Skip to main content
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2024 Mar 29;89(5):2867–2878. doi: 10.1111/1750-3841.17040

Determination of the process effect on cocoa butter crystallization by rheometer: Kinetic modeling by Gompertz equation

Esra Bölük 1, Esra Akdeniz 2, Recep Gunes 3, Ibrahim Palabiyik 1, Nevzat Konar 4,✉, Omer Said Toker 2
PMCID: PMC13281141  PMID: 38551060

Abstract

Abstract

In this study, the effects of temperature (22, 24, 26, 28, and 30°C) and strain (0.1%, 1%, and 5%) on cocoa butter (CB) crystallization were investigated by oscillatory test, and the four‐parameter Gompertz model was used to interpret the effect of parameters on pre‐crystallization, nucleation, and crystal growth stages of CB. Lag time and growth rate were calculated using the Gompertz model using time‐dependent storage modulus (G′) data. According to the results, CB crystallization at 26°C with a 1% strain value had the highest growth rate value, the shortest lag time, and the formation of β v polymorph type. Followingly, polymorphic types of the CB crystals were determined based on the melting points of polymorphs via the temperature ramp step, and the results obtained were correlated with a polarized light microscope. In conclusion, using a rheometer in both the observation of the pre‐crystallization process and the determination of polymorph types is very important for research and development studies in the chocolate industry for process and formulation optimization.

Practical Application

This study demonstrates the feasibility of a novel approach for investigating crystallization and oscillatory shear of CB using a rheometer, both for observing crystallization kinetics and determining polymorph type, accompanied by the Gompertz equation to model the crystallization kinetics. According to the results, the effect of process parameters (temperature and shear) on the crystallization behavior of CB can be observed by rheometer, which can provide a detailed perspective for chocolate manufacturers and researchers in research and development studies.

Keywords: chocolate, cocoa butter, crystallization, Gompertz model, polymorph

1. INTRODUCTION

Cocoa butter (CB) is one of the major areas of interest within the field of fat crystallization because of its polymorphic behavior. The main fatty acids present in CB are palmitic (C16:0, P), stearic (C18:0, S), and oleic (C18:1, O) acids in the ratio of about 26%, 36%, and 34% of the total fatty acid composition, respectively (Feng et al., 2024). Triglyceride compositions of CB are mainly composed of POP (∼19.0%), POS (∼39%), and SOS (∼28%). Origin and variety of cocoa influences fatty acid and triglyceride profile of CB. Triglycerides in CB can crystalize in several crystal packing configurations. As the form I melts around 16–18°C, form V around 32°C. CB has six crystal forms (I–VI) with different stabilities, densities, and melting points (Toro‐Vazquez et al., 2004). Thermal stability, melting temperature, and dense structure increase from forms I–VI. CB is the most decisive component for the quality of chocolate (melting characteristics, texture, mouthfeel, sensory properties, etc.) due to its physical properties. For instance, form V (β v) crystals are the most desirable for providing the best textural quality, resistance to fat bloom formation, and desired melting behavior considering the quality characteristics of chocolate products (Oh & Swanson, 2006; Wille & Lutton, 1966). The tempering process is performed for the pre‐crystallization of chocolate to obtain form V as a seed for further steps such as cooling and storage. Therefore, the formation of crystal types and their determination is crucial for chocolate science and technology. In this regard, CB crystallization has been studied by many researchers using tempermeter, DSC, NMR, XRD, and different microscope systems to understand the terms such as polymorphism, polymorphic transition, nucleation, crystal growth, and its kinetic (Brunello et al., 2003; Simoes et al., 2021; Sonwai et al., 2017). All of the mentioned methods have advantages or disadvantages considering the cost of the equipment, practical usage, labor requirements, and so on. Tempermeter is the most used equipment in the chocolate industry. However, the value measured with a tempermeter cannot give an idea about the polymorph types and proportions formed in the pre‐crystallization stage in chocolate. Similarly, it is not possible to get an idea about the pre‐crystallization time in DSC. On the other hand, DSC and NMR do not have the application of shear, and this is not suitable for the simulation of any dynamic conditions and practical applications for CB samples.

Rheology basically describes the flow or deformation behavior of materials under stress/strain forces. However, a rheometer can provide many different important parameters required for the optimization of process parameters due to its flexibility for process simulations. In this regard, rheometer has been used in the determination of CB crystallization rate and amount in recent years (Nemoto et al., 2021), and there are studies in which the crystallization of fats is mathematically modeled to enable the quantification of differences in crystallization behavior (Foubert et al., 2003; Kloek et al., 2000). Actually, the Gompertz equation is often used to describe bacterial growth but there are several similarities between fat crystallization and bacterial growth. Accordingly, the reproduction of bacteria by binary fission can be compared to the growth of crystals from a nucleus, and bacterial consumption of nutrients can be compared to the reduction of supersaturation. Therefore, the Gompertz model was used in many studies to estimate the parameters of crystallization of different fats and/or oils (Foubert et al., 2002; Mursalin, 2018; Szydłowska‐Czerniak et al., 2017).

