Physicochemical properties and sensory acceptance of Canavalia ensiformis tempeh energy bar

Zainal Abidin, N.A., Mohd Zin, Z., Abdullah, M.A.A. Rusli, N.D. and *Zainol, M.K. Faculty of Fisheries and Food Science, Universiti Malaysia Terengganu, 21030, Kuala Nerus, Terengganu Faculty of Science and Marine Environment, Universiti Malaysia Terengganu, 21030, Kuala Nerus, Terengganu Faculty of Agro Based Industry, Universiti Malaysia Kelantan, Jeli, Kelantan, Malaysia Institute of Food Security and Sustainable Agriculture, Universiti Malaysia Kelantan, Jeli, Kelantan, Malaysia


Introduction
Tempeh is a traditional fermented soy food, prepared from soaked and cooked soybeans by salt-free aerobic fermentation using Rhizopus oligosporus (Krisnawati and Adie, 2015). Freshly prepared tempeh is a cake-like substance that is fully coated and filled with white mycelium, with a clean, yeasty odour (Xiao, 2011). Tempeh is usually cooked before it is eaten, and preparation usually involves frying, deep-fat frying, or baking the product (Wilson, 1995). Tempeh is considered as a good source of protein, vitamins, antioxidants, phytochemicals, and other bioactive beneficial substances (Mani and Ming, 2017).
Canavalia ensiformis or Kacang Koro contain more than 32% protein, yet they also produce somewhat mild antinutrient content (protease inhibitors, lectins, saponins and tannins) in the form of anti-nutritional factors (Eke et al., 2007) which makes them not popular as human food. Therefore, few treatments have been conducted to render C. ensiformis edible to humans, such as heating, fermentation and extrusion, though obtaining favourable results (Tepal et al., 1994).
The concept behind energy bars is to provide anyone on the go with a quick snack or breakfast (Zhu and Labuza, 2010). The energy bar is a convenient and healthy ready to eat food that supplies balanced nutrients which contain protein, fat, minerals, vitamins and carbohydrates (Ho et al., 2016). Chitkara et al. (2017) reported that the primary purpose of energy bars is to satisfy hunger, replace a meal, and provide essential nutrition. The current market's energy bars are filled with a lot of fruit, nuts, granola, and sugar combinations (Norajit et al., 2011) to increase the product's sugar, protein, and fibre content. The consumers are concerned about getting healthier foods, and this has changed their eating habits which promoted growth in the energy bar market (Silva et al., 2016). Energy bar provides an effective sensory and nutritious snack because of their high carbohydrate, protein, lipid, and mineral content (Nadeem et al., 2012). Athletes and other physically active people also use energy bars to fulfil their nutritional needs, since they can have high amounts of protein, fats and carbohydrates (Norajit et al., 2011).
The protein content in energy bars was usually derived from soybeans and milk products as well as from other protein sources, such as peanut butter, nuts, to increase the protein quality (Aldrich, 2015). The development of the new product based on a 'modified version' of underutilised legumes such as C. ensiformis could produce a better form of energy bar hence increasing the usage of these underutilized legumes. The purpose of this research was therefore to evaluate the effect of C. ensiformis tempeh on the physicochemical properties and the sensory acceptability of the C. ensiformis energy bar.

Raw materials
C. ensiformis was harvested from Kuala Berang, Terengganu. The starter tempeh was obtained from Teluk Panglima Garang, Selangor. Other ingredients used in the production of C. ensiformis tempeh energy bar include butter, raisins, honey, and brown sugar which were purchased in Kuala Terengganu, Terengganu, Malaysia.

Preparation of Canavalia ensiformis tempeh
C. ensiformis tempeh was prepared according to the procedure described by Kustyawati et al. (2017) with some modification. C. ensiformis beans were washed and soaked in clean tap water at room temperature (28°C) for 48 hrs and the soaking water was changed 5 times, and then boiled for 30 mins (the ratio of water to C. ensiformis was 3:1). This was followed by dehulling to remove C. ensiformis skin from the cotyledon manually, and another boiling for 30 mins (Wan Mohamad Din et al., 2020). Subsequently, C. ensiformis was boiled, drained and air-dried. Then, 0.02 g of Rhizopus mold was inoculated for every 100 g cooked C. ensiformis, and packed into plastic bags, then incubated for 36 hrs at 32°C. The prepared tempeh was then kept in 4 o C prior to further usage.

