African Journal of Food Science and Technology

Optimization of ingredients for the production of noodles made from rice flour, soybean flour, corn and potato starch - I-Optimal Mixture Design

Ntukidem Ofonime James, Barber Lucretia Ifeoma^* and Emelike Nkechi Juliet T Department of Food Science and Technology, Rivers State University, Port Harcourt, Nigeria
^*Corresponding Author:
Barber Lucretia Ifeoma, Department of Food Science and Technology, Rivers State University, Port Harcourt, Nigeria, Email: barber.lucretia@ust.edu.ng, ofonntuks@gmail.com

Received: 14-Jun-2023, Manuscript No. AJFST-23-110349; Editor assigned: 16-Jun-2023, Pre QC No. AJFST-23-110349; Reviewed: 22-Jun-2023, QC No. AJFST-23-110349; Revised: 26-Jun-2023, Manuscript No. AJFST-23-110349; Published: 18-Jul-2023, DOI: 10.14303//ajfst.2023.031

Introduction

Noodles are dried thin strands of wheat dough and alkaline solution. They are ready-to-eat, safe and nutritious products that have become popular among all strata of the Nigerian populace. The term “Asian noodles” is used very broadly to describe noodle-like products from Eastern, Southeastern, or Pacific Asian countries that are made from non-wheat flour, rice flour, or other starch materials as the main structural ingredients (Lu & Nip, 2006). Wheat flour and water easily form a dough through appropriate mixing and kneading techniques that facilitate sheeting and slitting into strips or strands to produce noodles. This unique ability to form cohesive, elastic and extensible dough is attributed to gluten, the unique protein in wheat (Hoseney, 1990). Noodles made from rice flour or starches from several sources such as maize and potato differ from wheat noodles in many aspects such as nutritive and structural components (Buddenhagen & Persley 1978).

Rice and starch-based noodles are better alternatives to address the increasing demand for noodles and also reduce the increased demand for wheat importation in Nigeria. The characteristics of gluten-free noodles are heavily dependent on the functional properties of the starch as it undergoes one or two heat treatments during processing (Mestres, 1988; Young et al., 1996).

The lowland NERICA 6 is an improved high yielding, floodtolerant rice variety that is widely adopted by small-scale farmers in Nigeria. The inclusion of this variety in noodle production is expected to contribute to the income of these farmers and food security through value addition.

Soybean is one of the most nutritious legumes with 36.4 % protein. It also contains several essential amino acids and is important fortifying carbohydrate-rich diets in developing countries such as Nigeria. It also contains valuable phytochemicals and has an extraordinary capacity to nourish and prevent diseases. Soya is also known to be a good source of the trace elements copper, zinc, and manganese and can be said to contain all the nutrients needed in food (Ampofo, 2009). Acceptable noodles have been produced using soybean flour and cassava flour incorporation. Combing soybean flour and rice flour is expected to increase the protein content of rice noodles.

Starch is a primary material in the production of noodles. Owing to the absence of gluten compared to wheat flour, the physicochemical, thermal, and rheological properties of s of starch determine the quality of gluten-free noodles. Composite starches are more effective than single starches for the textural improvement of rice noodles (Sun, 2006). Composite starches can increase the tensile property of gluten-free noodles and decrease the percentage of broken noodles and cooking loss significantly (Sun, 2006). Corn starch is usually used to increase the hardness and decrease the adhesiveness of rice noodles. It is attributed to the high amylose content of corn starch (34.4%). Potato starch can also improve the transparency and elasticity of starch noodles. Sun (2006) also reported that the gelatinization temperature of potato starch is lower than that of corn starch but its swelling power is 48 times that of corn starch. Potato starch also has a slower retrogradation rate and is very effective in decreasing the percentage of broken noodles (Hou, 2010).