It is clear that the details of the crystallization processes of CB are still open to question, and oscillation conditions have not been investigated in detail. In this framework, two steps were presented for developing the basic concept to achieve and observe the crystallization of CB by rheometer. First, CB crystallization was carried out by applying oscillatory shear (strain), and then storage modulus (G′) versus time data were obtained. Second, these obtained data were adapted to the Gompertz model for the prediction parameters such as lag time, growth rate, and A‐value to measure the kinetics of CB crystallization. Then, polymorphic types of the CB crystals were determined based on the melting points of polymorphs via the temperature ramp step, and the results obtained were correlated with a polarized light microscope (PLM). The advantages of using oscillation measurement in determination of crystallization kinetics and polymorph type are as follows: (i) Rheometer can provide to observe shearing effect that directly affects the crystallization process (shearing effect cannot be observed with DSC, XRD, NMR, etc.), and (ii) crystallization kinetics and melting point can be detected using rheometer (it is not possible to observe both of them for the other equipments such as NMR, DSC, and XRD).

2. MATERIALS AND METHODS

2.1. Materials

CB used in the study was obtained from local chocolate manufacturer company in Türkiye. The main fatty acid profile of the CB used in the study is as follows: 36.3% stearic acid, 33.6% oleic acid, 26.2% palmitic acid, 2.2% linoleic acid, and 0.8% arachidic acid. Deodorized CB was used in the present study. It was extracted from forastero type cocoa, and its origin was Ghana.

2.2. Oscillatory tests by rheometer for the determination of crystallization kinetics

The effect of the process parameters such as temperature and strain on CB crystallization was investigated, and strain was used as a crystallization‐inducing factor using a stress–strain controlled rheometer (TA Instruments, Discovery HR 20). CB was left to crystallization via oscillatory time sweep test in between a 40 mm diameter parallel plate and a Peltier plate at a gap of 600 µm. Under these circumstances, a thermal process was applied to CB as follows after the sample was placed on Peltier plate at 50°C.

As the crystallization occurs G′ values increase at constant strain and temperature values. Experiments were carried out in two steps. In the first step, the effect of different temperatures (22, 24, 26, 28, and 30°C) and strain values (0.1%, 1%, and 5%, relative deformation) on the pre‐crystallization of CB was investigated at 10 Hz. G′ values were measured at every 6 s of analyses. For this purpose, melted CB was poured onto the plate of the rheometer device. The temperature of the plate was adjusted to 50°C for melting all polymorph types such as in the conventional tempering process where the chocolate samples were heated at 2°C/min for melting of all crystal types. After that, the temperature of the plate was decreased to pre‐crystallization temperature, and the test was started to observe the change in G′ values with respect to time at predetermined temperature/strain combinations for investigation of the pre‐crystallization kinetics of the samples. In the present study, oscillatory tests were applied to some samples for 15,000 s, whereas other samples were for 50,000 and 90,000 s as given in Figure 1 depending on the reaching of the G′ value to constant value. The reason for the changes in time of the G′ values of CB kept was followed until the G′ value was fixed as a sign of ending the CB crystallization. The G′ versus time data were used to fit of Gompertz model to calculate the corresponding model parameters. All of the analyses were conducted three times, and fitting was employed to average values of replicates.

FIGURE 1.

FIGURE 1

First step rheologic analyses of cocoa butter crystallized at 22, 24, 26, 28, and 30°C with 0.1%, 1%, and 5% strain values (a: 22°C, b: 24°C, c: 26°C, d: 28°C, e: 30°C).

2.3. Parameter estimation and calculating lag time and growth rate for crystallization kinetics

To describe the crystal behavior numerically, the Gompertz model was applied to fit the data obtained by the rheometer. The corresponding model was established by Zwietering et al. (1990) to explain microbial growth and then was adapted for crystallization behavior by Kloek et al. (2000). In our study, time sweep data (G′ crystallization) were taken from the rheometer and were fitted to Gompertz model. Time sweep data obtained within the scope of the study were applied to different algebraic models with nonlinear regression using Sigmaplot 2014 software (SPSS Inc.). This software uses the Marquardt–Levenberg algorithm to find the best fit parameters between the model and data. This algorithm makes the values of the parameters that minimize the sum of the squared differences between the observed and predicted values of the dependent variable. Equation (1) was applied to all time sweep data as four‐parameter Gompertz equation in Sigmaplot and adapted to Equation (2) in Gibson et al. (1987) as following Broughall et al. (1983), Broughall and Brown (1984), Jefferies and Brain (1984), and Gibson et al. (1987). The reparameterized equation of the model is as follows:

f=y0+aexp−exp−x−x0b (1)
L(t)=A+Cexp−exp−B(t−M) (2)

where L(t) is the log10 count at time t. This provides the growth curve to be outlined in four parameters B, M, C, and A, where A is the maximum amount of crystallization, C is the asymptotic amount of growth that occurs as t increases indefinitely, and B is the relative growth rate at M, where M is the time at which the absolute growth rate is a maximum. Parameters of Equation (1) were matched with Equation (2) where a is C; 1/b is B, x 0 is M, and y 0 represents A. Thus, B, M, C, and A were utilized by the Gompertz equation to obtain growth rate and lag time via the following equations (Gibson et al., 1987):

Growthrate=B×Ce (3)
lag(h)=M−1B (4)

All of the analyses were conducted three times, and fitting was employed to average values of replicates.

2.4. Polymorph determination with rheometer

After the pre‐crystallization period was completed, polymorph types formed were observed using the rheometer. Temperature ramp tests were carried out from hold temperature which was started in the first step of rheological analysis (FSRA) to 45°C with a 2°C/min thermal ramp. After melting the CB sample, the melting range was determined using the temperature versus G′ graph. Polymorph type was determined considering the sudden decrease in G′ value observed in the graph as the related polymorph started to melt. Temperatures for the polymorph melting were determined using the literature: γ(I): 16–18°C; α(II): 21–22°C; β′(III): 25.5°C; β′(IV): 27–29°C; β(V): 32–34°C; β(VI): 34–36°C (Garti & Widlak, 2012).

2.5. Investigation of crystal microstructure by polarized light microscopy

Polymorphic forms of CB were observed using a PLM (Carl Zeiss) with a 10× objective. For this purpose, first, solid CB was melted in the oven at 50°C to dissolve all the crystals present in the sample. The effect of different temperatures (22, 24, 26, 28, and 30°C) on the pre‐crystallization of CB was examined using the same PLM with the 10× objective equipped with a Peltier heating/cooling system. About 5 mL CB samples melted in an oven (50°C, 30 min) were added to the microscope cover and placed on the microscope table. Then, the samples were cooled to these temperature levels at 2°C/min and waited for 24 h to obtain the full formation of crystals. Followingly, the melting analysis was carried out by heating the samples from the pre‐crystallization temperature mentioned above to 45°C at 2°C/min where camera recording was conducted during this period. Polymorph types were obtained depending on the change of the microstructure based on the temperature variation considering the polymorph melting temperature mentioned above. Then, the differences in microstructure between the melting points of the polymorphs were investigated, and if there was a difference between these two points, it indicated that there was a melting of the polymorphs. The differences between microstructures at 24 and 26°C indicate melting of β′(III); 26 and 30°C indicate melting of β′(IV); 30 and 34°C indicate melting of β(V); 34 and 37°C indicate melting of β(VI).

3. RESULTS AND DISCUSSION

The present study is composed of two parts: (i) observation of pre‐crystallization kinetics of CB by oscillation analyses and modeling of the obtained data using the Gompertz model and (ii) determination of polymorph type formed during the pre‐crystallization period with rheometer and PLM.