Preparation of Canavalia ensiformis tempeh energy bar
All the ingredients were weighed based on the formulations presented in Table 1. The sliced 'C. ensiformis tempeh' was then roasted in the oven for 1 hr at 100℃. The C. ensiformis tempeh was then grind to small size using mortar and pestle. Prior to the mixing process, almonds were roasted for 10 mins in the oven, while oats also were baked for 10 mins. All the ingredients were cut to small sizes before mixing in a bowl. Brown sugar, honey, and butter were heated at low temperature (40°C) and then mixed with roasted ingredients. The prepared mixture was then pressed to produce a rectangular shape in a mould. The energy bar was then kept in the chiller (4 ο C) prior to further analysis (Wan Mohamad Din et al., 2020).

Colour profile analysis
The instrument that was used to measure the colour was Konica Minolta colourimeter (Konica Minolta, Tokyo, Japan) based on L*a*b* colour system, where L (lightness), a (redness) and b (yellowness). The colourimeter was calibrated by using a white calibration plate before analysis. The instrument was placed on the energy bar formulation and the values of L* (lightness), a* (redness) and b* (yellowness) for each energy bar was recorded and analysed .

Texture profile analysis
The analysis was conducted using a TA. XT the mechanical characteristics where material was subjected to a controlled force from a deformation curve of response. The sample was placed centrally under the 3 -point bend rig probe until the probe came to contact with the sample. Then, the deformation curves were recorded (Mamat et al., 2018).

pH analysis
The acidity and alkalinity of the energy bar samples were determined using a pH meter (Mettler Toledo, Ohio, USA). The samples (5 g) were finely mixed and homogenized with 20 mL of distilled water, and measured the pH value.

Toxicity analysis 2.7.1 Determination of oxalate content
Pulverized sample (1 g) was weighed into a conical flask, and added to 75 mL of 3 M sulphuric acid was added and stirred with a magnetic stirrer for an hour. This was filtered and 25 mL aliquot of the filtrate was collected and heated to 80-90℃. This filtrate was kept above 70℃ at all times. Next, the hot aliquot was titrated against 0.05 M of potassium permanganate oxide (KMnO 4 ) until an extremely faint pale pink colour persisted for 15-30 s (Agbaire, 2011). The oxalate content was calculated as followed: 1 mL of 0.05 M of KMnO 4 = 2.2 mg oxalate.

Determination of phytic acid content
Sample (2 g) was weighed into a 250 mL conical flask. Approximately 100 mL of 2% HCl was used to soak the sample for 3 hrs and then filtered through filter paper. A 25 mL aliquot of the filtrate was placed in a separate 250 mL conical flask and 5 mL of 0.3% ammonium thiocyanate solution was added. Approximately 53.5 mL of distilled water was added and this will be then titrated with standard iron (III) chloride solution which contained 1.95 mg of iron per mL until a brownish yellow colour persisted for 5 min (Unuofin et al., 2017). Phytic acid was calculated as follow: Phytic acid (%) = Titre value×0.00195×1.19×100.

Determination of tannins content
Tannin content was determined using the Folin-Ciocalteu method. One millilitre of energy bar sample was mixed with 7.5 mL of distilled water and 0.5 mL of Folin-Ciocalteu phenol reagent. Then, 1 mL of 35% sodium carbonate solution was added to the mixture and diluted to 10 mL with distilled water. The mixture was vigorously shaken and let it stand for 30 min at room temperature (25°C). Absorbance for test and standard solutions were measured against blank at 725 nm with a UV/Visible spectrophotometer. The tannin content was expressed in terms of mg of GAE/L of extract (Wan Mohamad Din et al., 2020).

Proximate analysis
The moisture, ash, fat, fibre and crude protein content of the C. ensiformis tempeh energy bar were determined according to AOAC International (2007) standard procedures. All analyses were carried out in triplicates.

