Praline Jurnall

Crack formation in chocolate pralines
Master of Science Thesis in the Master Degree Programme Biotechnology
KATARINA SLETTENGREN
Department of Chemical and Biological Engineering

Division of Food Science
CHALMERS UNIVERSITY OF TECHNOLOGY
Gteborg, Sweden 2010
KATARINA S. SLETTENGREN

CHALMERS UNIVERSITY OF TECHNOLOGY
KATARINA S. SLETTENGREN
KATARINA S. SLETTENGREN, 2010.
This master thesis was performed at Chalmers University of Technology in cooperation with
SIK The Swedish Institute for Food and Biotechnology.
Supervisors: Dr. Llia Ahrn, SIK and Lina Svanberg, SIK
Examiner: Prof. Ulf Svanberg, Chalmers University of Technology

Chalmers University of Technology
SE-412 96 Gteborg
Sweden
Telephone + 46 (0) 31-772 10 00
SIK The Swedish Institute for Food and Biotechnology

Box 5401
SE-402 29 Gteborg
Sweden
Telephone: +46 (0) 10 516 66 00
Cover:
Chocolate pralines produced at SIK with model filling aw 0.65 stored for 28 days.

Abstract
Chocolate pralines are complex food products that contain a soft filling surrounded by a
chocolate shell. One problem with pralines is that the filling often starts to migrate through
the shell which causes major structural changes that, for water or liquor based fillings, could
result in cracking of the shell. Cracking is one of the major issues that lead to reduced shelf
life in chocolate pralines. The main suggestions for crack formation are that cracks form due
to moisture or ethanol migration through the chocolate shell, or due to an unbalanced
distribution of moisture in the filling that causes some parts to shrink and other to expand.
In this work, two praline geometries and four different model fillings (with water activities
(aw) between 0.67 and 0.99) were analysed with respect to chocolate shell cracking during
storage. The pralines were stored for 28 days, during which high resolution images of the
pralines were collected and analysed by a digital colour imaging system. Furthermore, water
activity measurements in the shell, combined with weight changes in the pralines gave
information on migration properties during storage.
The results showed that pralines with aw 0.99 in the model filling cracked first and to the
highest percentage for both geometries. Pralines with model filling aw 0.86 and aw 0.78 had
quite similar results to each other, and cracked to a considerably lower percentage than
pralines with model filling aw 0.99. For model filling aw 0.67, square pralines did not crack at
all while round pralines cracked at a low percentage. Round pralines started to crack first for
all model fillings during the storage period.
The major conclusions that can be drawn from this work are that both water activity (aw) in
the model filling and praline geometry have an impact on crack formation in chocolate
pralines.
i
Sammanfattning
Chokladpraliner r ett komplext livsmedel som bestr av en mjuk fyllning omsluten av ett
chokladskal. Ett problem r att fyllningen ofta brjar migrera genom skalet, vilket orsakar
stora strukturella frndringar. Fr vatten- eller likrbaserade fyllningar kan detta leda till
sprickor i skalet. Sprickbildning r ett av de frmsta problemen som leder till reducerad
hllbarhet i chokladpraliner. De frmsta frslagen till sprickbildning r att sprickor uppstr p
grund av att vatten eller etanol migrerar genom skalet, eller p grund av en obalanserad
frdelning av vtska i fyllningen som medfr att vissa delar blir mindre medan andra blir
strre.
I det hr arbetet har tv pralingeometrier och fyra olika modellfyllningar (med vattenaktivitet
(aw) mellan 0,67 och 0,99) analyserats med avseende p tendens att spricka under lagring.
Pralinerna lagrades i 28 dagar, under vilket hgupplsta bilder av pralinerna togs och
analyserades med ett digitalt bildanalyssystem. Vidare gav vattenaktivitetsmtningar i skalet,
kombinerat med viktfrndringar i pralinerna information om migrationsegenskaper under
frvaringstiden.
Resultaten visade att fr praliner med aw 0,99 i modellfyllningen uppstod sprickbildning i

skalet tidigt under lagringsfrsket fr bda geometrierna. Praliner med modellfyllning aw
0,86 och 0,78 gav relativt lika resultat och hade en procentuellt lgre andel med
sprickbildning n praliner med modellfyllning aw 0,99. Fr modellfyllningen med aw 0,67 var
fyrkantiga praliner intakta, medan endast ngra procent av de runda pralinerna fick
sprickbildning. Fr samtliga modellfyllningar pverkades de runda pralinerna frst.
De frmsta slutsatserna som kan dras frn det hr arbetet r att bde vattenaktivitet i
fyllningen samt pralingeometri har en inverkan p sprickbildning i chokladpraliner.
ii
Acknowledgement
First of all I would like to thank my supervisors Lina Svanberg and Llia Ahrn for all their
help and support during this work. Furthermore, I would like to thank my examiner Ulf
Svanberg for his help and assistance.
I would also like to thank Martina Hellmansson for teaching me how to use the tempering
machine and tempermeter, Bjrn Forsgran for showing me how to do the water activity
measurements, Aurlie Chan Sui Ko for helping me with the DigiEye camera and Niklas
Lorn for introducing me on how to use Analysis Five program. In addition I would like to
thank everyone at the Process department at SIK for their help and encouragement.
Finally I would like to thank my family and friends for their support during this work.
iii
1 Contents
Abstract ....................................................................................................................................... i
Sammanfattning ......................................................................................................................... ii
Acknowledgement.....................................................................................................................iii
1. Introduction......................................................................................................................... 1
2. Objective ............................................................................................................................. 1
3. Theoretical background ...................................................................................................... 2
3.1 Chocolate pralines composition and origin .............................................................. 2
3.1.1 Chocolate shell composition ................................................................................ 2
3.1.3 Composition and structure of cocoa butter .......................................................... 3
3.1.2 Filling properties .................................................................................................. 5
3.1.3 Model filling properties........................................................................................ 5
3.2 Chocolate processing................................................................................................... 8
3.2.1 Tempering .......................................................................................................... 10
3.2.2 Seeding technique .............................................................................................. 11
3.2.3 Moulding of chocolate pralines.......................................................................... 12
3.2.6 Cooling process and storage............................................................................... 14
3.3 Crack formation in chocolate pralines....................................................................... 14
3.4 Crack formation due to migration ............................................................................. 15
3.4.5 Effects of migration............................................................................................ 15
3.4.6 Suggested mechanisms for migration ................................................................ 19
3.5 The impact of praline geometries on the migration rate............................................ 22
3.6 Analysing techniques................................................................................................. 22
3.6.1 Temper curves .................................................................................................... 22
3.6.2 Measurement of water activity........................................................................... 23
3.6.3 Digital colour imaging system ........................................................................... 24
4 Material and methods........................................................................................................ 25
4.1 Raw materials ............................................................................................................ 25
4.1.1 Chocolate............................................................................................................ 25
4.1.2 Filling thickeners................................................................................................ 25
4.1.3 Humectants......................................................................................................... 25
4.1.4 Antimicrobial inhibitor....................................................................................... 25
iv
4.1.5 Ultrapure water................................................................................................... 25
4.2 Methodology and equipment ..................................................................................... 25
4.2.1 Preparation of model filling ............................................................................... 25
4.2.2 Praline production .............................................................................................. 26
4.2.3 Characterisation techniques during storage........................................................ 30
5 Results............................................................................................................................... 35
5.1 Filling thickeners ....................................................................................................... 35
5.2 Humectants ................................................................................................................ 35
5.3 Preparation of model filling....................................................................................... 36
5.4 Preparation of pralines............................................................................................... 36
5.4.1 Removal of air bubbles (shaking time) .............................................................. 36
5.4.2 Shell thickness (pouring time)............................................................................ 36
5.4.3 Amount of filling................................................................................................ 36
5.4.4 Heating time ....................................................................................................... 36
5.5 Results from characterisation of the pralines during storage .................................... 36
5.5.1 Digital colour imaging ....................................................................................... 37
5.5.2 Weight differences ............................................................................................. 48
5.5.3 Water activity (aw) measurements...................................................................... 50
6 Discussion ......................................................................................................................... 52
6.1 Visual characterization of cracking ........................................................................... 52
6.1.1 The impact of model filling on cracking ............................................................ 52
6.1.2 The impact of praline geometry on cracking ..................................................... 52
6.1.3 Origin of the cracks ............................................................................................ 52
6.1.4 The impact of shell thickness on cracking ......................................................... 53
6.2 Weight measurements................................................................................................ 53
6.3 Water activity ............................................................................................................ 53
7 Conclusions....................................................................................................................... 55
8 Further work...................................................................................................................... 55
References ................................................................................................................................ 56
Appendix 1 Filling recipes .................................................................................................... 59
Appendix 2 - Shell thickness.................................................................................................... 61
Appendix 3 - Weight change.................................................................................................... 64
Appendix 4 Water activity .................................................................................................... 66
v
vi
1. Introduction
Chocolate is generally considered as an affordable luxury product. The fact that it is solid at
room temperature but melts in the mouth, giving a smooth delicate taste, makes it a unique
product. From a more technical point of view, chocolate is a complex food consisting of solid
particles of cocoa powder, sugar and milk powder in a continuous fat phase. The production
of chocolate is a multi-step process where the tempering is one of the crucial steps. During
this the temperature of the chocolate is varied to obtain the right crystalline form of the fat.
Recently a novel technique has been developed where seeding materials are added to the
chocolate for obtaining the right crystalline form instead of the more traditional tempering.
Chocolate pralines are an even more complex food product since they also contain a soft
filling which can interact with the chocolate shell. The filling often starts to migrate through
the shell which causes major structural changes that could result in cracking of the shell.
Another effect that can be seen is a greyish haze on the praline surface, known as fat bloom,
which can occur when liquid fat migrates through the shell and crystallize on the surface.
Cracking and fat bloom are two major issues that lead to reduced shelf life in chocolate
pralines. The main suggestions for crack formation are that cracks form due to moisture or
ethanol migration through the chocolate shell, or due to an unbalanced distribution of
moisture in the filling that causes some parts to shrink and other to expand.
In this work, two praline geometries and four different model fillings (with water activities
(aw) between 0.67 and 0.99) were analysed with respect to tendency to crack during storage.
The pralines were stored for 28 days, during which high resolution images of the pralines
were collected and analysed by a digital colour imaging system. Furthermore, water activity
measurements in the shell, combined with weight changes in the pralines gave information on
migration properties during storage.
This work was performed within the ProPraline project which is funded by the European
Commission in the 7th framework programme.
2. Objective
The aim of this project was to examine the impact of water based filling and different praline
geometry on the crack formation in pralines. Water based fillings with four different water
activities (aw 0.67- 0.99) was tested together with two different geometries of the pralines.
1
3. Theoretical background
In this chapter, a short literature review is presented on the composition and origin of
chocolate, chocolate processing, crack formation in chocolate pralines and different analysing
techniques.
3.1 Chocolate pralines composition and origin

Production of chocolate is a several hundred year old processing technique. The final result is
highly dependent on the initial quality of the cocoa beans and their processing, mixing and
refining as well as the crystallization of the cocoa butter. Chocolate is defined as a mixture of
cocoa butter, sugar, an emulsifier (e.g. lecithin or PGPR), and ground, roasted cocoa powder.
The typical characteristic of chocolate is that it is solid in room temperature whereas it melts
at body temperature, forming a smooth liquid in the mouth (Rousseau, 2007).
A praline is composed of a chocolate shell surrounding a soft filling (Figure 3.1), which can
be fat, water or ethanol based (Becket, 2008). Fat based fillings often contain nuts, such as
hazelnut, peanut or almond, while water and ethanol based fillings can be either fruit or
alcohol based or a combination of the two (Ghosh, Ziegler, & Anantheswaran, 2002)
(Benitez-Sanchez, Leon-Camacho, & Aparicio, 2003).
Chocolate shell
Soft filling
Figure 3.1: Image of a praline showing the two major components.
3.1.1 Chocolate shell composition

There are three main types of chocolate; dark, milk and white, which all have some notable
differences in composition. Dark chocolate consists mainly of cocoa butter, cocoa powder and
sugar, while milk chocolate also consists of added milk solids. White chocolate consists of
cocoa butter, sugar and milk solids but no cocoa powder (Rousseau, 2007). In Table 3.1
typical formulations for milk, dark and white chocolate are presented.
Table 3.1: Typical composition (%) for milk, dark and white chocolate (Rousseau, 2007).
Ingredient (%) Milk chocolate Dark chocolate White chocolate

