1 Introduction

The raw ingredients used for product development are mainly obtained from cereals. However, cereal-based ingredients have low protein and vitamin content. In recent years, many studies have indicated that including soybeans significantly enhanced the protein contents of cereal-based food products [1]. Moreover, research indicate that incorporating vegetable crops such as orange-fleshed sweet potato has also improved the vitamin A contents of the composite flours [2].

Finger millet is a small but mighty grain, playing a vital role in diets across India and Africa. Compared with other major cereals such as rice, wheat, and barley, its resilience to drought conditions and adaptability to challenging environments make it a best candidate for product development. It also contains a higher dietary fiber content than common grains like rice, wheat, and other millet varieties. Being gluten-free adds to its appeal, potentially benefiting those managing type 2 diabetes and influencing the glycemic index of foods [3]. Finger millet contains 7.94%, 1.39%, and 81.59% of protein, fat, and carbohydrate, respectively [4].

Orange-fleshed sweet potato (OFSP), the second most important root tuber globally, is a powerful contender in the fight against vitamin A deficiency [5]. Rich in beta-carotene, a precursor to vitamin A [45], OFSP offers an affordable and accessible source of this essential nutrient [2]. It has a strong chance of becoming a staple of the consumer food chain and resolving the issue of vitamin A deficiency. It is a significant source of beta-carotene and a cheap energy source [6]. The protein, fat, carbohydrate, and beta-carotene content in different OFSP verities varied from 2.4–4.2%, 1.2–1.8%, 82.7–87.1%, and 35.5–91.6 mg/100 g, respectively [7].

Soybean (Glycine max) is a nutritional powerhouse, offering a long history as a protein source. In addition to being a good source of proteins, it is also a good source of oils, carbohydrates, and essential minerals [8]. Soybean is one of the most adaptable food sources and provides essential nutrients for people of all ages. Soybeans also contain beneficial phenolic compounds with antioxidant properties that may contribute to cancer prevention [8]. Although The nutritional profile of soybeans varies across different varieties, they typically boast a significant amount of protein, fat, and carbohydrates, with protein in the range of 36.94–40.01%, 16.82–19.30%, and 34.97–39.86%, respectively, based on varieties [9].

When creating composite flours, carefully selecting ingredients and their blending ratios is vital in determining the final product's characteristics [10]. Previous research indicates significant changes in the blending ratios regarding composite flours' nutritional composition and sensory properties [11, 12]. For instance, Forsido et al. [13] have shown that increasing the proportion of soybean flour in blends can significantly boost the protein and fat content.

Therefore, this research aimed to enhance the nutritional values, especially the protein and vitamin A content, of finger millet flour by incorporating soybeans and orange-fleshed sweet potato and to optimize the blending ratios of composite flours. This will help produce optimized blending ratios of these ingredients to create nutritionally enhanced composite flours that can be used to develop new and improved food products.

2 Materials and methods

2.1 Experimental materials

The study utilized finger millet, Orange Fleshed Sweet Potato (OFSP), and soybean sourced from the Jimma Agricultural Research Centre in Jimma, Ethiopia. The specific varieties used for this study were Addis 01 for finger millet, Alamura for OFSP, and Clark 63 K for soybean. These varieties were selected for their known high yields and nutritional content.

2.2 Sample flour preparation

Each raw material underwent various processing steps to prepare them for flour production. To prepare soybean flour, the soybean grains were cleaned and boiled for 30 min, followed by husk removal and drying at 60 °C for 16 h, and milled (KARLKOLB D-6072 Dreich, West Germany) into flour (0.5-mm sieve) [14]. Similarly, finger millet was cleaned to remove impurities. It was then oven-dried at 60 °C for 10 h to facilitate milling into flour using a 0.5 mm sieve [4]. The OFSP was cleaned, washed, peeled, and sliced into 3 mm thick pieces, followed by soaking in NaCl solution for 30 min to prevent browning. The soaked OFSP was dried at 55 °C for 8 h and milled into flour using a 0.5 mm sieve [15]. The prepared flours were stored at 4 °C in airtight polyethylene bags until they were analyzed.

2.3 Experimental design and treatment combination

The experiment used Minitab® (Version 21.1, Minitab, Inc.) software to formulate thirteen different formulations of the composite flours using a mixture design with a range of finger millet 45–55%, OFSP 30–40%, and soybean 10–20%. The upper and lower values for finger millet, OFSP, and soybean were selected based on the previous research findings [16,17,18,19].