Calorie content analysis
The calorie content was determined using a bomb calorimeter (LECO, USA). The sample was placed in the crucible in the combustion chamber for the combustion process. Once the sample is ignited, a thermometer which is partially submerged in the water records the temperature changes that occur. The heat of combustion (cal/g), the change in temperature was recorded as the calories content of the sample. (Wan Mohamad Din et al., 2020).

Mineral content analysis
The determination of mineral nutrient was carried out using an Inductively Coupled-Plasma-Mass Spectrophotometry (ICP-MS) (Agilent Technologies, U.S.A). Dried sample (2 g) was placed into the crucible, burned in a furnace at 500 o C for 24 hrs. The crucible was then cooled at room temperature and 2 mL of concentrated HCl was added and was let to evaporate to dryness on a hot plate. Then, 10 mL of 20% HNO 3 was added and the crucible then placed in the water bath for 1 hr. Finally, the solution was analysed using ICP-Mass Spectrophotometer.

Sensory analysis
The sensory evaluation session was performed based on a 7-point hedonic scale (higher score indicates better quality attributes (1, dislike very much and 7, like very much)) (Mamat et al., 2018). The colour, aroma, appearance, crispiness and flavour of the energy bars were evaluated. All the attributes were independently judged by forty untrained panels based on their likeness. The sample was packed and coded with a 3-digit code. The mean score for each attribute was reported.

Statistical analysis
All data were analysed statistically using SPSS (Version 20) statistical software package. The results were expressed as mean±standard deviation. The significant difference at (p˂0.05) was performed by oneway analysis of variance (ANOVA) and Fisher's Least Significant Difference (LSD) test (Hau et al., 2018). Table 2 reveals that the L* value of the formulations was also increased with the increasing amount of C. ensiformis tempeh, suggesting the influence of tempeh on the lightness of the energy bar of colour. The findings also showed that formulation A had the highest a* rating, which shows the energy bar's redness. Formulation E, on the other hand, has the lowest value of a* (2.67). The energy bar's red colour may be attributed to the amounts of almond in the formulation as well as the presence of other almond ingredients. The formulation D's b* value (yellowness) was found to be the highest (25.23), while formulation A was found to be the lowest (20.62). For the other tests, the b* value was identical, suggesting that the butter mixture might be other ingredients that could have led to an increased value in yellowness of energy bar. The energy bar colours are quite similar to each formulation due to the colour of its ingredients, while the amount of C. ensiformis tempeh and the amount of almonds gives a small effect to the bars. Table 2 shows that the hardness of the C. ensiformis energy bar (formulation D) was found to be significantly the highest (7.71±0.37 N) while the control formulation had the lowest hardness value (6.44±0.73 N). In general, the low hardness of the formulation was observed when a low amount of tempeh was incorporated into the formulation. This could mean that when the proteins are used, the bar may have a better or worse texture than expected (Imtiaz et al., 2012). In addition, all formulation s showed a somewhat similar fracturability value indicating that the tempeh did not have an analytical effect on the texture profile value. However, due to the low amount of tempeh incorporated in the formulations, low fracturability was observed in the analysis. When proteins are blended together, they may have synergistic or antagonistic effects. Moisture migration, phase separation (Hogan et al., 2012) and fat oxidation could be the main reason for the texture value recorded in this study (Dan and Labuza, 2010). Mridula et al. (2013) stated that the hardness of the energy bar is significantly affected by the level of sweeteners and flaxseed. Table 3 shows the pH value of C. ensiformis tempeh energy bar ranges between 5.30 to 4.82 and the control energy bar was 5.44. The data reveal that the acidity of all formulation of energy bar was significantly different (p<0.05) between control, formulation A, B and C except for formulation D and E. Control sample, illustrated the highest pH value compared to other formulations (5.44±0.01), while formulation E exhibited the lowest pH value (4.82). According to Silva et al. (2016), the pH value of the energy bar incorporated with jeriva flour ranged from 6.78 to 6.92, where the pH of jeriva flour itself is 4.96. Compared to C. ensiformis tempeh energy bar, the pH value obtained was acidic. The ingredient that probably caused the pH above 5 is honey. This could be because the average pH of honey is 3.9, with a typical range of 3.4 to 6.1 (White et al., 1962). Table 3 shows the oxalate content of C. ensiformis tempeh energy bar formulation and control. The results showed that there was a significant difference (p<0.05) between formulation D and E except for formulation control A, B and C which showed no significant difference. The range of oxalate content of C. ensiformis tempeh energy bar was from 45.43 mg to 54.90 mg while the formulation control was 45.10 mg. The formulation E has the highest oxalate content which was 59.4 mg, meanwhile formulation A has the lowest which was 45.43 mg. Such a finding could be explained by the increasing amount of C. ensiformis tempeh in the energy bar which increased in the oxalate content. Fermentation also caused the changes of some organic acids such as acetic acid, oxalic acid, citric acid and succinic acid, which eventually influenced the pH. This finding was in agreement with previous studies on soybean fermentation using R. oligosporus starter (Vong et al., 2018) or Rhizopus starter (Moa et al., 2013). A possible explanation for these results may be the fermentation of C. ensiformis helps to reduce the amount of oxalate in the legume. Besides, fermentation helps to reduce the anti-nutritional and toxic factors in the raw materials by making the proteins and minerals which are complex with these phytochemicals readily available (Adegbehingbe, 2015).

