Use of locust bean flour as a substitute for cocoa in the production of chocolate spread: Quality attributes and storage stability

2.1. Materials

In producing chocolate spreads, cocoa butter, palm oil, powdered sugar, milk powder, whey powder, hazelnut paste, starch, lactose, cocoa, soy lecithin, hazelnut flavor, chocolate flavor, and vanillin were used. These materials were obtained from Nutpa Gıda Sanayi ve Ticaret A.Ş (Ordu, Turkey), where the chocolates were produced. The LBF used in the chocolate production was purchased from Aşçı Baharatları (Ordu, Turkey). All the chemicals used in the analyses were of analytical purity and obtained from Sigma-Aldrich (St. Louis, MO, USA).

2.2. Preparation and storage of chocolate spreads

Chocolate spreads were produced in Nutpa Gıda Sanayi ve Ticaret A.Ş., according to the procedure suggested by the company, using a ball mill with a 20-kg mixer. Four different chocolate spread formulations were included in the experimental design, and the first formulation, which contained 4.5% cocoa but not LBF, was taken as the control formulation (CON). The second formulation was composed of 3.0% cocoa + 1.5% LBF (CF15), the third formulation was 1.5% cocoa + 3.0% LBF (CF30), and the fourth formulation was 4.5% LBF (CF45) (Table 1). For the preparation of chocolate spreads, the cocoa butter was mixed in a ball mill at 42 °C until melted, and palm oil and lecithin, then sugar, milk powder, cocoa, whey powder, starch, hazelnut paste, and lactose were added. After mixing for one hour, the thickness of the mixture was measured, and when the particle size fell below 30 µm, hazelnut flavor, chocolate flavor, and vanillin were added. The chocolate spreads, prepared for 3 hours, were mixed for about 10 more minutes, filled into 350g glass jars, and after cooling to room temperature, stored for 12 weeks in warehouses at 22 and 35 °C. The chocolate spreads were analyzed for moisture, protein, fat, and sensory properties on the day of production, texture, water activity, color, free fatty acidity, and peroxide value at the beginning of storage and at the 4th, 8th, and 12th weeks.

2.3. Proximate composition and sensory evaluation

The proximate composition, including moisture, protein, and fat contents in CON, CF15, CF30, and CF45 formulations was measured using the official standard method (AOAC, 2000AOAC 2000. Official methods of analysis (16th ed.). Washington, DC: Association of Official Analytical Chemists. AOAC International.). Briefly, moisture percentage was calculated with a weight loss experiment in the sample maintained in the oven at 103 ± 2 °C until constant weight. According to Kjeldahl’s total nitrogen method, protein content was determined by multiplying the total nitrogen content by 6.25. For the determination of fat content, the samples were subjected to a liquid-solid extraction using diethyl ether in an extractor apparatus at 60 °C for six h. The fat content was obtained based on gravimetric difference. Proximate analyses were carried out in triplicate.

A panel consisting of faculty members and research assistants from the Food Technology Department of Ondokuz Mayıs University performed the sensory evaluation of chocolate spreads. The panel was asked to rate the nine attributes of the chocolate samples: appearance, granularity and roughness, spreadability, stickiness in the mouth, oiliness, odor, sweetness, color, and overall acceptability. Each attribute was scored from 1 (unacceptable) to 9 (excellent). All sensory evaluations were performed under fluorescent light and at room temperature.

2.4. Texture and water activity (a_w) measurement

Within the scope of the textural properties of chocolate spreads, the hardness and spreadability properties were determined using the TA XT Plus Texture Analyzer (Stable Micro Systems, UK) device equipped with an HDP/SR probe and connected with the computer program (Texture Expert Exceed 2.3., Stable Micro System, Godalming, Survey, UK). Analyses were made on a 15 g sample at room temperature (21 ± 2 °C), and 3 mm/s test speed and hardness values were given in kg and spreadability values in kg.s. Analyzes were performed in three replicates for each chocolate formulation.

The water activity of the chocolate samples was determined at 25 ºC using the Aqualab Dewpoint Water Activity Meter 4TE USA. Measurements were made in 3 replicates for each chocolate formulation.