2.4 Data collected

2.4.1 Proximate composition and beta carotene content determination

The various components of the proximate composition of the flours were analyzed using the standard methodology. Moisture, crude protein, crude fiber, crude fat, and ash content were determined using standard AOAC methods (925.09, 923.03, 979.09, 962.09, and 923.05, respectively) on a dry matter basis [20]. The sum of moisture, crude proteins, crude fat, moisture, crude fiber, and ash content was subtracted from 100% to calculate the carbohydrate content. β-carotene content was measured according to the method presented in Sadler, Davis, and Dezman [21] using a T-80 UV/Vis spectrophotometer (PG Instruments, China).

2.4.2 Functional properties determination

The method described by Oladele and Aina [22] was used to determine bulk density. It was calculated by dividing the weight (g) of the ground flour by its volume (ml) after settling. The water absorption index (WAI) was determined following the method presented by Anderson [23]. It was calculated as the weight of the sediment (gel) divided by the weight of the dry flour sample. The supernatant preserved from WAI measurement was dried for 5 h at 100 °C, and the Water solubility index was determined as the weight of the dried supernatant divided by the weight of the dry sample. Oil absorption capacity was calculated by subtracting the weight of the sample after centrifugation from the weight of the sample before centrifugation, dividing by the weight of the sample, and multiplying by one hundred [24].

2.5 Statistical analysis

The data was analyzed using Minitab Software version 21 and the significance of linear, quadratic, and interaction effects on each response was determined using the ANOVA. Mean separation was conducted at a significance level of 0.05 probe ability (p < 0.05).

3 Results and discussion

3.1 Proximate compositions of composite flours

3.1.1 Moisture content

ANOVA table from regression analysis showed there was a significant difference (p < 0.05) in the linear model, a highly significant variation (p < 0.01) in the quadratic model, and interaction among components except for interaction between OFSP and finger millet which was no significant (p > 0.05) effect on the moisture contents of the composite flours (Table 1 and Table 5).

Table 1 ANOVA p-values, R2, and coefficients of variation (CV %) for proximate composition and beta-carotene content
Full size table

The recorded moisture content of the composite flours ranged between 5.41% and 10.12% (Table 2). This suggested that the moisture content of current studies is found below the recommended level (< 14%) and enough to have an extended shelf life. Moisture content is related to the shelf life of food products. Low moisture content in food samples increases food product stability by inhibiting microbial growth and biochemical reactions [25].

Table 2 Proximate composition (%) and ꞵ-carotene content (mg/100 g) of composite flours and raw materials (DB)
Full size table

The highest moisture content (10.12%) was observed at 50% of finger millets, 40% OFSP, and 10% of soybean flours, and the lowest moisture contents (5.41%) was found at 50% of finger millets, 30% of OFSP and 20% of soybean flours. Increasing the proportion of OFSP led to a higher moisture content. This is likely due to the high water-binding capacity of the starch in sweet potatoes. However, increasing the amount of soybean flour resulted in a lower moisture content, which could be due to the low moisture content in soybean flour, as presented in Table 2. A similar study was observed by Alamu et al. [26], who stated that the moisture content decreases as the proportion of soybean flour increases in blending maize-soybean composite flours. On the other hand, Tadesse et al. [2] noted that enhancing the proportion of OFSP results in higher moisture content because starch in sweet potatoes has high water binding capacity [27].

3.1.2 Protein content

Analysis of variance indicated a significant difference (p < 0.05) in the quadratic model and a high significance variation (p < 0.01) in the linear model and interaction of finger millet with soybean between OFSP with soybean but no significant difference (p > 0.05) among finger millet with OFSP (Table 1 and Table 5). The protein contents of composite flours varied from 9.6% to 17.9% (Table 2). The maximum protein content (17.9%) was obtained at 50% finger millets, 30% OFSP, and 20% soybean flour. In comparison, the minimum protein content (9.6%) was documented at 50% of finger millets, 40% OFSP, and 10% of soybean flours. The current output showed a rise in the protein contents of the blended flour as the proportion of soybean flour increased. The increase in protein content of composite flour as soybean flour rises is due to the high protein content, which ranges from 36 to 40% [8]. The current studies were also consistent with the report of Ndife et al. [1], who documented increased protein contents of cookies made from wheat and soybean flour as the amounts of soybean flour increased.