Phytic acid content
The results showed that there was a significant difference (p>0.05) between formulation except for formulation control and C which has no significant difference similar to formulation A and B which also showed no significant difference (Table 3). The range of phytic acid content C. ensiformis tempeh energy bar was from 2.50% to 1.17% while the formulation control was 1.93%. The formulation A contained the highest tannin level which was 2.50%, meanwhile formulation E contained the lowest which was 1.17%. This may be due to the fact that the increased percentage of C. ensiformis tempeh in the energy bar reduced the amount of phytic acid content in the energy bar (Sridhar and Seena, 2006). Table 3 shows that there was a significant difference (p<0.05) between all formulations except for control and formulation A. Formulation E has the highest tannins content (0.03%), while formulation C has the lowest (0.00%). The highest content of tannins in formulation E still considers a safe percentage of tannins to be consumed. This is because high dietary levels (approximately 5%) can cause death (Sridhar and Seena, 2006). The level of content of tannins in the C. ensiformis tempeh energy bar is, however, below consumable level and can still be eaten. Table 4 indicates the moisture content of the energy bar C. ensiformis tempeh varying from 9.37% to 13.09% where the maximum moisture content was 13.09% and the lowest moisture content of the energy bar C. ensiformis tempeh was 9.37%, with just 9.29% moisture content in the control energy bar. Increasing the amount of C. ensiformis in the formulation thereafter increases the amount of moisture content in the energy bar of C. ensiformis. Similar findings were reported by Arbaje et al. (2016), who stated that granola bars moisture content was affected by the amount of granola in the formulations. The highest ash content of C. ensiformis tempeh energy bar was recorded in formulation A (1.48%) while the lowest percentage of ash content was found in the control sample. This shows that the addition of C. ensiformis tempeh in the formulation reduced the ash content in C. ensiformis tempeh energy bar. This may be due to the usage of the nuts, where the almonds itself contains higher ash content, which was 3.13% (National Germplasm Resources Laboratory, 2009). The ash content in this study could be traced to the nuts used, which could be considered as origin when minerals, such as calcium, iron and magnesium, are used (Fernandes et al., 2010).

