate formulations. It is well-known that emulsifiers

will also affect triacylglycerol (TAG) crystallization (Smith, Bhaggan, Talbot, & Van Malssen,

2011). When mixed with bulk fats, they may enhance crystal growth by crystallizing or self-

organizing prior to the TAGs themselves, thereby acting as a template for nucleation and

growth (Davis & Dimick, 1989; Ishibashi, Hondoh, & Ueno, 2017). Conversely, they may

retard crystallization by interfering with nuclei formation or slowing polymorphic transitions

(Smith et al., 2011).

Though lecithin remains the dominant choice in confectionery manufacturing, polyglycerol

polyricinoleate (PGPR), ammonium phosphatides (AP) and citric acid esters of mono- and di-

glycerides (CITREM) also see usage in commercial products (Ačkar et al., 2015; Beckett,

2008). There is no current consensus regarding the effects of commercial soy lecithin on

confectionery fat crystallization as, in some cases, it may enhance cocoa butter fat crystal

growth rate, particularly at lower concentrations (0.2 - 0.3 wt%), but inhibit it at higher

concentrations (0.5 – 2 wt%) (Man, Rahman, Yusoff, Aini, & Miskandar, 2007; Miyasaki,

Luccas, & Kieckbusch, 2016; Rigolle et al., 2015). PGPR, an oligomer of polyricinoleic acid

chains esterified onto a polyglycerol backbone, is generally thought to have little effect on the

crystallization of cocoa butter or hydrogenated fats (Miyasaki et al., 2016; Wang, Liu, Jin,

Huang, Meng, & Wang, 2011). AP is an alternative to lecithin based on the same molecular

architecture as phosphatidylcholine, except that AP is the ammonium salt of a phosphate ester

rather than a zwitterion. Reports to-date have presented conflicting conclusions stating that it

can either accelerate or slow fat crystallization (Rossi, 1998; MacMillan, 2000). Finally,

CITREMs are synthetic emulsifiers where fatty acids are esterified onto a glycerol moiety with

subsequent attachment of a polar citric acid group. CITREMs have shown improved

3
rheological and textural properties of compound chocolates with a reduced fat content, but little

is known regarding their effects on fat crystallization (Ashkezary, Yeganehzad, Vatankhah,

Todaro, & Maghsoudlou, 2018; Norn, 2014).

Research on confectionery fats often extrapolates crystallization from bulk materials to

complex systems while overlooking the important role of dispersed particles. Other than

emulsifiers, exploring how dispersed particles affect cocoa butter crystallization is important

as these often comprise ~ 70 % of a chocolate’s total mass. Of note, heterogeneous nucleation

of TAG crystals off the surface of added particles was observed with the addition of 1 % foreign

particulate (Yoshikawa, Kida, & Sato, 2014). Comparatively little is known about fat

crystallization when both sugar particles and emulsifiers are present.

The present study explores the effects of added sugar and emulsifiers relevant to the

confectionery industry on cocoa butter isothermal crystallization and its form IV-to-V

polymorphic transition. The governing hypothesis was that emulsifiers would alter cocoa butter

crystallization, with their effect(s) dependent on molecular structure. Dispersed sugar, other

than enhancing cocoa butter crystallization by acting as a site for heterogeneous nucleation,

would interfere with the emulsifier’s capacity to affect crystallization. Assessing such effects

is highly relevant for the confectionery industry given the dominant role of dispersed particles

in confectionery formulations.

2.0 MATERIALS AND METHODS

2.1 Materials

Commercial cocoa butter, canola oil and lecithin were provided by Mondelēz International

(East Hanover, New Jersey, USA), with the composition of cocoa butter provided in

Supplemental Table 1. Canola oil was purified on silica (60Å, 70-230 mesh, 63-200 µm)

purchased from Sigma Aldrich (St. Louis, Missouri, USA) to remove any partial acylglycerols

4
and free fatty acids. PGPR 4150, CITREM 4205 and AMP 4455 were acquired from Palsgaard

(Juelsminde, Denmark) and had average molecular weights of 1500, 465 and 700 g/mol,

respectively. CITREM 4205 and AMP 4455 are produced using sunflower oil and have a high

oleic content (75 – 90 %). All emulsifiers were used as received. Confectioner’s sugar (10x,

average particle size ~ 60 µm) was purchased from Domino Foods (Iselin, New Jersey, USA)

and used as received. Cocoa butter-emulsifier blends were prepared by adding 2 wt%

emulsifier to molten cocoa butter followed by magnetic stirring at 80 °C for 1 hour. Cocoa

butter sugar blends were prepared as follows: a 1:1 mixture of 10x confectioner’s sugar and

molten cocoa butter was magnetically stirred at 40 °C for 1 hour. If present, emulsifier was

premixed with molten cocoa butter as above, before addition of sugar. In samples containing

sugar, the mixture consisted of 1 wt% emulsifier (i.e., 2 wt% relative to the fat phase).

2.2 Solid fat content (SFC) analysis

SFC was determined using a pulsed NMR (Minispec mq20, Bruker, Karlsruhe, Germany). A

2.0 mL aliquot of sample was added to an NMR tube (10 mm O.D.). Samples were heated in

an 80 °C waterbath for 10 minutes to remove thermal history, then transferred to a 20 °C

waterbath for 1 minute to cool. The tube was immediately transferred to the NMR sample port

with the sample chamber set at 20 °C. Isothermal crystallization measurements were taken

every 60 seconds over 12 hours using an adapted protocol from the AOCS Official Method

Cd16b-93 for direct SFC measurements (American Oil Chemists' Society, 1998).