2.5. Determination of color properties

Color properties were measured on the surface of chocolate formulations at room temperature using a colorimeter (ColorFlex EZ spectrometer, Reston, VA, USA). Three readings were taken from each chocolate sample. Color measurement included CIE L*, a*, and b* parameters, where L* represents lightness with a scale from 0 (black) to 100 (white), a* represents redness with a scale from -60 (green) to +60 (red), and b* represents yellowness with a scale from -60 (blue) to +60 (yellow). Also, chroma (C*) and hue (h*) were calculated from a* and b* parameters according to the following Equations (Becerra et al., 2023Becerra LD, Quintanilla-Carvajal MX, Escobar S, Ruiz RY. 2023. Correlation between color parameters and bioactive compound content during cocoa seed transformation under controlled process conditions. Food Biosci. 53, 102526. https://doi.org/10.1016/j.fbio.2023.102526):

2.6. Determination of free fatty acid (FFA) and peroxide value (PV)

Firstly, the fats were extracted from the chocolate samples using a chloroform/methanol solvent system (2/1, v/v), and the lipid extract was used in FFA and PV analyses. The amount of FFA was determined for all samples according to the official standard method Ca 5-40 (AOAC, 2000AOAC 2000. Official methods of analysis (16th ed.). Washington, DC: Association of Official Analytical Chemists. AOAC International.). For this purpose, 5 g of lipid extract was dissolved in 50 mL of solvent mixture (diethyl ether/ethanol, 1/1, v/v) and titrated with 0.1 M KOH solution using a phenolphthalein indicator until the pink color disappeared. The percentage FFA content was calculated according to the Equation:

where % FFA represents the percentage of free fatty acids, V is the volume of solvent used, M corresponds to the molarity of the KOH solution, and m is the mass of the lipid extract. The results were expressed as % oleic acid, and all analyses were conducted in triplicate.

The PV was determined according to the official standard method Cd 8-53 (AOAC, 2000AOAC 2000. Official methods of analysis (16th ed.). Washington, DC: Association of Official Analytical Chemists. AOAC International.) and expressed as meq O₂/kg lipid. Briefly, 2 g of lipid extract were dissolved with 25 mL of a glacial acetic acid and chloroform mixture (3/2, v/v) and shaken vigorously to achieve complete dissolution. Then, 1 mL of saturated potassium iodide solution was added to the mixture and kept in the dark for 5 minutes. At the end of that time, 75 mL of distilled water and 1 milliliter of starch solution (1%, w/v) were added as indicator, and the oxidized iodine released from the potassium iodide as a result of the reaction was titrated with 0.002 M sodium thiosulphate solution and the POV was calculated according to the following Equation:

where V is milliliters of sodium thiosulfate solution (corrected to account for the blank test), M corresponds to the molarity of the sodium thiosulfate solution, and m is the mass of the lipid extract. All analyses were carried out in triplicate.

2.7. Statistical analysis

The whole trial was replicated twice, with each replication corresponding to a different production day. The data were analyzed with the SPSS 21 statistical software (IBM, Chicago, IL, USA) and normal distribution and homogeneity of variances were first checked. While data from texture, a_w, color, FFA, and PV analyses were analyzed using a randomized complete block design, those from proximate composition and sensory evaluation were analyzed by one-way ANOVA. Duncan’s multiple comparison tests were used to evaluate the differences between the mean values found to be significant (p < 0.05). All results were expressed as mean value ± standard deviation.

