Development of a nutri-rich roasted legume-based snack

. Horse gram and field bean legumes, although less consumed in conventional products, are mineral and fiber rich and have the potential to be included in value-added food formulations.

Materials and Methods

Six Legumes-Peanut (Arachis hypogaea), Green gram (Vigna radiata), Cowpea (Vigna unguiculata), Field bean (Lablab purpureus), Horse gram (Macrotyloma uniflorum), and Soybean (Glycine max) were procured from the local market of Pune. The procured legumes were cleaned, sorted, and packed in polythene bags for further process.

Standardization of product making process

The product was being prepared in 3 states: soaking, germination, roasting (Figure 1). Preliminary trials were undertaken to standardize the process for making/ production of the product.  Soaking: The soaking was carried out with various concentrations of solvents (0.2, 0.5 and 0.7%) each of salt and alum. The final concentrations were finalized based on sensory analysis.  Germination:

The germination was carried germinating for 48 hours at room temperature (27±1°C)  Roasting: Salt roasting was performed at a temperature range of 150-200°C to ensure uniform roasting.

Product Standardization

The product was formulated in two combinations. The formulation-1 contained equal proportions of all six legume seeds (1:1:1:1:1:1), whereas formulation-2 contained higher levels of Peanut, Field bean, Horse gram, and Soybean with lower proportion of Cowpea and Green gram (2:2:2:2:1:1). The formulation was standardized based on the physicochemical and sensory properties.

Physical Analysis

To assess the physical properties of legume seeds, routine procedures were used to measure seed weight, size measurements, geometric mean diameter, sphericity, bulk density, true density, and porosity. The experiments were performed at room temperature (27±1°C) and each measurement was made in triplicate for precision.  100-Seed Weight (g): A total of 100 seeds were manually counted and weighed on a digital analytical balance (Model: Contech CTL Series). 100 seed weight is an important measure of overall yield (Deshpande et al., 1982) [4] .  Seed Dimensions (Length, Width and Thickness):

Ten seeds of each legume were randomly selected to measure their length (L), width (W), and thickness (T) with help of digital vernier caliper (Model: Labin). Mean values were utilized for subsequent calculations (Mohsenin, 1986) [15] .  Geometric Mean Diameter (Dg): The geometric mean diameter, reflecting the mean seed size, was calculated as follows: (Mohsenin, 1986) [15] .

Where, L=Length W=Width T=Thickness Sphericity (Φ): Sphericity (Φ), indicating how spherical the seeds are, was calculated as per the method given by (Mohsenin, 1986) [15] .

Dg=geometric mean diameter L=length

Bulk Density (ρt) (g/cm³): Bulk density was measured by placing 10g of seeds into a 100 mL graduated cylinder without tapping and weighing them on the analytical balance. Bulk density was computed as per standard method (Singh and Goswami, 1996) [24] .

Bulk Density =

Mass of seeds

Volume of seeds

True Density (ρb), (g/cm³): True density was determined by the toluene displacement technique. Ten grams of seeds was added to a 100 mL measuring cylinder filled with toluene (a non-polar solvent that does not penetrate the seeds), and the volume displacement was recorded. True density was computed as per standard method (Sacilik et al., 2003) [23] .

True Density = Mass of seeds

Volume of toluene displaced

Porosity, that is a measure of void space in a bulk seed sample, was determined from the formula. (Sacilik et al., 2003) [23] .

Where, 𝜌𝑡=true density 𝜌𝑏=bulk density

Chemical Analysis

Moisture Content: Moisture level was measured using the oven drying method. About 5g of sample was taken on a pre-weighted moisture dish and dried under hot air in an oven (Model: i-therm AI-7981) at a temperature of 105± 2°C for 4-6 h until constant weight was obtained. Moisture content (%) was estimated as: (AOAC, 2019) [2] .