The direct method, when used to determine the SFC of sugar-containing blends, will result in

inflated SFC values as sugar crystals cannot be differentiated from solid fat. To circumvent

this problem, AOCS Official Method Cd 16-81 was used to determine the SFC of sugar-

containing samples (American Oil Chemists' Society, 1998). This indirect method compares

the liquid and solid signal from a sample and a control oil at the temperature of interest (20 °C)

and a temperature at which the fat is fully molten (60 °C). As a result, four separate

5
measurements are performed at two temperatures to determine SFC. This approach isolates the

contribution of sugar to the overall signal, which is then removed revealing the SFC of the fat

phase only. Here, after the melting and cooling treatments were applied to the direct method

samples above, the SFC of all cocoa butter-sugar-emulsifier blends were determined after 15

minutes and 1, 3 and 5 hours of isothermal crystallization, and compared to the results of the

direct method. The Avrami rate and onset parameters of the bulk SFC vs. time data obtained

via the direct and indirect methods were fitted to the linearized Avrami equation:


SFCt  SFCeq 1  e
 K ( t Tlag ) n
 (1)

where SFCt and SFCeq represent the SFC at time t and at equilibrium and Tlag corresponds to

the onset time prior to the primary and secondary crystallization onsets. The value of n, the

Avrami exponent, gives an indication of the dimensionality of crystal growth and is sensitive

to the time dependence of nucleation. The parameter K, the Avrami constant, is an indicator of

crystal growth and nucleation rate. A program was written in Microsoft Excel v.14 (Redmond,

WA, USA) that allows for determination of relevant parameters (n, SFCeq, K and Tlag).

2.3 Melting temperature and enthalpy

Cocoa butter and cocoa butter-sugar blend melting behavior in the absence and presence of the

emulsifiers was characterized using a differential scanning calorimeter (DSC) (Q2000, TA

Instruments, New Castle, DE, USA). Molten fat samples (~ 5 mg, corrected for the presence

of sugar) were placed in an aluminum DSC pan, hermetically sealed, and heated in an 80 °C

oven for 10 minutes, with subsequent ageing in a 20 °C incubator. Measurements were taken

at 24 hour intervals over a period of 96 hours. Thermal profiles were evaluated from 20 °C to

80 °C at a heating rate of 5 °C/min. Peak fitting and integration were performed using

OriginPro 2015 software (Northampton, Massachusetts, USA). Supplemental Figure 1

6
illustrates the general approach, where curve fitting is used to differentiate the respective

enthalpic contributions of forms IV and V. The percentage enthalpy associated with form V

was determined from the total enthalpy and was used to demonstrate the relative polymorphic

ratio over 96 hours of ageing.

2.4 X-ray diffraction

Wide-angle x-ray diffraction (WAXD) was investigated using a powder X-ray diffractometer

(Miniflex 600, Rigaku, Tokyo, Japan), where molten samples (2.5 mL) were placed in 5 cm

diameter, circular aluminum pans, heated in an 80 °C oven for 10 minutes, and subsequently

aged in a 20 °C incubator. Measurements were taken after 24 or 96 hours in an aluminum

sample holder. Only cocoa butter-emulsifier blends were characterized using X-ray diffraction

given the interference of sugar.

2.5 Crystal morphology

Fat crystal morphology following storage at 20 °C for 24 or 96 hours was observed by confocal

laser scanning microscopy (CLSM) (LSM800 Zeiss Inc., Toronto, ON, Canada) with a 63 ×

oil immersion objective. Each 1 mL sample was melted at 80 °C for 10 min and stored at 20

°C. Morphology was evaluated after 24 and 96 hours by gently scraping ~ 5 mg onto a glass

slide along with ~ 0.1 mL of a 0.1 wt% solution of Fluorol Yellow (Sigma Aldrich, St. Louis,

Missouri, USA) in canola oil which was then gently covered with a glass cover slip, and imaged

at a 488 nm excitation wavelength.

2.6 Statistical analyses

All reported results are the arithmetic mean (± standard deviation) of triplicate experiments.

Analyses were performed using one-way analysis of variance (ANOVA) with a post-hoc Tukey

HSD test to compare multiple groups. A pairwise comparison was performed using the Student

t-test. Differences were considered significant at p < 0.05.

7
3. RESULTS AND DISCUSSION

Rosen and Dahanayake proposed that emulsifiers should serve specific functions. One function

is that, when added to a system, they should adsorb to a surface or interface and change its

properties (Rosen & Dahanayake, 2000). The other function is that they should aggregate in

the solvent in which they are dissolved and change the properties of the solution. The ability

of emulsifiers to influence fat crystallization has, as one of its fundamental tenets, molecular

complementarity in the molten state (Smith et al., 2011).

3.1 Emulsifier effects on isothermal crystallization

In chocolate, ~ 0.6 wt% emulsifier is added to the final mass of chocolate to facilitate flow

However, that quantity of emulsifier is relative to the entire chocolate mass, which takes into

account sugar crystals, cocoa particles, milk powder or other particles, if present. Using a

model system with only sugar and emulsifier, this relative ratio must be taken into

consideration. Here, 2 wt% emulsifier in cocoa butter was chosen relative to cocoa butter which

represents ~ 30 % of the finished chocolate mass.

Characteristic two-stage cocoa butter crystallization was observed in all blends, with a primary

stage where higher-melting TAGs crystallized and a secondary stage, where mass

crystallization of the dominant TAGs in cocoa butter occurred (Supplemental Figure 2)

(Dewettinck, Foubert, Basiura, & Goderis, 2004). The addition of either emulsifier or oil

shortened nucleation times and enhanced crystal growth rates, with the largest effects conferred

by the low molecular weight emulsifiers. Figure 1 reports the primary and secondary onsets

and rates of crystallization of cocoa butter in the presence/absence of sugar and emulsifier.

Addition of lecithin alone shortened the onset of nucleation from 25 minutes to 4 minutes.

Compared to the control cocoa butter, secondary growth occurred ~ 40 minutes earlier and its

rate increased from 2.2 × 10-6 s-1 to 2.2 × 10-4 s-1 - an increase of 100 ×. This was similar to

other efforts that have shown that lecithin accelerates cocoa butter crystallization (Ashkezary

8
et al., 2018; Miyasaki et al., 2016; Svanberg et al., 2011). In the high-melting fraction of cocoa

butter that crystallizes during primary nucleation, there tends to be a greater fraction of

phospholipids that are native to the cocoa butter, and it has been suggested that they facilitate

nuclei formation (Davis & Dimick, 1989; Rigolle et al., 2015). However, this depends on the

lecithin composition, as acceleration is facilitated more so by phosphatidylcholine rather than

lysophosphatidylcholine and phosphatidylinositol, which both hinder crystallization (Savage

& Dimick 1995). Given the observed effects, it was phosphatidylcholine that dominated the

behavior of lecithin.

