Improving the crystallization and melting characteristics of cocoa butter substitute by blending with krabok seed fat

Abstract

This work investigated the crystallization and melting behavior of a commercial cocoa butter substitute (CBS) blended with 10–80% (by weight) of a hard lauric fat called krabok seed fat (KSF). The aim was to find CBS-KSF blends with improved crystallization and melting characteristics from that of the CBS. It was found that the addition of 10–80% KSF to CBS improved the melting properties of the CBS. However, 10–20% KSF resulted in too high solid fat content (SFC) values at the body temperature (37 °C) which would lead to waxy mouth feel. Adding 30–40% KSF resulted in better melting profiles than 10–20% KSF with SFC values < 3% at 37 °C and SFC curves most similar to cocoa butter. However, 40% KSF led to a significant decrease in the crystallization rate from that of CBS and a significant increase in the average crystal size. With 60–80% KSF, although the blends melted completely at the body temperature, their crystallization rates were significantly reduced. All CBS-KSF blends crystallized into β′ structure. Therefore, the addition of 30% KSF to the CBS is recommended for industrial use to obtain compound chocolate with improved quality.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13197-022-05513-1.

Introduction

Cocoa butter substitutes (CBS) are produced from lauric fats such as palm kernel oil and coconut oil (Lipp and Anklam 1998). They are commonly used in chocolate and confectionery industry as a substitute for cocoa butter (CB) in compound chocolate or compound coating. Although CBS are different from CB in terms of fatty acid (FA) and triacylglycerol (TAG) compositions, their physical properties as well as their melting characteristics are quite similar to CB (Smith et al. 2004; Sonwai et al. 2015). The advantages of CBS are no tempering requirement and gloss retention because they tend to crystallize directly from the melt into a stable β′ polymorph. However, many commercial CBS suffer from possessing the following properties: (1) too low solid fat content (SFC) at room temperature, making compound chocolate too soft with unpleasant stick surface especially in tropical countries, and (2) too high SFC at body temperature (37 °C), making it waxy when consumed. The SFC should be high enough at room temperature in order to prevent oil migration which leads to detrimental effects on the surrounding matrix (Danthine et al. 2015). Adding of hard fats consisting mainly of saturated or high-melting FAs to confectionery fats can help to improve the melting properties of the final products. In addition, hard fats can act as structuring agents, preferential nuclei in the crystal lattice ordering and inducers of specific polymorphic forms (Ribeiro et al. 2013). For example, the addition of fully hydrogenated vegetable oils improved the crystallization kinetics, thermal behavior and SFC of CB and palm oil (Ribeiro et al. 2013). However, the addition of hard fats could lead to waxy mouthfeel because the resulting fat blends do not melt completely or almost completely at the body temperature.

Krabok (Irvingia malayana) is widely grown in tropical countries. Krabok seeds contain approximately 70.3% fat (dry basis) (Bandelier et al. 2002) with lauric and myristic acids as the main FA components. Krabok seed fat (KSF) has a melting point of ~ 42 °C and SFC above 80% at 30 °C (Sonwai et al. 2015). It is mostly use in confectionery and pharmaceutical applications. In our previous work, an attempt was made to improve the melting characteristic of a commercial CBS for using as a fat phase in heat-resistant compound chocolates by adding KSF (Podchong et al. 2020). The CBS exhibited high SFC (> 90%) at ≤ 25 °C but showed too low SFC at 30–35 °C, indicating that it would not be able to resist high room temperatures especially in tropical countries. Adding 10, 20 and 60% (by weight) of KSF to the CBS increased its SFC within that temperature range and the blends still melted almost completely at 37 °C. This allowed it to resist heat better without leaving waxy mouthfeel when consumes.

In this research, KSF was added to another commercial CBS with different properties from the previous one to improve its crystallization and melting characteristics for industrial use. The CBS exhibited SFC > 80% at ≤ 20 °C. Its SFC values were relatively low at 25–32.5 °C but were too high at 37 °C, making it unable to withstand high temperatures during storage and at the same time too waxy when consumed.

