Crystallization of cocoabutter
Polymorphism of cocoa butter
In the confectionery industry, crystallization of cocoa butter (alone or in chocolate) is
carried out in two steps:
(1) pre - crystallization, or tempering;
(2) crystallization by cooling (moulding etc.) and in storage.
Since cocoa butter has six crystal modifi cations, the purpose of pre - crystallization is to produce the necessary amount of crystal seeds of the least unstable modifi cation, β (V) (see below). Here, ‘ least unstable ’ means that this modifi cation remains unchanged over several months.
To produce the stable modifi cation β (VI) directly needs sophisticated technology, and such technology is not yet in everyday use, but investigations aimed at solving this problem are in progress. If the correct technology is used, the proportion of the β (V) modifi cation generated by tempering is about 1 – 5%, and the proportion generated by cooling is about 45 – 60%. Crystallization is finished in storage, when the proportion of crystals of the β (V) modifi cation increases to 60 – 80%.
Several authors have discussed the crystalline forms of cocoa butter polymorphs (Duck 1964 , Wille and Lutton 1966 , Huyghebeart and Hendrickx 1971, Lovegren et al. 1976, Dimick and Davis 1986, Jovanovic et al. 1995). For the determination of the melting points of cocoa butter, see IOCCC Analytical Method 4 ( 1961 ). The data on the melting points are rather different for the various crystal modifi cations.
In the confectionery industry, the data provided by Wille and Lutton ( 1966 ) are perhaps the most often used, although in a ‘ mixed ’ form (Greek letter + numbering):
γ = I, 17.3 ° C;
α 1 = II, 23.3 ° C;
α 2 = III, 25.5 ° C;
β ′ = IV, 27.3 ° C;
β (V), 33.8 ° C;
β (VI), 36.3 ° C.
The idea of polymorphic crystalline forms of cocoa butter – as well as of other fats refers not to the external microscopic or macroscopic geometrical appearance of the fat crystals but to the internal structure of the crystals at a molecular level, i.e. the packing of the triglycerides in the molecular crystal lattice.
Figure 10.17 represents the transition β ′ (IV) → β (V) of the crystal modifi cations of cocoa butter. The transition β ′ → β is stimulated by shearing, which is caused by strong mixing of chocolate mass. The characteristic feature of the α (II and III) modifi cations is that the TAGs start to align along the axis of the fatty acids and a chair - type arrangement is formed. (The modifi cation I is designated by γ in the literature.) The β ′ (IV) modification is more compact, its consistency is harder and two chairs form one bond (DCL arrangement). The characteristic feature of the β (V) modifi cation is a compact structure in which three chairs form one bond (TCL arrangement). In the modifi cation β (VI), which is the stable one and evolves over weeks or months, the consistency becomes more compact through the development of a curved tuning - fork
shape of the parts of the TAGs where the oleic acid groups are located.
Tempering of cocoa butter and chocolate mass
The crystallization of cocoa butter or chocolate mass, usually containing about 28 – 38% cocoa butter, means the solidifi cation of the material in such a way that the cocoabutter is crystallized in the form of the β (V) modifi cation. The series of operations starts with tempering, the next operation is the shaping of the cocoa butter or chocolate mass and, finally, the operation of cooling finishes this series.
In the following descriptions, the tempering of cocoabutter and of chocolate mass are presented together. However, there is an important difference: since the contraction of cocoabutter in a chocolate mass is proportional to the volume ratio of cocoabutter, the contraction of a cocoa butter bar is about three times higher than that of a chocolate bar, assuming that they are of the same volume. Therefore, the moulding of cocoa butter bars, which is a relatively rare task, needs more cautious cooling because the bars can crack. The risk of such a phenomenon is less in the case of the moulding of chocolate mass.
From the point of view of the technology, the control of the transitions α (III) → β ′ (IV) → β (V) plays an essential role. This is the tempering operation. At the end of tempering, all of the β ′ (IV) modifi cation has to be melted and, at the same time, tempering must provide a seed concentration of the β (V) modifi cation of 0.1 – 1.15% of the cocoabutter mass according to Loisel et al. ( 1997 ). Jewell ( 1972 ), however, reported that larger amounts of seeds, 2 – 5% of the cocoa butter, were needed for good temper. According to Lonchampt and Hartel ( 2004 ), this difference may be due to differences in seed size, which affects the number of seed crystals. Von Drachenfels et al. ( 1962 ) specifi ed the importance of crystal size. The smaller and more regular the size of the seed crystals, the glossier the chocolate and the greater its bloom resistance. On the other hand, if the crystal size is too large, the crystals tend to recrystallize during storage. It was mentioned above that the transitions from modifi cation I to modifi cation VI are increasingly slow. At the beginning of the cooling of cocoa butter, the γ (I) and α (II and III) modifi cations occur but they change rapidly to the β ′ (IV) and β (V) modifi cations. For details, see Ziegleder ( 1988 ).
Since the crystallization of cocoa butter follows monotropic polymorphism, the direction of the changes is exclusively γ (I) → α (II) → α (III) → β ′ (IV) → β (V) → β (VI). Moreover, under the usual conditions all the modifi cations can be crystallized directly from molten cocoa butter except for β (VI), which crystallizes slowly from the β (V) modifi cation (Fig. 10.18 ).