CITREM and AP both decreased the onset of nucleation of cocoa butter from 25 minutes to ~3

minutes and decreased the onset of secondary growth by ~ 40 minutes, similar to lecithin

(Figures 1A and 1C). One postulated mechanism that describes enhancement of fat nucleation

is the formation of mesophase-ordered regions by polar emulsifiers that ultimately do not

incorporate into the crystal lattice due to unsaturation of their constituent fatty acids (Rigolle

et al., 2015). As lecithin, CITREM and AP all displayed similar effects on nucleation, it is

likely that their polar headgroups were responsible for the shortened onset times. Other liquid-

state emulsifiers with similar unsaturation in the TAG moieties such as sorbitan monooleate

and sorbitan trioleate have shown minimal effects on cocoa butter nucleation despite bearing

the same fatty acid moieties, which further supports that headgroup polarity influences

nucleation onset (Sonwai, Podchong, & Rousseau, 2017).

The presence of PGPR reduced the primary onset of crystallization to ~ 11 minutes rather than

~ 25 minutes in cocoa butter alone (Figure 1A). Secondary growth also occurred earlier, after

~ 86 minutes instead of ~ 111 minutes, with the secondary rate increasing to a lesser extent

than with any other emulsifier (Figure 1D). It was surmised that PGPR’s oligomeric nature

would result in dissimilar behavior to the other emulsifiers, given its inability to form ordered

organized assemblies in solution. For comparison, the effects of 2 wt% canola oil on cocoa

9
butter crystallization were investigated and closely resembled that of PGPR. Canola oil is

primarily comprised of triolein, with a cis configuration in the olefin at the n-9 position. Due

to its kinked structure, it is also unlikely that canola oil would integrate itself into cocoa butter

crystal nuclei (Rizzo, Norton, & Norton, 2015). Given their structural dissimilarity to the

crystallizing TAGs, any changes in cocoa butter crystallization were attributed to a dilution

effect as both canola oil and PGPR increased the liquid fraction of cocoa butter, mimicking

softer cocoa butters, which also have faster nucleation rates (Foubert, Vanrolleghem, Thas,&

Dewettinck, 2004). In contrast to our results, PGPR was previously shown to have no tangible

effect on cocoa butter crystallization rate (Miyasaki et al., 2016). This discrepancy may have

arisen from differences in concentration, where up to 0.8 % PGPR was used by Miyasaki et al.

compared to 2 wt% in the present study. As well, the present samples were isothermally

crystallized at 20 °C vs. at 15 °C in their work. With a greater degree of undercooling, the

driving force for crystallization is greater which would potentially diminish the effect of an

additive (Smith et al., 2011), however this would greatly depend on cooling rate (Bowser,

2006).

The Avrami constant n describes the dimensions that crystal growth occurs, where lower values

of n indicate growth in fewer directions, i.e., rod-like crystal growth, whereas higher values of

n indicate disc-like or spherulitic growth (Avrami, 1939, 1940; Wright, Hartel, Narine, &

Marangoni, 2000). The addition of emulsifier and oil did not have a significant effect on the

primary growth mode with all values remaining at n ~ 1, indicating rod-like growth from

instantaneous nuclei. Secondary growth n values in the presence of lecithin, AP and CITREM

were also ~ 1.0 whereas cocoa butter alone and in the presence of PGPR or canola oil yielded

n ~ 1.4.

Early-stage SFCeq showed a statistically insignificant increase from 4.1 to 4.5% in the presence

of emulsifiers, with no increase upon addition of canola oil (p ≥ 0.05). Second-stage SFCeq

10
showed similar trends, as canola oil showed no increase in SFC whereas the emulsifiers

enhanced equilibrium from ~ 45 % SFC to ~ 55 %. As the emulsifiers themselves contributed

no crystallizable species, the increase in SFC was strictly due to the solidification of the TAGs

present in the cocoa butter.

Table 1 highlights isothermal crystallization SFC values acquired at 15 min (representative of

the primary growth phase) after 1 and 3 hours (representative of the secondary growth phase)

and after 5 hours (representative of SFCeq). Emulsifier presence modestly increased the SFC

of the cocoa butter after 15 min and 1 hour. After 3 hours, the emulsifiers greatly increased

cocoa butter’s SFC of ~ 17 % to values of 50 - 53% for the low molecular weight emulsifiers

whereas PGPR and canola oil enhanced it to 32 – 36 %. At 5 hours, cocoa butter SFC was ~

40 %. This contrasted with all small-molecule emulsifiers, which increased SFC to 56 – 58 %

and PGPR as well as canola oil SFC values of 47 – 51 %. These results showed that all

emulsifiers promoted cocoa butter crystallization.

3.2 Effect of sugar and emulsifier on cocoa butter isothermal crystallization

The presence of sugar lowered the cocoa butter primary and secondary nucleation onsets from

25 to 9 minutes and from 111 to 64 minutes, respectively (‘CB + Sugar’ results in Figures 1A

and 1C). Secondary growth rate increased over 280 ×, demonstrating that sugar greatly

enhanced cocoa butter crystallization, likely through the promotion of heterogeneous

nucleation (Figures 1B and 1D) (Dhonsi & Stapley, 2006; Svanberg et al., 2011). Interestingly,

sugar did not impact cocoa butter’s primary rate of nucleation, which suggested that the higher-

melting TAGs in cocoa butter were not affected by the presence of sugar.

It was expected that the presence of sugar would limit the ability of emulsifier molecules to

associate with each other and template fat crystallization, owing to their preferential interaction

with the sugar crystal surface. Across all emulsifiers tested, sugar muted any effect added

emulsifier had on crystallization kinetics (using the direct method), resulting in similar Avrami

11
K values regardless of the emulsifier added (p ≥ 0.05). The onset of primary and secondary

growth across all samples occurred within 3 - 8 minutes and 69 - 81 minutes, respectively. The

rate constant, K, showed similar values across samples, 9 × 10-4 s-1 for primary growth and ~ 4

– 10 × 10-4 s-1 for secondary growth. These results suggested that the surface effects of sugar

were dominating crystallization events, and the ability of any emulsifier to affect crystallization

was diminished either through their immobilization at the surface of sugar or simply through

the magnitude of the effect of sugar. Presence of sugar did not change the calculated mode of

growth in the primary stage where, similar to samples without sugar, values suggested rod-like

growth from instantaneous nuclei, i.e., n ~ 1.0.