Materials and methods

Materials

A commercial CBS (Melarin 40) was purchased from Sino-Pacific Trading (Bangkok, Thailand) Co., Ltd. Krabok seeds were purchased from a local fruit and vegetable market in Bangkok, Thailand. Crude KSF was extracted from the krabok seeds and purified using methods provided elsewhere (Sonwai et al. 2015). The standard fatty acid methyl esters for FA analysis using gas chromatography (GC) were purchased from AccuStandard Inc. (New Haven, CT, USA). All other chemicals were obtained commercially and were of analytical grade.

Preparation of fat blends

KSF was added to CBS at different weight concentrations (10, 20, 30, 40, 60 and 80%). The mixtures were melted at 80 °C with stirring for 10 min to ensure complete melting and to obtain the homogenous blends. All fat blends were kept at 4 °C until analysis.

Fatty acid composition

The two original fats (CBS and KSF) and the CBS-KSF blends were converted to fatty acid methyl esters (FAME) following the Association of Official Analytical Chemists (AOAC) official method 969.33 (AOAC 2000). The FAME were analyzed by GC (Shimadzu GC-2010, Shimadzu Corp, Kyoto, Japan) with flame ionization detector and a VertiBond™ wax capillary column (50 m long, 0.25 mm internal diameter and 20 μm film thickness). The injection and detector temperatures were 210 °C and 280 °C, respectively. The oven temperature program was: (1) hold at 120 °C for 3 min; (2) increase to 220 °C at 10 °C/min and hold for 30 min, and (3) increase to 240 °C at 5 °C/min and hold for 30 min. Helium was used as the carrier gas with a flow rate of 1 ml/min and a controlled initial pressure of 93.2 kPa at 120 °C. Nitrogen gas and air were makeup gases. The injection volume was 1 μl with a split ratio of 100:1. FAs were identified by comparing their retention times with external standards. The FA content was given in percentage area.

Slip melting point

Slip melting points (SMP) were analyzed following the American Oil Chemists’ Society (AOCS) official method Cc 3–25 (AOCS 2003). A capillary glass tube with 1 mm internal diameter (Vitrex, Denmark) was dipped in melted sample until the sample rose about 10 mm high in the tube. The sample tube was then stored for 16 h in a refrigerator at 4 °C. Then, the sample tube was attached to a thermometer and immersed in a water bath set at 20 °C. The bath temperature was increased at a rate of 1 °C/min, slowing the rate to 0.5 °C/min as the expected SMP was approached. SMP was expressed as the temperature at which the fat softened and became sufficiently fluid to slip in an open capillary tube.

Thermal analysis

The crystallization and melting thermograms were determined by differential scanning calorimetry (DSC) (PerkinElmer DSC 8000, PerkinElmer Co., Norwalk, CT, USA) following the AOCS official method Cj 1–94 (AOCS 2003). The DSC was calibrated with indium (m.p. 156.6 °C) as a reference standard. Fat samples (3–5 mg) were placed in an aluminum pan (20 μl capacity) and hermetically sealed. An empty pan served as reference. The samples were heated from room temperature to 80 °C at 30 °C/min and maintained for 10 min and then cooled to − 60 °C at 5 °C/min to obtain the crystallization thermograms. After holding at − 60 °C for 30 min, samples were reheated to 80 °C at 5 °C/min to obtain the melting thermograms.

Solid fat content

SFC was determined by pulse-nuclear magnetic resonance (p-NMR) spectrometer (Minispec-mq20, Bruker, Karlsruhe, Germany) using the AOCS official method Cd 16b-93 (AOCS 2003). Three SFC standards (0, 31.4, and 75.6%) were used for calibration. Molten fat samples were put into 10 mm outer diameter p-NMR tubes to a height of ~ 4 cm. The fat samples were melted at 80 °C for 15 min and then placed at 60 °C for 10 min, followed by storing at 0 °C for 60 min, and at each measuring temperature for 30 min prior to analysis. The measurement of SFC was performed at 5 °C intervals from 10 to 45 °C. The compatibility between CBS and KSF was presented in an iso-solid phase diagram of the fat blends constructed using the SFC data from the p-NMR (Metin and Hartel 2005) at the SFC range of 5–60%.

