Accepted Manuscript

Development of an extraction protocol for the removal of the fat phase within
chocolate

Dérick Rousseau, Aliénor Coutouly, Patrick Hendricks, Shane Hodge, Nicole L. Green

PII: S0023-6438(15)00368-0
DOI: 10.1016/j.lwt.2015.05.019
Reference: YFSTL 4675

To appear in: LWT - Food Science and Technology

Received Date: 19 February 2015

Revised Date: 5 May 2015
Accepted Date: 8 May 2015

Please cite this article as: Rousseau, D., Coutouly, A., Hendricks, P., Hodge, S., Green, N.L.,
Development of an extraction protocol for the removal of the fat phase within chocolate, LWT - Food
Science and Technology (2015), doi: 10.1016/j.lwt.2015.05.019.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.
 ACCEPTED MANUSCRIPT

Development of an extraction protocol for the removal of the fat

phase within chocolate
Dérick Rousseaua,1, Aliénor Coutoulya, Patrick Hendricksa, Shane Hodgea, Nicole L. Greena
a Department of Chemistry and Biology, Ryerson University, 350 Victoria Street, Toronto, Ontario, Canada M5B

PT
2K3

RI
Abstract

An extraction method to remove the cocoa butter fat phase from chocolate while leaving

SC
the dispersed particulate network intact was developed. Three parameters were evaluated:
i) the method of addition of small amounts of water to the chocolate during its preparation;
ii) the contact method of a solvent (petroleum ether) with the fat phase; and iii) the
duration of fat phase extraction. The addition of water was necessary to retain the

U
structure of the backbone particulate network during extraction. The absence of water
AN
resulted in dispersed particulates that fell apart, suggesting that the water aided in
network preservation. Water in the form of a emulsifier-stabilized water-in-oil emulsion
was the most efficient means of addition. Optimal fat extraction and structure retention
were achieved with a combined capillary/vapour phase method. Extraction times upwards
M

of 24 hours were necessary to extract sufficient cocoa butter from the chocolate for
particulate network analysis. Characterization with scanning electron microscopy and
confocal microscopy confirmed the adequacy of the developed extraction method and
D

demonstrated that the backbone structure consisting of dispersed particulates could be

made into a self-supporting network.
TE

Keywords: chocolate, cocoa butter, solvent extraction, particulate network, microscopy

EP

1. Introduction

Plain chocolate consists of a mixture of micron-scale particulates (sugar crystals and cocoa
C

powder) dispersed within a continuous fat phase. This cocoa butter (CB) phase is central to
AC

the enjoyment of chocolate, allowing chocolate to remain solid at ambient temperature and
melt at body temperature (Beckett, 2000). Much like industrial concrete, the CB fat phase
in chocolate is thought to act as the cement holding the dispersed particulates in place.
Particulates are typically 5 − 30µm in diameter, small enough to avoid a gritty mouthfeel
when chocolate is eaten (Afoakwa et al., 2007; Soulié et al., 2006). The sensory properties

1 Corresponding author. Tel.: 1+416-979-5000 x2155 Fax: 1+416-979-5044 Email

address: [email protected] (Dérick Rousseau)
Preprint submitted to LWT - Food Science and Technology April 20, 2015
 ACCEPTED MANUSCRIPT

of chocolate such as a glossy surface, melting profile, and texture depend on the quality of
the tempering process. Tempering is a highly controlled solidification protocol of the fat
phase that promotes the crystallization of CB into the desired polymorphic form and
morphology (Beckett, 2000).
The purpose of this study was to develop an effective extraction method to remove the CB

PT
fat phase from chocolate while leaving the dispersed particulate network intact. Such
efforts would permit examination of the role of the dispersed particulates on the
rheological and textural properties of heat resistant chocolate. Preliminary experiments

RI
indicated that the addition of moisture was necessary for successful fat phase extraction.
The effect of moisture on granular materials has been studied extensively (Iveson et al.,

SC
1996; Hornbaker et al., 1997; Willett et al., 2000; Bika et al., 2001; Kohonen et al., 2004),
where it has been observed that increasing the moisture content will consolidate loose
agglomerates through adhesive forces associated with interstitial liquid bridges between

U
grains. These liquid bridges are responsible for capillary bonding between grains. As a
result, when wet, the mechanical properties of the granular material change compared to
AN
its dry state. To illustrate, when sand becomes wet, it is able to form shapes and structure
(i.e., sandcastles) that would otherwise be unstable in dry sand. These capillary bridges
have been of recent interest with potential in engineering low fat or heat resistant
M

chocolate (Stortz & Marangoni, 2011; Hoffmann et al., 2014).