3.1. Oscillatory tests by rheometer for determination of crystallization kinetics

3.1.1. Observation of pre‐crystallization kinetics

First, the pre‐crystallization stage was carried out using the rheometer and then the melting of the polymorphs was investigated which was formed in the pre‐crystallization step for the determination of polymorph type formed. As can be seen in Figure 1, the G′ values of the samples increased with time. In other words, as the crystallization developed in CB, the G′ values also increased. It was seen that there were significant differences in G′ values of CB due to pre‐crystallization process conditions (temperature and strain value). When the effect of strain values applied at different temperatures was examined, it was seen that generally the lowest G′ values were reached in the condition where 5% strain was applied. In line with previous study, Tran and Rousseau (2016) found that shear needs to be in a suitable range; if applied shear resulting mechanical effect over the upper limit, network of crystallines can be broken or if it is below the lower limit, nucleation cannot be induced. Considering the analysis carried out, it was seen that the structure formed as a result of crystallization and the deformation caused by the applied strain had a simultaneous effect on the G′ value of the samples. If crystallization dominates the deformation of the network structure present in the CB sample, the G′ value increases, and if strain deformation dominates the CB sample network structure formed as a result of pre‐crystallization, the G′ value decreases. When the results are examined, it is seen that the G′ value increases to a certain level under all conditions studied. After a certain time, fixation and decreases in G′ value of some samples are observed. The point at which the G′ value of the samples is fixed can be regarded as the point at which crystallization formation is completed. Obtaining the smallest G′ values at a 5% strain value is related to the excessive deformation as a result of mechanical forces applied. When 0.1% and 1% strain values were compared considering the deformation effect, it was expected that the 1% strain would be lower, whereas the G′ value of the samples with 1% strain was generally higher than the samples exposed to 0.1% strain. This indicated that strain was also important for accelerating the crystallization. Therefore, the acceleration of the crystallization and deformation of the crystal network due to applied strain should be considered simultaneously during process optimization. After some period of crystallization zigzag (especially observed in 26°C at 1% strain) was observed in some figures, indicating the solidification of CB as a result of crystallization, which was also observed visually during analyses. Depending on the crystallization process, crystal structures of TAG are formed, and therefore, G′ values increase. Formation of crystallization in further steps resulted in the solidification of the CB. When rheometer probe contact with the solid particles, zigzags are observed in the analyses. Atik et al. (2020) have found that a rheometer might be a suitable and effective device to determine type polymorph type and its proportional amount. In the previous study, the pre‐crystallization step was carried out by seeding technique, and the crystallization was tracked with the rheometer by observing the change of G′ values with respect to time during the pre‐crystallization. The ratio of polymorph types was correlated with the G′ values.

3.1.2. Curve fitting by Gompertz model

In order to deeply interpret the results obtained from the time‐dependent oscillatory test, the obtained data were fitted to the Gompertz model. Gompertz model results were obtained with R2 values higher than 0.96, and it was a positive sign of a suitable fitting. Supplementary files, including detailed information about fitted models, were prepared. R2 , Adj R2 , and p values for model coefficients, analysis of variance results, and number of iterations performed for fitting were presented in the file. Generally, all of the model parameters mentioned were found as significant (p < 0.01). Adj R2 values were very close to R2 values. Number of iterations changed between 9 and 27. All of the presented data in the file indicated that the established model successfully represented the relation between dependent and independent parameters. From those fitted curves, B, M, C, and A parameters were derived, and by using these parameters, lag (h) and growth rate (log10 increase/h) were calculated for interpreting CB crystallization as shown in Table 1. To date, various methods have been developed to measure crystallization kinetics and revealed the superiority of the four‐parameter Gompertz model over the Avrami model (Foubert et al., 2002). Foubert et al. (2002) studied the crystallization kinetics of 14 different CB crystallization experiments; a desired nonlinear equation was established manually and compared with three equations: Avrami with induction time, Avrami without induction time, and Gompertz model. After the comparison, it was observed that it achieved a high level of harmony with the manually created nonlinear equation and the four‐parameter Gompertz model. For this reason, the Gompertz model was selected, considering that the crystallization experiments were carried out within the scope of the study, and the best fit was achieved with the four‐parameter Gompertz model as a result of the matching studies performed with many nonlinear equations. Equation (1) was fitted to the time versus G′ data series, and four model parameters were calculated for each crystallization trial. Equation (2) was also a four‐parameter Gompertz model which was used in the study of Gibson et al. (1987) by adapting the model to microbiology for measuring bacterial growth rate. General model behavior is given in Figure 2, and the Gompertz equation showed one of the sigmoidal kinetic models (Zwietering et al., 1990). In this study, the Gompertz model applied to bacterial growth rate was adjusted to crystallization kinetics as mentioned in literature where the model was reparameterized as declared by Zwietering et al. (1990) and adapted for fat crystallization by Kloek et al. (2000) was fitted to the one step crystallization curves as given in Equation (1).

TABLE 1.

Parameters estimated from the Gompertz equation.

Temp (°C) Strain (%) C B M A R2
22 0.1 18,009 0.0041 5833 210 0.98
1 12,097 0.4075 18 137 0.99
5 127,299 0.5796 20 2221 0.97
24 0.1 8457,375 0.0002 20,442 −34,867 0.99
1 41,413 0.1117 62 −39 0.96
5 39 0.3282 9 −0.487 0.99
26 0.1 1863 0.00006 12,515 −296 0.99
1 1391,367 1.1593 15 2922 0.99
5 26,794 0.2333 45 2.4 0.97
28 0.1 151 0.3334 12 −2.15 0.99
1 22,765 0.0735 80 −67 0.98
5 24 0.1906 6 −4.74 0.98
30 0.1 556 0.3106 7 1.68 0.98
1 179 0.00009 1541 −56 0.99
5 4 0.5832 7 2.66 0.96
FIGURE 2.

FIGURE 2

General model behavior of Gompertz equation (A can be one of the parameters be modelled such as G′ in this study).