3.1. Proximate composition and sensory evaluation

The proximate composition of chocolate spreads produced with different proportions of LBF is presented in Table 2. Replacing cocoa with LBF significantly affected the chocolates’ moisture and fat content (p < 0.05), whereas its effect on the protein content was not significant (p > 0.05). The highest moisture content was determined in the CF15 chocolates as 0.73% (p < 0.05), and the fat content in the chocolates partially decreased as the addition of LBF increased (p < 0.05). These differences in the moisture and fat content of the chocolate formulations could be due to the proximate composition of LBF. Despite these differences, the proximate composition of all the chocolate formulations was similar to those reported by Cağındı and Otles (2007)Cağındı O, Otles S. 2007. Determination of some physical and sensory properties of milk, dark and white chocolate at different storage temperatures. EJPAU, #01. in dark chocolate and by Ali et al. (2021)Ali AME, Shekib LA, Elshimy NM, Sharara MS. 2021. Producing and quality attributes of low calories dark chocolate using different intense sweeteners and wheat fiber isolate. Am. J. Food Sci. Technol. 9, 1–7. https://doi.org/10.12691/ajfst-9-1-1. in low-calorie dark chocolate using different intense sweeteners and wheat fiber isolate.

The popularity of hedonic foods, like chocolate, mainly depends on their sensory properties, which constitute the key to the acceptance of food products on the market (Roda and Lambri, 2019Roda A, Lambri M. 2019. Changes in antioxidants and sensory properties of Italian chocolates and related ingredients under controlled conditions during an eighteen-month storage period. Nutrients 11, 2719. https://doi.org/10.3390/nu11112719.). Replacing cocoa with LBF significantly affected the chocolates’ appearance, odor, sweetness, color, and overall acceptability scores (p < 0.05). In contrast, its effect on other sensory attributes was insignificant (p > 0.05), as shown in Figure 1. Generally, the highest appearance, odor, sweetness, color, and overall acceptability scores were observed in the CON and CF15 containing 1.5% LBF. The scores decreased with more LBF addition. However, the use of up to 3.0% LBF did not affect the appearance, odor, sweetness, color, and overall acceptability scores compared to the CON formulation, and the lowest appearance, odor, sweetness, color, and overall acceptability scores were noted for the CF45 formulation due to an astringent taste and a lighter color formation after high-level LBF addition. Because the whole LB fruit contains a high level of tannins, causing excess astringency, this factor limits its use in human consumption (Avallone et al., 1997Avallone R, Plessi M, Baraldi M, Monzani A. 1997. Determination of the chemical composition of carob (Ceratonia siliqua): protein, fat, carbohydrates, and tannins. J. Food Compost. Anal. 10. 166–172. https://doi.org/10.1006/jfca.1997.0528.). Similarly, Akdeniz et al. (2021)Akdeniz E, Yakışık E, Rasouli Pirouzian H, Akkın S, Turan B, Tipigil E, Toker OS, Ozcan O. 2021. Carob powder as a cocoa substitute in milk and dark compound chocolate formulation. J. Food Sci. Technol. 58, 4558–4566. https://doi.org/10.1007/s13197-020-04943-z. reported that chocolates produced using 40% LBF as a cocoa substitute had the highest odor, aroma, appearance, sweetness, melting, and overall acceptability scores, but the use of higher rates of LBF reduced sensory acceptability. Also, Rosa et al. (2015)Rosa CS, Tessele K, Prestes RC, Silveira M, Franco F. 2015. Effect of substituting cocoa powder for carob flour in cakes made with soy and banana flour. Int. Food Res. J. 22, 2111–2118. observed that the cakes with up to 75% replacement of cocoa powder by LBF showed no difference in flavor, odor, or texture, demonstrating that replacement up to this level did not influence sensory attributes. In another study, Aydın (2012)Aydın N. 2012. Keçiboynuzu unu ilavesinin bisküvinin bazı kalite kriterlerine etkisi. Pamukkale Üniversitesi Fen Bilimleri Enstitüsü, Yüksek Lisans Tezi, Denizli, 42 p. reported that adding LBF at up to 20% to biscuits did not negatively affect general appeal and taste. Still, the sensory scores generally decreased at higher rates when adding LBF. The current study and literature findings show that the use rate of LBF alone or as a cocoa substitute varies depending on the product type, and up to 3% LBF can be used with minimal compositional and sensory changes as a cocoa substitute in the production of chocolate spread.