In this case, we work in the reverse direction - removing fat from preexisting chocolates.
D

Prior research has used a model chocolate system of just sugar and fat (Killian & Coupland,
2012), while we are interested in the effectiveness of our extraction method in virtually
TE

unadulterated chocolates. In developing an extraction method to remove the CB fat phase

from chocolate, three factors were assessed: solvent contact method, experimental
duration, and water addition technique. To extract the fat phase within the chocolate and
EP

expose the particulate network for subsequent microstructural characterization, petroleum

ether (PET) was used. PET is a non-polar solvent often used for lipid extraction in food
(Timms, 2003). More polar solvents were not as effective for fat phase extraction. The
C

method of contact between the solvent and chocolate was examined, namely direct contact,
vapour, and a combination of the two. The duration of CB extraction was determined as
AC

well, with the fat phase extracted for up to a week. Finally, the method of water addition
(up to 4 wt% of the chocolate) during its preparation was considered, using either a direct
spray method or incorporation via a water-in-oil (W/O) emulsion.
2. Materials and methods

2.1. Materials

The chocolate used for this study was provided by Cadbury Chocolates (Toronto, ON,
Canada). Its ingredients included sugar, CB, unsweetened chocolate, and soy lecithin. The

2
 ACCEPTED MANUSCRIPT

CB fat content for milk chocolate was 30%. Cadbury also provided raw CB. RO water was
used throughout this work. Petroleum ether (PET) was used as received from Fisher
Scientific (Toronto, ON, Canada). Polyglycerol polyricinoleate (PgPr) was provided by
Nealanders (Mississauga, ON, Canada).
2.2. Sample preparation

PT
By employing a lab-scale temperer (Tenon Engineering LTD, Leatherhead, UK), the
chocolate was tempered to make the entire chocolate crystallize into the form V polymorph
using the protocol shown in Figure 1. For each experiment, 400 g of chocolate was melted

RI
at 50°C for 1 hour, then the temperature was reduced to 27°C and held there under
continuous stirring for 10 minutes to induce crystallization. The chocolate was then heated

SC
to 32°C for 5 minutes to ensure that the unstable polymorphs melted and only form V seeds
remained.
At this stage, water was added in one of two approaches: bulk addition or incorporation via

U
a W/O emulsion. For the former, a spray bottle yielding finely-dispersed droplets was used
to add water at the end of the tempering stage. The spray bottle was weighed regularly to
AN
determine the amount of water added. Water was also added to the chocolate as a W/O
emulsion consisting of CB (47.5 wt%), water (47.5 wt%), and PgPr (5 wt%). Emulsification
took place in a two-stage valve homogenizer (APV-1000, SPX Corporation, Brockville, ON,
M

Canada) operating at 68.9 MPa, and the resulting emulsion was stored at 50°C to prevent
undesirable CB crystallization. Droplet size distribution of the emulsified water was
D

determined at 25°C using a Bruker Minispec Mq pulsed field gradient nuclear magnetic
resonance (pfg-NMR) unit (Bruker Canada, Milton, ON, Canada) that allows unimodal
TE

characterization of emulsion droplet size distributions via restricted diffusion

measurement.
The reported droplet size was averaged over three data sets. Amounts equivalent to 14
EP

wt% water were added to the chocolate. For example, for a chocolate mass of 400 g to
contain 4 wt% water, 34 g of emulsion were added. Beyond 4 wt% water, its negative
impact on flavor and texture were too great. Dispersion of the emulsified water was
C

ensured by pouring the post-tempered chocolate into a laboratory-scale blender and

blended for 10 seconds.
AC

The mixture was then poured onto a plastic kitchen cutting board and manually spread to
remove air bubbles. Lastly, the chocolate mass was poured into cylindrical moulds (h = 4
mm, D = 35 mm) followed by cooling at 4°C for 30 minutes and storing at 17°C until use (24
h).
2.3. Extraction protocol

For the fat phase extraction, chocolate disks were placed on a piece of circular Whatman
filter paper with 47 mm diameter, 8µm pore size. This was then placed on a second piece of