Initial weight × 100

Crude Protein Content: Protein was estimated by the Kjeldahl method. A portion of approximately 1g of finely powdered sample was digested with concentrated H₂SO₄ in a Kjeldahl digestion unit (Model: Tulin Kjeltron). The digest was then distilled and titrated for nitrogen content, and it was calculated to protein by multiplying with a factor of 6.25: (AOAC, 2019) [2] .

w

• V sample = Volume of acid used for sample titration (mL)

• V blank = Volume of acid used for blank (mL)

.4007 is a constant derived from molar mass of nitrogen (14.007 g/mol) and unit conversions.

Crude Fat Content: Crude fat was analyzed by Soxhlet extraction. About 3-5 g of dried material was put in a thimble and extracted with petroleum ether (boiling point 40-60°C) for 6-8 hours in a Soxhlet apparatus (Model: Tulin SOXTRON Sox-4). The solvent was evaporated after extraction, and the fat content was estimated by weight difference. (AOAC, 2019) [2] .

Crude Fat (%) = Weight of extracted fat

Weight of sample × 100

Crude Fiber Content: Crude fiber was estimated by acidbase digestion method. About 2g of defatted material was boiled with 1.25% sulfuric acid followed by boiling with 1.25% sodium hydroxide solutions. The residue was filtered, dried, and incinerated in a muffle furnace (Model: Tulin FIBROTRON FRB-2) at 550°C. Crude fiber was estimated by difference between residue weight before and after ashing. (AOAC, 2019) [2] .

Crude Fiber (%) = Weight of residue before ashing -Weight after ashing

Weight of sample × 100

Mineral (Ash) Content: The overall mineral content was estimated by ash analysis. Around 2-3 g of sample was transferred to a pre-weighed crucible and incinerated in a muffle furnace at 550°C for 5-6 hours until the production of white ash. Ash content was stated as a percentage: (AOAC, 2019) [2] .

Initial weight of sample × 100

Anti-nutritional factors

 Phytic Acid Reduction: Phytic acid content was determined using the Wade reagent method, which involves extracting the phytic acid with 0.2 N HCl, followed by reaction with Wade reagent (a mixture of FeCl₃ and sulfosalicylic acid). The absorbance was measured at 500 nm using a UV-Vis spectrophotometer. Reduction (%) was calculated by comparing treated samples with the untreated control (Reddy et al., 1982; Latta and Eskin, 1980) [21,9] .  Tannin Estimation: Tannin content was measured using the Folin-Denis method. Extracts were prepared with 70% aqueous acetone, and the reaction was carried out with Folin-Denis reagent and sodium carbonate.

Absorbance was read at 700 nm. The percentage reduction was calculated relative to the control (Price et al., 1978) [19] .  Trypsin Inhibitor Activity (TIA): Trypsin inhibitor activity was assessed based on enzyme-substrate reaction using BAPNA (Nα-Benzoyl-DL-arginine-pnitroanilide) as the substrate. The decrease in absorbance at 410 nm indicated inhibitor activity.

Results were expressed in TIU (Trypsin Inhibitory Units), and reduction percentage was calculated accordingly (Kakade et al., 1974) [7] .  Lectin Activity: Lectin content was estimated using hemagglutination assay. Serial dilutions of the legume extract were mixed with human or animal erythrocytes to determine the minimum concentration that causes visible agglutination. The reduction in lectin activity was expressed as a percentage compared to the control (Liener et al., 1980) [10] .  Saponin Content: Saponins were quantified using the spectrophotometric method involving vanillin-sulfuric acid reaction. Extracts were reacted with vanillin and sulfuric acid, and absorbance was measured at 544 nm. The reduction in saponin content was calculated in percentage terms in comparison to the untreated control (Obadoni and Ochuko, 2001) [18] .

Sensory Evaluation

Sensory evaluation of the treated legumes as well as legume snacks formulations was undertaken using the 9-point Hedonic scale, comprising of scores designated between 1 = dislike extremely and 9 = like extremely. Semi-trained panel of 10 members assessed the sensory quality of the product to standardize the formulations on grounds of determining the consumer acceptability. The sensory properties analyzed were colour and appearance, flavour, texture, crispiness, taste, and overall acceptability (Meilgaard et al., 2007) [13] .