Cocoa liquor (=cocoa particles 12 40 -
suspended in cocoa butter)
Cocoa butter 19 12 23
Milk powder 20 - 30
Sugar 48.5 47.5 46.5
Lecithin 0.5 0.5 0.5
2
Traditionally chocolate contain about 50% sugars. The sugar in the chocolate is mostly in the
form of sucrose, a disaccharide containing one glucose and one fructose monomer, which is
shown in Figure 3.2 (Becket, 2008).
Figure 3.2: Sucrose, a disaccharide containing one glucose and one fructose monomer (Coultate, 2009).
The size and distribution of the particles in the chocolate matrix, together with the
polymorphic form and morphology of the fat, strongly affect the sensory properties and
quality of the chocolate (Rousseau, 2007). In Figure 3.3 a schematic image of the
microstructure in milk chocolate is presented.
Figure 3.3: Schematic image of the microstructure in milk chocolate, where cocoa solids and crystalline sugar
are surrounded by a continuous cocoa butter phase. The cocoa butter phase consists of both liquid and crystalline
fat that are in contact with an emulsifier (enlarged to the right in the image). The size of the compounds is not
correlated (Aguilera, Michel, & Mayor, 2004).
3.1.3 Composition and structure of cocoa butter

Cocoa butter consists of a mixture of about 40-50 different triacylglycerides (TAGs) (Schenk
& Peschar, 2004). A triacylglyceride, also called triglyceride or triacylglycerol, consists of a
glycerol backbone and three fatty acids of which the most common in chocolate are palmitic
acid (P), oleic acid (O) and stearic acid (St) (Rousseau, 2007). These can be bound to the
glycerol backbone in different orders and the TAG composition varies between cocoa beans
from different growing areas (Becket, 2008). The three dominating TAG combinations in
chocolate are POSt with 35%, StOSt with 23% and POP with 15% (Schenk & Peschar, 2004).
The atomic structure of a POSt triacylglyceride is shown in Figure 3.4.
3
H O
H C O C (CH2)14 CH3
O
H O C (CH2)7 CH CH (CH2)7 CH3

O
H C O C (CH2)16 CH3
H
Figure 3.4: Atomic structure of a POSt triacylglyceride, where P stands for palmitic acid at the top, O for oleic
acid in the middle, and St for stearic acid in the bottom.
Cocoa butter is polymorphic, which means that even if its fatty acid and TAG composition is
constant, it can exist in different crystal forms, depending on the arrangement of TAGs within
the solid fat matrix. Each crystal form has different thermodynamic stability. There are three
primary polymorphic forms in food fat, , and , where is the least and the most stable.
In cocoa butter there are six polymorphs, which can be noted with either Greek or Roman
nomenclature. The desired form is the Form V polymorph, also called V, which gives a
stable product with high resistance to fat bloom. Furthermore, it gives positive sensory
attributes as its melting point is above room temperature and slightly below body temperature.
In Table 3.2 the different polymorphic forms of cocoa butter and their melting points are
shown (Rousseau, 2007).
Table 3.2: Melting points of the polymorphic forms of cocoa butter according to Duck (1964) and Wille and
Lutton (1966). The more unstable and loosely packed forms are at the top, with an increasing stability and more
dense packing for the forms down the table (Lonchampt & Hartel, 2004).
Polymorphic form Melting point (C) Polymorphic form Melting point (C)
(Wille and Lutton) (Duck)
I 17.3 17.0
II 23.3 21-24
III 25.5 '' 28.0
IV 27.3 ' 33.0
V 33.8 34.4
VI 36.3
Cocoa butter is a difficult fat to process and there are strong efforts to replace cocoa butter for
chocolate production, either partly or wholly, both for technological and economic reasons.
These cocoa butter alternatives (CBAs) are mostly mixtures of various vegetable fats, often
modified, and can consist of palm and palm kernel oil, illip fat, shea butter, sal fat and
kokum butter. Such fats are referred to as cocoa butter equivalents (CBEs), cocoa butter
substitutes (CBSs) and cocoa butter replacers (CBRs). In several countries within the EU
other fats than cocoa butter is only allowed in chocolate up to 5% (Lipp & Anklam, 1998).
4
3.1.2 Filling properties
As mentioned earlier the filling can be fat, water or ethanol based (Becket, 2008). Fat based
fillings often contain nuts, such as hazelnut, peanut or almond, while water and ethanol based
fillings can be either fruit or alcohol based or a combination of the two. A distinct difference
that can be observed between fat based and water/ethanol based fillings are that fat based
fillings have a high liquid fat content in combination with a low water activity, while water
and ethanol based fillings do not have any liquid fat content at all and a much higher water
content (Ghosh, Ziegler, & Anantheswaran, 2002) (Benitez-Sanchez, Leon-Camacho, &
Aparicio, 2003). A problem that can be seen in pralines with water or ethanol based fillings is
cracking of the shell, while a more common problem in pralines with fat based filling is so
called fat bloom (Ghosh, Ziegler, & Anantheswaran, 2002).
3.1.3 Model filling properties

Fillings are a complicated product to investigate. In order to reduce the different parameters
that could affect the final results, a model filling can be used. To obtain a model filling with
desired properties, different humectants can be used to vary the water activity, together with
different thickeners to obtain good viscous properties.
3.1.3.1 Humectants
Water activity, aw, gives good information about the stability of a food product since it
accounts for the availability of water for degradation reactions. It is also of great importance
for controlling microbial growth. Water activity is defined as the ratio of partial pressure of
water vapour in the product, p, to the partial pressure of pure water, p0 (Mathlouthi, 2001)
(Coultate, 2009):
The water activity for some food products can be seen in Table 3.3. Below aw 0.5 there are no
bacterial growth (Coultate, 2009). As can be seen the water activity for pure chocolate is very
low.
Table 3.3: Water activity for some different foods (Coultate, 2009) (Baylis, MacPhee, Robinson, Griffi, Lilley,
& Betts, 1994).
Food Typical water activity

Fresh meat 0.98
Cheese 0.97
Preserves 0.88
Salami 0.83
Dried fruit 0.76
Honey 0.75
Dried pasta 0.50
Milk chocolate ~ 0.40
The water activity can be lowered with water activity depressors, so called humectants, such
as sugar, salt, citric acid and glycerol. Different substrates have a varying ability to depress
5
water activity, called humectancy. If sucrose is used as a reference standard with a
humectancy of 1, the humectancy of different substrates can be compared. This can be seen in
Table 3.4 (Mathlouthi, 2001).
Table 3.4: Comparison of humectancies of some substrates with sugar taken as reference standard
(humectancy=1) (Mathlouthi, 2001).
Substrate Humectancy
Sugar (sucrose) 1
Salt 11
Tartaric and citric acids 3
Glycerol 4
3.1.3.2 Thickeners
To stabilise fillings and other food products various thickening agents can be used. Examples
of thickening agents are different polysaccharides, such as xanthan gum or carrageenan, or
animalic extracts such as gelatin. Below follows a short description of the thickeners used in
this project.
3.1.3.2.1 Xanthan gum

Xanthan gum is a natural polysaccharide produced by the bacterium Xanthomonas
campestris. The xanthan polymer has a backbone of glucose units joined by -1,4 linkages,
such as for cellulose. Every fifth glucose unit carries a trisaccharide side-chain, including one
or two charged carboxyl groups that give its high affinity for water (Figure 3.5). The
distinctive feature for gums is their high affinity for water and the high viscosity of their
aqueous solutions. They do not form gels, instead their solutions contain their plasticity even
at relatively high concentrations. The essential structure for all gums is extensive branching,
which do not leave any length of ordered backbone to form the characteristic junction zones
of polysaccharide gels. The branches can still entrap a large amount of water and interactions
between them gives a high viscosity even for dilute solutions. A special feature for xanthan
gum is how easily the polymer chains associate together and are reversibly broken. When the
gum is cool and stationary, the polymer chains form a zigzag shape of sufficient length for
them to come together in junction zones and form a gel-like network, but just a small physical
stress is enough to disrupt them (Coultate, 2009).
6
Figure 3.5: Xanthan gum, produced by X. campestris (Garca-Ochoaa, Santosa, Casasb, & Gmez, 2000).
Xanthan is approved as an emulsifier/stabilizer additive in Europe since 1980, as item E415.

Due to its ability of emulsion stabilization, temperature stability, compatibility with food
ingredients, and pseudo plastic rheological properties, xanthan has been used in a wide variety
of foods. Xanthan gum is also used in pharmaceutical formulations, cosmetics, and
agricultural products due to its properties in thickening aqueous solutions, as a dispersing
agent, and stabilizer of emulsions and suspensions (Garca-Ochoaa, Santosa, Casasb, &
Gmez, 2000).
3.1.3.2.2 Carrageenan
Carrageenan is the general name for a family of sulphated polysaccharides extracted from red
sea weeds. Depending on the number and position of anionic sulphate groups, several types of
carrageenans can be recognized. The three most commercially exploited are kappa-, iota-, and
lambda-carrageenans (GAO, CHEN, XU, CHEN, & YAN, 2009). The repeating units of iota-
carrageenan are shown in Figure 3.6. Different monosaccharides predominate in particular
carrageenans, but the difference between different carrageenans in particular seaweeds is not
always clear-out. Commercial carrageenans used as gelling or thickening agents in food are
blends designed to have the right properties for a specific application (Coultate, 2009).
Figure 3.6: The repeating units of iota-carrageenan (Coultate, 2009).
Gelatination of carrageenan involves first the formation of double helices of galactan chains
wound around each other. These chains are not sufficient to form gels by themselves, but
7
must first associate together and form so-called superjunctions, showed in Figure 3.7. To form
these superjunctions neutralising cations are required, such as potassium (K+) for kappa-
carrageenan and calcium (Ca2+) for iota-carrageenan (Coultate, 2009).
Figure 3.7: Carrageenan superjunctions (Coultate, 2009).
3.1.3.2.3 Gelatin
Gelatin is extracted from the protein collagen, which is the main constituent of connective
tissues and bones (Eysturskar, Haug, Ulset, & Draget, 2009). Gelatin is widely used as a
gelling agent, especially in confectionary, desserts, and in meat products. Like gel forming
polysaccharides, gelatin is insoluble in cold water but dissolves in hot water. When the
solution cools the gelatin polymer revert from a disorganised random coil to a helical
arrangement similar to that in original connective tissue. Like this, parts of the polymer chains
associate together and form junction zones based on hydrogen bonding, as occurs in
polysaccharide gels (Coultate, 2009).
3.2 Chocolate processing

Cocoa beans are collected from cacao trees, a small tree growing in tropical climate. The tree
originates from South and Central America but now it is commercially grown around the
equatorial line. The three major cocoa growing regions are West Africa, South-East Asia and
South America (Becket, 2008). There are three main types of cocoa beans used in chocolates;
Criollo, from the Caribbean and Central and South America, has beans with white cotyledons
and a mild flavour, Forastero, dominating the West African cocoa crop, is more vigorous and
Trinitario, representing a small fraction of the worldwide cocoa production, is a hybrid of the
other two. There exist also a number of other varieties of beans, of which each bean will give
a unique flavour and aroma to the chocolate (Rousseau, 2007).
Tiny flowers on the branches and the trunk of the trees grow into small pods called cherelles
(Figure 3.8), which each contain about 30-45 cocoa beans. After the beans have been
harvested they are fermented, which is essential to produce characteristic flavour compounds
of the chocolate. After fermentation the beans must be dried before transportation to the
manufacturing countries. The beans must be stored dry so that they do not take up water.
They will become mouldy if the moisture level rises above 8% (Becket, 2008).
8
Figure 3.8: A cocoa tree with cocoa pods (Adhistory).
Before chocolate can be processed, the beans must be processed into cocoa mass, cocoa butter
and cocoa powder. This process involves cleaning the beans, roasting, cocoa bean shell
removing, also called winnowing, and cocoa butter extraction. The first step in chocolate
manufacturing is then the blending of ingredients, especially sugar, cocoa powder, the
emulsifier and cocoa butter, to a paste. After blending the paste is ground, while more cocoa
butter is added. The particle size of the solid particles is not allowed to be larger than 30 m
to give chocolate its smooth texture. After grounding the blend is conched in a scraped-
surface mixer (Rousseau, 2007). During conching the liquid chocolate is stirred continuously
for several hours, which allow some undesirable off-flavour compounds to be driven off
(Coultate, 2009). A flowchart of the chocolate manufacturing process is shown in Figure 3.9.
9
Figure 3.9: Flowchart of the chocolate manufacturing process (Rousseau, 2007).
3.2.1 Tempering
An important step in the production of chocolate is the tempering. By a well-defined
temperature programme the crystallization of the cocoa butter is controlled to form the desired
V form (Rousseau, 2007). A tempering sequence during lipid crystallization in chocolates is
shown in Figure 3.10.
Figure 3.10: Tempering sequence during lipid crystallization in chocolates (Afoakwa, Paterson, & Fowler,
2007).
At the first stage of tempering the chocolate is maintained at 45-50C to fully melt all fat
crystals. In the second stage the chocolate is cooled to around 27C under agitation, which
promote formation of both stable and unstable forms of crystal nucleus, also called seeds,
10
such as III, IV and V. The chocolate is then reheated to around 30C to melt unstable
polymorphs and ensure that only V crystal nucleus remains. Crystal growth is then promoted
under agitation and shearing for the last stage (Rousseau, 2007).
Tempering of the chocolate is made in a tempering machine, which is basically a heat

exchanger that heat and cool the chocolate as it passes through it. Since chocolate is a poor
heat conductor it must be mixed well so that all of it comes in contact with the cold contact
surface of the tempering machine. This is achieved with a rotating shaft on which is attached
scrapers or discs that force the chocolate to flow alternatively to the wall and to the centre,
instead of just passing through the machine. The faster the shaft rotates, the higher is the shear
rate and further the crystallization rate. What limits the shear rate is the size of the motor and
the heat formed by the mixing, which could start to melt the stable crystals as well. The
temperature must be carefully controlled which is done within so called cooling zones. Figure
3.11 shows a simplified image of a tempering machine.
Reheating and homogenisation ~30C 3