Isothermal crystallization

The study of isothermal crystallization at 25 °C was performed by the DSC. The samples were heated from room temperature to 80 °C at 30 °C/min and held for 10 min, then cooled to 25 °C with a cooling rate of 20 °C/min and held for 2 h to observe crystallization. Partial areas under the exothermic curve were calculated as percentages of solid fraction during the crystallization period by integration (Rashid et al. 2012). The data was then fitted to the Gompertz equation (Eq. 1):

$$F(t)=A{e}^{-e((\mu e(\tau -t)/A)+1)}$$ 1

where F(t) is the relative percentage of solid fraction crystallized at time t, A is the maximum percentage of solid fraction when t approaches infinity, μ is the maximum crystallization rate (%/min) and τ is the crystallization induction time (min). Estimation of the Gompertz parameters was performed on the experimental data by non-linear regression and the parameters were used to describe the crystallization kinetics of the samples (Foubert et al. 2004).

Crystal morphology

Observation of the crystal network microstructure was performed by the polarized light microscopy (ZEISS Primo Star, Carl Zeiss Microscopy, LLC, Jena, Germany) equipped with a digital camera (Canon EOS 700D, Canon, Inc., Taipei, Taiwan). At 25 °C, the samples contained too much crystallized mass that made it difficult to obtain clear images of the samples, hence, 30 °C was used in this part of the study. The samples were melted at 80 °C for 10 min. 10 μl of each melted sample was placed on a preheated glass slide and covered with a preheated coverslip. The samples were incubated at 30 °C for 24 h prior to observation. A magnification of 100× was used.

Polymorphism

Polymorphic form was determined by an X-ray diffractometer (MiniFlexII, Rigaku, Tokyo, Japan). The samples were melted at 80 °C for 10 min and crystallized at 25 °C for 24 h. The crystallized samples were put on a sample holder before analysis. Scans were made in wide angle X-ray scattering (WAXS) from 15°2 to 30°2θ with a scan rate and a step width of 2°2θ/min and 0.02°2θ, respectively.

Statistical analysis

All experiments were performed in triplicate. The results were analyzed by analysis of variance with least significant difference (ANOVA/LSD) at 95% confidence interval.

Results and discussion

Fatty acid composition

FA composition of fats and oils has strong influence on the properties such as melting point, crystallization behavior and polymorphism. The FA composition of the CBS, KSF and CBS-KSF blends is shown in Table 1. CBS and KSF were composed mainly of saturated FAs (> 95%). The major FAs in CBS were lauric (~ 44%), stearic (~ 30%), myristic (~ 14%) and palmitic (~ 9%) acids. KSF consisted mainly of myristic (~ 47%) and lauric (~ 44%) acids. The amount obtained were within the range reported in the literature for KSF (Bandelier et al. 2002; Sonwai and Ponprachanuvut 2012; Sonwai et al. 2015). As the content of KSF in the CBS-KSF blends increased, the content of myristic and oleic acids increased (p < 0.05) but the content of stearic, palmitic and capric acids decreased (p < 0.05). The content of lauric acid remained unchanged (p > 0.05) due to the fact that CBS and KSF had similar lauric acid content.

Table 1.

Fatty acid composition and slip melting point (SMP) of CBS, KSF and CBS-KSF blends