The stable form β (VI) cannot be produced directly from melted chocolate except by the addition of β (VI) cocoa butter seeds and under very well - controlled conditions (Giddey and Clerc 1961 , van Langevelde et al. 2001 ). It should be emphasized that the target of tempering is to bring about the β (V) modifi cation, which is unstable, although its transition to the stable β (VI) modifi cation is very slow: it needs weeks or months. During these monotropic changes the Gibbs free enthalpy decreases continuously; its minimum is reached in the β (VI) modifi cation.
However, if the tempering results in a majority of crystals of the β ′ (IV) modifi cation, the transition β ′ (IV) → β (V) will take place in the chocolate product within hours or days, and the consequence of such a transition will be the appearance of fat bloom on the surface of the chocolate product. This is a severe quality defect, called blooming . Taking into account all the considerations above, the principle of the tempering process is to produce the β ′ (IV) and β (V) modifi cations, and then to melt the β ′ (IV) modifi cation while the β (V) modifi cation is retained. Although the β (V) modifi cation can be produced directly from a molten chocolate mass, such a direct method cannot exclude the development of crystals of the β ′ (IV) modifi cation. A warming period necessary in the tempering operation which destroys the crystals of the β ′ (IV) modifi cation – this is the way to avoid fat bloom.
Figure 10.19 shows the temperature profi le of a correct tempering operation for chocolate mass, which consists of three steps: two steps of cooling and one step of warming between them. The traditional tempering machine is similarly partitioned in the direction of advance of the chocolate mass. It is evident that a simple conical double - jacketed chocolate tank with a mixer is hardly suitable for performing tempering correctly, because it is diffi cult to carry out the warming phase. Strong mixing of the chocolate mass during tempering promotes the development of crystals of the β (V) modifi cation by the shearing effect. The measurement of tempering, for which the ‘ temperimeter ’ is a practical instrument, provides important technological parameters. This instrument includes a small vessel, which the tempered chocolate mass is poured into. The vessel is placed in an ice – water bath, and the temperature of the chocolate mass is measured as a function of time. The resulting temperature vs time plots are represented in Fig. 10.20 .
the amount of crystals of the β (V) modifi cation is suffi cient, and in the time interval represented by this line the latent heat generated by crystallization (an exothermic effect) and the cooling effect of the bath (an endothermic effect) are in balance. Consequently, the temperature does not change in this interval.
When the chocolate mass is undertempered, too many crystals of the β ′ (IV) modifi cation develop, which rapidly transform to the β (V) modifi cation. Consequently, the latent heat dissipated by their crystallization exceeds the cooling effect of the bath. Therefore, an increase in temperature occurs. When the chocolate mass is overtempered, too many crystals of the β (V) modifi cation develop, which melt too slowly to compensate the cooling effect of the bath. Consequently, the temperature decreases continuously. In many publications, bloom in chocolate is often described as a process involving the migration by capillary action of a liquid fat to the surface (Kleinert 1962 ). Loisel et al.
( 1997 ) considered chocolate as a porous material and were able to determine, by mercury porosimetry, the porosity volume of well - tempered dark chocolate [ β (V)], undertempered chocolate [ β (IV)] and overtempered chocolate [a mixture of β (V) and β (VI)]. The volume of air bubbles due to the process was determined by X - ray radiography to be less than 0.1% of the sample volume. The porosity of normal chocolate was about 1% of the total volume, and this increased to 2% for the undertempered chocolate and 4% for the overtempered chocolate. The results did not allow determination of the precise pore diameter, but suggested that the chocolate did not have open, interconnected pores with a mean diameter larger than 0.1 μ m at the surface. Moreover, it seems that the pores were fi lled by the liquid fraction of cocoa butter at room temperature. As a result, it is better to talk about empty cavities rather than pores. Khanet al. ( 2003 ) highlighted the presence of pores at the surface of milk chocolate by scanning the surface with an atomic force microscope. These authors estimated the concentration of pores to be thousands/cm 2 ; the pores, 1 – 2.5 μ m in depth, were randomly distributed on the surface. As mentioned previously, a preferable method of crystallization from melts is to add crystal seeds of the stable modifi cation to the molten substance, which start an overall crystallization in the stable modifi cation. This is the principle of the Seedmaster tempering machine manufactured by Bindler, in which crystals of the stable β (VI) modifi cation are produced by intensive shearing ( ‘ Seedmaster cryst ’ ), and the pre - tempered chocolate mass is seeded by these stable crystals in the Seedmaster mix.
Beside cocoa butter, several types of chocolate may contain milk fat (milk chocolate) and/or oils derived from added almonds or hazelnuts (dark and milk chocolate) if these nuts are refi ned together with the chocolate mass. Since the properties of these fats/oils are essentially different from those of cocoa butter, they can exert an important effect on the crystallization of cocoa butter in chocolate. As a rule, it can be stated that in the case of milk fat, almond oil or /hazelnut oil the end point of cooling will be ∼ 26 ° C instead of 27 ° C, and the end point of warming will be 29 – 31 ° C instead of 30 – 32 ° C. The decrease in temperature that is to be used is dependent on the amount of these fats/oils. For further details, see Kniel ( 2000 ) and McGauley ( 2001 ).
Confectionery and Chocolate Engineering / Principles and Applications
Professor Ferenc Á. Mohos, PhD