~50
~27C 2 Cooling zones
~32C 1
Figure 3.11: Simplified image of a tempering machine. The chocolate is heated to around 50C to erase all
crystal memory. In the cooling zones the chocolate is first cooled to a temperature where crystal nucleus can start
to form (zone 1 and 2), varying for different types of chocolate, before its temperature is raised again to melt out
unstable polymorphs (zone 3).
3.2.2 Seeding technique

Seeding is an alternative technique that can be used to obtain the right crystalline form of the
cocoa butter. Instead of creating crystal nucleus by a temperature treatment, as in
conventional tempering, it involves addition of cocoa butter in the VI form as seeds directly
to the pre-melted chocolate. The chocolate is melted to about 48C to erase all crystal
memory. The temperature is then directly lowered to 32-33C while mixing and 0.2-2 w% of
cocoa butter crystal suspension, containing VI seeds, is added. The paste is mixed to ensure
that the seeds dissolve evenly through the chocolate. When cooling, the VI seeds act as
nucleus to form the desired V crystals. To summarize, the main difference between
conventional tempering and seeding is that for tempering, seeds in form V and/or VI is
produced by temperature treatment and for seeding, the VI seeds is added to the chocolate
(Zeng, Braun, & Windhab, 2002).
11
3.2.3 Moulding of chocolate pralines
Moulding of chocolate pralines can mainly be done by three different moulding techniques;
conventional, cold-stamp or one-shot. They all have their different benefits depending on the
conditions and what type of product that is to be produced. In this work the pralines are
produced by conventional moulding.
3.2.3.1 Moulds used for chocolate production

When moulding solid chocolate tablets, most moulds used today are made of plastic,
compared to before when metal moulds were used. Plastics have the advantage of being
lighter and making less noise, as well as being flexible, which can ease to remove the solid
chocolate when sticking occurs. It is important that the moulds are preheated to within a few
degrees of the temperature of the tempered chocolate before the moulding process begins.
This is because if the tempered chocolate would come in contact with a too warm surface, the
crystals within it would melt and there would no longer be enough seeds for the chocolate to
crystallize properly. If the chocolate instead would come in contact with a cold surface, this
could cause some of the fat to crystallize in the wrong form (Becket, 2008).
3.2.3.2 Conventional moulding

In conventional moulding, the chocolate can be fed into the mould either by hand or by
equipment. The chocolate is then spread evenly in the mould and air bubbles are removed by
vigorously vibrating it. This is due to that chocolate is a non-Newtonian liquid and it has a
yield value, meaning that energy must be applied for it to start to move. This yield value will
also stop air bubbles from rising through the chocolate. By vibrating it, the vibration
decreases the viscosity of the chocolate, and thereby eases the removal of bubbles (Becket,
2008).
The mould is then turned upside down and vibrated, causing the liquid chocolate in the centre
to pour out. There are several critical features for this step, such as to achieve a good viscosity
of the chocolate. If the chocolate is not vibrated enough, it could lead to that no chocolate
come out of the shell, while if it is vibrated to vigorously, too much chocolate could come out
giving a too thin shell. Furthermore, the time that the mould is turned upside down and the
chocolate is let to pour out is of importance. The mould is then turned again and the shell is
let to cool before it is filled. It is important that the filling is not too hot so that it melts the
chocolate. For fondants and fat based fillings this is quite easy to achieve, since they can be
relatively fluid at about 30C. This is more difficult for caramel, for which the viscosity
changes rapidly with temperature (Becket, 2008).
Once the filling is in place a chocolate base is added to the shell. This is usually done by
having a scraper blade pressing against the mould and having tempered chocolate poured into
the mould in front of it. Like this any remaining indents or holes are filled while the scrape
takes away any excess chocolate. Finally the mould is cooled in a cooling tunnel, to fully
solidify the shell and base, before the praline is packed. Moulded pralines like these are
usually much shinier than sweets produced using an enrobing process. This is because the fat
at the outside of the shell crystallizes in contact with the mould surface, which gives a very
12
smooth and glossy praline surface (Becket, 2008). In Figure 3.12 a schematic diagram of a
conventional moulding process is illustrated.
Figure 3.12: Schematic diagram of a conventional moulding process.
3.2.3.3 Cold stamp process

Cold stamp process, also called frozen cone moulding or cool core among other names,
allows the production of thin and precise shells. Tempered chocolate is fed into a mould with
a volume equivalent to the required shell size. A cooled stamp is lowered down into the
moulds that presses out the chocolate and forms the shells. The cool surface induces the
chocolate to partially solidify and when the shell is strong enough to not collapse the stamp is
removed again. This only takes some seconds. The shell is then run through the usual cooling
process and can be filled and covered with a base. Figure 3.13 shows a photo of a cold stamp
process (Bhler Bindler) (Rossettini & Pellegrini, 2006).
Figure 3.13: Photo of a cold stamp process (Bhler Bindler).
3.2.5.3 One shot technology

With the one-shot technology the moulds are filled with both chocolate and filling at the same
time. An inner nozzle shots out the filling while an outer nozzle shots out the chocolate that
enrobes the filling. Figure 3.14 shows an image of a one-shot depositor. The technique has
13
been strongly refined in the last few years thanks to an improved electronic control of the
dosing parameters. A recent development is one-shot products with double filling and with
small inclusions mixed in the filling, which opens for new goals of this technology (Rossettini
& Pellegrini, 2006).
Figure 3.14: Image of a one-shot depositor (Bhler Bindler).
3.2.6 Cooling process and storage

After moulding most of the fat in the chocolate is still in a liquid state. To receive pralines that
are hard enough to handle and package, most of the fat must solidify in the right crystalline
form (Becket, 2008). This requires a cooling process where the temperature is strictly
controlled, to not have unwanted polymorphic transformations (Rousseau, 2007). The cooling
process is usually performed in a cooling tunnel, where cool air is blown over the pralines so
that heat is removed by convection. The temperature of the surrounding tunnel should be
maintained above the dew point, meaning the temperature at which moisture just starts to
condense. If the temperature is too low the fat can set in the wrong form. Besides lowering the
quality of the chocolate this can cause moulded chocolate to not contract properly so that they
are difficult to remove from the moulds. A too low temperature can also cause the air to
condense on cold surfaces and then drip onto the chocolate. This will dissolve some of the
sugar within it, leaving a white powdery surface called sugar bloom on the chocolate when it
is rewarmed and the water evaporates again (Becket, 2008).
Cooling tunnels are usually divided into different temperature zones. The initial cooling is
quite gentle. This is followed by the coolest part of the process, of about 13C. The
temperature can be lower on condition that the air moves fast enough to not condensate. The
time it takes for the chocolate to set depends on the amount of already existing crystals in the
chocolate, as well of its type and amount. A large piece of chocolate naturally takes longer
time to set than a small piece. Usually the cooling time vary between 10 and 20 minutes to
obtain a high quality chocolate (Becket, 2008).
Since chocolate is very sensitive to temperature and humidity, storage should be maintained at
a temperature of 15-17C and a relative humidity (RH) below 50% (Rousseau, 2007).
3.3 Crack formation in chocolate pralines

Crack formation in chocolate is thought to be one of the major causes for shelf-life reduction
in chocolate pralines. Cracking can be caused by different mechanisms, also depending of the
type of filling. A reason for crack formation in pralines with water based fillings is migration
of water from the filling into the chocolate shell. Similar observations have also been made
with ethanol in liquor fillings.
14
In cookie confectionary combinations a factor that causes cracking is redistribution of
moisture within the cookie/wafer centre. This causes the centre to both contract and expand at
the same time at different parts. Water vapour can also be absorbed through holes in the
coating (Minson, 1990) (Barron, 1977).
Furthermore, cracking can occur during process. The fact that tempered chocolate shrinks
upon cooling of the cocoa butter, which is essential for moulding, must be taken into account
for enrobing. If the enrobed centre is allowed to cool too quickly it could result in that the
shell cracks due to that it contracts more than the filling. On the other hand, could the
chocolate be discoloured and soft if it cools too slowly (Man & Jones, 1994).
3.4 Crack formation due to migration

As mentioned earlier, a suggested reason for crack formation in pralines with water based
fillings is migration of water from the filling into the chocolate shell. In this chapter, that
phenomenon is explained further.
3.4.5 Effects of migration

Migration in chocolate can be divided into three main types; moisture, ethanol and fat, of
which moisture and ethanol migration are quite similar and fat migration differ more
significantly to the other two. Fat, in a fat based filling with e.g. nuts, will migrate into the
coating mainly through the fat phase, while the non-fat particles are considered impermeable
to the migrating fat. For moisture and ethanol migration, the non-fat particles are not
impermeable, and water and ethanol molecules migrate both through and along the surfaces of
these hydrophilic particles. Since the hydrophilic particles in chocolate are embedded in a
continuous fat phase, migrating water and ethanol molecules must migrate through the fat
phase to come in contact with the hydrophilic particles. However, migration of moisture will
be slowed when increasing the fat content and fat migration will be slowed when increasing
the hydrophilic part (Ghosh, Ziegler, & Anantheswaran, 2002).
3.4.5.3 Moisture migration

Moisture that migrates through chocolate will get adsorbed by the hydrophilic particles within
the chocolate. As a result this might cause swelling of the particles, which in turn might lead
to crack formation and an acceleration of the moisture migration (Figure 3.15). Water
adsorption could also lead to surface dissolution, accumulation of sugar and crystallization of
amorphous milk powder, which could lead to sugar bloom (Ghosh, Ziegler, &
Anantheswaran, 2002).
15
Direction of the migration
Figure 3.15: A model for the migration of moisture through chocolate. The moisture content is represented by
the different background shading. The swelling of hydrophilic particles is marked as halos around the particles
(Ghosh, Ziegler, & Anantheswaran, 2002).
Moisture migration occurs when regions with different moisture content are brought together,
such as for a chocolate praline when a filling with high water content, e.g. fruit or liquor
based, comes in contact with the chocolate shell. The migration does not necessarily take
place from the region with the higher moisture concentration but from the region with the
higher water activity (section 3.1.3.1) (Ghosh, Ziegler, & Anantheswaran, 2002).
Chocolate consists of hydrophilic particles that are embedded in a continuous fat phase. The
amount of migrating moisture that will be absorbed by the hydrophilic particles in the
chocolate can be obtained from a moisture sorption isotherm. A moisture sorption isotherm is
a plot of the water content of a food, in mass of water per unit mass of dry material, towards
the water activity at a given temperature. The amount of water that is adsorbed is less at a
higher temperature for the same water activity, due to that molecular motion increases with
temperature. In Figure 3.16, a moisture sorption isotherm for dark chocolate at different
temperatures is shown (Ghosh, Ziegler, & Anantheswaran, 2002).
16
Figure 3.16: Moisture sorption isotherm for dark chocolate at different temperatures (Ghosh, Ziegler, &
Factors influencing moisture migration through chocolates are water activity or vapour
pressure difference, storage temperature, composition and structure, thickness of the shell and
solid fat content (Ghosh, Ziegler, & Anantheswaran, 2002). These factors are explained
further below:
Shell thickness: According to Ficks law, which relates the molar flux of a component
to the concentration gradient, the water diffusion rate is suggested to decrease with
increasing shell thickness (Ghosh, Ziegler, & Anantheswaran, 2002).
Structure of the cocoa butter: There have not been done any systematic study on the
effect of moisture migration through chocolate, but a close-packed fat crystal network
could be expected to allow less moisture migration than a loose-packed one (Ghosh,
Ziegler, & Anantheswaran, 2002). Each polymorph has a different morphology, where
-crystals, which are the least stable, also are the least densely packed compared to the
and forms. -crystals, which have the highest melting point, also have a more
dense crystal structure than the -form (Rousseau, 2007). A study by Landmann et al.
showed that the permeability through a film of cocoa butter was much higher for
cocoa butter with low melting polymorphs than for cocoa butter that had been properly
tempered (Landmann, Lovegren, & Feuge, 1960).
Surrounding water activity (aw): An increased water vapour gradient or water activity
increases the water vapour transmission rate and an increase of the relative humidity to
100% gives a large change in the permeability constant. The second can be explained
by that e.g. sucrose and cocoa beans tend to absorb a significant amount of moisture at
high humidity. This will swell the coating and change its structure, which will affect
17
the permeability behaviour of the coating (Ghosh, Ziegler, & Anantheswaran, 2002)
(Landmann, Lovegren, & Feuge, 1960).
Chocolate composition: At low vapour pressures the presence of non-fat particles does
not influence the migration rate or the permeability constant, but at high relative
humidity it increases the migration rate drastically (Ghosh, Ziegler, & Anantheswaran,
2002) (Landmann, Lovegren, & Feuge, 1960).
Solid fat content: Another important factor influencing water migration is the solid fat
content. A fat with high solid fat content is a good moisture barrier since fat in solid
state are more densely packed and allow less moisture to migrate (Ghosh, Ziegler, &
Anantheswaran, 2002) (Landmann, Lovegren, & Feuge, 1960).
Temperature: Temperature has a great influence on moisture migration. When

changing the temperature it gives rise to phase and structural changes in the chocolate.
An increase in temperature increases the diffusivity of the migrating molecule and
decreases the solid fat content, which booth increases the migration rate. Also the
thermal history of the chocolate is of great importance since it determines the structure
of the coating (Ghosh, Ziegler, & Anantheswaran, 2002).
With these factors in mind, there are some methods to control moisture migration. By
tempering the chocolate properly moisture migration can be reduced to a great extent, since it
increases the amount of stable and more densely packed fat crystals, and further the solid fat
content. Selection of proper emulsifiers and sweeteners are also aspects to consider. Since
non-fat particles tend to adsorb a large amount of moisture and swell the chocolate, the
selection of a sweetener that adsorbs less moisture could decrease the migration rate.
Furthermore, shell thickness and storage conditions, such as temperature and relative
humidity, are factors that can be controlled (Ghosh, Ziegler, & Anantheswaran, 2002)
(Rousseau, 2007).
3.4.5.4 Ethanol migration