3.2. Texture and water activity (a_w)

The effect of the formulation, temperature, and time on the hardness, spreadability, and a_w values for chocolate spreads produced with different proportions of LBF is shown in Table 3. As seen, LBF formulation and storage time affected the hardness (p < 0.05), whereas storage temperature effect on hardness was not significant (p > 0.05). Among the chocolate formulations, the highest hardness values were determined for CF30 and CF45 formulations, but the differences between CF30 and CF45 formulations and CON formulation were insignificant (p > 0.05). The lowest hardness value was determined for the CF15 formulation. As a result, adding LB at a high ratio to chocolate caused a slight increase in hardness. Hardness, which depends on the content and type of fat, sugar, and cocoa particles used in chocolate production (Konar et al., 2014Konar N, Özhan B, Artık N, Dalabasmaz S, Poyrazoglu E S. 2014. Rheological and physical properties of inulin-containing milk chocolate prepared at different process conditions. CYTA – J. Food 12, 55–64. https://doi.org/10.1080/19476337.2013.793214.; Barisic et al., 2021Barisic V, Petrovic J, Loncarevic I, Flanjak I, Subaric D, Babic J, Milicevic B, Doko K, Blazic M, Ackar D. 2021. Physical properties of chocolates enriched with untreated cocoa bean shells and cocoa bean shells treated with a high-voltage electrical discharge. Sustainability 13, 2620. https://doi.org/10.3390/su13052620.), is one of the essential mechanical properties that affects the sensory acceptability of chocolate (Machalkova et al., 2015Machalkova L, Hrivna L, Nedomova S, Juzl M. 2015. The effect of storage temperature on the quality and formation of blooming defects in chocolate confectionery. Potravinarstvo 9, 39–47. https://doi.org/10.5219/425.; Hrivna et al., 2021Hrivna L, Machalkova L, Buresova I, Nedomova S, Gregor T. 2021. Texture, color, and sensory changes occur in chocolate bars with filling during storage. Food Sci. Nutr. 9, 4863–4873. https://doi.org/10.1002/fsn3.2434). In the present study, LBF replaced some of the cocoa mass in the chocolate samples, resulting in lower total fat content (Table 2). This may be the reason for the increase in the hardness of chocolates containing high ratios of LBF. In addition, the differences in hardness can also be associated with the internal structure and more intermolecular bonds in chocolate samples prepared with LB powder. Since LB powder addition enhanced the fiber and carbohydrates of the produced samples, these compounds may act with the other compounds in chocolate and make more intermolecular bonds (Akdeniz et al., 2021Akdeniz E, Yakışık E, Rasouli Pirouzian H, Akkın S, Turan B, Tipigil E, Toker OS, Ozcan O. 2021. Carob powder as a cocoa substitute in milk and dark compound chocolate formulation. J. Food Sci. Technol. 58, 4558–4566. https://doi.org/10.1007/s13197-020-04943-z.). As in our findings, Akdeniz et al. (2021)Akdeniz E, Yakışık E, Rasouli Pirouzian H, Akkın S, Turan B, Tipigil E, Toker OS, Ozcan O. 2021. Carob powder as a cocoa substitute in milk and dark compound chocolate formulation. J. Food Sci. Technol. 58, 4558–4566. https://doi.org/10.1007/s13197-020-04943-z. also reported that adding LBF affects chocolates’ hardness values. The hardness values for chocolates produced using LBF increased until the 8th week of storage, then decreased, but this decrease was insignificant (p > 0.05).

Spreadability is a key textural attribute affecting consumer acceptance (Aydın and Özdemir, 2017Aydın S, Özdemir Y. 2017. Development and characterization of carob flour based functional spread for increasing use as nutritious snack for children. J. Food Qual. 2017, 1–7. https://doi.org/10.1155/2017/5028150.). As observed in Table 3, only LBF formulation affected the spreadability values for chocolate spreads (p < 0.01). As in the hardness values, the highest spreadability values were determined for the CF30 and CF45 formulations. However, the differences between the spreadability values of CF45 and the CON formulations were insignificant (p > 0.05), and the lowest values were obtained for the CF15 formulation. This situation could be attributed to the chemical composition of the spreadable chocolates.