Storage Studies

Accelerated shelf-life test (ASLT) was performed to analyze the stability and storage properties of the legume-based snack under high-temperature conditions. Samples were sealed in laminated pouches and stored at a high temperature of 45±2°C and relative humidity of 65± 5% for 30 days to mimic prolonged storage conditions.

Measurements were taken at 0, 15, and 30 days to observe sensory properties (appearance, flavour, texture, crispiness, taste, and overall acceptability) and visible spoilage or moisture absorption. Loss of crispness and moisture uptake were the important parameters because they affect consumer acceptability and the product's deterioration in low-moisture snack foods. Accelerated conditions were selected for anticipating the shelf-life and the observation of initial changes in quality deterioration, hence enabling the fast evaluation of the

International Journal of Agriculture and Food Science https://www.agriculturaljournals.com product within a small testing period (Manohar and Rao, 2000) [11] .

Statistical Analysis

All experiments and evaluations in this study were conducted in triplicate to make the results more accurate and reliable. The data obtained was statistically analyzed using standard statistical methods described by Snedecor and Cochran (1990). Mean values and standard deviations were calculated for each parameter. To find out if the differences between treatments were meaningful, Duncan's Multiple Range Test was used, with probability set at p≤0.05.

Results and Discussion

Standardization of product making process:

Preliminary trials were conducted to standardize the soaking medium for treatment of legumes. The legumes were separately soaked in varying levels of salt and alum solutions (0.2, 0.5 and 0.7%). The treated legumes were evaluated for their sensory properties and the results were recorded (Figure 2). Based on sensory evaluation scores it was clearly observed that among the concentrations, 0.2% salt whereas 0.5% alum found to be more suitable as a soaking medium for legumes compared to other concentrations, as indicated with the high scores for overall acceptability of peanut (7.8 and 6.8, respectively), green gram (7.7 and 7.1, respectively), cowpea (7.6 and 6.9, respectively), field bean (7.9 and 7.2, respectively), horse gram (7.6 and 7.0, respectively) and soybean (7.8 and 7.1, respectively). The high concentration of salt affected the taste and crispiness of the product and improved colour as well as texture.

Fig 2:

Sensory analysis of processed legumes

Physical analysis 3.2.1 Weight

Based on the data comparing the weight of 100 seeds under different treatments 0.2% salt, 0.5% alum, and a combination of both (0.2% salt + 0.5% alum), distinct trends emerge across legumes and processing methods. It is clearly observed from table no. 1, that soaking resulted in increase in the weight of seeds due to imbibition of water in the dry seeds making the seeds rehydrated (Bewley and Black, 1994) [3] . After germination, the weights of seeds slightly decreased as germination is an exothermic process, causing loss of moisture due to internal heat (Mayer and Poljakoff-Mayber, 1989) [12] . Whereas roasting being the heat treatment, caused great reduction in the weights of seeds due evaporation of moisture at high temperature (Tiwari, et al., 2011) [28] . In general, the combination treatment (0.2% salt + 0.5% alum) often resulted in a slight increase in seed weight compared to individual treatments, particularly in peanuts and green gram during soaking and germination. For instance, peanut seeds showed a marginal increase from 82.98 g (0.2% salt) and 82.83 g (0.5% alum) to 83.35 g under the combined treatment during soaking. Similarly, green gram had a progressive increase in weight across treatments in both soaking and germination. However, in roasting, the trend was less consistent; for example, green gram seed weight slightly declined under the combined treatment, while in Cowpea and Field bean, roasting showed more notable weight increases under the combination treatment. These variations suggest that while combined salt and alum treatments may enhance hydration and possibly preserve or enhance mass in some legumes and processes, the effect is not uniform across all types and conditions.