Chocolate pralines with ethanol based filling have the highest tendency of quality loss since
they contain both ethanol and moisture, which are very mobile liquids. The changes start
usually at the early stages of migration, where loss of liquid in the filling causes a decrease in
volume of the centre. A continuous migration causes cracks in the chocolate surface, which
accelerate the migration of ethanol and moisture even further, and eventually leads to a
complete loss of liquid filling (Ghosh, Ziegler, & Anantheswaran, 2002).
The mechanism for migration of ethanol is believed to be quite similar to that of moisture. A
difference is that ethanol is more soluble in the fat, why a faster migration of ethanol through
the fat phase could be expected compared to that of moisture (Ghosh, Ziegler, &
18
3.4.5.5 Fat migration
Fat migration often occurs in chocolate when there is a compositional difference in the
product, i.e. some parts contain more liquid fat than others. This can be seen in chocolate
confectionary with nut-based centres, chocolate coated biscuits and cream filled chocolates,
where the liquid oil from the filling or biscuit tend to migrate into the chocolate coating. The
driving force for fat migration is the difference in TAG (triacylglycerol) concentration but
since it is easier to measure the liquid fat content (LFC), fat migration is often modelled as a
function of the difference in LFC (Ghosh, Ziegler, & Anantheswaran, 2002).
Migration of fat can result in softening of the coating and hardening of the filling, which both
can decrease the perceived quality of the product. Softening of the coating depend on the
compatibility of the fat in the filling and the fat in the coating. If they are incompatible they
form a so called eutectic that contain a lower solid fat content than in either of the two fats
separately (Ghosh, Ziegler, & Anantheswaran, 2002).
Fat migration occurs mostly through the liquid phase of the fat and the presence of non-fat
particles can decrease the migration rate by increasing the tortuosity of the diffusion pathway.
With tortuosity means the deviation from the linear migration path, which is illustrated in
Figure 3.17. Migrating oil may also dissolve part of the solid fat in the coating, which
increases the migration rate. This fat can re-crystallise on the surface of the coating and give
rise to the whitish grey of so called fat bloom (Ghosh, Ziegler, & Anantheswaran, 2002) (Lee,
McCarthy, & McCarthy, 2010) (Rousseau, 2007).
Chocolate shell
Filling
Cocoa particle Sugar crystal
Figure 3.17: Illustration of what is called the tortuosity of chocolate. Diffusing fat from the filling cannot
migrate in a straight path due to solid particles that force the fat to move around them.
3.4.6 Suggested mechanisms for migration

The mechanism for migration has been debated mainly between diffusion and capillary
forces, which are explained further in the following section.
3.4.6.3 Diffusion
Diffusion is the process by which matter is transported from one part of a system to another
by random molecular motion, and takes place when there is a concentrations gradient between
two points in the system. The molecules then migrate from the point with the higher
concentration to the point with the lower concentration until the system reaches
19
thermodynamic equilibrium (Ghosh, Ziegler, & Anantheswaran, 2002). In chocolate pralines
with e.g. water based filling there is a concentration gradient of moisture between the filling
and the shell that makes moisture diffuse from the centre through the shell.
Mathematically diffusion can be expressed with Ficks first and second law. Ficks first law
states that in a continuous medium and under the presence of a concentration gradient, the net
migration of solute molecules due to random movement occurs from a region of high
concentration to one of lower concentration (Aguilera, Michel, & Mayor, 2004).
For diffusion in only z direction, Ficks first law can be written (Welty, Wicks, Wilson, &
Rorrer, 2001):
, (Equation 1)
Where JA is the molar flux of component A, DAB is the diffusion coefficient for component A
diffusing through component B, c is the concentration of component A and dcA/dz is the
concentration gradient in the z direction.
Furthermore, Ficks second law states that the rate at which this process proceeds at a point in
space, where one solute is present in a large portion of solvent, is proportional to the variation
of the slope of the concentration gradient (Aguilera, Michel, & Mayor, 2004). Unsteady-state
diffusion according to Ficks second law can be written (Welty, Wicks, Wilson, & Rorrer,
2001):
, (Equation 2)
Where cA/ t is the change in concentration of component A with time, DAB is the diffusion
2
coefficient for component A diffusing through component B and cA/ z2 is the slope of the
concentration gradient in the z direction.
3.4.6.4 Capillary forces

Since models based on the so-called Fickian diffusion fail to fully describe the migration of
fat in chocolate, other mechanisms have been suggested, such as capillary pressure
(Galdmez, Szlachetka, Duda, & Ziegler, 2009). Capillary flow is a spontaneous process
driven by an interfacial pressure gradient. It occurs omnipresent in nature in fine pores and
cracks. For instance capillary rise is what brings up water to the surface from the soil, allows
dyeing and wetting of textile fabrics, and is the mechanism in wetting of cookies with hot
coffee or tea that improves flavour release (Aguilera, Michel, & Mayor, 2004). An illustration
of capillary flow is shown in Figure 3.18.
20
Air
Capillary tube
Water
Figure 3.18: Capillary flow. Water is drawn into the tube due to a pressure gradient between the air and water.
The Lucas-Washburn equation is the most common expression for capillary rise and assumes
that the capillary pressure in a cylindrical capillary in contact with an infinite liquid reservoir
is compensated by viscous drag and gravity. The equation can be written:
(Equation 3)
Where h is the distance the fluid is drawn into the capillary, the surface tension of the fluid,
the contact angle between the fluid and the capillary wall, r the radius of the capillary, and
the density respectively the viscosity of the liquid, and g the acceleration due to gravity
(Aguilera, Michel, & Mayor, 2004).
An ideal model for capillary flow has cylindrical uniform pores, infinite liquid reservoir, and
so on. The physical situation within a chocolate is quite different from that within an ideal
model, but the Lucas-Washburn equation has proven to hold for several physical situations
(Aguilera, Michel, & Mayor, 2004).
3.4.6.5 Diffusion versus capillary forces

Molecular diffusion has been a preferred mechanism to explain fat and moisture migration in
chocolate. As mentioned earlier, diffusion of fat is thought to be due to a gradient in TAG
concentration within the chocolate. In plain chocolate that has a homogeneous liquid phase of
cocoa butter, this is less likely to occur (Aguilera, Michel, & Mayor, 2004).Whether the
capillary rise theory is significant or not is debated. A study by Afoakwa et al. (2009) supports
the capillary rise theory. They found that fat migration and bloom occurred faster in plain
dark chocolate samples that had a greater particle size, 50 m compared 18 m, and
hypothesized that larger particles give larger capillaries where fat can migrate faster
(Afoakwa E. O., Paterson, Fowler, & Joselio, 2009). However, recent studies on oil migration
through chocolate using magnetic resonance imaging have shown that diffusion is the
dominating mechanism for fat migration. They conclude that if capillary pressure is involved
it is a minor contributor (Galdmez, Szlachetka, Duda, & Ziegler, 2009).
21
3.5 The impact of praline geometries on the migration rate
According to Ficks law, the thickness of the chocolate layer influences the migration rate of
moisture and fat. The higher amount of filling and the smaller amount of chocolate, the more
intense will the migration be. Furthermore, the geometry of the product will influence the
shell thickness. With a given chocolate content of 50% and a given weight, the coating in a
half bowl shape would be thicker than in a cylindrical form, and a thick quadratic bar is
advantageous compared to a flat, rectangular one (Ziegler, 1997).
3.6 Analysing techniques

Several techniques can be used to analyse chocolate pralines and other food products. In the
following chapter follows a short description over the most important techniques used within
this project.
3.6.1 Temper curves

To be sure that the chocolate have been well tempered and that there are enough of the V
crystals in the chocolate, X-rays can be used to determine the exact type of the present
crystals. This is a very expensive and slow technique, which works in the laboratory but not in
industry. Instead the fact that heat is released when fat crystallizes is used. This effect is
caused by that the energy required to melt the fat, the latent heat, is very much larger than the
energy required to change its temperature by 1C, the specific heat. This is also the case for
chocolate when solidifying it, except that the latent heat is given out when the fat changes
from a liquid to a solid. The latent heat for cocoa butter is 157 J/g, which can be compared to
the specific heat of 2.0 J/g. By measuring a so called cooling or temper curve, it can be
determined if the chocolate have the right amount of crystals present which is often correlated
to the presence of desired V crystals (Becket, 2008).
A temper curve can be done with a tempermeter, which is made up of a metal tube with a cup
in the top to hold the chocolate sample. The temperature is measured by a thermometer probe,
which fits through a cap, which in turn fits tightly on to the cup. The temperature can be
recorded on a chart recorder or on a computer database. The tube fits through a holder in a
thermos flask, where the cup is held above the water in the flask. In the flask a mixture of
water and ice is placed and then the tube is put in position. Tempered chocolate is poured into
the cup and the thermometer cap is pushed on top. Finally a plot is made up of the
temperature in the centre of the sample against time (Becket, 2008).
At first the temperature drops slowly and uniformly. If the chocolate is correctly tempered
there is enough crystal nucleus distributed in the chocolate to make it crystallize quickly.
During the crystallization latent heat is released that will compensate for the cooling and
cause the temperature to remain constant for some time. This gives rise to a plateau in the
temper curve that can be seen in Figure 3.19 (Becket, 2008).
22
Figure 3.19: Temper curves for under tempered, correctly tempered and over tempered chocolate. The red
arrows show the inflection point (Talbot).
If there are not enough crystal nucleuses in the chocolate, the chocolate is said to be under-
tempered, and it will take longer time for the chocolate to crystallize. On the other hand, a
smaller amount of crystal nucleus gives a larger amount of fat to crystallize during the cooling
and further more latent heat to be released. However, due to the lack of crystal nucleus this
will cause a longer initial drop in temperature, but when the fat starts to crystallize so much
latent heat is released that it causes the temperature to rise again. When all latent heat has
come out the temperature will continue to fall (Becket, 2008). During cooling new crystal
nucleus can spontaneously be formed in an unstable polymorph, which can give rise to re-
crystallization problems and a lower quality of the chocolate (Lonchampt & Hartel, 2004).
If almost none, or either far too many, crystal nucleus are present the chocolate is said to be
over tempered and it will not crystallize correctly. With far too many crystal nucleus the latent
heat will already have been removed and mainly specific heat will slowly be released (Becket,
2008). The presence of a high amount of seeds can be due to an excessive tempering time.
With no or very few crystal nucleus present unstable polymorphs can form as with the case
for under tempered chocolate (Lonchampt & Hartel, 2004).
At which temperature that the inflection of the curve occurs is also important since it indicates
what polymorphic form the chocolate has. Higher temperature at the inflection point means
that more stable crystals are present. The inflection point is shown with red arrows in Figure
3.19 (Becket, 2008).
3.6.2 Measurement of water activity

There are different techniques to measure water activity, of which one is by measuring the
dew point by a so called chilled-mirror-technique. The sample is put in a chamber together
with a closed air volume, where it is brought to moisture and temperature equilibrium. A
mirror in the chamber is chilled to the dew point of the surrounding air. The temperature at
which saturation of water is observed and the beginning of condensation, obtained by the
chilled mirror, is related to vapour pressure and further to water activity. Photo detection of
the condensation on the cooled mirror together with a precise measurement of the mirror
temperature gives the relative humidity of the chamber where the food sample is disposed. In
equilibrium the relative humidity of the air is equal to the water activity of the sample. This is
23
a rapid and precise method but that in reality only accounts for the water activity of the
surface of the sample (Decagon Devices Inc., 2000) (Mathlouthi, 2001).
3.6.3 Digital colour imaging system