3.1.3 Fat content

ANOVA table p-value showed a highly significant difference (p < 0.01) in both linear and quadratic models and the interaction between finger millet with soybean and OFSP with soybean. However, no significant variation (p > 0.05) was observed in the interaction between OFSP and finger millets (Table 1 and Table 5). The lower and upper values of fat contents of composite flours obtained from current studies were between 3.45 and 7.65% (Table 2). The study indicates that as the level of soybean flour rose from 10 to 20%, the fat contents of the composite flour improved from 3.45% to 7.65%. The reason for increasing the fat contents of composite flour as the level of soybean flour rises is the presence of large amounts of fat in the soybean flour. The related research finding indicated that the fat content of rice-soybean composite flour increased as soy flour supplementation increased [11]. The present result was also similar to the result of Ayo et al. [28], who reported an increase in the fat content of acha-based bread fat as the amounts of soybean flour increased.

3.1.4 Fiber content

There was highly significant variation (p < 0.01) in linear and quadratic models and the interaction between finger millet with OFSP and finger millet with soybean flours. In contrast, no significant difference (p > 0.05) was observed in the interaction of OFSP with soybean flours (Table 1 and Table 5). The recorded maximum and minimum values of fiber for composite flours varied from 3.5% to 5.4% (Table 2). The current finding showed that raising a finger millet flour increased the fiber contents of composite flours. The lowest values of fiber contents of formulated flours were obtained at the lowest proportion of finger millets. In contrast, the highest values of fiber contents were registered at the highest proportion of finger millet flour. Similarly, Feyera et al. [29] noted an enhancement in fiber content as the proportion of finger millet increased. This increase in fiber content is because finger millet is considered a rich source of fiber [30].

3.1.5 Ash content

ANOVA result indicated that there was a high significance difference (p < 0.01) in linear and a significant difference (p < 0.05) in quadratic models, as well as the interaction between finger millet with soybean and OFSP with soybean. However, no significant (p > 0.05) was affected by the interaction of OFSP and finger millets (Table 1 and Table 5). In this research, the ash value of composite flours was documented in the range of 2.09–3.70% (Table 2). The lowest value was recorded at the lowest levels of soybean flour, and the highest result was recorded at the highest proportion. This might be due to the availability of more minerals in soybean flour. The results noted by different authors agree with the present study, indicating a rise in ash contents with the increase in soybean flour supplementation in the composite flour. Ndife et al. [1] also observed that upgrading the percentage of soya bean flour in cookies increased the ash content of cookies.

3.1.6 Carbohydrate content

There was a high significance difference (p < 0.01) in linear, quadratic, and among interaction of finger millet with soybean, finger millet with OFSP, and OFSP with soybean (Table 1 and Table 5). The carbohydrate contents of the composite flours ranged from 61.22% to 70.66% (Table 2). The carbohydrate content of formulated flour decreased as the proportion of soybean flour increased. These findings agreed with Mezgebo et al. [31], who documented that increasing malted soybean flour to red teff-based flour declined the carbohydrate contents of the composite flours. On the other hand, carbohydrate content increases as the proportion of OFSP and finger millet increases. Malavi et al. [12] also stated that the carbohydrate contents of bread made from OFSP and wheat flour enhanced as the proportion of orange-fleshed sweet potato supplementation rose. Sweet potato has been considered to be rich in carbohydrates; this indicates that flour is a source of high-energy and nutrient-dense food for consumers [32].