	CBS	10%KSF	20%KSF	30%KSF	40%KSF	60%KSF	80%KSF	KSF
Fatty acids (area %)
Capric acid	3.0 ± 0.1a	2.9 ± 0.1ab	2.8 ± 0.1abc	2.7 ± 0.1bc	2.6 ± 0.1d	2.4 ± 0.1e	2.2 ± 0.1f	1.9 ± 0.1g
Lauric acidns	44.1 ± 0.2	44.1 ± 0.1	44.1 ± 0.0	44.1 ± 0.1	44.1 ± 0.1	44.2 ± 0.3	44.2 ± 0.4	44.3 ± 0.6
Myristic acid	13.7 ± 0.2h	17.0 ± 0.1g	20.3 ± 0.1f	23.9 ± 0.2e	26.9 ± 0.3d	33.5 ± 0.5c	40.1 ± 0.7b	47.0 ± 0.9a
Palmitic acid	8.7 ± 0.2a	7.8 ± 0.2b	7.0 ± 0.2c	6.2 ± 0.1d	5.3 ± 0.1e	3.7 ± 0.1f	2.0 ± 0.1g	0.3 ± 0.0h
Stearic acid	29.8 ± 0.4a	26.9 ± 0.4b	23.9 ± 0.3c	21.0 ± 0.3d	18.0 ± 0.2e	12.1 ± 0.2f	6.2 ± 0.1g	0.3 ± 0.0h
Oleic acid	nd	0.4 ± 0.0g	0.8 ± 0.1f	1.2 ± 0.1e	1.6 ± 0.1d	2.4 ± 0.2c	3.2 ± 0.2b	3.9 ± 0.2a
Arachidic acid	0.2 ± 0.0b	0.4 ± 0.1ab	0.6 ± 0.3ab	0.8 ± 0.4ab	1.0 ± 0.5ab	1.4 ± 0.8ab	1.8 ± 1.1ab	2.1 ± 1.2a
Behenic acidns	0.3 ± 0.4	0.3 ± 0.4	0.3 ± 0.4	0.3 ± 0.3	0.3 ± 0.3	0.3 ± 0.2	0.3 ± 0.1	0.2 ± 0.1
Ligoceric acidns	0.3 ± 0.3	0.3 ± 0.3	0.2 ± 0.3	0.2 ± 0.2	0.2 ± 0.2	0.1 ± 0.1	0.1 ± 0.1	nd
SMP (°C)	39.1 ± 0.3a	38.1 ± 0.4b	37.5 ± 0.1c	35.7 ± 0.5d	34.7 ± 0.4e	34.2 ± 0.2e	35.2 ± 0.2d	37.3 ± 0.1c

Different letters in the same row are significantly different (p < 0.05). ns, not significantly; nd, not detected

Slip melting point

SMP of all fat samples are presented in Table 1. The SMP of CBS and KSF were 39.1 and 37.3 °C (p < 0.05), respectively. All CBS-KSF blends exhibited lower SMP than CBS (p < 0.05). SMP of 10% KSF blend (38.1 °C) was higher than that of KSF (p < 0.05) whereas SMP of 20% KSF blend (37.5 °C) was similar to KSF (p > 0.05). On the contrary, SMP of fat blends with 30–80% KSF (34.2–35.7 °C) were lower than that of KSF (p < 0.05), suggesting that these blends were softer than both CBS and KSF due to some degree of incompatibility between CBS and KSF as will be demonstrated later.

Crystallization and melting thermograms

The DSC crystallization and melting thermograms are given in Fig. 1. CBS started to crystallize at 29.7 °C and exhibited two broad exothermic peaks; a small peak at 26.1 °C and a main peak at 17.1 °C (Fig. 1a and Table S1). This suggests that CBS crystallized into two distinct fractions of higher- and lower-melting TAGs. The crystallization thermogram of KSF started at 24.8 °C and exhibited one sharp exothermic peak at 17.5 °C due to the fact that ~ 90% of the FAs found in KSF were medium chain (e.g., lauric and myristic acids). With the addition of KSF to CBS, the crystallization onset temperature decreased from 29.7 °C for CBS to 22.6 °C for 80% KSF blend (p < 0.05) owing to the decrease in the content of long-chain FAs in the blends (i.e., stearic and palmitic acids). With 10–60% KSF, the blends exhibited two crystallization peaks, the locations of which moved gradually towards lower temperatures as the KSF content in the blends increased, suggesting that the addition of KSF affected both lower- and higher-melting fractions of CBS and that the difference in FA and TAG compositions of these two fats retarded the arrangement of TAGs during the crystallization (Jahurul et al. 2014). With 80% KSF, the blend also showed two crystallization peaks, however, both peaks moved slightly towards higher temperature (from 17.7 and 14.9 °C for 60% KSF blend to 19.2 and 16.1 °C, respectively, for 80% KSF blend, p < 0.05). These results were different from our previous work which showed that the addition of KSF to a different CBS gradually changed the locations of the crystallization peaks towards higher temperatures (Podchong et al. 2020). The crystallization enthalpy, which is affected by the amount of crystallized mass, the microstructure, and the stability of the crystalline polymorph formed (Ribeiro et al. 2013), increased with the content of KSF in the blend (from 101.7 J/g for 10% KSF blend to 113.4 J/g for 80% KSF blend, p < 0.05).