LBF formulation and time affected the a_w values for chocolate spreads (p < 0.01), whereas the storage temperature effect on a_w was not significant (p > 0.05) (Table 3). The CF15 and CF30 formulations showed higher a_w values than the CON formulation (p < 0.05), and the differences between the a_w values of CF45 and CON formulations were not significant (p > 0.05). The highest a_w value determined at the beginning of the storage decreased to the lowest value in the 4th week of storage and tended to increase again from the 12th week. Despite these changes, the a_w values during storage did not reach 0.6, thus prohibiting microbial growth. However, several factors, such as the raw materials used, the surface area of the materials, and the temperature and humidity of refining and conching, may influence this parameter. Chocolate is protected from external water by its fatty surface, which makes it difficult to uptake moisture. However, the presence of amorphous sugars in chocolates should be considered. Amorphous sugar is in metastable form and tends to crystallize under several factors, mainly temperature and moisture (Rossini et al., 2011Rossini K, Norena CPZ, Brandelli A. 2011. Changes in the color of white chocolate during storage: potential roles of lipid oxidation and non-enzymatic browning reactions. J. Food Sci. Technol. 48, 305–311. https://doi.org/10.1007/s13197-010-0207-x.).

3.3. Color properties

The color of chocolate is the first characteristic that influences consumers and their desire to consume chocolate (Barisic et al., 2021Barisic V, Petrovic J, Loncarevic I, Flanjak I, Subaric D, Babic J, Milicevic B, Doko K, Blazic M, Ackar D. 2021. Physical properties of chocolates enriched with untreated cocoa bean shells and cocoa bean shells treated with a high-voltage electrical discharge. Sustainability 13, 2620. https://doi.org/10.3390/su13052620.). The effects of the formulation, temperature, and time on the color properties of chocolate spreads produced with different proportions of LBF are shown in Table 4. As seen, LBF formulation and storage time affected all measured (L*, a*, and b*) and calculated color properties (C* and h*) at the p < 0.01 level, while the storage temperature affected the L* value at p < 0.05. However, the effect of storage temperature on the h* value was insignificant (p > 0.05). All color values generally increased with the addition of LBF, and the highest values were determined for the CF45 formulation containing 4.5% LBF. High values for a* and C* indicated chocolates with a dark color. As the substitution of LBF for cocoa increased, a* and C* values increased, and accordingly, the darkening of the chocolates occurred. It can be concluded that the compositional differences play a crucial role in the color properties of the final product as they directly affect the product’s color. Absorptivity and scattering factors can also affect the color properties of foods (Rad et al., 2019Rad AH, Pirouzian HR, Konar N, Toker OS, Genc Polat T. 2019. Effects of polyols on the quality characteristics of sucrose-free milk chocolate produced in a ball mill. RSC Advances 9, 29676. https://doi.org/10.1039/C9RA04486H.). Many researchers also reported that substituting LBF for cocoa affects the color properties of different products. For example, Rosa et al. (2015)Rosa CS, Tessele K, Prestes RC, Silveira M, Franco F. 2015. Effect of substituting cocoa powder for carob flour in cakes made with soy and banana flour. Int. Food Res. J. 22, 2111–2118. showed that using LBF (0, 25, 50, 75, and 100%) as a substitute for cocoa in the production of gluten-free cake increased the C* and a* values, resulting in darker cakes.

Chocolate spreads stored at 22 °C exhibited higher L* and C* values than those held at 35 °C, while those kept at 35 °C showed higher a* and b* values than those stored at 22 °C (p < 0.05). Similarly, it was also reported by Machalkova et al. (2015)Machalkova L, Hrivna L, Nedomova S, Juzl M. 2015. The effect of storage temperature on the quality and formation of blooming defects in chocolate confectionery. Potravinarstvo 9, 39–47. https://doi.org/10.5219/425. that storage temperature affected the color properties of chocolates.