Table 1 contrasts the differences in SFC of the emulsifier-containing cocoa butter-sugar blends

at 15 min and after 1, 3 and 5 hours. The presence of sugar alone enhanced cocoa butter

crystallization most notably at 1, 3, and 5 hours, with increases of 12, 44, and 25 %,

respectively. The combination of sugar and emulsifier did yield greater SFCs in the early stages

of crystallization, with values ranging from ~ 11 % with PGPR to 15 % with AP.

Addition of lecithin to the cocoa butter-sugar blend increased SFC to a lesser extent than either

AP and CITREM, but all three showed a more substantial effect than canola oil and PGPR.

The largest difference was seen at the 1 hour mark, with the emulsifiers increasing SFC from

~ 15 % in the cocoa butter-sugar blend to 19 – 30 %, with CITREM displaying the greatest

increase. By 3 hours, all SFC values were 57 – 62 % and by 5 hours, all samples were at 64 –

66 % SFC, demonstrating that the emulsifier effects were more prevalent during the initial

period of stage 2 crystallization. Due to the speed of primary crystallization and the use of

standard solutions involved in indirect measurements, it was difficult to obtain primary rate

information from this method.

3.3 Fat crystal morphology

12
Figure 2 contrasts the fat crystal morphology of bulk cocoa butter isothermally crystallized for

96 hours at 20 °C with that observed in a 1:1 cocoa butter:sugar blend solidified under the same

conditions. In the bulk, cocoa butter crystallized in an unimpeded fashion based on the

spherulitic morphology and presence of fine needles jutting into surrounding molten fat (Figure

2A). Addition of any emulsifier to the cocoa butter did not appreciably alter the resulting fat

crystal morphology or size, with spherulitic crystals dominating all samples (Supplemental

Figure 3).

The typical spherulite size of the cocoa butter fat crystals in Figure 2A (D ~ 50 μm) was

significantly reduced in the presence of sugar, where the average crystal size was well-below

~ 50 μm (Figure 2B). After storage for 96 hours, the network of the cocoa butter-sugar blends

were composed of haphazardly organized sugar and needle-shaped fat crystals (Figure 2B).

Restricted fat crystal growth was apparent, and presumably resulted from the limited physical

space available, as newly-formed fat crystals were now forced to grow within a physically-

confined environment, given the significant volume fraction occupied by the dispersed sugar

(shown as black in Figure 2B).

3.4 Effects on polymorphic transitions

Cocoa butter alone showed a majority of form IV crystals with d-spacings of 4.32 and 4.13 Å

after 24 hours of isothermal storage at 20 °C, similar to Chapman et al. (Chapman, Akehurst,

& Wright, 1971) (Supplemental Figure 4). Storage for 96 hours revealed a gradual transition

from form IV to V, given the presence of a d-spacing at 4.53 Å. The form V polymorph is

associated with a strong peak at 4.5 – 4.6 Å accompanied by fingerprint peaks in the 3.7 – 4.0

Å region. The melting points for all fat-emulsifier blends ranged from 26.4 – 27.2°C for form

IV and from 30.7 – 31.4°C for form V, irrespective of emulsifier type.

Emulsifiers that share structural similarities with the solidifying fat may integrate themselves

into the fat crystal lattice, allowing them to modify the properties of the crystal rather than that

13
of the oil phase. Conversely, dissimilar emulsifiers will not be able to change the properties of

fat crystals directly and may instead alter crystallization by increasing the liquid fraction and

molecular motion via dilution, or by changing the polarity of the mother liquor (Garti,

Schlichter, & Sarig, 1986). The present results strongly suggested that all emulsifiers were not

integrated within the cocoa butter crystal lattices, which was expected given their molecular

dissimilarity and/or molecular weight difference (Norton, Lee-Tuffnell, Ablett, & Bociek,

1985). Elsewhere, Ueno and coworkers showed that mixing palm mid-fraction with 5%

sorbitan tripalmitate, which share the same palmitic acid fatty acid moiety, resulted in mixed

melting behavior given the emergence of a melting peak distinct from either palm mid-fraction

or sorbitan tripalmitate (Ishibashi, Hondoh, & Ueno, 2018). By contrast, sorbitan tristearate

and sorbitan tribehenate, showed no influence on the melting of palm mid-fraction given their

molecular dissimilarity.

Figure 3 shows the melting profile of cocoa butter isothermally aged at 20 °C for 96 hours.

Under these conditions, cocoa butter will initially crystallize in the form II polymorph and

transition to form IV within a few hours (Dewettinck et al., 2004). Here, a combination of

deconvolution, curve fitting, and integration applied to the thermograms was used to determine

the relative enthalpies of the form IV and V polymorphs (Figure 4). The enthalpies were then

converted into a ratio of polymorphs, as they relate to ageing and composition. By following

the transition from form IV to form V over a period of 96 hours, we determined the ability of

the emulsifier and sugar, and combinations thereof, to alter this polymorphic transition.

Figure 4A compares the evolution in the proportion of form V in the cocoa butter-emulsifier

blends as a function of storage time following isothermal storage at 20 °C over 96 hours. Cocoa

butter alone showed the lowest proportion of form V after 96 hours as all emulsifiers, as well

as the canola oil, promoted the form IV to V transition. Lecithin and AP promoted form V

formation in a non-linear manner, given the increase from ~ 5-10 % at 24 hours to ~ 65-70 %

14
after 96 hours, with a large increase between 24 and 48 hours. The CITREM and PGPR

promoted the formation of form V in a similar manner (from 25-30 % after 24 hours to ~ 80 %

after 96 hours).