Digital colour imaging system is a computer controlled digital camera that can be used to
measure colour and capture high quality repeatable images. The camera can obtain reflectance
or colorimetric data from samples also with a very small area or a curved surface. Besides
colour information it can capture the total appearance of an object including surface details,
such as gloss, texture and surface characteristics. Captured images can for instance be
coloured to specific standard shades to simulate final products, and can also be used for
quality control etc. (DigiEye Plc., 2007). In this work it is used to analyse crack formation in
the pralines.
The digital colour imaging system consists of an illumination cabinet that uniformly
illuminates objects for capturing repeatable images, a digital camera that allows previewing
and capturing of the image of the sample and enables targeting of selected areas, and a
desktop personal computer running a software system with a colour monitor (DigiEye Plc.,
2007).
Other methods for visual assessment exist. An advantage with the digital colour imaging
system is that it is a non-destructive method to analyse visual quality characteristics of an
object. It is cost-effective, quick, permits a permanent record for later examination and
comparison, and the images are captured under controlled and reproducible conditions. The
computer vision system was developed under the 1960s and became popular within the food
industry during the 1990s. Earlier, it has for example been used in meat quality evaluation
(Tan, 2004), to assess oriental noodles appearance and colour (Hatcher, Symons, &
Manivannan, 2004), to determine bubble size distribution in beer (Heptworth, Hammond, &
Varley, 2004), to inspect pizza topping percentage and distribution (Sun, 2000), and to
evaluate fat bloom in chocolate (Chan Sui Ko, 2010).
24
4 Material and methods
In this chapter, is presented a description of the materials and methods used in the project.
4.1 Raw materials

In this section, is presented a description of the raw materials used in the project.
4.1.1 Chocolate
Dark chocolate was received from Aarhus/Karlshamn A/S. It was stored at 15 C and 30%
RH. Before usage it was heated in a heating oven (Heraeus Kendro Laboratory Products) at
50 C and poured into the tempering machine. The recipe for the chocolate is shown in Table
4.1.
Ingredients Content (%)

Cocoa mass 40
Cocoa butter 10
Sugar 50
Table 4.1: Recipe of dark chocolate from Aarhus/Karlshamn A/S.
4.1.2 Filling thickeners

GENUGEL carrageenan type CJ, KELTROL TF Food grade xanthan gum from CP
Kelco, and gelatin were stored at 20C before they were used as thickeners in the model
fillings.
4.1.3 Humectants
85% glycerol from Apoteket, 95% glycerol and 99.3 % glycerol from VWR, among with
sugar from Danisco, calcium chloride hexa hydrate and citric acid were stored at 20C before
they were used as humectants in the model fillings.
4.1.4 Antimicrobial inhibitor

NATAMAX SF Natural Antimicrobial from Danisco was used in the model fillings to prevent
mould growth in the pralines.
4.1.5 Ultrapure water

Ultrapure water that was used in the model fillings was produced with a NANOpure
ultrapure water system.
4.2 Methodology and equipment

In this section, is presented the methodology and equipments used during the project.
4.2.1 Preparation of model filling

Xanthan, carrageenan and gelatin were tested as thickeners in the filling to achieve a suitable
viscosity, in combination with different humectants to lower the water activity. The
humectants tested were glucose, citric acid, salt and glycerol (85%, 95% and 99.3%).
25
4.2.1.1 Model filling recipes
A model filling as thick as possible but that still was fluid enough to be pumped into the
praline shells was desired, in combination with different water activities. Several
combinations of thickeners and humectants were tested. An example of some combinations is
presented in Table 4.2. All combinations are presented in Appendix 1.
Table 4.2: An example of tested amounts (w%) of thickeners with 30w% glycerol (85%).
Thickener w% Humectant w%
Xanthan 5 Glycerol(85%) 30
Carrageenan 4 Glycerol(85%) 30
Carrageenan 3,5 Glycerol(85%) 30
Gelatin 3 Glycerol(85%) 30
4.2.1.2 Xanthan and carrageenan filling

The xanthan and carrageenan fillings were made according to the following scheme:
1. All the ingredients (humectant, thickener, ultrapure water) were measured in weight
percent.
2. The humectant and the thickener were mixed together.
3. The mix was placed on a RCT basic IKA LABORTECHNIK (Tamro MED LAB)
magnet stirrer and heated to 50C.
4. Water was added during mixing.
4.2.1.3 Gelatin filling

The gelatin filling was prepared according to the following scheme:
1. All the ingredients (humectant, gelatin, Natamax and ultrapure water) were measured
in weight percent.
2. The ingredients were mixed and the gelatin was let to swell for 10 min.
3. The mix was heated to 70C in a water bath during stirring on a magnet stirrer of type
RCT basic IKA LABORTECHNIK (Tamro MED LAB).
4. The mix was left to cool to 30C in a second water bath on a magnet stirrer of type
Heidolph MR 3003 control.
4.2.2 Praline production
4.2.2.1 Tempering of chocolate

The chocolate was tempered in a tempering machine of model AMK 10 (Aastedt Microverk
A/S, Farum, Denmark) (Figure 4.1) and controlled with a MultiTherm Tempermeter (Bhler
AG, Uzwil, Switzerland) (Figure 4.2).
26
(a)
(c)
(b)
Figure 4.1: Photo of the tempering machine of model AMK 10, with (a) a big holder for the chocolate, three
cooling zones and a swinging arm where the chocolate pours out, (b) monitors showing the chocolate and water
temperatures in the different cooling zones, and (c) a cooling system.
The cooling water temperatures were set to 26.0C, 14.7C and 33.0C for zone 1, 2 and 3
respectively. The Bhler tempermeter measured the temper curve of the sample, which was
controlled by the slope of the curve and the TI-value (Temperature Index). According to
Bhler the chocolate was considered well tempered with a slope as close to zero as possible
and a TI-value ~5. A lower TI-value indicated that the chocolate was under tempered and a
higher TI value indicated that the chocolate was over tempered (Bhler group).
Figure 4.2: Photo of the MultiTherm Tempermeter, monitoring a temper curve. The slope of the temper curve
should be as close to zero as possible and the TI-value should be ~5 for the chocolate to be well tempered
(Bhler group).
27
4.2.2.2 Geometries
Pralines with two different geometries, squares and rounds (Figure 4.3), were produced and
analysed at SIK.
Figure 4.3: Illustration of the two praline geometries tested, square and round shaped.
4.2.2.3 Moulds
Plastic moulds with round and square shapes were kindly received from Aarhus/Karlshamn
A/S and were used when moulding the pralines (Figure 4.4).
Figure 4.4: Photo of the round and square shaped plastic moulds from Aarhus/Karlshamn A/S.
4.2.2.4 Preparation of pralines

Pralines were manufactured manually at SIK according to the following scheme:
1. Tempered chocolate was filled into a plastic mould (Figure 4.5).

2. Excess chocolate was scraped off with a slick (Figure 4.5).
3. The filled mould was shaken on a Heidolph shaking table of type DSG302 for 30
seconds to reduce the air bubbles in the chocolate (Figure 4.5).
4. The filled mould was turned upside down and was slightly shaken for 1 minute and 20
seconds by hand to let chocolate pour out (Figure 4.6).
5. Excess chocolate were again scraped off with a slick (Figure 4.6).
6. The chocolate shells were let to cool next to a fan for 15 minutes according to the
method developed by Martina Hellmansson (Figure 4.6).
7. Filling was pumped into each shell with a dispenser (3 ml in the round shaped and 5
ml in the square shaped pralines) (Figure 4.7).
a. Xanthan and carrageenan fillings was spread out and let to cool/stabilize for 10
minutes next to the fan.
b. Gelatin filling was left to cool for an hour in a refrigerator at 5-8C (Figure 4.7).
8. The top of the shells were heated with a Bosch PHG600-2 CE thermo-electronic
heating gun for 10 seconds each to have a better attachment of the base (Figure 4.7).
9. More chocolate was added to form the base of the praline (Figure 4.8).
10. Excess chocolate was scraped off with a slick (Figure 4.8).
28
11. The pralines were let to dry for 15 minutes next to the fan (Figure 4.8).
12. The pralines were carefully knocked out of the moulds (Figure 4.9).
13. The pralines were stored in a constant room with 20 C and 50 % RH for five weeks
(Figure 4.10).
Four different parameters were tested during production, which were:
Removal of air bubbles (shaking time): Different shaking times at the shaking table
were tested to have a good removal of air bubbles (step 3).
Shell thickness (pouring time): Different pouring times were tested to achieve an
optimal shell thickness (step 4).
Amount of filling: Different amounts of filling were tested (step 7).
Heating time: Different heating times with the thermo-electronic heating gun were
tested to have a proper addition of the base (step 10).
Once the process had been optimized with respect to the four parameters listed above, 240
pralines were produced, with 40 pralines of each filling and geometry, and in addition ten
empty shells of each geometry were produced as reference.
Figure 4.5: (1) Photo of chocolate being poured into a plastic mould, (2) a mould filled with chocolate where
excess chocolate just have been scraped off, (3) and a chocolate filled mould being shaken on a shaking table to
reduce the air bubbles in the chocolate.
Figure 4.6: (1) Photo of chocolate shells being formed by turning the chocolate filled mould upside down and
letting the centre chocolate pour out, leaving only a thin layer on the sides of the mould. (2) Excess chocolate
have been scraped off, (3) before the shells are left to cool in front of a table fan.
29
Figure 4.7: (1) Photo of the dispenser when filling the chocolate shells with model filling, (2) shells filled with
model filling left to stabilise in a refrigerator, (3) and heating of the chocolate shells with a Bosch PHG600-2 CE
thermo-electronic heating gun before adding the base.
Figure 4.7: (1) Photo of chocolate being added to the filled shells to form the base, (2) excess chocolate having
been scraped off, (3) before letting the pralines cool in front of the fan.
Figure 4.8: Photo of ready produced pralines, square and round shaped.
4.2.3 Characterisation techniques during storage

The pralines were stored in a constant chamber with 20C and 50 % RH for four weeks
(Figure 4.10 and 4.11), during which physical appearance, weight, and water activity in the
shell were analysed. All samples were analysed every 7th day with respect to weight changes
and physical appearance. However, additional measuring occasions were added based on
when the different samples were believed to crack, at which also water activity in the shell
was measure. The pralines with model filling aw 0.98 were analysed every day for five days,
the pralines with model filling aw 0.85 were analysed every second day for nine days, the
30
pralines with model filling aw 0.75 were analysed every fourth day for sixteen days, pralines
with model filling aw 0.65 were analysed once a week for four weeks, and the empty shells
used as reference were analysed every second week for four weeks, at only three measuring
occasions. A full analysing scheme is shown in Table 4.3.
Table 4.3: Scheme of the analysing measuring occasions for all samples. D = Digital colour imaging (DigiEye),
W = Weight, aw = water activity.
aw 0.99 0.86 0.78 0.67 ref

Day D W aw D W aw D W aw D W aw D W aw
0 x x x x x x x x x x x x x x x
1 x x x
2 x x x x x x
3 x x x
4 x x x x x x
5 x x x
6 x x x
7 x x x x x x x x x x x x
8 x x x x x x
9
10
11
12 x x x
13
15 x x x
16
17
18
19
20
22
23
24
25
26
27
31
Figure 4.10: Photo of pralines being stored in a constant chamber at 20C and 50 % RH.
Figure 4.11: Photo of the constant chamber with 20C and 50 % RH where the pralines were stored.
4.2.3.1 Digital colour imaging

The pralines were photographed and analysed by a DigiEye camera of version 2.43 (Figure
4.12). For each filling and geometry the same 10 replicates as for weighing were
photographed and analysed. For the reference samples the same four replicates from each
geometry were photographed and analysed at each measuring occasion and every 7th day.
32
Figure 4.12: Photo of the DigiEye camera of version 2.43 when taking pictures of some praline samples.
4.2.3.1.1 Shell thickness

At the end of the storage period, after that all samples had been analysed, each praline was cut
in two and photographed so that the shell thickness could be measured. The shell thickness
was measured with Analysis Five program (Soft Imaging System GmbH) by measuring the
shell thickness of the top, side and base of each praline (Figure 4.13). Four measurements
were made on each side, giving in total 16 length measurements per praline.
Figure 4.13: Intersection of a square and round praline where the shell thickness has been measured with
Analysis Five program (Soft Imaging System GmbH).
4.2.3.2 Weighing of pralines