Optimization experiments have the advantage of predicting desired responses without the need for repeated laboratory experiments at every variation of I the ingredients of a product. I-optimal split-plot designs, minimize the average prediction variation. I-optimal designs aim to reduce the whole design space's average prediction variance (Goos & Jones, 2012). Additionally, compared to their estimationoriented D-optimal counterparts, the prediction-oriented I-optimal split-plot designs offer more accurate factor impact estimates (Goos and Jones, 2012).

The objective of this study was to establish the optimum levels of rice flour, soybean flour, corn and potato starch for the production of rice soy noodles and determination of ingredients levels on protein, carbohydrate, colour, texture, appearance, taste and overall acceptability of rice soy noodles.

Materials and Methods

White rice (Lowland Nerica 6) was purchased from Taraba State, Nigeria, and soybean (Glycine max) was purchased from mile one market, Port Harcourt, River State, Nigeria. Other ingredients needed for the preparation and analysis of rice noodles were purchased from Merck chemical limited, Lagos state, Nigeria

Preparation of rice flour

Local white rice flour was processed using the method described by Eke-Ejiofor and Nwiganale, with slide modification as shown in Figure 1. Rice grains were sorted and cleaned, known quantity of the rice was fermented in a known amount of water in a ratio of 1:3 for 16 hours, then crush and dry at 60°C for 4 hours (partial drying) to 24% moisture content, milled and finally dried at 60°C for 12 hours then sieved with a mesh size of 150 microns (Wang et al., 2004).

Preparation of soybean flour

Soybean flour was prepared by the method described by (Adekoyeni et al., 2013). Soybeans was sorted to remove particles and washed in clean tap water. The seeds were boiled for 30mins and drained so as to inactivate the trypsin inhibitors followed by dehulling using a manual method i.e. hand rubbing within two palms, after dehulling, the soybean seeds were dried in a hot air oven at 70°c for 10 hours. After drying the soybean hulls were removed by winnowing, and the dried samples were milled to fine powder and sieved through a standard sieve of 200μm particle size.

Alkaline formulation

Alkaline was prepared following the method described by Hou (2010), with modification. The alkaline ingredient contained 3.6g potassium carbonate, 2.3g sodium carbonate (2.3), 0.00015g riboflavin, 3.57g polyrinsan, 2.86g gaur gum, 3.0g iodized salt and 1 liter of treated water.

Experimental design

Optimal Custom Mixture Design (Design-Expert version 13.0.5.0 Stat-ease, Inc, 2021) was used to design the research. the design generated 20 runs by mixing four (4) components at two (2) levels which were rice flour X₁ (70- 85%), soybean flour X₂ (5-15%), corn starch X₃ (5-15%) and potato starch X₄ (5-12%) with five (5) replicate points. Table 1 shows the formulated mixture runs generated from the design.

Where: Σ X _i = X₁ + X₂ + X₃ + X₄ = 100        (1)

Table 1: Formulation of composite flour blends (%) for rice -soy noodle production.

Production of gluten free noodles

Rice -soy noodle was produced according to the method described by Sun (2006). 100g of composite flour blend was transferred into the mixer followed by the addition of 28 ml alkaline solution and 50 ml warm water (75°C), with continuous mixing, with steady mixing to form a dough. Strand formation was carried out through the help of an extruder with 1.5 mm aperture diameter, produced strand was steam at a temperature of 100 °C for 15 min, followed by another second of extruding to give a fine and smooth strand surface to control the thickness. Steamed rice noodles were dried at 70 °C for two hours. The dried glutenfree noodles were allowed to cool followed by packaging.

Proximate composition and amylose content

The methods described by AOAC (2012) were used to determine protein, fiber, fat, moisture and ash contents of rice, soybeans, potato and corn starches. carbohydrate was determined by difference.The amylose content of these samples was also determined by (AOAC 2012) methods. The proximate composition of the rice-soy noodle samples was determined by (AOAC 2012) methods while their carbohydrate contents were determined by difference. The determinations were carried out in triplicates.