Canola oil also promoted the form VI to V transition likely through a dilution mechanism, with

only slight changes in form V content over 96 hours, as after 24 hours ~ 59 % of the cocoa

butter had transition to form V. This result was expected as fat crystallization from a solvent

will generally lead to the formation of more stable polymorphs, typically the β polymorph

(Arishima & Sato, 1989). By considering the canola oil as a solvent, albeit at only 2 wt% of

the cocoa butter, it diluted the cocoa butter enough to provide greater space for the TAGs to

transition to form V more readily.

Neither the addition of emulsifier or oil affected the total melting enthalpy of cocoa butter (p

< 0.05). When encompassing the contributions of both the form IV and V polymorphs, cocoa

butter melting enthalpy rose from 102 to 131 J/g over 96 hours. The addition of all emulsifiers

resulted in similar trends in enthalpy, ranging from 99 – 102 J/g after 24 hours to 124 - 140 J/g

after 96 hours. The only outlier was AP, which yielded melting enthalpies of 124 J/g after 3

and 4 days, suggesting a hindrance of the form IV to V transition. With canola oil, the overall

melting enthalpy after 24 hours was ~ 116 J/g and 130 J/g after 96 hours.

Presence of sugar alone hindered the form IV to V polymorphic transition, both in rate and

extent (Figure 4B). After 24 hours, the cocoa butter-sugar blend solely consisted of form IV,

which gradually increased to ~ 49 % after 96 hours. Similar to the nucleation and growth results

above, sugar generally minimized the effect of added emulsifier, with most emulsifiers yielding

proportions of 65 - 75 % form V after 96 hours. The same effect was seen with canola oil,

which was surprising, as it was expected that the oil would remain free in the fat phase as it is

not surface-active. A potential explanation might be the presence of a micro-viscous

15
environment at the surface of the sugar crystals, which would reduce molecular motion

(Acevedo, Block, & Marangoni, 2012; Arishima & Sato, 1989; Lishchuk & Halliday, 2009).

The only exception to the rule was the mixture of AP + sugar, which completely blocked the

form IV to V transition over 96 hours. In the absence of sugar, AP behaved similarly to lecithin

with a minimal effect on the cocoa butter form IV-to-V polymorphic transition. Yet, when

combined sugar, there was a synergistic effect between the sugar and AP. Figure 4B shows that

AP prevented the transition from form IV to form V, perhaps by limiting the fluidity of the

continuous cocoa butter phase (Garti & Sato, 2001). Previously, AP was shown to limit the

absorption of water to sugar crystals in chocolates submerged in alcohol syrups (Holdgaard &

Wikman, 2003). This was attributed to a tight adsorption of AP to the surface of the sugar

crystals, but also supports the concept of a less plastic continuous phase, hindering the transfer

of water to the surface of the sugar crystals. Why AP caused this only in the presence of sugar

is unknown. Amongst the emulsifiers tested, AP had the smallest headgroup and lowest

compositional variability, which both likely contributed to its ability to efficiently pack on the

available sugar crystal surface and organize in the continuous phase around these. This ability

would increase the local micro-viscosity around the sugar crystals unlike the other emulsifiers,

thereby slowing the rate of cocoa butter form IV towards form V. Clearly, this topic warrants

further scrutiny.

The combined effect of sugar and emulsifier on the total enthalpy of melting of cocoa butter

crystals was limited. As noted, cocoa butter alone showed melting enthalpies of 102 - 131 J/g

over a period of four days. In the presence of all emulsifiers, the cocoa butter-sugar blend

showed similar enthalpies ranging from 90 – 120 J/g after 24 hours, and 101 – 139 J/g after 96

hours. Canola oil did reduce the overall enthalpy slightly, but this was not surprising as it

increased the liquid TAG fraction in the samples, thereby reducing the amount of fat that could

crystallize. Finally, the presence of sugar had no effect on the melting point of the form IV or

16
V polymorphs either alone or in the presence of emulsifier, as melting points ranged from 26.8

– 27.2 °C for form IV and from 30.3 – 31.0 °C for form V, respectively (p > 0.05).

3.5 Mechanistic considerations

Figure 5 depicts possible mechanisms relevant to the role of emulsifiers and sugar in the

systems studied. Emulsifier clustering in solution forming templates for crystallization to occur

has previously been reported for phospholipids in cocoa butter (Figure 5A) (Davis & Dimick,

1989). Rizzo, Norton and Norton also described the formation of mono and di-acylglycerol

micelles in a coconut - sunflower oil blend, which enhanced crystallization (Rizzo et al., 2015).

The ability for lecithin, CITREM and AP to enhance cocoa butter crystallization may be

attributed to this clustering and templating phenomenon. The PGPR and canola oil were

unlikely to participate in this behavior due to their structures and polarity, respectively, as they

are not conducive to the formation of an ordered region that would promote fat crystallization

(Figure 5B). Norton and coworkers described the exclusion of triolein during the crystallization

of tristearin and tripalmitin, with the presence of triolein also promoting the formation of

higher-melting polymorphs, similar to the present effects of PGPR and canola oil (Norton et

al., 1985). Sonwai et al. saw similar enhancement of cocoa butter crystallization and the

promotion of higher melting polymorphs in the presence of canola oil, and attributed the result

to a dilution effect (Sonwai et al., 2017).

The presence of sugar introduced a surface that may induce TAG heterogeneous nucleation

(Figure 5C). Yoshikawa and coworkers saw crystallization accelerated with the introduction

of 1% dispersed solid particles in cocoa butter (Yoshikawa et al., 2014). Sugar surfaces also

provide sites for emulsifiers to adsorb to, which changes the lipophilicity of the coated sugar

crystals (Rousset, Sellappan, & Daoud, 2002). Though the Avrami-based values for the

crystallization kinetics suggested a limited effect of the emulsifiers on cocoa butter

solidification in the presence of sugar, SFC values in the presence of sugar (with or without

17
emulsifier) were similar to, or greater, than that of cocoa butter and emulsifier alone at

intermediate times. However the form IV to V polymorphic transition was retarded, suggesting

the existence of a synergistic effect between the sugar and emulsifier that merits further

investigation. West and Rousseau saw similar trends in palm oil whereby sugar enhanced

crystallization and hindered polymorphic evolution (West & Rousseau, 2017). Similarly,

Acevedo, Block and Marangoni observed that increased viscosity decreased molecular

mobility during crystallization and promoted the formation of less stable polymorphs (Acevedo

et al., 2012).