Fig. 1.

DSC crystallization (a) and melting (b) thermograms of CBS, KSF and CBS-KSF blends

The melting thermogram of CBS displayed one very broad endothermic region with multiple melting peaks locating at between 20 and 45 °C (Fig. 1b), due to the high variety in its FA and TAG compositions (Tan and Che Man 2000). The fat showed a main melting peak at 33 °C and a melting completion temperature (T_mc) of 45.1 °C (Table S1). The melting range of this CBS was much wider than the melting range of the CBS used in our previous study (Podchong et al., 2020). KSF exhibited a sharp endothermic peak at 39.6 °C and T_mc at 41.8 °C. The addition of KSF resulted in narrower endothermic peak ranges than CBS. The thermograms showed multiple melting peaks in blends with up to 40% of KSF. At higher KSF content, the thermograms exhibited only one endothermic peak. As the content of KSF increased, the location of the melting peaks moved to higher temperatures but with a concurrent decrease in T_mc from 44.6 °C for 10% KSF blend to 39.9 °C for 80% KSF blend (p < 0.05) which was in line with the SMP results (Table 1). In general, the melting enthalpy increased with the content of KSF in the blends (p < 0.05). This was due to an increase in the proportion of myristic acid which led to the blends containing mostly lauric and myristic acids, thereby reducing the packing imperfection and resulting in a more order molecular packing (Himawan et al. 2006).

Solid fat content and iso-solid diagram

SFC can define properties and functionality of confectionery fats. The SFC obtained at temperatures below 25 °C indicate hardness of the fats, while the values measured above 30 °C relate to their melting behavior and mouthfeel properties (Norberg 2006; Torbica et al. 2006; Jahurul et al. 2014). The SFC profiles of all samples are shown in Fig. 2a. KSF showed the highest SFC within the temperature range of 10–39 °C. The fat began to melt rapidly at ~ 25 °C with a rapid drop in SFC value (Bandelier et al. 2002) which then reached 0% at 40 °C, indicating a complete melting at this temperature. CBS showed a much less steep SFC curve than KSF and this was consistent with its broader DSC melting thermogram (Fig. 1b). CBS started to melt rapidly at ~ 18 °C and did not melt completely until the temperature reached 45 °C. With the addition of 10–40% KSF to CBS, SFC of the blends increased with increasing KSF content at the temperatures < 32.5 °C. This effect was similar to what was observed when KSF was added to a different CBS (Podchong et al. 2020). At 32.5 °C, there was an SFC cross over (arrow) where the blend with highest KSF content (40%) became the one with lowest SFC and melted completely at lower temperature and vice versa. At 35 °C, SFC values of these blends were lower than those of both CBS and KSF, suggesting that they were softer than the original fats and that at 35 °C CBS and KSF were incompatible at these blend ratios. In addition, at 40–45 °C the SFC of these CBS-KSF blends decreased with the increasing of KSF content. The higher SFC in the blends with higher CBS content (or lower KSF content) within this temperature range was due to the higher content of saturated long-chain FAs (palmitic and stearic acids) presented in the blends (Maheshwari and Yella Reddy 2005; Zaidul et al. 2007; Jahurul et al. 2014). The blend with 60% KSF displayed higher SFC than the 10–40% KSF blends at temperatures < 35 °C whereas the blend with 80% KSF exhibited highest SFC among all blends at temperatures ≤ 36 °C.

Fig. 2.

a Changes in solid fat content (SFC) with temperature of CBS, KSF and CBS-KSF blends. The SFC curve of cocoa butter (CB) was taken from the literature (Norberg 2006), b Iso-solids diagram for binary mixtures of CBS and KSF. SFC in each line are 5% intervals from 5% SFC (top line) to 60% SFC (bottom line)

The temperature at which the SFC of a fat begins to reduce sharply is associated with the resistance of the fat to heating (Torbica et al. 2006). As mentioned above, SFC of KSF and CBS began to decrease rapidly at ~ 25 and ~ 18 °C, indicating that KSF would have higher temperature resistance than CBS. As the KSF content in the blends increased, the temperature at which the SFC began to decrease rapidly also increased (from ~ 19 °C for 10% KSF blend to ~ 22.5 °C for 80% KSF blend), suggesting an increase in heat resistance of the blends.