Sensory evaluation of rice-soy noodles

Twenty (20) semi-trained panelists who are familiar with noodles were randomly selected from students and staff of the Department of Food Science and Technology Rivers State University, Nigeria to evaluate the products. Evaluation of all the samples took place in a sensory booth. Panelists were instructed to evaluate appearance, texture, taste and overall acceptability of the noodles. A 9-Point hedonic scale was used with 1=dislike extremely, 5=neither like nor dislike and 9=like extremely (Ocheme et al., 2018). Samples were identified with 3-digit code numbers and presented in a random sequence to the panellists. Glass of water was provided for the panelists to rinse their palate after each sample and expectoration bowls with lids were provided for assessors in the event they did not wish to swallow the samples. To prevent interfering with other panelists, the panelists were advised to remain silent throughout the evaluation. They were also asked to comment freely on samples on the form given to them.

Statistical analysis

Statistical analysis of variance (ANOVA), correlation coefficient determination and fit regression analysis were carried out on the data generated from protein, carbohydrate, texture, appearance and overall acceptability of rice soy noodles using statistical software (Design-Expert version 13.0.5.0 Stat-ease, Inc, 2021). Mean separation and one sample T-test were carried out using Minitab software version 20.0.

Numerical optimization

Optimization of formulation Numerical values of independent variables and simultaneous optimization of the multiple responses were carried out by choosing the desired goals for each variable and response by using Design Expert software. Among the ingredients rice flour, soybean flour, corn starch and potato starch (Hoseney, 1990) were kept in the range so as to obtain formulation of blend ratio which produces gluten-free noodles with good nutritional and functional qualities. The pre-determined goal for protein and carbohydrate was maximized while texture, appearance and overall acceptability were set in range (Singh et al., 1989).

Results and Discussion

Proximate composition of rice flour, soybean flour potato and corn flours

Potato starch had the highest mean moisture content of 11.92 % (Figure 1). Rice and soybean flours had similar moisture contents of 5.24 and 5.14%, respectively. Porter & Jones (2003) reported moisture content of 5.20% for soy flour. Soybean protein content of 40.41% was significantly (p< 0.05) higher than those of all the other ingredients.

Protein was not detected in the potato and maize starches. This could either be due to the complete absence of protein or they were present in amounts lower than the limits of detection of the procedure used. Carbohydrate is the most abundant nutrient in rice flour, potato and maize starches. The values obtained in this study (Figure 1) are similar to those reported by with carbohydrate content ranging from 76.55 to 82.48% in new rice for African flour. The amylose content of rice flour (Figure 1), the main ingredient in the formulated rice soy noodles, was less than the 27% minimum recommended for production of noodles with acceptable texture. However, inclusion of corn and potato starches with 26 and 30% amylose compensated for the shortfall (Young et al., 1996).

Protein and carbohydrate contents of rice- soy noodles

Table 2 shows the mean values of protein, carbohydrate, texture, appearance and overall acceptability of rice soy noodles. Experimental runs had a significant (p<0.05) effect on the crude protein and carbohydrate contents, texture, appearance and overall acceptability of the noodles. Crude protein content ranged from 9.51 to 14.46 %. Sample 4 had the lowest protein content with 5% soy flour addition while sample 15 had the highest protein content with 15% soy flour addition. The increase was attributed to the level of substitution of other ingredients with soybean flour. Soybeans were considered in the formulation because of their high protein content and quality amino acid profile and their content of minerals, such as calcium and iron (Plahar et al., 2003). Adding rice flour and corn starch also decreases the protein content of rice soy noodles. Okpalanman & Chukwu, (2020) also reported that increasing the quantity of cassava flour in the blend decreases the protein content and that increase in both soybean and wheat flour in the blend increased the protein content of the biscuit. Protein content in rice soy noodles was similar to those quoted by Nagao (1996) for alkaline noodle flours (10.5–12%). Crosbie et al., (1999) recorded protein contents ranging between 10 and 12.2% in 21 experimental noodle samples, while the commercial Top Raman noodle contained 11% protein. Baik & Lee (2003); Wang et al. (2004) reported that the protein values were between 10.5-16.4 and 10.1-19.3% respectively for cooked white salted noodles.