Finally, canola oil’s ability to influence cocoa butter crystallization was likely due to a dilution

effect rather than surface activity. Thus, the presence of sugar would presumably not impact

its ability to accelerate polymorphic transitions. Since sugar hindered polymorphic evolution

even in the presence of canola oil, its effects were likely due to an increase in viscosity and a

decrease in molecular motion, rather than the association of additives to the surface of sugar

(Figure 5D).

4. CONCLUSION

We have observed the effects of 2 wt% emulsifier (on a fat basis) on cocoa butter nucleation,

crystal growth and polymorphism both in the presence and absence of sugar. We demonstrated

that all low molecular weight emulsifiers promoted crystallization of bulk cocoa butter during

the early stages of nucleation and growth to a similar extent despite the variety in polar

headgroup. Both PGPR and canola oil modestly enhanced nucleation and growth possibly

suggesting dilution as a consequence of their presence rather than an integration into the cocoa

butter crystal lattice. The low molecular weight emulsifiers demonstrated a minimal effect on

the form IV to V transition, given their liquid state and lack of molecular complementarity

whereas PGPR behaved similarly to added oil, accelerating the form IV to V transition, and

18
behaved as a solvent. The presence of sugar accelerated cocoa butter nucleation and growth

but muted the effect of emulsifiers, which was surmised as resulting from their adsorption to

the sugar crystal surface. The sugar retarded the form IV to V transition both in the presence

and absence of emulsifiers, potentially through a form IV stabilization mechanism with this

phenomenon enhanced in the presence of AP. Despite the variety of emulsifiers explored, there

was no significant impact on the sub-cell packing due to the inability of an emulsifier to become

integrated within the crystal lattice of the cocoa butter crystals as corroborated by the thermal

analysis. We demonstrated that, though the presence of sugar does indeed limit the ability of

emulsifiers to impact cocoa butter crystallization, there does exist and unknown cooperative

mechanism between AP and sugar that will retard the kinetics of polymorphic transition in

cocoa butter, from form IV to form V. Future work will continue to explore the mechanistic

relationship between emulsifier structure, presence of sugar and effects on fat crystallization

in greater depth.

ACKNOWLEDGMENTS

The Natural Sciences and Engineering Research Council of Canada (NSERC) is gratefully

acknowledged for funding support.

The authors declare no conflict of interest.

19
REFERENCES

Acevedo, N. C., Block, J. M., & Marangoni, A. G. (2012). Unsaturated emulsifier-mediated

modification of the mechanical strength and oil binding capacity of a model edible fat

crystallized under shear. Langmuir, 28(46), 16207–16217.

https://doi.org/10.1021/la303365d

Ačkar, D., Škrabal, S., Šubarić, D., Babić, J., Miličević, B., & Jozinović, A. (2015).

Rheological properties of milk chocolates as influenced by milk powder type,

emulsifier, and cocoa butter equivalent additions. International Journal of Food

Properties, 18(7), 1568–1574. https://doi.org/10.1080/10942912.2014.923440

Arishima, T., & Sato, K. (1989). Polymorphism of POP and SOS III. Solvent crystallization

of β2 and β1 polymorphs. Journal of the American Oil Chemists Society, 66(11), 1614–

1617. https://doi.org/10.1007/BF02636187

Ashkezary, M. R., Yeganehzad, S., Vatankhah, H., Todaro, A., & Maghsoudlou, Y. (2018).

Effects of different emulsifiers and refining time on rheological and textural

characteristics of compound chocolate. Italian Journal of Food Science, 30(1), 26–36.

Avrami, M. (1939). Kinetics of phase change. I: General theory. The Journal of Chemical

Physics, 7(12), 1103–1112. https://doi.org/10.1063/1.1750380

Avrami, M. (1940). Kinetics of Phase Change . II TransformationTime Relations for Random

Distribution of Nuclei Kinetics of Phase Change . II Transformation-Time Relations for

Random Distribution of Nuclei, 212(1940). https://doi.org/10.1063/1.1750631

Beckett.T.Stephen. (2008). “The Science of chocolate.” The Royal Society of Chemistry.

https://doi.org/10.1089/jam.2007.0571

Bowser, A. (2006). Crystallization of Cocoa Butter. 60th PMCA Production Conference,

124–127.

Chapman, G. M., Akehurst, E. E., & Wright, W. B. (1971). Cocoa butter and confectionery

20
 fats. Studies using programmed temperature X-ray diffraction and differential scanning

calorimetry. Journal of the American Oil Chemists Society, 48(12), 824–830.

https://doi.org/10.1007/BF02609292

Davis, T. R., & Dimick, P. S. (1989). Lipid composition of high-melting seed crystals formed

during cocoa butter solidification. Journal of the American Oil Chemists Society, 66(10),

1494–1498. https://doi.org/10.1007/BF02661979

Dewettinck, K., Foubert, I., Basiura, M., & Goderis, B. (2004). Phase behavior of cocoa

butter in a two-step isothermal crystallization. Crystal Growth and Design, 4(6), 1295–

1302. https://doi.org/10.1021/cg049772n

Dhonsi, D., & Stapley, A. G. F. (2006). The effect of shear rate, temperature, sugar and

emulsifier on the tempering of cocoa butter. Journal of Food Engineering, 77(4), 936–

942. https://doi.org/10.1016/j.jfoodeng.2005.08.022

Foubert, A.I.; Vanrolleghem, P.A.; Thas, O.; Dewettinck, K. (2004). Influence of Chemical

Composition on the Isothermal Cocoa Butter Crystallization. Journal of Food Science,

69(9), 478–487.

Garti, N., & Sato, K. (2001). Crystallization rocesses in fats and lipid systems. Marcel

Dekker. https://doi.org/10.1017/CBO9781107415324.004

Garti, N., Schlichter, J., & Sarig, S. (1986). Effect of food emulsifiers on polymorphic

transitions of cocoa butter. Journal of the American Oil Chemists’ Society, 63(2), 230–

236. https://doi.org/10.1007/BF02546144

Holdgaard, J., & Wikman, H. H. (2003). Alcohol resistant chocolate. Palsgaard A/S.