3
 ACCEPTED MANUSCRIPT

Whatman filter paper with 150 mm diameter and 20 − 25µm pore size. The larger filter
paper was cut and folded to optimize extraction efficiency (Figure 2). The chocolate and
filter papers were set on a rectangular metal grid inside a 600 mL beaker. 200 mL of PET
was poured into the beaker without contacting the filter paper atop the metal grid. The end
of the cut section of the large filter paper was immersed in PET, leading to chocolate-
solvent contact via capillarity. This allowed for more gentle fat extraction while retaining

PT
the particulate networks structure. A watch glass and parafilm were used to cover the
beaker thereby diminishing solvent loss, leaving only a small opening near the mouth.

RI
Sample weight was measured every 30 minutes.

2.4. Scanning electron microscopy

SC
Scanning electron microscopy (SEM) was used to characterize the morphology and spatial
distribution of the dispersed particulates prior to and following fat extraction. Chocolates
were sputter-coated with gold and examined using a JEOL JSM-6380LV scanning electron

U
microscope. Samples were typically imaged using an accelerating voltage of 10 kV and
mixed backscattered/secondary electron imaging (3:1). Magnifications were 50, 100, 250,
AN
500, and 1000×. All images shown are of the top (moulded) surface or broken face of the
chocolate and are representative of the microstructure following set extraction times.
2.5. Confocal laser scanning microscopy
M

In-situ characterization of the particulate network was performed via confocal laser
scanning microscopy (Zeiss LSM-510, Zeiss, Toronto, ON, Canada) (Auty et al., 2001). The
D

fluorescent stains fluorol yellow 088 (Sigma-Aldrich, MO, USA) and rhodamine B (Acros
Organics, NJ, USA) were both added into molten chocolate to better visualize the dispersed
TE

particulates in the chocolate. Multitrack images were taken with a 63× (1.2 NA) objective at
1024 × 1024 pixels, yielding a final resolution of 0.14 µm/pixel. Pinhole diameters were set
to collect optical slices of < 2.0 µm. Each was line-averaged four times, and the three
EP

resulting frames were overlaid using ImageJ software (National Institute of Health,
Bethesda, MD, USA, http://imagej.nih.gov/ij/). Light sources were an Argon laser (488 nm)
and a Helium/Neon laser (543 nm and 633 nm). Excitation wavelength and emission filter
C

combinations are detailed in Table 1.

AC

3. Results

3.1. Chocolate tempering

Thermal analysis confirmed that all chocolates (with and without added water) were
correctly tempered based on the presence of a large endothermic peak at 32 − 33°C in all
samples, corresponding to the form V polymorph (not shown). Hence, the presence of
water added as a W/O emulsion did not disrupt the crystallization of the fat phase.

4
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE

Figure 1: Tempering procedure to generate the required fat phase crystal polymorph.
EP

Table 1: Excitation and emission filter configurations for CLSM

excitation wavelength (nm) dichroic mirror(s) emission

C

filter
AC

HFT 488, NFT 545

Ar 488 NFT 635 VIS LP 505
NFT 545

He/Ne 543 HFT 488/543 LP 560

He/Ne 633 HFT UV/488/543/633 LP 650

5
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M

Figure 2: Schematic of fat phase extraction method. Chocolate disk sits atop a smaller filter paper used to
measure fat extraction and a larger filter paper cut such that a strip hangs into the PET solvent.
D

3.2. Mode of water addition on macrostructure

TE

The appearance of chocolates containing 0, 1, 2, or 4 wt% water incorporated through a

W/O emulsion is shown in Figure 3 following extraction via vapour and limited PET
contact. Samples with 0 or 1 wt% incorporated water did not retain their form, whereas
EP

greater water content allowed the chocolate samples to maintain their macrostructure
during CB extraction. As stated earlier, water was added to the chocolates in two ways; the
water spray approach was not as effective, leading to large surface droplets that did not
C

penetrate the chocolate sample. This is in contrast to similar work in which water
consistently stabilized a model fat/sugar network regardless of the mode in which it was
AC

added (Killian & Coupland, 2012). Our results may have been due to the macroscopic
droplets from the spray being too large to be incorporated. The volume-weighted
geometric mean diameter of the emulsified water droplets was 8.7 ± 0.1 µm with 95% of all
droplets between 3 and 23 µm in diameter.
CB extractions were still conducted for both water incorporation methods; however, only
the emulsified water yielded defatted samples with structural integrity (Figure 4). As a
result, the water spray approach was not used for further experiments.