Length, width, thickness, geometric mean diameter, sphericity

The influence of different treatment concentrations (0.2% salt, 0.5% alum, and a combination of 0.2% salt with 0.5% alum) on the physical dimensional properties of legumes was evaluated in terms of length, width, thickness, geometric mean diameter, and sphericity. In peanuts, the 0.5% alum treatment generally resulted in marginal increases across all measured dimensions compared to the 0.2% salt treatment, indicating a potential enhancement in water uptake or structural preservation. The combined treatment (0.2% salt + 0.5% alum) further supported these observations, particularly with slight increases in geometric mean diameter and sphericity, suggesting a synergistic effect of alum and salt. For green gram, the physical parameters remained relatively stable across treatments, with only minor variations noted. Notably, the combined treatment yielded a modest increase in sphericity compared to individual treatments. These findings indicate that the combination of alum and salt may contribute to improved retention of physical integrity in legumes during processing, potentially enhancing their quality and functionality in food applications.

Bulk density, true density, porosity

The bulk density, true density, and porosity of legumes are significantly influenced by soaking, germination, and roasting treatments. Soaking generally maintains or slightly increases bulk density due to water absorption, while true density remains relatively stable or shows marginal variation. Germination tends to slightly reduce both densities as enzymatic activity alters the seed structure. Roasting, in contrast, significantly reduces bulk density across all legumes, likely due to moisture loss and puffing, while true density also drops, reflecting structural degradation. Porosity tends to increase with roasting, indicating greater internal voids, while germination and soaking show moderate changes. These trends suggest roasting causes the most drastic physical changes, followed by germination and soaking (Mohsenin, 1986; Kulthe et al., 2025) [15,8] .

Chemical analysis

The data demonstrates how the different concentrations influence the nutritional composition of the six legumes. Moisture content generally showed a slight increase with the addition of alum, especially in field bean and cowpea, while the combination treatment had a moderate effect. Protein levels remained relatively stable across treatments, with minor increases under 0.5% alum, particularly in peanut and horse gram. Fat content exhibited minimal variation, although slight increases were noted with alum and combination treatments in cowpea and soybean. Fiber content remained consistent throughout, with slight increases in soybean and cowpea under combination treatment. Mineral content showed minor fluctuations, with a trend of increase under 0.5% alum and combination treatment especially in soybean and field bean.

Carbohydrate content generally remained stable or decreased slightly in most samples with combination treatment, especially in soybean and field bean. Overall, the combination of salt and alum had variable effects across different nutrients and legumes, suggesting ingredient specific interactions with these substances. (Agoreyo, et al., 2011; Kulthe et al., 2025) [8] .

Anti-nutritional factors

The application of varying salt and alum concentrations affected the reduction of anti-nutritional components in all six legumes examined. 0.5% alum treatment recorded more reductions in phytic acid, tannins, trypsin inhibitors, lectins, and saponins than 0.2% salt treatment alone. The combination treatment (0.2% salt + 0.5% alum) had the highest overall reduction for all anti-nutritional factors in every legume, suggesting synergism. For example, in soybean, the combined treatment decreased phytic acid by 33.77%, tannins by 37.68%, trypsin inhibitors by 31.35%, lectins by 27.82%, and saponins by 23.88%, much greater than decreases with single treatments. These trends were also observed in other legumes field bean and horse gram, in which combination treatments produced above 30% decreases for most anti-nutrients. These findings indicate that using a combined treatment of alum and salt is more efficient in reducing anti-nutritional compounds, thus possibly enhancing the nutritional quality and digestibility of legumes (Kulthe et al., 2025) [8] .

Product Standardization

Sensory analysis of two different formulations identified that formulation 2 was far more acceptable in terms of all parameters except one against formulation 1. Formulation 2 scored higher on flavour (8.0), texture (7.1), crispiness (7.7), taste (8.1) and overall acceptability (7.7) and showed better consumer preference. Conversely, formulation 1 ranked lower in texture (6.0), taste (6.5), and overall acceptability (6.9), although the two formulations were ranked equally in colour and appearance (7.1 and 7.3, respectively). The improved scores for formulation 2 indicate that adjustments in the ingredient formulation or processing conditions successfully improved product quality, which was more desirable to the panelists.