3.1.7 Beta-carotene contents of composite flours

Statistical analysis indicates that the proportion of finger millet influenced the beta-carotene content of the formulated flours, orange-fleshed sweet potato (OFSP), and soybean flour. As shown by ANOVA analysis, no significant difference (p > 0.05) was observed in the linear model, suggesting a linear relationship between the ingredients and beta-carotene content was insignificant. However, a highly significant difference (p < 0.01) was found in the quadratic models, indicating that the relationship between the ingredients and beta-carotene content is more complex, likely involving a curved or quadratic relationship. Moreover, significant interaction (p < 0.01) was observed between finger millet and OFSP and between OFSP and soybean Flour. This suggests that the combined effect of these flours on beta-carotene content is not simply additive (Table 1 and Table 5). The beta-carotene content of the composite flours varied from 6.3 to 9.015 mg/100 g (Table 2). The result indicates that increasing the inclusion of OFSP flour resulted in higher beta-carotene content. This is because OFSP is a rich source of beta-carotene, compared to soybean and finger millet, as presented in Table 2. Similarly, the research of Kindeya et al. [33] observed that the beta-carotene content of biscuits developed from wheat, haricot bean, and OFSP increased as the OFSP level supplementation improved. The current findings also agreed with the output of Laelago et al. [34], who reported that increasing the proportion of OFSP flour in composite flour increased the beta-carotene content of biscuits made from wheat and OFSP. Many authors have reported similar results, confirming the positive relationship between OFSP content and beta-carotene levels in formulated foods [2, 12, 35]. Overall, the research highlights the importance of OFSP in enriching the beta-carotene content of composite flours, providing a valuable contribution to the dietary intake of this important nutrient.

3.2 Functional properties of composite flours

3.2.1 Bulk density

The bulk density of composite flours was a significant difference (p < 0.05) in linear and a highly significant difference (p < 0.01) in quadratic models. Significant variation (p < 0.05) was observed between finger millet and OFSP, and a high significant difference (p < 0.01) was shown in the interaction of finger millet with soybean. However, no significant difference was observed between OFSP and soybean (Table 3 and Table 5).

Table 3 ANOVA p-values, R2, and coefficients of variation (CV) for Functional properties of composite flour
Full size table

As indicated in Table 4, the results of the bulk density of composite flours ranged from 0.62 to 0.89 g/ml. The result illustrates that the bulk density of the composite flour was improved as the supplementation of finger millet flour enhanced. The highest bulk density was recorded at the highest proportion of finger millets and the lowest proportion of soybean flours. The high bulk density is due to the high fiber contents in finger millet flour, which enhances the bulk density of the composite flours. The output of current results was in agreement with Siddhart [36], who stated that the higher bulk density might be due to the availability of more crude fiber in the composite flour sample. Alamu et al. [26] also indicated that the bulk density of the composite flour of maize and soybean increases as the substitution of soybean flour decreases.

Table 4 Mean Values of Functional properties of composite flours and raw materials
Full size table

Bulk density is important when developing and using formulated flours, particularly for weaning foods. A flour's bulk density directly affects the packaging material needed, which indicates easy handling, as lower bulk density generally means lighter flour, making it easier to handle, transport, and package [37]. Moreover, flours with lower bulk density are often perceived as easier to digest, which benefits young children. Additionally, Lower bulk density can be advantageous for preparing nutrient-dense weaning foods as it allows for a higher concentration of nutrients within a given volume [38]. However, high bulk density in flours and starches indicates good thickening properties, making them suitable for food products requiring thickening.

3.2.2 Water absorption index

Analysis of Variance (ANOVA) found a highly significant difference (p < 0.01) between linear and quadratic models for water absorption in composite flours. This difference was also observed in interactions between finger millet, soybean, OFSP, and soybean. However, no significant interaction (p > 0.05) was found between finger millet and OFSP (Table 3 and Table 5). The composite flour's water absorption index (WAI) ranged from 2.01 to 3.08 g/g (Table 4), increasing in proportion to the amount of soybean flour. This aligns with previous research by Ojinnaka et al. [39], who attributed this increase to soybean protein's ability to bind water molecules due to its complex structure.

Table 5 Estimated regression coefficients for proximate composition, beta-carotene content, and functional properties of composite flours
Full size table

WAI measures the amount of water absorbed by a specific amount of starch at a given temperature. It indicates starch gelatinization, a process where starch granules absorb water and swell, creating a gel-like consistency [40]. It helps determine a flour's ability to absorb water and swell, which is crucial for achieving the desired consistency in food products [41]. This flour property is desirable for enhancing the product yield and consistency by improving food products' texture and mouthfeel. The findings indicate soybean flour's significant impact on composite flours' water absorption properties, ultimately influencing their application and functionality in various food products.