Parameters were calculated by fitting obtained time versus G′ data to the Gompertz model, and the corresponding model parameters are presented in Table 1. The lag time and growth rate were calculated according to the formulations given by Gibson et al. (1987) who calculated the combinations of the best parameters for bacterial growth rate. In the present study, CB crystallization growth rate and lag time were calculated based on microbiology parameters. Kloek et al. (2000) used a modified Gompertz equation to explain the crystallization kinetics of fully hydrogenated palm oil in sunflower oil. Indeed, there are several similarities between fat crystallization and bacterial growth. Although the production of bacteria is comparable to nucleation and growth of crystals, bacterial consumption of nutrients can be compared to the reduction of supersaturation. In this regard, optimum strain values were chosen based on lag time and growth rate results which were calculated by Equations (3) and (4), and their graphics are given in Figure 3. Lag time represents the nucleation time of crystals, and the growth rate correlates with the crystal growth rate. To select the optimum combination, lag time should be lower value, whereas growth rate should have higher value. CB samples exposed to 1% strain‐applied had the highest growth rate for 24, 26, and 28°C. The growth rates in crystallization performed at 22 and 30°C which were the highest, and the lowest temperatures were found to be lower values. When 22, 24, and 26°C temperature levels were examined at 0.1% strain value, their lag times were extremely high. Although CB crystallization at 24°C with 0.1% strain value displayed a high amount of total crystalline based on growth rate, it should not be selected as an optimum pre‐crystallization combination because of its exceeding lag time. Moreover, increased lag time is an undesirable situation for cost‐effective industrial processes as if it increases the required time to complete, the process also increases, which is undesirable for the industry in terms of many aspects such as the lower amount of production, more energy expenditure, more carbon footprint, more production cost, and so on. The results, as shown in Figure 3, indicate that CB crystallization at 26°C with a 1% strain value had the highest growth rate value and the shortest lag time, and it can be concluded that CB crystallization at 26°C with a 1% strain value was optimum combination.

FIGURE 3.

FIGURE 3

The growth rate and lag time results based on strain (x‐axis shows the temperature values).

3.1.3. Determination of polymorph types with oscillatory parameters

After the pre‐crystallization process, a temperature sweep test was performed to determine polymorph types observed considering the melting points. Several studies found that rheological parameters such as yield stress, viscosity, loss, and storage modulus did not depend on crystal polymorph. For this reason, only G′ cannot be correlated with the polymorph type but the total crystal amount can be calculated by using the Gompertz model. Differently from the literature, in this study, the second rheologic step was carried out which is called temperature ramp and it was thought that the melting point of CB polymorph can also be determined by rheometer based on the melting point of polymorphs. Melting analyses were carried out to determine which polymorph was formed after pre‐crystallization of CB under different conditions. Polymorph types observed under different CB crystallization conditions determined by oscillatory parameters are summarized in Table 2. The melting graphic of CB pre‐crystallized at 22°C at different strain values is given in Figure 4A. Two different reductions of G′ values were observed between 26–29°C and 29–31°C in CB at 22°C with a 0.1% strain combination. These results showed that considering the melting of CB polymorphs mentioned in Section 2.4, β III and β IV forms exist. On the other hand, in the CB samples crystallized at 1% and 5% strains, there was only a decrease in the G′ value between 29 and 31°C, and this temperature range is an indication of the formation of β IV at these strain values (Loisel et al., 1998; Wille & Lutton, 1966). Dhonsi and Stapley (2006) studied the effect of shear rate and temperature parameters on the crystallization of CB and presented that keeping CB at low temperatures resulted in quick crystallization and formation of polymorphs with lower melting points, regardless of different shear rate applications in parallel to Figure 4. When the CB crystallized at different strain values at 24°C were examined, it was observed that β ııı, β ıv, β v, and β vı were formed at 0.1% strain value. Formation of β ıv and β v polymorphs was observed in samples pre‐crystallized with 1% and 5% strain (Figure 4B). The presence of melting peaks at about 29 and 32°C in CB at 26°C with 1% strain indicated the formation of β ıv and β v polymorphs under these conditions. The peak seen around 32 and 36°C in CB with 0.1% strain showed the formation of β v and β vı. At 5% strain value, as the G′ values were quite low compared to the other two strain values, the melting peaks were seen around 26, 29, and 32°C although not as clear as the others. This situation also showed the presence of β ııı, β ıv, and β v polymorphs. The decrease around 35°C in CB at 28°C with 0.1% strain showed the formation of β vı, and the decrease around 33°C and 36°C observed in strain applications of 1% and 5% indicated the formation of β v and β vı polymorphs. As can be seen from the figure, when the CBs pre‐crystallized at 30°C with 0.1%, 1%, and 5% strain values were examined, it was seen that β vı polymorph was formed. It can be concluded that CB at 26°C with 1% strain was found in β v melting range. For determining these conditions, the maximum decrease in G′ value was carried out in the SSRA (temperature ramp) where the melting range of β v polymorph was also consistent with the results mentioned in the previous section. As known, the tempering process also includes a heating process of about 30°C after crystallization for melting of the unstable polymorphs like β ıv. Depending on the findings of lag time and growth rate, 26°C and 1% strain were selected as the optimum condition where β ıv and β v polymorphs. Formation of β ıv is not a problem for this condition as they can easily melt during the re‐heating process.