6
 ACCEPTED MANUSCRIPT

3.3. Extraction efficiency

Chocolate disks were placed on two pieces of filter paper that sat on a metal grid in a
beaker containing PET. The grid was shaped to keep the chocolate above the solvent, with
the lower filter paper cut such that a filter paper ‘finger’ could absorb the solvent.
Chocolates were weighed prior to extraction and at various time intervals to assess the

PT
extraction rate and removed fat percentage. With the optimized protocol, 88% of the fat
phase was removed, thus permitting visualization of the exposed particulate network
backbone.

RI
Figure 5 shows the time-dependent extraction of the fat phase. All water-containing
chocolates demonstrated similar CB extraction patterns. The weight of the control

SC
chocolate did not decline as sharply as the others. In the absence of added water, 80% of
the fat phase was extracted after 5 days, with a plateau value after only 3 days. Presence of
added water accelerated and improved fat phase extraction efficacy. With 1 wt% added

U
water, a plateau value of 80% was reached after 24 hours. Small gains in extraction were
AN
achieved with up to 4 wt% added water, achieving a maximum extraction of 88% after 120
hours. Extraction times upwards of one month did not lead to appreciable gains in fat
extraction.
M
D
TE
EP

Figure 3: Top and side views of chocolate structure after fat extraction at varying water content: (A) 0 wt%, (B)
1 wt%, (C) 2 wt%, and (D) 4 wt%. All water added as a PgPr-stabilized W/O emulsion.
C
AC

Figure 4: Impact of the water addition method on the chocolate structure after fat extraction: 4 wt% water
added (left) as a fine mist and (right) as an emulsion.

7
 ACCEPTED MANUSCRIPT

It was not possible to remove the remaining fat irrespective of the various methods
attempted. Solvent vapour extraction alone did not perform as well as extraction via vapor
and limited solvent contact the former resulted in 65% fat extraction after 120 hours.
Based on these findings, the fat extraction method using both PET vapour and limited 170
solvent contact was used for all further experiments.

PT
3.4. Native chocolate structure

Differential staining of the dispersed particulates in chocolate permitted clear delineation

RI
of the fat and non-fat components (Figure 6). The 488 nm excitation illuminated the fluorol
yellow 088 staining of the continuous fat phase (blue), while the 543 nm excitation

SC
illuminated the rhodamine B staining of the milk protein particulates (yellow). The 633 nm
laser exploited the autofluorescence of the cocoa solids (red), and the dark regions were
the nonfluorescent sugar crystals.

U
AN
M
D
TE
C EP
AC

8
 ACCEPTED MANUSCRIPT

Figure 5: Fat phase extraction yield reported as a percentage of overall fat content as a function of water added
via W/O emulsion.

3.5. Chocolate microstructure with SEM

To provide visual information on pre- and post-extraction chocolate, SEM images of the

PT
chocolate samples containing 4 wt% moisture before and after 120 hours of extraction are
shown in Figure 7. Prior to PET extraction, embedded dispersed particulates are vaguely
visible in the fat network. Upon completion of extraction, the particulate network is

RI
distinctly visible, thereby verifying the effectiveness of CB extraction via vapour and
limited solvent contact.

SC
4. Discussion

We achieved a maximum CB extraction of 88% after 120 hours. More severe methods (e.g.

U
Soxhlet extraction) are capable of nearly 100% fat extraction (Simoneau et al., 2000);
however, in such methods, preserving the structure is not paramount. We suggest that the
AN
remaining 12% fat either was tightly bound within the particle network or present within
cocoa powder fibers. It has been shown that the internal cocoa powder structure is a
porous network capable of trapping fat (Do et al., 2011). Further, our SEM results lack
M

evidence of the fat network present in standard chocolate (Figure 7). We thus conclude
that our extraction technique has successfully removed the fat network that was originally
D

stabilizing the dispersed particulates. The post-extraction collapse at 0 wt% added water
indicated that any residual fat alone is not enough to maintain the structural integrity. We
TE

performed similar experiments with only PgPr; in those, an amount of PgPr equivalent to
that present in the W/O emulsion was added. These did not result in network formation,
showing that both water and emulsifier are needed for defatted network stabilization.
EP