The pralines were weighed with a scale of type Mettler Toledo PR8002 (Figure 4.14). For
each model filling and geometry the same 10 pralines were measured at five measuring
occasions and every 7th day. For the reference samples four praline shells of each geometry
were measured every 7th day.
Figure 4.14: Photo of the scale of type Mettler Toledo PR8002 measuring the weight of a praline.
33
4.2.3.3 Water activity measurements
Decagons AquaLab Series 3 (Figure 4.15) was used for water activity measurements. The
water activity of the model fillings was measured before they were poured into the pralines
and the water activity of the chocolate shell was measured during storage. To measure the
water activity of the filling it was spread out in the sample cup so it covered the bottom. The
chocolate shell was first shredded of the praline with a shredder before it was put in the
sample cup (Figure 4.16). The sample was then measured by putting the cup in the chamber.
For each filling and geometry five replicates were made for each measuring occasion. For the
references of empty shells two replicates were made for each geometry at every measuring
occasion.
Figure 4.15: Photo of Decagons AquaLab Series 3.
Figure 4.16: Photo of chocolate being shredded of a praline, filled in a sample cup and being measured in
Decagons AquaLab Series 3.
34
5 Results
5.1 Filling thickeners
Desired properties for the model filling was for it to be as easy as possible to pour into the
chocolate shells but at the same time stiff to being able to resist the pressure that was applied
on it when adding the base.
Xanthan and carrageenan gave quite similar results but the problem with using them as
thickeners was that it was difficult to achieve a stiff gel that was still fluid enough to easily
being poured into the shells. This gave an uneven bottom, thicker on one side and thinner on
the other, forming a sort of slope. This can be seen in Figure 5.1.
Figure 5.1: Intersection of a praline with carrageenan filling (3.5 w%). The filling is not rigid enough to
withstand the pressure applied on it when adding the base why a chocolate slope is formed.
Gelatin turned out to be an outstanding thickener compared to xanthan and carrageenan. The
benefit with gelatin was that it was liquid when heated so it could easily be pumped into the
shells with the dispenser, and it gelatinised when cooled down and was then able to withstand
the pressure performed on it when adding the base. The properties of gelatin were quite
similar to that of a nougat filling. An intersection of a praline with a gelatin filling compared
to nougat filling can be seen in Figure 5.2.
Figure 5.2: Intersection of a praline with gelatin filling (3 w%) to the left and nougat filling to the right. Since
both gelatin and nougat are fluid when heated they are easy to pour into the shells, giving a smooth surface that
becomes stiff enough when cooled to withstand the pressure performed on it when adding the base.
5.2 Humectants
Different humectants were tested to lower the water activity (aw) of the model filling.
Glycerol showed to be the absolute best humectant compared to sugar, salt and citric acid.
Salt was very difficult to dissolve even in small amounts. Sugar was suitable for higher water
activities but was difficult to dissolve in large amounts, thus it was not possible to reach lower
water activities. Citric acid was tested in combination with sugar but did not improve the
result why it was discarded.
35
5.3 Preparation of model filling
Glycerol (99.3%) turned out to be the best humectant and gelatin (3 w%) the best thickener
for the model filling. Amount of glycerol, gelatin and Natamax used for the four fillings with
different aw are shown in Table 5.2. For each aw 400 ml model filling was prepared.
Table 5.2: Amount (w%) of 99.3 % glycerol, gelatin and Natamax added in the final model fillings.
Wateractivity(aw) Glycerol(w%) Gelatin(w%) Natamax(w%)

0.99 10 3 0.04
0.86 40 3 0.04
0.78 50 3 0.04
0.67 62 3 0.04
5.4 Preparation of pralines

During the praline preparation four parameters were tested, with the following result.
5.4.1 Removal of air bubbles (shaking time)

During the praline preparation different shaking times (step 3 in section 4.2.3.4) were tested to
have a good removal of air bubbles. It was found that shaking time for 30 seconds gave an
acceptable removal of air bubbles.
5.4.2 Shell thickness (pouring time)

The shell thickness varied depending on the time the chocolate was let to pour out from the
filled moulds (step 4 in section 4.2.3.4). The amount of chocolate that poured out, and thus the
shell thickness, varied also significantly with how vigorously the mould was shaken. When
producing the pralines by hand this parameter was difficult to measure but special attention
was paid to perform this in the same manner each time. The pouring time was measured and
kept constant. The desired properties of the shell were to have it as thin and even as possible,
and still being able to remove it from the mould and without the filling leaking out. A pouring
time of 1 minute 20 seconds gave a good shell thickness for both square and round pralines.
5.4.3 Amount of filling

Different amounts of filling were tested in the pralines (step 7 in section 4.2.3.4). It was found
that 5 ml and 3 ml were suitable amounts in the square and round shaped pralines
respectively.
5.4.4 Heating time

Different heating times with the thermo-electronic heating gun (step 8 in section 4.2.3.4) were
tested. It was found that 10 seconds was enough to get the base to attach well to the shell.
5.5 Results from characterisation of the pralines during storage

In this section, the results from characterisation of the pralines during storage are presented.
The pralines were characterised by three methods; digital colour imaging system, weight
measurements and water activity measurements. The results from these three methods are
presented both with aspect to the different model fillings and to the two geometries.
36
Furthermore, the shell thickness of the pralines was measured to investigate its impact on the
results.
5.5.1 Digital colour imaging

Crack formation in the pralines were registered and analysed with a digital colour imaging
system. In this section, the results are presented according to the different model fillings
followed by the two geometries. Furthermore, the results of a deeper analysis of from where
the cracks originated are presented.
5.5.1.1 Tendency to crack with different model filling

A suggestion for crack formation in pralines is that water from the centre migrates through the
chocolate shell due to a concentration difference. The pralines with higher aw in the model
filling is therefore expected to experience a higher level of migration due to a higher
concentration difference and hence an increased tendency to crack.
In Figure 5.4 an example of cracks in a square and round praline is shown. As expected, the
pralines with model filling aw 0.99 cracked first and to the highest percent both for square and
round pralines. The results for model filling aw 0.86 and 0.78 were quite similar. Square
pralines with model filling aw 0.86 cracked a few days earlier than square pralines with model
filling aw 0.78. Round pralines with model filling aw 0.86 cracked to a slightly higher
percentage compared to round pralines with model filling aw 0.78. Pralines with model filling
aw 0.67 had no (square) or very few (round) cracks and the reference of empty shells had no
cracks at all. The results are presented in Figure 5.5 and 5.6.
Figure 5.4: Example of a cracked square and round praline with model filling aw 0.99.
37
Figure 5.5: Total cracked square pralines (%) for each model filling and reference. Pralines with aw 0.99 started
to crack after 2 days and 100% had cracked after 4 days. Pralines with aw 0.86 started to crack after 5 days and
after 28 days 40% had cracked. Pralines with aw 0.78 start to crack after 21 days and after 28 days 40% have
cracked. The pralines with aw 0.67 and the reference of empty shells had not cracked at all after 28 days.
Figure 5.6: Total cracked round pralines (%) for each model filling and reference. Pralines with aw 0.99 started
to crack after 1 day and 100% of the pralines had cracked after 21 days. Pralines with aw 0.86 started to crack
after 2 days and after 28 days 70% had cracked. Pralines with aw 0.78 had already cracked to 10% from day 0
due to inconsistency during production but had cracked further to 30% after 4 days and after 28 days 60% had
cracked. The pralines with aw 0.67 started to crack after 7 days and after 28 days 20% had cracked. The reference
of empty shells had not cracked at all after 28 days.
38
5.5.1.2 Geometries
Round pralines cracked first for all model fillings and to highest percent except for model
filling aw 0.99 where both geometries cracked to 100%. In Figure 5.7-10, the total percent of
cracked pralines for the two geometries is presented.
Figure 5.7: Total amount cracked pralines (%) for square and round pralines with model filling aw 0.99. The
square pralines started to crack after 2 days and had cracked to 100% after 4 days, while the round pralines
started to crack after 1 day and had cracked to 100% after 14 days.
started to crack after 2 days and had cracked to 70% after 28 days.
39
started to crack after 4 days and had cracked to 60% after 28 days.
square pralines did not crack at all, while the round pralines started to crack after 7 days and had cracked to 20%
after 28 days.
40
In Figure 5.11 and 5.12, the crack formation in a square and round praline with model filling
aw 0.99 can be followed from day 0 to day 28. The square praline cracked after 2 days and the
round praline after 1 day. The whitish on the square praline could be sugar from the chocolate
or glycerol from the filling that have crystallised, which then is called sugar bloom.
Alternatively it could be fat bloom if the chocolate was not optimally tempered. This was not
studied further since it was not the aim with the project.
Day 0 Day 1 Day 2
Day 3 Day 4 Day 7
Day 14 Day 21 Day 28
Figure 5.11: Square praline with model filling aw 0.99 after 0 days, 1 day, 2, days, 3 days, 4 days, 7 days, 14
days, 21 days and 28 days. The praline cracked after 2 days (marked red) in the corner of the top (red arrow).
41
Day 0 Day 1 Day 2
Day 3 Day 4 Day 7
Day 14 Day 21 Day 28
Figure 5.12: Round praline with model filling aw 0.99 after 0 days, 1 day, 2, days, 3 days, 4 days, 7 days, 14
days, 21 days and 28 days. The praline cracked after 1 day (marked red) in the middle of the top by that the top
imploded.
42
5.5.1.3 Origin of the cracks
To better understand the rise and reason for cracks, all pralines were analysed according to
where the crack originated. The pralines were categorised in three zones; top, side and base,
and further in corner/edge and middle (top), sealing, middle and corner (side) and sealing and
middle (base) (Figure 5.13a and 5.14a).
Square pralines cracked mainly at the corners at the top/side of the praline and in the middle
of the top. Round pralines cracked mainly at the top by that the top imploded and at the
sealing between the shell and base. In Figure 5.13b and 5.14b, is presented the origins of
cracks on square and round pralines with model filling aw 0.99 that have been stored for 28
days. Figure 5.15 shows an illustration of where on the pralines most samples cracked.
Top/middle Base/middle
Side/corner Side/middle (a)
Side/sealing
Top/corner Base/sealing
(b)
Figure 5.13: (a) From the left: top, side and base of a square praline. Coloured areas indicate the categorisation
positions for cracks on the praline while red arrows point out examples of actual cracks. For square pralines the
categorisation positions were corner and middle at the top; corner, middle and sealing at the side; middle and
sealing at the base. (b) Origin of the cracks (%) on square pralines that have been stored for 28 days, out of a
total number of 18 cracked pralines. Note that some pralines cracked on several places.
43
Top/middle Base/sealing
(a)
Side/edge Side/middle
Top/edge Side/sealing Base/middle

(b)
Figure 5.14: (a) From the left: top, side and base of a round praline. Coloured areas indicate the categorisation
positions for cracks on the praline while red arrows point out examples of actual cracks. For round pralines the
categorisation positions were edge and middle at the top; edge, middle and sealing at the side; middle and sealing
at the base. (b) Origin of the cracks (%) on round pralines that have been stored for 28 days, out of a total
number of 25 cracked pralines. Note that some pralines cracked on several places.
Figure 5.15: Illustration of a square and round praline. The red marks show where on the praline most of the
pralines cracked. Square pralines cracked mainly at the middle of the top and at the corners on the top and sides.
Round pralines cracked mainly in the middle and edge of the top and in the sealing between the shell and base.
5.5.1.4 Shell thickness measurements