Chemical analysis as per formulation

Chemical analysis of the two formulations showed minimal difference in nutritional content. Formulation 1 contained slightly more moisture (4.74±0.36%) than formulation 2 (4.48±0.3%), which would impact shelf life. Protein levels were slightly higher in formulation 2 (22.22±0.13%) than in formulation 1 (21.72±0.25%), reflecting modest improvement in nutritional content. Fat levels also experienced a slight rise in formulation 2 (2.37±0.06%) compared to formulation 1 (2.23±0.38%). Fibre and mineral levels were similar between the two formulations at 8.09±0.37% and 8.01±0.49% for fiber, and at 3.71±0.07% and 3.52±0.06% for minerals in Formulations 1 and 2, respectively. Carbohydrate content was also constant, at 59.35±0.19% for Formulation 1 and 59.16±0.21% for Formulation 2. These data imply that the two formulations were chemically equivalent with slight nutritional advantages seen in Formulation 2 (FAO, 2003; Kulthe et al., 2025) [6,8] .

Anti-nutritional factors based on formulation:

The anti-nutritional evaluation of the two formulations shows that Phytic acid reduction in Formulation 2 was slightly more at 30.91±0.06%, whereas in Formulation 1 it was 30.7±0.05%. Likewise, trypsin inhibitor reduction was enhanced from 27.62±0.29% in Formulation 1 to 27.85±0.27% in Formulation 2, and lectin reduction was boosted from 24.3±0.16% to 24.74±0.32%. While tannin and saponin decreases were similar in both formulations, with Formulation 1 indicating a slightly greater reduction of tannin (34.63±0.04% vs. 34.44±0.47%) and Formulation 2 with a slightly reduced saponin decrease (20.78±0.16% vs. 21.07±0.19%), overall, Formulation 2 indicated a consistent pattern of slightly improved reduction of most antinutritional factors. This implies that Formulation 2 has a better option in terms of improved nutritional quality through better reduction of anti-nutrients. (Mubarak, 2005; Kulthe et al., 2025) [8] . like colour and appearance, flavour, texture, crispiness, taste, and general acceptability. The sensory scores at 0 days were high, with a general acceptability of 8.08, which reflects outstanding quality and freshness of the product. After 15 days, a slight reduction was seen in all the parameters, but the overall acceptability remained very high at 7.86. At 30 days, a minor further reduction lowered the overall acceptability to 7.7. These figures reflect good product stability under stress conditions for long-term storage. The deterioration in all quality attributes during 30 days is rather slow and restricted, with all ratings still well above 7.5, which is typically equivalent to a "liked very much" category on a 9-point hedonic scale. Based on the minor loss in quality during the 30-day test, it is reasonable to project that the product will be of acceptable sensory quality for not less than 6 months under standard storage conditions.

Conclusion

The results obtained from above research helped to develop a nutri rich, roasted legume based snack out of local legumes and minimal processing methods such as soaking, germination, and roasting with salt. The 0.2% salt and 0.5% alum combination was the most effective in enhancing sensory acceptance and minimizing anti-nutritional factors without compromising nutrient composition. Of the two formulations that were tried, formulation 2, with increased levels of peanut, field bean, horse gram, and soybean, showed superior sensory acceptability, better nutritional parameters, and maximum reduction of anti-nutrients. Accelerated storage studies also proved the sensory stability of the product for 30 days under stress. Overall, the snack is a promising, healthy, and shelf-stable option compared to traditional RTE snacks, especially in filling dietary needs in protein and fiber using sustainable, plant-based sources.