We speculate that the addition of moisture enhanced the yield strength of the particulate
network by the formation of liquid capillary bridges and/or the dissolution and
recrystallization of sugar crystals into a solid network in a process similar to sintering.
C

Focusing first on the liquid capillary bridges, it is known that the addition of moisture leads
AC

to attractive forces in granular systems. The driving force behind this interparticle
attraction is the surface tension of the liquid. These wet granular systems forming capillary
bridges can further be divided into four states (pendular, funicular, capillary, and slurry)
based on the amount of liquid in the system (Mitarai & Nori, 2006). We hypothesize that
our system falls into the lowest moisture content category, pendular, based on the fact that
we moved from unstable at 1 wt% added water to stable at and above 2 wt%. Indeed, slight
changes in the liquid content of wet granular systems have been shown to cause sharp
changes in the number of contacts between particles and resultant tensile strength as the

9
 ACCEPTED MANUSCRIPT

system transitions from coating individual dry particles with moisture to connecting
adjacent particles (Kohonen et al., 2004; Scheel et al., 2008).

PT
RI
U SC
AN
M
D
TE
C EP
AC

Figure 6: CLSM of stained chocolate: (A) fluorol yellow 088 staining of continuous fat phase, (B) rhodamine B
staining of milk particles, (C) autofluorescence of cocoa particles, and (D) overlay of three images. Scale bar
represents 10µm.

10
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M

Figure 7: SEM of chocolate (A) prior to and (B D) following solvent extraction of the fat phase. Scale bars
D

represent 10µm.
TE

As for the role of sugar dissolution, sugar is highly soluble in water, and its solubility
decreases with temperature. Our sugar/water concentration was well into the
supersaturated regime so solid sugar crystals were present at even at 4 wt% added water.
EP

The presence of emulsifier and water has been shown to enhance adhesion between sugar
crystals in oil (Johansson & Bergenst˚ahl, 1992). As the samples cooled, the solubility of
sugar in water would decrease, leading to the precipitation of sugar crystals potentially
C

across these contact points. After 24 hours, water is able to create solid sugar bridges with
greater strength than the liquid bridges from which they formed (Billings et al., 2006). It
AC

seems thus feasible that the sugar crystals further strengthened the aforementioned
particulate network bonds, resulting in a stable defatted particulate network.
5. Conclusions

A gentle, efficient method for the removal of the fat phase in chocolate was developed. A
means of incorporating water into chocolate via a W/O emulsion stabilized with PgPr was
also established. Addition of 4 wt% water proved optimal as it allowed the microstructure
of chocolate to remain intact during CB extraction, and it was responsible for the highest

11
 ACCEPTED MANUSCRIPT

percentage of CB extraction. More than 50% CB was removed within the first 2 hours of
extraction (56%), yet 100% CB was not extracted even after a month, suggesting the
presence of tightly-bound or solvent-inaccessible fat (e.g., within the cocoa powder). The
most effec tive extraction configuration was based on the combined PET vapour and limited
contact with the solvent. With the successful development of this extraction protocol,

PT
subsequent research efforts will focus on the role of the dispersed particulates on the
rheological and textural properties of chocolate as well as further contributions of added
water on structure.

RI
Acknowledgments

SC
Funding from the Natural Sciences and Engineering Research Council of Canada is
acknowledged.

U
References AN
Afoakwa, E. O., Paterson, A., & Fowler, M. (2007). Factors influencing rheological and textural qualities in
chocolate - a review. Trends in Food Science & Technology, 18, 290–298. doi:10.1016/j.tifs.2007.02.002.
Auty, M. A. E., Twomey, M., Guinee, T. P., & Mulvihill, D. M. (2001). Development and application of confocal
scanning laser microscopy methods for studying the distribution of fat and protein in selected dairy products.
M

Journal of Dairy Research, 68, 417–427. doi:10.1017/S0022029901004873.