Ishibashi, C., Hondoh, H., & Ueno, S. (2017). Epitaxial growth of fat crystals on emulsifier

crystals with different fatty acid moieties. Crystal Growth and Design, 17(12), 6363–

6371. https://doi.org/10.1021/acs.cgd.7b01039

Ishibashi, C., Hondoh, H., & Ueno, S. (2018). Influence of Fatty Acid Moieties of Sorbitan

21
 Esters on Polymorphic Occurrence of the Palm Mid-Fraction. Journal of the American

Oil Chemists’ Society, 95(6), 709–720. https://doi.org/10.1002/aocs.12062

Lishchuk, S. V., & Halliday, I. (2009). Effective surface viscosities of a particle-laden fluid

interface. Physical Review E - Statistical, Nonlinear, and Soft Matter Physics, 80(1), 1–

7. https://doi.org/10.1103/PhysRevE.80.016306

MacMillan, S. D. (2000). Studies of the crystallisation of mixed confectionery fats under

sheared conditions using on-line synchrotron radiation saxs techniques. Heriot-Watt

University (United Kingdom), Ann Arbor. Retrieved from

http://ezproxy.lib.ryerson.ca/login?url=https://www.proquest.com/docview/301586980?

accountid=13631

Man, Y. B. C., Rahman, R. A., Yusoff, M. S. A., Aini, I. N., & Miskandar, M. S. (2007).

Effects of Emulsifiers on Crystal Behavior of Palm Oil Blends on Slow Crystallization.

Journal of Food Lipids, 14(1), 1–18. https://doi.org/10.1111/j.1745-4522.2006.00064.x

Miyasaki, E. K., Luccas, V., & Kieckbusch, T. G. (2016). Modified soybean lecithins as

inducers of the acceleration of cocoa butter crystallization. European Journal of Lipid

Science and Technology, 118(10), 1539–1549. https://doi.org/10.1002/ejlt.201500093

Norn, V. (2014). Emulsifiers in food technology (2nd ed.). Wiley.

Norton, I. T., Lee-Tuffnell, C. D., Ablett, S., & Bociek, S. M. (1985). A Calorimetric, NMR

and X-Ray Diffraction Study of the Melting Behavior of Tripalmitin and Tristearin and

their Mixing Behavior with Triolein. Journal of the American Oil Chemists’ Society,

62(8), 1237–1244. https://doi.org/10.5962/bhl.title.51366

Rigolle, A., Gheysen, L., Depypere, F., Landuyt, A., Van Den Abeele, K., & Foubert, I.

(2015). Lecithin influences cocoa butter crystallization depending on concentration and

matrix. European Journal of Lipid Science and Technology, 117(11), 1722–1732.

https://doi.org/10.1002/ejlt.201400555

22
Rizzo, G., Norton, J. E., & Norton, I. T. (2015). Emulsifier effects on fat crystallisation. Food

Structure, 4(2004), 27–33. https://doi.org/10.1016/j.foostr.2014.11.002

Rosen, M. J., & Dahanayake, M. (2000). Industrial Utilization of Surfactants: Principles and

Practice.

Rossi, A. (1998). An Examination of the inter-relationship between Structure and Kinetics

associated with Crystallisation of Single and Mixed Chocolate Confectionery Fats.

University of Strathclyde.

Rousset, P., Sellappan, P., & Daoud, P. (2002). Effect of emulsifiers on surface properties of

sucrose by inverse gas chromatography. Journal of Chromatography A, 969(1–2), 97–

101. https://doi.org/10.1016/S0021-9673(02)00370-9

Savage, C. M. (1995). Influence of phospholipids during crystallization of hard and soft

cocoa butters. The Manufacturing Confectioner, 75(5), 127–132. Retrieved from

http://www.gomc.com/firstpage/199505127.pdf

Smith, K. W., Bhaggan, K., Talbot, G., & Van Malssen, K. F. (2011). Crystallization of fats:

Influence of minor components and additives. Journal of the American Oil Chemists’

Society, 88(8), 1085–1101. https://doi.org/10.1007/s11746-011-1819-7

Sonwai, S., Podchong, P., & Rousseau, D. (2017). Crystallization kinetics of cocoa butter in

the presence of sorbitan esters. Food Chemistry, 214, 497–506.

https://doi.org/10.1016/j.foodchem.2016.07.092

Svanberg, L., Ahrné, L., Lorén, N., & Windhab, E. (2011). Effect of sugar, cocoa particles

and lecithin on cocoa butter crystallisation in seeded and non-seeded chocolate model

systems. Journal of Food Engineering, 104(1), 70–80.

https://doi.org/10.1016/j.jfoodeng.2010.09.023

West, R., & Rousseau, D. (2017). Modelling sugar, processing, and storage effects on palm

oil crystallization and rheology. LWT - Food Science and Technology, 83, 201–212.

23
 https://doi.org/10.1016/j.lwt.2017.05.021

Wright, A. J., Hartel, R. W., Narine, S. S., & Marangoni, A. G. (2000). Effect of minor

components on milk fat crystallization. Journal of the American Oil Chemists’ Society,

77(5), 463–475. https://doi.org/10.1007/s11746-000-0075-8

Yoshikawa, S., Kida, H., & Sato, K. (2014). Promotional Effects of New Types of Additives

on Fat Crystallization. Journal of Oleo Science, 63(4), 333–345.

https://doi.org/10.5650/jos.ess13155

24
FIGURE LEGEND

Figure 1 Avrami model results of the effect of 2 wt% emulsifier on the isothermal

crystallization of cocoa butter. (A) Primary onset; (B) Primary rate; (C) Secondary

onset (C); (D) Secondary rate.

Figure 2 Morphology of cocoa butter alone (A) and mixed with sugar (B) following 96

hours of isothermal crystallization at 20 °C. Dark areas in (B) are sugar crystals.

Size bars = 10 μm.

Figure 3 Evolution in DSC melting profile of cocoa butter isothermally crystallized at 20 °C

for up to 96 hours. For each time point, the left endotherm is associated with form

IV and the right endotherm with form V. Peak integration was used to determine

the relative amounts of the two polymorphs.