TABLE 2.

Polymorph types formed in different crystallization processes determined by rheometer and microscope.

Temp. (°C) Strain (%) Rheometer Microscope
β I β ıı β ııı β ıv β v β vı β I β ıı β ııı β ıv β v β vı
22 0.1 √ √ √ √
1 √
5 √
24 0.1 √ √ √ √ √ √ √
1 √ √
5 √ √
26 0.1 √ √ √ √
1 √ √
5 √ √ √
28 0.1 √ √ √
1 √ √
5 √ √
30 0.1 √ √
1 √
5 √
FIGURE 4.

FIGURE 4

Second step rheologic analyses of cocoa butter crystallized at 22, 24, 26, 28, and 30°C with 0.1%, 1%, and 5% strain values for the determination of polymorph type (a: 22°C, b: 24°C, c: 26°C, d: 28°C, e: 30°C).

As a result of the present study, it was observed that both crystallization kinetics and polymorph type formed can be determined using a rheometer. Determination of those with rheometer easily is very important for the food industry and science in terms of research and development studies, including process and product optimization. Requirements of very few samples during rheometer analysis and adaptation of rheometer to process simulation by adjusting shear and temperature‐based parameters can also provide advantages. Moreover, a rheometer can be used instead of different equipment like DSC and XRD for many different purposes simultaneously like determination of rheological parameters, observation of butter crystallization, and polymorph type detection, which is also important for preventing device waste and improving the efficiency of the equipment.

3.2. Determination of polymorph type by PLM

PLM equipped with Peltier system was also employed for the determination of polymorph type formed under different conditions. The images of CB pre‐crystallized at five different temperatures (22, 24, 26, 28, and 30°C) are shown in Figure 5. These temperature levels were selected considering the melting temperature of the polymorphs. At the end of the 24 h of crystallization, microstructural features of CB crystallized at different conditions were different depending on the polymorph type and crystal amount. After pre‐crystallization of CB at different temperature levels, the video was recorded during their melting with Peltier system of the microscope. Polymorph types observed depending on the obtained video are summarized in Table 2. As the CB pre‐crystallized at 22°C was examined, major differences in images from 22 to 30°C were not observed. In addition, the clustering of crystals is observed in these temperature ranges. The differences between 28–30 and 30–34°C indicate that β ıv and β v polymorphs were formed as a result of pre‐crystallization at 22°C. The lack of a difference between 34 and 37°C means that form β vı is not formed as a result of keeping CB at 22°C. When comparing the results of the microscope and rheometer, the β v polymorph type was identified with both methods. In the study of Marangoni and McGauley (2003), the aggregation of crystal was observed in samples kept at 20°C for 1 day. Figure 5a,b shows that the aggregation of the crystal of CB was also observed after the pre‐crystallization process. Differences between images taken at 28–30, 30–34, and 34–37°C temperatures were observed in CB pre‐crystallized at 24°C. This case showed the formation of β ıv, β v, and β vı polymorphs in a sample maintained at 24°C under stationary conditions. However, polymorphs of CB had a maximum granular structure at 26°C and showed a less granular appearance at 28 and 30°C. Although there was no difference between 26 and 30°C when looking at Figure 5C, variations of microstructure were observed between 30–34 and 34–37°C, indicating the formation of β ıv and β v as a result of CB crystallization at 26°C. According to the results of microstructure, β v and β vı polymorphs were formed in CB crystallized at 28°C and β vı was observed in CB crystallized at 30°C. The differences between the microscope and rheometer results considering polymorph type detection can be the result of strain value applied during the pre‐crystallization process conducted by the rheometer. However, in the microscope‐related studies, pre‐crystallization was conducted under stationary conditions.

FIGURE 5.

FIGURE 5

Polarized light microscope images of cocoa butter statically crystallized at 22°C (a), 24°C (b), 26°C (c), 28°C (d), and 30°C (e) for a day.