The blends containing ≥ 30% KSF exhibited the SFC of ≤ 3% at body temperature, indicating that they would give no waxy mouthfeel. On the contrary, the SFC of 10% and 20% KSF blends were 6.5 and 5%, respectively, at body temperature, suggesting that they would give waxy mouthfeel. In addition, a comparison between the SFC curves of all CBS-KSF blends and that of CB showed that the blends with 30–40% KSF displayed most similar SFC curves to CB, especially within 20–35 °C range.

The CBS-KSF iso-solid diagram is shown in Fig. 2b. Each iso-solid line represents a unique SFC value, with the topmost line being the 5% SFC line and the subsequent lines representing 5% SFC increments up to 60% SFC. The compatibility between different fats has a strong effect on the processing conditions, qualities and shelf lives of their blends. The diagram portrayed a eutectic behavior at the temperatures > 32 °C as can be observed with 5–15% SFC lines where the SFC of the blends fell below the SFC of either of the two component fats at any given temperature. This indicated the incompatibility between CBS and KSF due to the difference in the TAG compositions of KSF, which were mainly saturated medium-chain TAGs, and CBS, which contained mainly long-chain FAs (palmitic and stearic acids), resulting in a rather weak co-crystallization between the two fats (da Silva et al. 2017). The observed eutectic behavior was useful in this case because it led to a decrease in the SFC of the blends at body temperature.

Isothermal crystallization

The isothermal crystallization curves of the samples obtained at 25 °C are shown in Fig. 3. The crystallization kinetic parameters acquired from the Gompertz equation (Eq. 1) fitting are presented in Table S2. All samples exhibited sigmoidal curve pattern containing an induction period with no crystallization followed by rising of solid fraction due to nucleation and crystal growth until the maximum solid fraction was reached (Wang et al. 2011). CBS and KSF started to crystallize at 1.9 and 0.4 min and reached the maximum solid fraction at ~ 7 and ~ 3 min with crystallization rate of 47.3 and 84.1%/min (p < 0.05), respectively. The higher crystallization rate of KSF was in agreement with its sharp DSC crystallization peak (Fig. 1a). The blends with the addition of 10–30% KSF exhibited the crystallization curves with similar characteristics to that of CBS but with significantly increased crystallization rates (56–57.5%/min) from that of CBS (p < 0.05). Ribeiro et al. (2013) reported that the crystallization of CB was accelerated with the addition of hard fats such as fully hydrogenated palm oil and fully hydrogenated soybean oil. On the contrary, when 40–80% KSF was added, the crystallization of the blends was decelerated with a decrease in the crystallization rates to 26.2–40.3%/min due to the significant decrease in the content of palmitic and stearic acids in those blends from that of CBS. This resulted in the competition between similar TAG molecules (medium-chain TAGs) at the crystal surface sites, hence the growth rates were slow (Walstra et al. 2001).

Fig. 3.

Fitting of Gompertz model to the crystallization data of CBS, KSF and CBS-KSF blends. The data was obtained during isothermal crystallization at 25 °C

Crystal morphology

The microstructure information of fat crystals including shape, size, packing density and interactions of the crystalline particles affect the macroscopic properties of the fat system (Shi et al. 2005; Domingues et al. 2015). The crystal morphology of the samples is shown in Fig. 4. CBS existed as small and densely-packed granular crystals with an average size of 21 ± 2.5 μm. The crystal morphology of KSF was spherulites with a significantly larger average crystal size (787 ± 138 μm) and less crystal number than CBS. Blending of CBS with 10–30% KSF led to no significant changes in the crystal microstructure in terms of shape, size and packing density compared to that of CBS. The blend containing 40% KSF showed spherulites (305 ± 59 μm) consisting of fine needle-like crystals with some dark fields of liquid oil between the fat crystals. With 60 and 80% KSF, the fat blends exhibited extra-large spherulites similar to the morphology of KSF crystals. The isothermal crystallization study showed that the blends with 40–80% KSF had lower crystallization rates compared to the original fats (Table S2), indicating that the crystallization of these blends proceeded more slowly and this allowed the blends to crystallize into larger crystals. In addition, the crystal morphology of these blends also showed the impingement effect with the colliding of the neighboring crystal faces, resulting in the deceleration of crystal growth and crystallization process (Himawan et al. 2006).