Beckett, S. T. (2000). The Science of Chocolate. (2nd ed.). Cambridge, UK: Royal Society of Chemistry. 245 Bika,
D. G., Gentzler, M., & Michaels, J. N. (2001). Mechanical properties of agglomerates. Powder 246 Technology, 117,
D

98–112. doi:10.1016/S0032-5910(01)00318-7.
TE

Billings, S. W., Bronlund, J. E., & Paterson, A. H. J. (2006). Effects of capillary condensation on the caking of
bulk sucrose. Journal of Food Engineering, 77, 887–895. doi:10.1016/j.jfoodeng.2005.08.031.
Do, T. A. L., Vieira, J., Hargreaves, J., Mitchell, J., & Wolf, B. (2011). Structural characteristics of cocoa particles
and their effect on the viscosity of reduced fat chocolate. LWT - Food Science and Technology, 44, 1207–1211.
EP

doi:10.1016/j.lwt.2010.10.006.
Hoffmann, S., Koos, E., & Willenbacher, N. (2014). Using capillary bridges to tune stability and flow 253
behavior of food suspensions. Food Hydrocolloids, 40, 44–52. doi:10.1016/j.foodhyd.2014.01.027. 254
C

Hornbaker, D. J., Albert, R., Albert, I., Barabasi, A. L., & Schiffer, P. (1997). What keeps sandcastles 255 standing?
Nature, 387, 765–765. doi:10.1038/42831.
AC

Iveson, S. M., Litster, J. D., & Ennis, B. J. (1996). Fundamental studies of granule consolidation. I. Effects of
binder content and binder viscosity. Powder Technology, 88, 15–20. doi:10.1016/0032-5910(96)03096-3.
Johansson, D., & Bergenst˚ahl, B. (1992). The influence of food emulsifiers on fat and sugar dispersions in oils.
III. Water content, purity of oils. Journal of the American Oil Chemists Society, 69, 728–733.
doi:10.1007/BF02635907.
Killian, L. A., & Coupland, J. N. (2012). Manufacture and application of water-in-oil emulsions to induce the
aggregation of sucrose crystals in oil: A model for melt-resistant chocolate. Food Biophysics, 7, 124–131.
doi:10.1007/s11483-012-9249-0.

12
 ACCEPTED MANUSCRIPT

Kohonen, M. M., Geromichalos, D., Scheel, M., Schier, C., & Herminghaus, S. (2004). On capillary bridges in wet
granular materials. Physica A-Statistical Mechanics and Its Applications, 339, 7–15.
doi:10.1016/j.physa.2004.03.047.
Mitarai, N., & Nori, F. (2006). Wet granular materials. Advances in Physics, 55, 1–45.
doi:10.1080/00018730600626065.

PT
Scheel, M., Seemann, R., Brinkmann, M., Di Michiel, M., Sheppard, A., Breidenbach, B., & Herminghaus, S.
(2008). Morphological clues to wet granular pile stability. Nature Materials, 7, 189–193.
doi:10.1038/nmat2117.

RI
Simoneau, C., Naudin, C., Hannaert, P., & Anklam, E. (2000). Comparison of classical and alternative extraction
methods for the quantitative extraction of fat from plain chocolate and the subsequent application to the
detection of added foreign fats to plain chocolate formulations. Food Research International, 33, 733–741.

SC
doi:10.1016/S0963-9969(00)00063-6.
Soulié, F., El Youssoufi, M. S., Cherblanc, F., & Saix, C. (2006). Capillary cohesion and mechanical strength of
polydisperse granular materials. European Physical Journal E, 21, 349–357. doi:10.1140/epje/i2006-10076-2.

U
Stortz, T. A., & Marangoni, A. G. (2011). Heat resistant chocolate. Trends in Food Science & Engineering, 22,
201–214. doi:10.1016/j.tifs.2011.02.001.
AN
Timms, R. E. (2003). Confectionery Fats Handbook: Properties, Production and Application. Oily Press.
Bridgwater, UK: Oily Press.
Willett, C. D., Adams, M. J., Johnson, S. A., & Seville, J. P. K. (2000). Capillary bridges between two spherical
M

bodies. Langmuir, 16, 9396–9405. doi:10.1021/la000657y.

D
TE
C EP
AC

13
Highlights ACCEPTED MANUSCRIPT

Development of an extraction protocol for the removal of the fat

phase within chocolate

• We extract the continuous cocoa butter phase from nonfat dispersed particulates.
• Small amounts of added water during tempering stabilizes the particulate network.
• Without added water, defatted chocolate cannot form a self-supporting structure.

PT
• Adding water allows the removal of 88% fat without loss of structural integrity.

RI
U SC
AN
M
D
TE
C EP
AC