The effect of the storage time on the color properties differed depending on the color properties. For example, a* values decreased (p < 0.05) until the 8th week of storage and then did not change (p > 0.05), while b* and C* values decreased until the 8th week of storage and then increased slightly (p < 0.05). On the other hand, L* values increased until the 4th week of storage and then decreased, while after the 4th week, h* values first reduced and then increased (p > 0.05). A high fat percentage probably made the chocolates more susceptible to oxidation during storage, resulting in color changes. Similar to our findings, in a study by Bui and Coad (2014)Bui LTT, Coad R. 2014. Military ration chocolate: The effect of simulated tropical storage on sensory quality, structure, and bloom formation. Food Chem. 160, 365–370. https://doi.org/10.1016/j.foodchem.2014.03.084, it was also noted that some color properties of chocolate changed during storage. The authors found that storage time affected the L* and b* values but not the a* value.

3.4. Free fatty acid (FFA) and peroxide value (PV)

The effects of the formulation, storage temperature, and storage time on the FFA and PV values for chocolate spreads produced with different proportions of LBF are shown in Table 5. As observed, LBF formulation and storage time affected the FFA values for chocolates (p < 0.01). While CF15 and CF45 formulations showed lower FFA values than the CON formulation, the CF30 formulation showed a higher value (p < 0.05), and FFA values increased slightly during storage (p < 0.05). The increase in FFA during storage can be attributed to the activity of the lipase enzyme, which is activated due to the high temperature of the storage medium. Similar to our findings, Yadav et al. (2011)Yadav P, Pandey JP, Garg SK. 2011. Biochemical changes during storage of chocolate. Int. Res. J. Biochem. Bioinforma. 1, 242–247. and Ali et al. (2021)Ali AME, Shekib LA, Elshimy NM, Sharara MS. 2021. Producing and quality attributes of low calories dark chocolate using different intense sweeteners and wheat fiber isolate. Am. J. Food Sci. Technol. 9, 1–7. https://doi.org/10.12691/ajfst-9-1-1. also reported that the FFA values for dark chocolates increased during storage.

The formation of primary oxidation products was monitored by PV, and similar to FFA, only LBF formulation and storage time affected the PV values for chocolates (p < 0.01). As observed in Table 5, the CF45 formulation showed the highest PV value. In contrast, the CF30 formulation exhibited the lowest PV value (p < 0.05), and the differences between the PV values for CON and CF15 formulations were not significant (p > 0.05). The PV values for chocolate formulations increased during storage, and the highest value was determined to be 1.06 meq O₂/kg in the 12th week. Despite this increase, it was still lower than 10 meq O₂/kg, indicating that alterations related to lipid degradation were still in the initial stage (Rossini et al., 2011Rossini K, Norena CPZ, Brandelli A. 2011. Changes in the color of white chocolate during storage: potential roles of lipid oxidation and non-enzymatic browning reactions. J. Food Sci. Technol. 48, 305–311. https://doi.org/10.1007/s13197-010-0207-x.). A high percentage of fat probably made the control group and chocolates produced with different proportions of LBF more susceptible to oxidation during storage. Similar results were obtained by Mexis et al. (2010)Mexis SF, Badeka AV, Riganakos KA, Kontominas MG. 2010. Effect of active and modified atmosphere packaging on quality retention of dark chocolate with hazelnuts. IFSET 11, 177–186. https://doi.org/10.1016/j.ifset.2009.09.001. in dark chocolate with hazelnuts during 12 months of storage at 20 °C, by Rossini et al. (2011)Rossini K, Norena CPZ, Brandelli A. 2011. Changes in the color of white chocolate during storage: potential roles of lipid oxidation and non-enzymatic browning reactions. J. Food Sci. Technol. 48, 305–311. https://doi.org/10.1007/s13197-010-0207-x. in white chocolates during ten months of storage at 20 and 28 °C and by Ali et al. (2021)Ali AME, Shekib LA, Elshimy NM, Sharara MS. 2021. Producing and quality attributes of low calories dark chocolate using different intense sweeteners and wheat fiber isolate. Am. J. Food Sci. Technol. 9, 1–7. https://doi.org/10.12691/ajfst-9-1-1. in low-calorie dark chocolates during the storage period of up to 90 days.