It was difficult to achieve an even shell thickness all over the praline and for all samples. To
investigate if the thickness of the shell had affected the results, the top, side and base was
measured on each sample. The results showed that for square pralines, the base became much
44
thicker and had larger variation than the top and sides for pralines with all model fillings. The
shell thickness for top and sides was otherwise quite similar between the square pralines. For
the round pralines the shell thickness differed more between the pralines with different model
fillings. The round pralines with model filling aw 0.67 had much thicker top and sides, but
thinner base, than the pralines with other model fillings. Mean shell thickness and standard
deviation for all pralines are shown in Figure 5.17 and 5.18.
Figure 5.17: Mean shell thickness with standard deviation for square pralines with different model filling.
Figure 5.18: Mean shell thickness with standard deviation for round pralines with different model filling.
To further investigate the impact of shell thickness on crack formation, samples that differed
from the other pralines, either by cracking first or resisting from cracks, was looked at
individually. No clear correlation was seen between cracking and shell thickness. Neither did
the pralines that cracked first have particularly thinner shell thickness, nor did the pralines that
45
did not crack have particularly thicker shell thickness. For example, for square pralines with
aw 0.99 sample 1, 5 and 6 cracked first (top and sides), but as can be seen in Figure 5.19 there
is no large difference in shell thickness for the top and sides among the samples. The base is
on the other hand significantly thicker for all samples.
Figure 5.19: Shell thickness (mm) for square pralines with model filling aw 0.99. All samples had cracked after
28 days. Sample 1, 5 and 6 cracked first.
Furthermore, for square pralines with aw 0.86 samples 1, 4, 5 and 10 had cracked after 28
days, and sample 2, 3, 6, 7, 8 and 9 had not. As can be seen in Figure 5.20 there is no large
difference in shell thickness for the top and sides where the pralines cracked. A slightly
thicker base can be seen for all samples that cracked compared to the ones that did not, except
for sample 7 that did not crack but still had a thicker base. All results are presented in
Appendix 2.
46
Figure 5.20: Shell thickness (mm) for square pralines with model filling aw 0.99. After 28 days sample 1, 4, 5
and 10 had cracked, while sample 2, 3, 6, 7, 8 and 9 had not.
Besides shell thickness, some other parameters differed between the two geometries. The
square pralines were somewhat larger than the round ones. This means that the square pralines
both weighed more and had a higher filling content. Based on the shell thickness
measurements, the mean chocolate volume could be calculated and showed that square
pralines also had a larger chocolate volume. Furthermore, square pralines had a larger inner
chocolate surface that comes in contact with the filling. What is interesting to look at is the
ratio of contact surface to chocolate volume, and the amount of filling per amount of
chocolate. Even though these ratios did not show any pronounced difference, both ratios were
slightly larger for the round pralines. All parameters are presented in Table 5.3.
Table 5.3: Parameters that varied for square and round pralines. The numbers are mean values with standard
deviation.
Square pralines Round pralines
Weight (g) 9.14 (0.62) 6.40 (0.22)
Filling volume (Vf) (mm3) 3668 (603) 2495 (652)
Chocolate volume (Vc) (mm3) 4938 (603) 3193 (652)
Contact surface (Ac) (mm2) 1408 (144) 1052 (201)
Ac/Vc 0.29 (0.07) 0.35 (0.12)
Vf/Vc 0.77 (0.22) 0.85 (0.33)
47
5.5.2 Weight differences
In this section, results from the weight measurements are presented. Moisture in the filling
was suggested to migrate from the centre through the shell, hence a weight decrease in the
pralines was assumed to occur due to loss of moisture. A crack in the shell would lead to that
the filling evaporated directly to the air and the migration can be assumed to decrease due to
that the aw difference decreases as the filling dries out. Moisture loss due to evaporation from
a crack was further assumed to lead to a much faster weight loss than moisture loss due to
migration through the shell.
5.5.2.1 Change in weight as a function of model filling

In Figure 5.21 and 5.22 the mean weight change for square and round pralines is presented
with respect to each model filling. The weight change in percentage was calculated for each
praline compared to its original weight and then the mean value and standard deviation was
calculated for each model filling and praline geometry. Furthermore, the weight change for
pralines that had not cracked was calculated to investigate the migration phenomenon. The
graphs show measurements made after 0 days, 7 days, 14 days, 21 days and 28 days.
Figure 5.21: Weight change (%) for square pralines for each model filling and reference. The measurements are
made after 0 days, 7 days, 14 days, 21 days and 28 days.
48
Figure 5.22: Weight change (%) for round pralines for each model filling and reference. The measurements are
made after 0 days, 7 days, 14 days, 21 days and 28 days.
In both Figure 5.23 and 5.24 it can be seen that pralines with model filling aw 0.99 have a
major weight decrease compared to the pralines with model fillings with lower water
activities. This correlates well to that pralines with model filling aw 0.99 were the pralines that
cracked first. A smaller weight change was seen for pralines with model filling aw 0.86 and
aw0.78, which cracked somewhat later than model filling aw 0.99 and not to the same extent.
Pralines with model filling aw 0.67 did not crack at all (square) or to a low extent (round) and
have a very low weight difference. The same was true for the reference that did not contain
any model filling.
It was expected that pralines that had not cracked should show a small weight decrease due to
loss of moisture that had migrated from the centre through the shell and evaporated from the
surface. Pralines with model filling aw 0.86, which had the highest aw of the pralines with
samples that did not crack, were mainly expected to show this. However, neither square nor
round pralines that had not cracked showed any major weight difference that confirms this.
The weight change for square and round pralines with model filling aw 0.86 that had not
cracked is shown in Figure 5.23. Results for all model fillings are presented in Appendix 3.
49
Figure 5.23: Weight change (%) for square and round pralines with model filling aw 0.86 that had not cracked.
5.5.2.2 Geometries
Without adjustment for pralines that had cracked, round pralines had a slightly higher weight
decrease, although not significant, compared to square pralines. That correlates to that the
round pralines cracked first. Since no significant weight difference could be seen for pralines
that had not cracked, no comparison between the geometries can be made concerning the
migration phenomenon.
5.5.3 Water activity (aw) measurements

The water activity (aw) in the chocolate shell was measured at five different occasions
(between day 0 and day 28) for the various model fillings depending on when they were
assumed to crack. Moisture from the filling was assumed to migrate into the shell during
storage, why the water activity in the shell was expected to increase with time. Since the
water activity was measured at different storage times, the result for each model filling is
presented separately.
The water activity did not increase with time as expected for any model filling. Instead it both
increased and decreased during the storage period. The water activity for pralines with model
filling aw 0.78 is shown in Figure 5.24. The results for the other model fillings are quite
similar and are presented in Appendix 4.
50
Figure 5.24: Mean water activity in the chocolate shell of pralines with model filling aw 0.78 at various
measuring occasions.
A reason for that the results was not as expected could be due to the methodology used to
measure water activity. To exclude faults from the water activity measuring equipment it was
controlled regularly during the measurements with specified salt solutions to see that it
measured correctly. However, it was quite difficult to shred off chocolate from the pralines
without breaking them. Therefore, the chocolate used for analysis was mainly chocolate from
the outermost layer of the shell. During the storage time it is quite probable that the outermost
layer was more affected by the relative humidity (RH) in the surrounding air than the filling.
The chocolate pralines were stored in a constant chamber at 50% RH and 20C, where the
temperature and RH were controlled with a temperature and RH logger. Both temperature and
RH were stable during the testing period, why the water activity then ought to have stayed
constant. It is however possible that the shredding, which resulted in a larger surface area of
the chocolate, enabled water to quickly evaporate into the surrounding before it could be
placed in the Decagons AquaLab for aw measurements.
51
6 Discussion
In this chapter the results from characterisation of the pralines during storage are discussed.
6.1 Visual characterization of cracking

6.1.1 The impact of model filling on cracking
The pralines with model filling with the highest aw was expected to crack first since it contain
more free water that can migrate into the shell, and a suggested mechanism for cracking is
migration of moisture. This is confirmed by the results, which showed that pralines with
model filling aw 0.99 cracked first and to the highest percent for both square and round
pralines. Since, there was almost no difference in cracking behaviour between the pralines
with model filling aw 0.86 or 0.78, although much lower than aw 0.99, one could hypothesise
that it should be possible to find a threshold value for water activity in the filling where
cracking is significantly reduced. Further studies on model fillings with different aw would be
interesting to see if such a threshold value could be found.
6.1.2 The impact of praline geometry on cracking

When looking on the impact of praline geometry on cracking, it was found that the round
pralines cracked first and to the highest percentage for all model fillings, besides for model
filling aw 0.99 where both geometries cracked to 100%. This is a very interesting result which
can not solely be explained by shell thickness. One explanation could be that the round
pralines showed a trend of a somewhat larger inner shell surface to chocolate volume ratio
compared to the square pralines. This will expose the moisture in the filling to a larger surface
area per amount of chocolate to initiate migration. Furthermore, the filling volume compared
to the chocolate volume was slightly larger, although not pronounced, for round compared to
square pralines. The higher amount of filling and the smaller amount of chocolate, the more
intense will the migration be. Another reason could be due to production defects. The round
pralines had a much higher percentage that cracked in the sealing between the shell and the
base compared to the square pralines. This could be due to that the shell was not heated
properly before adding the base or that some filling had come in between so that the base did
not attach as it should. Worth to note though is that a high percent of the round pralines
cracked at the top, which cannot be explained by production defects.
6.1.3 Origin of the cracks

On square pralines most of the cracks originated from the middle of the top or from the
corners at the top and sides. Corners were expected to be a weak point due to forces induced
by swelling of the shell in different directions. That such a high percentage cracked in the
middle was more surprising. The measurements on the shell thickness did not show any large
differences even though the mean thickness of the top is slightly thinner than the sides and
base for all model fillings.
52
Round pralines mainly cracked in the middle of the top and in the sealing between the shell
and base. At the top the shell cracked due to tensions the top caused by imploding towards the
centre. That a high percentage of the pralines cracked in the sealing was expected since it was
a critical step to properly attach the base to the shell. It was more surprising that only 10% of
the square pralines cracked in the sealing. The sealing was assumed to be a weak point due to
forces induced by swelling of the shell in different directions, similar as in the corners of the
square pralines. The same reason can be used to explain that a relatively high percentage of
the round pralines also cracked at the edge of the top.
6.1.4 The impact of shell thickness on cracking

The shell thickness varied somewhat between the samples, both between different model
fillings and different geometries. Moisture was suggested to migrate through the shell in a
shorter time in a thinner shell than through a thicker. Furthermore, pralines with thinner shells
were assumed to crack first due to that the thinner the shell the less pressure it can withstand
before it cracks. However, when examining the shell thickness on individual pralines no such
correlation could be found.
The base on the shell was generally much thicker for all samples compared with the top and
sides. Due to that the base is added separately to the shell after that the filling is added, it is
not surprising that it had a variable thickness. The fact that the base was much thicker than the
rest of the shell could have influenced that no praline cracked in the middle of the base. For
the top and sides of the shell, where most of the cracks occurred, no major difference in shell
thickness could be observed between square and round pralines.
6.2 Weight measurements

Weight was assumed to decrease with time due to loss of moisture, either by migration of
moisture through the shell, or by evaporation of moisture through a crack. As expected, the
weight measurements that included cracked pralines, showed a higher decrease in weight for
the pralines with model filling aw 0.99 than pralines with model fillings with lower aw. This
correlates well to the cracking ability of the pralines. Those with model filling aw 0.99 cracked
first and to the highest percentage, and hence had a higher evaporation of moisture from the
filling. Furthermore, round pralines showed a somewhat, but not significant, higher weight
loss than square pralines that correlates to that the round pralines cracked first.
For the pralines that had not cracked, there was no weight loss that indicated moisture
migration through the shell. Pralines with model filling aw 0.86 were most likely to show this
due to the highest aw of pralines that had not cracked. However, no significant weight change
could be seen for any of the geometries or model filings. A reason could be that the storage
period for the pralines in this project was too short or that the scale that was used only had a
precision of 0.01 grams.
6.3 Water activity

The water activity in the chocolate shell was assumed to increase with time due to that
moisture from the filling migrates into the shell. However, no such correlation was observed
during the experiment. Instead the water activity both increased and decreased. A reason for
53
this could be due to the methodology that was used to measure water activity. To calibrate the
equipment for aw measurements it was controlled regularly during the chocolate sample
measurements with specified salt solutions. However, it was quite difficult to shred off
chocolate from the pralines without breaking them. Therefore the chocolate used for analysis
was mainly chocolate from the outermost layer of the shell. During the storage time, it is quite
probable that the outermost layer was more affected by the relative humidity (RH) in the
surrounding air than the filling. The chocolate pralines were stored in a constant chamber at
50% RH and 20C, where the temperature and RH were controlled with a temperature and
RH logger. Both temperature and RH were stable during the testing period, why the water
activity then ought to have stayed constant. A reason that the water activity still varied could
be due to that the samples were left in room temperature for variable time periods before
measurement. It is also possible that the shredding, which resulted in a larger surface area of
the chocolate, enabled water to quickly evaporate into the surrounding before it could be
placed in the Decagons AquaLab for aw measurements. However, further development and
verification of the methodology used to measure water activity in the shell is required in order
to eliminate the possible sources of error mentioned above.
54
7 Conclusions
A major conclusion that can be drawn from this work is that the water activity in water based
filling have a pronounced impact on crack formation in chocolate pralines. To produce
pralines that do not crack, the water activity should therefore be held below 0.7. Square
pralines mainly cracked at the middle of the top or at the corners at the top and sides, while
round pralines mainly cracked in the middle of the top and in the sealing between the shell
and base. For the model fillings with lower aw, round pralines cracked to a higher percent than
square pralines. This could be due to a somewhat larger inner shell surface to chocolate
volume ratio, and hence a larger amount of filling per amount of chocolate, for round
compared to square pralines, but further tests are required to confirm this. Another reason
could be due to production defects. To conclude, this work shows that praline geometry is of
importance for crack formation, but further studies on different praline geometries is needed
to better understand its impact. Shell thickness could not be confirmed to have an impact on
cracking in this work.
DigiEye was a good technique to visually analyse crack formation in pralines, even though
the method could be developed further so that cracks could be measured after e.g. a standard.
To measure moisture migration by weight changes, a more accurate scale is needed than the
one used in this project. Furthermore, the method to measure aw in the chocolate shell needs
to be developed.
8 Further work
As one could hypothesise that it should be possible to find a threshold value for aw in the
filling where cracking is significantly reduced, further studies on model fillings with different
aw would be interesting to see if such a threshold value could be found. It would also be
interesting to do more studies on different praline geometries to further investigate its impact
on crack formation.
In this work the pralines were produced by conventional moulding and traditional tempering.
It would be interesting to try different moulding techniques to see if the processes have an
impact of the final quality of the chocolate pralines. Cold stamp process would be interesting
to use to have a more precise shell thickness in the pralines, where also the shell thickness
could be varied in a controlled way to investigate its influence on cracking. Furthermore, it
would be interesting to compare the quality in chocolate that have been tempered with
conventional tempering towards seeding.
55
References
Adhistory. (n.d.). Retrieved April 26, 2010, from
http://adhistory.wikispaces.com/file/view/Cocoa_tree.jpg/73393531/Cocoa_tree.jpg
Afoakwa, E. O., Paterson, A., & Fowler, M. (2007). Factors influencing rheological and
textural qualities in chocolate - a review. Trends Food Science & Technology , 18, 290-298.
Afoakwa, E. O., Paterson, A., Fowler, M., & Joselio, V. (2009). Fat bloom development and
structure-appearance relationships during storage. Journal of Food Engineering , 91, 571-581.
Aguilera, J. M., Michel, M., & Mayor, G. (2004). Fat Migration in Chocolate: Diffusion or
Capillary Flow in a Particulate Solid? -A Hypothesis Paper. Journal of Food Science , 69 (7),
167-174.
Barron, L. F. (1977). The expansion of wafer and its relation to the cracking of chocolate and
"bakers" chocolate coatings. Journal of Food Technology , 12, 73-84.
Baylis, C. L., MacPhee, S., Robinson, A. J., Griffi, R., Lilley, K., & Betts, R. P. (1994).
Survival of Escherichia coli O157:H7, O111:H - and O26:H11in artificially contaminated
chocolate and confectionery products. International Journal of Food Microbiology (96), 35-
48.
Becket, S. T. (2008). The science of chocolate (2 ed.). Cambridge: The Royal Society of
Chemistry.
Benitez-Sanchez, P. L., Leon-Camacho, M., & Aparicio, R. (2003). A comprehensive study