Fig. 4.

Crystal morphology of CBS, KSF and CBS-KSF blends crystallized at 30 °C for 24 h

Polymorphism

Fats can crystallize into different crystal polymorphs. The three basic polymorphs are α, β′ and β. The different crystal lattice structures of these polymorphs influence the macroscopic physical properties of the fat-based products (D’Souza et al. 1990; deMan and deMan 2001; Sato 2001). The X-ray diffraction patterns in WAXS of all samples are shown in Fig. 5. CBS had two d-spacings at 3.78 and 4.23 Å, a characteristic of β′ polymorph (D’Souza et al. 1990). KSF also crystallized into β′ form with two strong d-spacings at 3.84 and 4.27 Å and two medium-strong d-spacings at 4.06 and 4.44 Å. All CBS-KSF blends also crystallized into β′ structure with the peaks at 3.78 and 4.23 Å that appeared to shift towards 3.84 and 4.27 Å, respectively, as the KSF content in the blend increased. This effect was in agreement with what was observed when KSF was added to a different CBS (Podchong et al. 2020). Another previous study also found that a CBS obtained from KSF-coconut oil blend crystallized into β′ polymorph (Sonwai et al. 2015). Lauric fats generally crystallize into the β′ form directly from the melts upon cooling. The stable β′ form confers pleasant functionality, e.g., good aeration and creaming properties for margarines and shortenings, and no tempering need for confectionery fats (Shukla 2005; Norberg 2006; Domingues et al. 2015; Yamoneka et al. 2015).

Fig. 5.

Wide angle X-ray diffraction patterns of CBS, KSF and CBS-KSF blends after crystallization at 25 °C for 24 h

The results from this study have shown that addition of 10–80% KSF to CBS improved the heat resistance of the CBS. However, 10–20% KSF gave too high SFC values at body temperature which would lead to waxiness. Adding 30–40% KSF increased heat resistance further from 10 to 20% KSF with a decrease in SFC at 37 °C and SFC curves most similar to CB. However, 40% KSF led to a significant decrease in the crystallization rate and a substantial increase in the crystal size from that of CBS. Adding 60–80% KSF although increased the heat resistance significantly and the blends melted completely at the body temperature, their crystallization rates were greatly reduced. In addition, adding that high content of KSF might alter the organoleptic properties of the final product as perceived by the consumers. Therefore, 30% KSF was the best quantity to be added to the CBS to enhance its crystallization and melting properties.

Conclusion

KSF was added to a commercial CBS in order to improve the crystallization and melting characteristics of the CBS. The addition of 30% KSF resulted in a CBS-KSF blend with an improved melting characteristic and a reduced SFC value at the body temperature. The crystallization rate also increased with no significant effect on crystal microstructure and polymorphism. In addition, the blend exhibited an SFC curve similar to that of CB. Therefore, addition of 30% KSF to the CBS is recommended for industrial use to obtain compound chocolate with better quality.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CB

Cocoa butter

CBS

Cocoa butter substitute

FA

Fatty acid

FAME

Fatty acid methyl esters

GC

Gas chromatography

KSF

Krabok seed fat

SFC

Solid fat content

SMP

Slip melting point

TAG

Triacylglycerol

WAXS

Wide angle X-ray scattering

Author contributions

PP conceived, carried out the experiments and wrote the manuscript; PI carried out the experiments; UK edited the manuscript; SS supervised the work and edited the manuscript.

Funding

The Department of Food Technology, Silpakorn University.

Declarations

Conflict of interest

The authors have declared no conflicts of interest.