4. CONCLUSION

A rheometer is generally used for the determination of flow behavior and viscoelastic properties of the materials. In the present study, pre‐crystallization of CB was investigated by oscillatory tests. For this aim, time‐dependent oscillatory test was employed to obtain time versus G′ values. The Gompertz equation was fitted to the obtained data for obtaining the data about the nucleation time of crystals and crystal growth rates. Optimum temperature and strain combination of CB were selected based on three basic points: brief lag time, high amount of crystallines in FSRA (pre‐crystallization), and suitable melting range of desired polymorph in SSRA. FSRA was mostly illuminative for the selection of optimum strain value, whereas the temperature ramp was effective in choosing the optimum temperature value to crystallize. The 5% strain value was found devastating for all temperatures in the experimental pattern. The pre‐crystallization temperature of 26°C was found the best temperature for eliminating unstable polymorphs to crystallize CB at 1% strain value due to its short lag time, high growth rate, and resulting in the formation of β v polymorph. The findings of the present study highlighted that a rheometer can be used as an effective tool to obtain pre‐crystallization‐related parameters and polymorphs formed in the pre‐crystallization process, which is very important for the chocolate industry for the optimization of the tempering process in terms of formulation and process‐based research and development studies. In the future, this method can be applied to model chocolate systems, and the results of this method can be correlated with industrial applications.

AUTHOR CONTRIBUTIONS

Esra Bölük: Formal analysis. Esra Akdeniz: Formal analysis. Recep Gunes: Formal analysis; methodology. Ibrahim Palabiyik: Conceptualization; methodology; writing—review and editing; writing—original draft. Nevzat Konar: Conceptualization; writing—original draft; methodology; Writing—review and editing. Omer Said Toker: Conceptualization; methodology; writing—original draft; writing—review and editing; project administration.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest and no competing financial interest.

Supporting information

Supporting Information

JFDS-89-2867-s001.docx (43.7KB, docx)

ACKNOWLEDGMENTS

This work was funded by the Scientific and Technological Research Council of Türkiye (TUBITAK), Project No: TOVAG‐119R006.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