References

1. Agoreyo BO, Obansa ES, Obanor EO. Comparative nutritional and phytochemical analysis of two varieties of Cajanus Cajan. Afr J Food Sci. 2011;5(6):323-329.
2. AOAC. Official Methods of Analysis of AOAC International. 21 st Ed. AOAC International, 2019.
3. Bewley JD, Black M. Seeds: Physiology of Development and Germination. Springer Science and Business Media, 1994.
4. Deshpande SS, Sathe SK, Salunkhe DK. Functional properties of lupin seed proteins and protein concentrates. J Food Sci. 1982;47(2):491-497.
5. Duranti M. Grain legume proteins and nutraceutical properties. Fitoterapia. 2006;77(2):67-82.
6. FAO. Food Energy-Methods of Analysis and Conversion Factors. Food and Nutrition Paper 77. Rome: Food and Agriculture Organization of the United Nations, 2003.
7. Kakade ML, Rackis JJ, McGhee JE, Puski G. Determination of trypsin inhibitor activity of soy products: A collaborative analysis of an improved procedure. Cereal Chem. 1974;51:376-382.
8. Kulthe AA, Khapre AP, Rana K. Processing driven impact on functional and nutritional profile of pearl millet. Asian J Dairy Food Sci. 2025;44(2):268-272.
9. Latta M, Eskin M. A simple and rapid colorimetric method for phytate determination. J Agric Food Chem. 1980;28(6):1313-1315.
10. Liener IE. Toxic constituents of plant foodstuffs. Proc Nutr Soc. 1970;29(1):56-57.
11. Manohar RS, Rao PH. Storage studies on functional and rheological properties of blends of wheat flour and finger millet flour. Eur Food Res Technol. 2000;211:272-276.
12. Mayer AM, Poljakoff-Mayber A. The Germination of Seeds. Elsevier, 1989.
13. Meilgaard M, Civille GV, Carr BT. Sensory Evaluation Techniques. 4th ed. CRC Press, 2007.
14. Messina M. Legumes and soybeans: overview of their nutritional profiles and health effects. Am J Clin Nutr. 1999;70(3 Suppl):439S-450S.
15. Mohsenin NN. Physical properties of plant and animal materials. 2 nd Ed. Gordon and Breach Science Publishers, 1986.
16. Mubarak AE. Nutritional composition and antinutritional factors of mung bean seeds (Phaseolus aureus) as affected by some home traditional processes. Food Chem. 2005;89(4):489-495.
17. Mudryj AN, Yu N, Aukema HM. Nutritional and health benefits of pulses. Appl Physiol Nutr Metab. 2014;39(11):1197-1204.
18. Obadoni BO, Ochuko PO. Phytochemical studies and comparative efficacy of the crude extracts of some haemostatic plants in Edo and Delta States of Nigeria. Glob J Pure Appl Sci. 2001;8(2):203-208.
19. Price ML, Scoyoc VS, Butler LG. A critical evaluation of the vanillin reaction as an assay for tannin in sorghum grain. J Agric Food Chem. 1978;26(5):12141218.
20. Rasane P, Jha A, Sabikhi L, Kumar A, Unnikrishnan VS. Nutritional advantages of oats and opportunities for its processing as value added foods: A review. J Food Sci Technol. 2015;52(2):662-675.
21. Reddy NR, Pierson MD, Sathe SK, Salunkhe DK. Phytates in cereals and legumes. Crit Rev Food Sci Nutr. 1982;17(3):273-335.
22. Rochfort S, Panozzo J. Phytochemicals for health, the role of pulses. J Agric Food Chem. 2007;55(20):79817994.
23. Sacilik K, Ozturk R, Keskin R. Some physical properties of hemp seed. Biosyst Eng. 2003;86(2):191198.
24. Singh KK, Goswami T. K. Physical properties of cumin seed. J Agric Eng Res. 1996;64(2):93-98.
25. Singh RP, Heldman DR. Introduction to Food Engineering. Gulf Professional Publishing, 2001.
26. Sreerama YN, Neelam DA, Sashikala VB, Pratape VM. Distribution of nutritional components and antinutritional factors in milled fractions of chickpea and horse gram: Seed coat, cotyledon, and germ. LWTFood Sci Technol. 2012;49(1):108-113.
27. Tang Y, Li X, Zhang B, Chen PX, Liu R, Tsao R. Characterisation of phenolics, betanins and antioxidant activities in seeds of three Chenopodium quinoa Willd. genotypes. Food Chem. 2014;166:380-388.
28. Tiwari BK, Gowen A, McKenna B. Pulse Foods: Processing, Quality and Nutraceutical Applications. Academic Press, 2011.