Figure 4 Evolution in the proportion of cocoa butter that has transitioned from form IV to V

during isothermal storage at 20 °C over 96 hours as affected by emulsifier presence

(A) or the combined presence of sugar and emulsifier (B).

Figure 5 Possible mechanisms influencing cocoa butter crystallization through: templating

(A), dilution (B), heterogeneous nucleation (C) or viscosity (D) effects. Yellow

triangles or diamonds - fat crystals; A) Emulsifier supramolecular aggregates are

micelles; B) Canola oil and PGPR are triskelion-like graphics; C) and D) Sugar

crystals are large, white polygons.

25
Figure 1:

30 14
A B
12
25
Primary Onset (minutes)

Primary Rate (10-6 s-1)

10
20

8
15
6

10
4

5
2

0 0
CB CITREM AP Lecithin PGPR Canola oil CB + Sugar CB CITREM AP Lecithin PGPR Canola oil CB + Sugar

125 800
C D
700
Secondary Onset (minutes)

100

Seondary Rate (10-6 s-1)

600

500
75

400

50
300

200
25
100

0 0
CB CITREM AP Lecithin PGPR Canola oil CB + Sugar CB CITREM AP Lecithin PGPR Canola oil CB + Sugar

26
Figure 2:

A B

27
Figure 3

24 Hours
2.5
48 Hours
72 Hours
2.0 96 Hours
Heat Flow (W/g)

1.5

1.0

0.5

0.0

20 25 30 35 40
Temperature °C

28
Figure 4

29
Figure 5

30
Table 1: SFC values at set time points showing the combined effect of sugar and emulsifier on

cocoa butter crystallization. SFC values were measured with either the direct method (a) or the

indirect method (b) (n = 3).

Sample 15 minutes 1 hour 3 hours 5 hours

CB(a) 0.6 ± 0.2 3.6 ± 0.2 17.4 ± 0.6 39.8 ±1.9
CB + Lec(a) 3.2 ± 0.2 5.3 ± 0.8 50.2 ± 1.7 55.9 ± 0.6
CB + CITREM(a) 3.4 ± 0.1 5.3 ± 0.2 52.1 ± 0.2 57.7 ± 0.2
CB + AP(a) 3.2 ± 0.3 5.7 ± 0.1 53.0 ± 0.1 58.2 ± 0.1
CB + PGPR(a) 1.5 ± 0.5 4.4 ± 0.3 36.4 ± 2.9 51.0 ± 1.2
CB + Canola oil(a) 1.5 ± 0.2 4.0 ± 0.2 31.7 ± 3.5 46.7 ± 2.2
CB + Sugar(b) 2.6 ± 0.7 15.0 ± 3.8 61.8 ± 1.0 65.5 ± 1.1
CB + Lec + Sugar(b) 12.2 ± 1.6 18.9 ± 1.3 57.2 ± 2.9 64.6 ± 0.9
CB + CITREM + Sugar(b) 14.3 ± 2.4 29.7 ± 2.6 62.3 ± 0.6 65.8 ± 0.5
CB + AP + Sugar(b) 15.0 ± 1.4 24.3 ± 3.5 60.6 ± 0.5 63.8 ± 1.1
CB + PGPR + Sugar(b) 10.9 ± 0.3 24.0 ± 2.3 59.9 ± 1.4 63.9 ± 0.4
CB + Canola + Sugar(b) 11.5 ± 1.3 25.5 ± 5.9 57.9 ± 4.6 64.6 ± 1.2

31
Supplemental Figure 1:
Heat Flow (W/g)

Form IV Form V

Temperature (°C)

Figure 1 Illustration of the curve fitting used to separate the contributions of the form IV

(blue) and V (red) polymorphs to melting enthalpy.

32
Supplemental Figure 2:

80

70

60
Solid Fat Content (%)

50

40
CB
CB+Lec
30
CB+PGPR
CB+CITREM
20 CB+AP
CB+canola

10

0
0 2 4 6 8 10 12
Time (hours)

Figure 2 Cocoa butter isothermal crystallization at 20 °C for 12 h using pulsed NMR in the

presence of 2 wt% emulsifier or canola oil. Error bars removed for clarity (n=3).

33
Supplemental Figure 3

Figure 3 Morphology of cocoa butter alone (top) and mixed with sugar (bottom) following 96

hours of isothermal crystallization at 20 °C. Dark, geometric areas are sugar crystals.

Micrographs taken at 100x magnification, and are part of, at least, triplicate experiments that

are representative of the bulk sample.

34
Supplemental Figure 4

35
Figure 3 Wide-angle X-ray diffraction of isothermally-aged cocoa butter at 20 °C after 24

hours (A) and 96 hours (B) of storage.

36
Supplemental Table 1

The TAG composition of the cocoa butter used in this study.

Triacylglycerol Abbreviation Percentage (%)

1-Palmitoyl-2-oleoyl-3-stearoylglycerol POS 29.3

1,3-Stearoyl-2-oleoylglycerol SOS 28.8

1,3-Palmitoyl-2-oleoylglycerol POP 24.4

1-Stearoyl-2,3-oleoylglycerol SOO 4.2

1,2-Stearoyl-3-linoleoylglycerol SSL 3.7

1-Palmitoyloyl-2,3-oleoylglycerol POO 3.4

1-Palmitoy-2-linoleoyl-3-stearoylglycerol PLS 3.0

1,2-Palmitoyl-3-linoleoylglycerol PPL 2.0

1-Stearoyl-2-oleoyl-3-arachidonoylglycerol SOA 0.8

Simoes et al 2020 - Author contributions

This manuscript was prepared through the collaboration of all authors, who collectively discussed,
designed and wrote the manuscript. All experiments were performed by authors El-Aooiti and Lelaj.

Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.

☐The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests:

37
HIGHLIGHTS
 Low molecular weight emulsifiers promote cocoa butter crystallization
 Emulsifiers do not alter cocoa butter polymorphic behavior
 Sugar accelerates cocoa butter crystallization, but slows its form IV to V transition
 The presence of sugar mutes the effects of emulsifiers on crystallization

38