3.2.3 Water solubility index

Analysis of Variance (ANOVA) indicates that the WSI of composite flours showed a highly significant difference (p < 0.01) based on the proportion of ingredients used. The linear quadratic models showed a strong interaction between finger millet with OFSP and OFSP with soybean flour. However, no significant (p > 0.05) effect was observed on the interaction of finger millet and soybean flours (Table 3 and Table 5). The upper and lower values obtained from composite flours ranged from 14.38% to 19% (Table 4). An increase in the water solubility index was observed as the proportion of OFSP level enhanced. This is likely due to the abundance of soluble polysaccharides (starch) in sweet potato flour, as supported by Adeola et al. [42], who observed a similar trend when adding OFSP to pigeon pea flour.

WSI quantifies the amount of soluble polysaccharides released from starch granules and dissolved in water. It is often used to indicate the quality and functionality of starch-based products, including their processing and storage stability [43]. The finding indicates that incorporating OFSP significantly improved the water solubility of composite flours, a crucial factor to consider when developing and optimizing food products.

3.2.4 Oil absorption capacity

ANOVA table has shown a highly significant difference (p < 0.01) in the linear model and a significant difference (p < 0.05) in the quadratic model; interaction between finger millets and OFSP and finger millet with soybean with no significant variation between OFSP and soybean flours (Table 3 and Table 5). The recorded values of oil absorption capacities of composite flour varied from 89 to 141% (Table 4). The highest values of oil absorption capacity were recorded at the highest levels of finger millet flour. The lowest results of oil absorption capacity were obtained at the lowest proportion of finger millet flours. So, increasing the proportion of soybean decreases the OAC of the composite flours. This results in agreement with the findings of Feyera et al. [29], who reported an improvement in oil absorption capacities with the inclusion of finger millet flour rose. However, the oil absorption capacity decreased with the increase in soybean flour. This decline is due to the high contents of soybean oil, which decreases oil absorption capacity. Since soybeans contain high amounts of oil, their absorption capacity is low.

Analysis of Variance results revealed a significant relationship between the composition of composite flours and their oil absorption capacity (OAC). Finger millet flour significantly increased OAC (p < 0.01 linear model, p < 0.05 quadratic model), while soybean flour had the opposite effect, leading to a decrease in OAC (Table 3). This finding aligns with previous research [29] and can be due to soybean flour's high oil content, which inherently limits its ability to absorb additional oil. The OAC of our composite flours ranged from 89 to 141% (Table 4), highlighting the significant impact of ingredient proportions on this characteristic. Oil absorption capacity is vital in food product development, influencing texture, flavor retention, and overall sensory appeal [44].

3.2.5 Optimization of the blending ratios of composite flour

Graphical optimization using design expert version 13 software programs was used to optimize composite flours, as presented in Fig. 1. This optimization aimed to obtain the optimum blending ratio for response values. It was performed by setting a goal from the given criteria for every response based on the desired response. Finger millets, OFSP, and soybean have been set in range while protein, fat, ash, carbohydrate, and beta-carotene contents were maximized. The optimum desirable blending ratios for nutritional and functional properties were found at 47.46% of finger millets, 34.54% of OFSP, and 17.99% of soybean flour with a desirability of 0.58.

Fig. 1
figure 1

Graphical optimization of the response and the optimum levels of the variables X1, X2, and X3 represent finger millet, OFSP, and soybean, respectively

Full size image

4 Conclusion

This study was conducted to improve the protein and vitamin A content of finger millet flour by mixing with OFSP and soybean flours and to optimize the blending ratios of composite flours. A mixture design was used to generate thirteen treatment combinations of composite flour. The ANOVA table shows that changes in blending ratios of finger millet, OFSP, and soybean flours had significant effects (p < 0.05) on the measured parameters. The result indicates that the addition of soybean and OFSP flour into the blend significantly upgraded the protein, fat, and beta-carotene contents of composite flours from 9.6% to 17.9%, 3.45 to 7.65%, 6.3 to 9.015 mg/100 g. Increasing the soybean flour ratio into the blend significantly (p < 0.05) enhanced the water absorption capacity; in contrast, increasing the OFSP level significantly (p < 0.05) upgraded the water solubility index of composite flours. From this study, the blending ratios obtained at 47.46% of finger millet, 34.54% of OFSP, and 17.99% of soybean flour give optimum values of response variables. Therefore, the optimized composite flour obtained from this finding will be used to manufacture different nutrient-rich products such as snack foods. This research suggests blending vegetable and legume crops into cereal will improve composite flours' nutritional values and other characteristics.