Chemical composition of Canavalia ensiformis tempeh energy bar
The highest protein content was found in formula E (10.34%), while the lowest protein content was found in formula B (4.93%). There was no significant difference between formulations A, C and D (5.34%), (5.25%) and (5.95), respectively. The higher value of protein content FULL PAPER  in energy bars may be due to the utilization of C. ensiformis tempeh in the energy bar production which contains higher protein compared to almonds, which was 34.47% and 21.06% respectively (Nimenibo-Uadia, 2017).
Table 4 also depicts that there was no significant difference (p>0.05) in the fibre content between the sample tested. The crude fibre in C. ensiformis recorded in this study was higher than that of soybean (4.28%) and kidney beans (4.2%) (Apata and Ologhobo, 1994). The observed increase in crude fibre could be attributed to the presence of tempeh of C. ensiformis which provides more fibre than the C. ensiformis itself. The results also showed that there was no significant difference (P>0.05) in carbohydrate content between all formulations. The range of C. ensiformis tempeh energy bar was from 58.91% to 64.94%, while the carbohydrate content of control energy bar formulation was 60.59%. The carbohydrate source might be from the usage of brown sugar and honey in the production of C. ensiformis tempeh energy bars (E'zzati, 2019). Table 5 shows that there was a significant difference in the total caloric value between all formulations. The total calorie content of the energy bars ranged from 4742.5 cal to 5267.80 cal. Control sample contained significantly the highest calorie content (5276.80 cal) while formulation E exhibited the lowest calorie value (4627.55 cal). The decreasing trend of calorie value of the energy bar could be due to the reduced amount of cereals and nuts used in the bars (Okoye and Eke-Ejiofor, 2018). Table 5 also shows the mineral content that presents in five formulations of C. ensiformis tempeh energy bars and control. The data showed that Iron (Fe), Magnesium (Mg), Calcium (Ca), Phosphorus (P), and Potassium (K). Mineral level range in C. ensiformis tempeh energy bar was 25.32 -58.63 mg/kg for potassium, 9.26 -13.89 mg/kg for calcium, 35.55 -83.44 mg/kg and 8.91 -18.24 mg/kg of magnesium. The iron level shows quite similar between each formulation which was in the range 1.00 -2.82 mg/kg in 10 mL solution of sample. This shows that iron levels are not affected by an increasing amount of C. ensiformis tempeh and a decreasing amount of almonds. The mineral content in the energy bars was typically lowered as the amount of C. ensiformis tempeh increased. This result is in line with the study by Ho et al. (2016), which stated that ash content of energy snack bars is reduced when glutinous rice content increases banana puree as the main ingredient. Table 6 showed that there was some variation yet no significant difference (p>0.05) between all formulations. The highest value of sensory colour acceptance was formulation B and the lowest colour acceptance was exhibited in formulation E. This might be due to the ingredients such as the amount of brown sugar, almond and butter that provide colour to the energy bar used in the same amount. According to Khouryieh and Aramouni (2012), changes in flaxseed bar colour may be associated with the possibility of browning Maillard reactions between flaxseed protein and other ingredients such as honey and brown sugar. Ho et al. (2016) demonstrated that the bright brown colour of the energy snack bar may be attributed to caramelisation occurring with the presence of sugar during heating and boiling.

Sensory analysis
Formulation B exhibited the highest value for odour acceptance while the lowest was found in formulation E. The caramelization process that occurs on brown sugar may difficult the panel to distinguish the odour unless the smell of the C. ensiformis tempeh was strong. Ho et al. (2016) stated that the strong odour of the "energy" snack bar might be attributed by the caramelization that occurred during the heating and boiling process with the presence of sugar. Formulation A showed to the best hardness liked by the untrained panellist with a value of 5.30. Meanwhile, formulation E was not favoured by the panellist. Brahim et al. (2017)  study also showed that the degree of likeness for C. ensiformis tempeh energy bar decreased as the hardness decreased.
The results also showed no significant difference (p<0.05) in flavour acceptance between all samples except formulation E. The results was affected by the different amount of C. ensiformis tempeh incorporated in the product. However, formulation A showed the best formulation of C. ensiformis tempeh energy bar liked by the panellists while formulation E was the least.
The colour, odour, hardness, fracturability and taste, the panellist selected the most agreed formulation A compared to other formulations. This might be due to the different percentage of C. ensiformis in the formulation that affected the characteristics of the energy bar.

Conclusion
This study exhibited that C. enformis tempeh may serve as an excellent candidate for the development of an alternative energy bar using rich protein raw materials which have been under exploited. Formulation A was mostly accepted by the panellist. This product is safe to be consumed as the level of the toxicity is still within a safe range. This study shows that C. ensiformis tempeh can be used in energy bar production, thereby encouraging its use in the food industry.
(2017). Effect of pear apple and date fibres incorporated on the physico-chemical, sensory and nutritional characteristics and the acceptability of cereals bars.