REFERENCES

  1. Atik, D. S. , Bölük, E. , Toker, O. S. , Palabiyik, I. , & Konar, N. (2020). Investigating the effects of lecithin‐PGPR mixture on physical properties of milk chocolate. Lebensmittel‐Wissenschaft & Technologie—Food Science and Technology, 129, 109548. 10.1016/j.lwt.2020.109548 [DOI] [Google Scholar]
  2. Broughall, J. M. , Anslow, P. A. , & Kilsby, D. C. (1983). Hazard analysis applied to microbial growth in foods: Development of mathematical models describing the effect of water activity. Journal of Applied Bacteriology, 55(1), 101–110. 10.1111/j.1365-2672.1983.tb02653.x [DOI] [PubMed] [Google Scholar]
  3. Broughall, J. M. , & Brown, C. (1984). Hazard analysis applied to microbial growth in foods: Development and application of three‐dimensional models to predict bacterial growth. Food Microbiology, 1(1), 13–22. 10.1016/0740-0020(84)90005-4 [DOI] [Google Scholar]
  4. Brunello, N. , McGauley, S. E. , & Marangoni, A. (2003). Mechanical properties of cocoa butter in relation to its crystallization behavior and microstructure. Lebensmittel‐Wissenschaft & Technologie—Food Science and Technology, 36(5), 525–532. 10.1016/S0023-6438(03)00053-7 [DOI] [Google Scholar]
  5. Dhonsi, D. , & Stapley, A. G. F. (2006). The effect of shear rate, temperature, sugar and emulsifier on the tempering of cocoa butter. Journal of Food Engineering, 77(4), 936–942. 10.1016/j.jfoodeng.2005.08.022 [DOI] [Google Scholar]
  6. Feng, C.‐H. , Otani, C. , & Hoshina, H. (2024). Characterization of different types of crystallization from cocoa butter by using terahertz spectroscopy. Applied Sciences, 14, app14010035. 10.3390/app14010035 [DOI] [Google Scholar]
  7. Foubert, I. , Dewettinck, K. , & Vanrolleghem, P. A. (2003). Modelling of the crystallization kinetics of fats. Trends in Food Science & Technology, 14(3), 79–92. 10.1016/S0924-2244(02)00256-X [DOI] [Google Scholar]
  8. Foubert, I. , Vanrolleghem, P. A. , Vanhoutte, B. , & Dewettinck, K. (2002). Dynamic mathematical model of the crystallization kinetics of fats. Food Research International, 35(10), 945–956. 10.1016/S0963-9969(02)00157-6 [DOI] [Google Scholar]
  9. Garti, N. , & Widlak, N. R. (Eds.). (2012). Cocoa butter and related compounds. Elsevier. [Google Scholar]
  10. Gibson, A. M. , Bratchell, N. , & Roberts, T. A. (1987). The effect of sodium chloride and temperature on the rate and extent of growth of Clostridium botulinum type A in pasteurized pork slurry. Journal of Applied Bacteriology, 62(6), 479–490. 10.1111/j.1365-2672.1987.tb02680.x [DOI] [PubMed] [Google Scholar]
  11. Jefferies, C. J. , & Brain, P. (1984). A mathematical model of pollen‐tube penetration in apple styles. Planta, 160(1), 52–58. 10.1007/BF00392465 [DOI] [PubMed] [Google Scholar]
  12. Kloek, W. , Walstra, P. , & Van Vliet, T. (2000). Nucleation kinetics of emulsified triglyceride mixtures. Journal of the American Oil Chemists' Society, 77(6), 643–652. 10.1007/s11746-000-0104-7 [DOI] [Google Scholar]
  13. Loisel, C. , Keller, G. , Lecq, G. , Bourgaux, C. , & Ollivon, M. (1998). Phase transitions and polymorphism of cocoa butter. Journal of the American Oil Chemists' Society, 75(4), 425–439. 10.1007/s11746-998-0245-y [DOI] [Google Scholar]
  14. Marangoni, A. G. , & McGauley, S. E. (2003). Relationship between crystallization behavior and structure in cocoa butter. Crystal Growth & Design, 3(1), 95–108. 10.1021/cg025580l [DOI] [Google Scholar]
  15. Mursalin, M. (2018). Crystallization kinetics of coconut oil based on Gompertz model. Indonesian Food Science & Technology Journal, 1(1), 1–7. 10.22437/ifstj.v1i1.4269 [DOI] [Google Scholar]
  16. Nemoto, F. , Yamada, N. L. , Takabatake, F. , & Seto, H. (2021). Installation of a rheometer on neutron reflectometer SOFIA at J‐PARC toward Rheo‐NR and observation of the crystallization behavior of cocoa butter in chocolate. JPS Conference Proceedings, 33, 011077. 10.7566/JPSCP.33.011077 [DOI] [Google Scholar]
  17. Oh, J. H. , & Swanson, B. G. (2006). Polymorphic transitions of cocoa butter affected by high hydrostatic pressure and sucrose polyesters. Journal of the American Oil Chemists' Society, 83(12), 1007–1014. 10.1007/s11746-006-5155-2 [DOI] [Google Scholar]
  18. Simoes, S. , Lelaj, E. , & Rousseau, D. (2021). The presence of crystalline sugar limits the influence of emulsifiers on cocoa butter crystallization. Food Chemistry, 346(1), 128848. 10.1016/j.foodchem.2020.128848 [DOI] [PubMed] [Google Scholar]
  19. Sonwai, S. , Podchong, P. , & Rousseau, D. (2017). Crystallization kinetics of cocoa butter in the presence of sorbitan esters. Food Chemistry, 214, 497–506. 10.1016/j.foodchem.2016.07.092 [DOI] [PubMed] [Google Scholar]
  20. Szydłowska‐Czerniak, A. , Rabiej, D. , & Pawłowicz, R. (2017). Comparison of the crystallization behaviors of different types of cocoa butters and chocolates. Journal of Food Processing and Preservation, 41, e13154. 10.1111/jfpp.13154 [DOI] [Google Scholar]
  21. Toro‐Vazquez, J. F. , Pérez‐Martínez, D. , Dibildox‐Alvarado, E. , Charó‐Alonso, M. , & Reyes‐Hernández, J. (2004). Rheometry and polymorphism of cocoa butter during crystallization under static and stirring conditions. Journal of the American Oil Chemists' Society, 81(2), 195–202. 10.1007/s11746-004-0881-z [DOI] [Google Scholar]
  22. Tran, T. , & Rousseau, D. (2016). Influence of shear on fat crystallization. Food Research International, 81, 157–162. 10.1016/j.foodres.2015.12.022 [DOI] [Google Scholar]
  23. Wille, R. L. , & Lutton, E. S. (1966). Polymorphism of cocoa butter. Journal of the American Oil Chemists' Society, 43(8), 491–496. 10.1007/BF02641273 [DOI] [PubMed] [Google Scholar]
  24. Zwietering, M. H. , Jongenburger, I. , Rombouts, F. M. , & Van't Riet, K. (1990). Modeling of the bacterial growth curve. Applied and Environmental Microbiology, 56(6), 1875–1881. 10.1128/aem.56.6.1875-1881.1990 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information

JFDS-89-2867-s001.docx (43.7KB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.


Articles from Journal of Food Science are provided here courtesy of Wiley

RESOURCES