of hazelnut oil composition with comparisons to other vegetable oils, particularly olive oil.
European Food Research Technology , 218 (1), 13-19.
Bhler Bindler. (n.d.). Retrieved July 13, 2010, from http://www.bindler.com/en/cool.html
Bhler group. (n.d.). Retrieved October 7, 2010, from

http://www.buhlergroup.com/global/en/products/multitherm-t.htm
Chan Sui Ko, A. (2010). Evaluation of Fat Bloom and Cracking in Chocolate by Digital
Colour Imaging System .
Coultate, T. P. (2009). Food The chemistry of its Components (5 ed.). Cambridge: The Royal
Society of Chemistry.
Decagon Devices Inc. (2000, November 7). AquaLab modell Series 3 och 3TE - Manual
Version 1.5.
DigiEye Plc. (2007, January). DigiEye User Guide Version 2.43.
56
Eysturskar, J., Haug, I. J., Ulset, A.-S., & Draget, K. I. (2009). Mechanical properties of
mammalian and fish gelatins based on their weight. Food Hydrocolloids , 23, 23152321.
Galdmez, R. J., Szlachetka, K., Duda, L. J., & Ziegler, G. R. (2009). Oil migration in
chocolate: A case of non-Fickian diffusion. Journal of Food Engineering , 92, 261268.
GAO, Y., CHEN, H.-M., XU, J.-L., CHEN, D.-Y., & YAN, X.-J. (2009). Analysis of Oligo-
kappa-carrageenan by Reversed Phase Ion-pair Ultra Performance Liquid Chromatography
Coupled with Electrospray Ionization-Time of Flight-Mass Spectrometry. CHINESE
JOURNAL OF ANALYTICAL CHEMISTRY , 37 (11), 15901595.
Garca-Ochoaa, F., Santosa, V. E., Casasb, J. A., & Gmez, E. (2000). Xanthan gum:
production, recovery, and properties. Biotechnology Advances , 18, 549579.
Ghosh, V., Ziegler, G. R., & Anantheswaran, R. C. (2002). Fat, Moisture, and Ethanol
Migration through Chocolates and Confectionary Coatings. Critical Reviews in Food Science
and Nutrition , 42 (6), 583-626.
Hatcher, D., Symons, S., & Manivannan, U. (2004). Developments in the use of image
analysis for the oriental noodle appearance and colour. Journal of food engineering (61), 109-
119.
Heptworth, N., Hammond, J., & Varley, J. (2004). Novel applications of computer vision to
determine bubble size distribution in beer. Journal of food engineering (61), 119-125.
Landmann, W., Lovegren, N. V., & Feuge, R. O. (1960). Permeability of Some Fat Products
to moisture. The Journal of the American Oil Chemists' Society , 37, 1-4.
Lee, W. L., McCarthy, M. J., & McCarthy, K. L. (2010). OilMigration in 2-Component.

Journal of Food Science , 75, 83-89.
Lipp, M., & Anklam, E. (1998). Review of cocoa butter and alternative fats for use in
chocolate - Part A. Compositional data. Food Chemistry , 62, 73-97.
Lonchampt, P., & Hartel, R. W. (2004). Fat bloom in chocolate and compound coatings.
European Journal of Lipid Science and Technology , 106, 241-274.
Man, C. M., & Jones, A. A. (1994). Shelf life evaluation of foods. Glasgow: Blackie
Academic and Professional.
Mathlouthi, M. (2001). Water content, water activity, water structure and the stability of
foodstuffs. Food Control (12), 409-417.
Minson, E. (1990). Cookie Confectionary Combinations. The Manufacturing Confectioner ,

121-126.
Rossettini, A., & Pellegrini, R. (2006). CONFECTIONERY: Chocolate Molding Demystified.

Retrieved July 13, 2010, from Asia Food Journal: http://www.asiafoodjournal.com/article-
3719-confectionerychocolatemoldingdemystified-Asia.html
57
Rousseau, D. (2007). The microstructure of chocolate. In D. J. McClements, Understanding
and controlling the microstructure of complex foods (pp. 648-687). Cambridge: Woodhead
Publishing Limited.
Schenk, H., & Peschar, R. (2004). Understanding the structure of chocolate. Radiation
Physics and Chemistry , 71, 829-835.
Sun, D.-W. (2000). Inspecting pizza toping percentage and distribution by a computer vision
method. Journal of food engineering (44), 245-249.
Talbot, G. (u.d.). Chocolate temper.
Tan, J. (2004). Meat quality evaluation by computer vision. Journal of food engineering (61),
27-37.
Welty, J. R., Wicks, C. E., Wilson, R. E., & Rorrer, G. (2001). Fundamentals of Momentum,
Heat, and Mass Transfer (4 uppl.). United States of America: John Wiley and Sons.
Zeng, Y., Braun, P., & Windhab, E. J. (2002). Tempering - Continuous precrystallization of
chocolate with seed cocoa butter crystal suspension.
Ziegler, G. (1997). Fat Migration and Bloom. The Manufacturing Confectioner , 43-44.
58
Appendix 1 Filling recipes
In Appendix 1, the different combinations of thickeners and humectants that were tested are
presented (Table A1.1-A1.3).
Table A1.1: Different combinations that was tested with xanthan as thickener.
Thickener w% Substrate w%
Xanthan 1 Sucrose 10
Xanthan 2,5 Sucrose 20
Xanthan 0,5 Sucrose 65
Xanthan 0,4 Sucrose+Citricacid 80/0.4
Xanthan 1 Calciumchloride 3
Xanthan 2,5 Glycerol(95%) 10
Xanthan 3.5 Glycerol(85%) 30
59
Table A1.2: Different combinations that was tested with carrageenan as thickener.
Table A1.3: Different combinations that was tested with gelatin as thickener.
Gelatin 3 Glycerol(99,3%) 10
60
Appendix 2 - Shell thickness
In Appendix 2, the shell thickness for square pralines with model filling aw 0.78 and aw 0.67,
and round pralines with model filling aw 0.99, aw 0.86, aw 0.78 and aw 0.67 is presented Figure
A2.1-A2.6).
Figure A2.1: Shell thickness (mm) for square pralines with model filling aw 0.78. Sample 2, 5, 6 and 7 had
cracked after 28 days, while sample 1, 3, 4, 8, 9 and 10 had not.
Figure A2.2: Shell thickness (mm) for square pralines with model filling aw 0.67. No samples had cracked after
28 days.
61
Figure A2.3: Shell thickness (mm) for round pralines with model filling aw 0.99. All samples had cracked after
28 days. Sample 1, 3, and 8 cracked first.
Figure A2.4: Shell thickness (mm) for round pralines with model filling aw 0.86. Sample 3, 4, 6, 7, 8, 9 and 10
had cracked after 28 days, while sample1, 2 and 5 had not.
62
Figure A2.5: Shell thickness (mm) for round pralines with model filling aw 0.78. Sample 2, 3, 5, 6, 7 and 9 had
cracked after 28 days, while sample 1, 4, 8 and 10 had not.
Figure A2.6: Shell thickness (mm) for round pralines with model filling aw 0.67. Sample 6 and 7 had cracked
after 28 days, while sample 1, 2, 3, 4, 5, 8, 9 and 10 had not.
63
Appendix 3 - Weight change
In Appendix 3, the weight change for pralines with model filling aw 0.78 and 0.67 that had
not cracked, and reference is presented (Figure A3.1-A3.3).
Figure A3.1: Weight change (%) for square and rounds pralines with model filling aw 0.78 that did not crack.
Figure A3.2: Weight change (%) for square and rounds pralines with model filling aw 0.67 that did not crack.
64
Figure A3.3: Weight change (%) for square and round reference pralines without model filling after different
storage time.
65
Appendix 4 Water activity
In Appendix 4, the water activity measurements for pralines with model filling aw 0.99, 0.86,
0.67 and the reference of empty shells is presented (Figure A4.1-A4.4).
Figure A4.1: Mean water activity of the chocolate shell for pralines with model filling aw 0.99 at various
66
Figure A4.4: Mean water activity of the reference chocolate shells at various measuring occasions.
67

Praline Jurnall

Uploaded by

Copyright:

Available Formats

Praline Jurnall

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Praline Jurnall

Uploaded by

Copyright:

Available Formats

Crack formation in chocolate pralines

Master of Science Thesis in the Master Degree Programme Biotechnology

Department of Chemical and Biological Engineering

Department of Chemical and Biological Engineering

KATARINA S. SLETTENGREN, 2010.

Supervisors: Dr. Llia Ahrn, SIK and Lina Svanberg, SIK

Examiner: Prof. Ulf Svanberg, Chalmers University of Technology

Department of Chemical and Biological Engineering

SIK The Swedish Institute for Food and Biotechnology

Department of Chemical and Biological Engineering

Resultaten visade att fr praliner med aw 0,99 i modellfyllningen uppstod sprickbildning i

3.1 Chocolate pralines composition and origin

Figure 3.1: Image of a praline showing the two major components.

3.1.1 Chocolate shell composition

Ingredient (%) Milk chocolate Dark chocolate White chocolate

3.1.3 Composition and structure of cocoa butter

H O C (CH2)7 CH CH (CH2)7 CH3

3.1.3 Model filling properties

Food Typical water activity

3.1.3.2.1 Xanthan gum

Xanthan is approved as an emulsifier/stabilizer additive in Europe since 1980, as item E415.

Figure 3.6: The repeating units of iota-carrageenan (Coultate, 2009).

Figure 3.7: Carrageenan superjunctions (Coultate, 2009).

3.2 Chocolate processing

Tempering of the chocolate is made in a tempering machine, which is basically a heat

Reheating and homogenisation ~30C 3

3.2.2 Seeding technique

3.2.3.1 Moulds used for chocolate production

3.2.3.2 Conventional moulding

Figure 3.12: Schematic diagram of a conventional moulding process.

3.2.3.3 Cold stamp process

Figure 3.13: Photo of a cold stamp process (Bhler Bindler).

3.2.5.3 One shot technology

Figure 3.14: Image of a one-shot depositor (Bhler Bindler).

3.2.6 Cooling process and storage

3.3 Crack formation in chocolate pralines

3.4 Crack formation due to migration

3.4.5 Effects of migration

3.4.5.3 Moisture migration

Temperature: Temperature has a great influence on moisture migration. When

3.4.5.4 Ethanol migration

Cocoa particle Sugar crystal

3.4.6 Suggested mechanisms for migration

3.4.6.4 Capillary forces

3.4.6.5 Diffusion versus capillary forces

3.6 Analysing techniques

3.6.1 Temper curves

3.6.2 Measurement of water activity

3.6.3 Digital colour imaging system

4.1 Raw materials

Ingredients Content (%)

4.1.2 Filling thickeners

4.1.4 Antimicrobial inhibitor

4.1.5 Ultrapure water

4.2 Methodology and equipment

4.2.1 Preparation of model filling

4.2.1.2 Xanthan and carrageenan filling

4.2.1.3 Gelatin filling

4.2.2 Praline production

4.2.2.1 Tempering of chocolate