Table 2: Means of protein, carbohydrate, texture, appearance, taste and overall acceptability of gluten-free noodles .

The Carbohydrate content for the 20 runs ranged from 73.55 to 83.20%. The carbohydrate content was influenced by the combined effect of rice flour, corn and potato starch. The carbohydrate content decreased significantly (p<0.05) with an increase in soybean flour substitution, 83.20% with 5% soybean flour addition in sample 4 and 73.55 % carbohydrate with 15% soybean flour addition in sample 15.

Mean scores for texture, appearance, taste and overall acceptability of rice- soy noodle

Texture depicts the chewiness, firmness and/or softness, springiness, elasticity, and other bite characteristics of instant rice soy noodles. Texture score ranged from 5.6 to 8.8 which indicated neither like nor dislike to like very much. Development of paste during instant noodles preparation was observed in samples 2, 5, 10,11,13,14,15 and 18, this was due to the high soybean flour percentage of above 9.51%, Rice soy noodles texture score increased with the increase in the addition of corn and potato starch this was due to their high amylose content of 32 and 26% respectively Soybean negatively impacted the texture of rice soy noodles with a correlation coefficient of -0.841 (Table 3) Wu & Corke (2005) also reported that due to the protein structure, related to the potential for crosslink formation, can affect noodle texture. According to Hou (2021) in improving the processing and texture of noodles, starch or vital gluten is added.

Table 3: ANOVA results of responses as  linear, quadratic, special cubic and interaction terms on each response.

Appearance plays a major role in determining consumer acceptance of food products. The highest appearance score of 8.80 (like very much) of rice soy noodles was observed in sample 19 which was due to the low content of soybean flour. Soybean flour also exhibited and negative correlation coefficient of -0.843 (Table 3) which indicated that an increase in rice-soy noodles resulted in a reduction in the appearance score. The reduction of the degree of likeness of appearance is due to the Maillard reaction that occurred during the drying of rice soy noodles thereby leading to browning. This study reveals that an increase in rice flour, corn and potato starch resulted in a significant (p<0.05) increase in the degree of likeness of the appearance of the rice soy noodles. Sharma & Chanhan (2000) reported a decrease in the appearance score of composite flour baked products as substitution increased. Similarly, Emeka et al., (2020) also reported a reduction in appearance (darker and brownish colour) score with an increase in the addition of soybean flour and cassava flour in the production of biscuits. The main senses stimulated by oral processing of food are taste (Pereira & Van der Bilt, 2016). Sample 8 had the highest taste score of 8.80 indicating “like very much” and the lowest taste score of 6.70 was observed in sample 6 indicating “like slightly”. The low score rating observed in sample 6 may be due to the increased addition of soybean flour. Some panelists observed beany taste in the sample. Oknalanma et al., (2020), also reported limiting sensory attributes of a beany taste in biscuits with the addition of soybean flour. This report agrees perfectly with Sabanis & Tzia (2009) and Mohammed, Ahmed & Senge (2003). The overall acceptability score of rice soy noodles ranged from 6.57 (like slightly) to 8.63 (like very much). Sample 12 had the highest overall acceptability, which is attributed to the increase in the composite combination of rice flour, corn and potato starch. Nevertheless, soybean flour exhibited a negative correlation of -0.829 (Table 3) with the overall acceptability. Savita, et al., (2018) also reported that the overall acceptability of multigrain gluten-enriched instant noodles reduced with an increase in soy flour. Collins and Pangloli (1997) also reported that the sensory acceptability of sweet potato and soy flour incorporated noodles decreased with increased levels of soy flour.

Fitting the model

Protein content of rice-soy noodles

The ANOVA for protein indicated a quadratic model with a p-value of 0.0001 it also indicates that there was an interactive effect of soybean flour on the protein content of the noodles in Table 4, this shows the individual effect (A, B, C and D) and interactive effect AB, BC and BD terms were significant (p≤0.05) while AC, AD and CD were not significant to the model. The coefficient estimate of protein for the components are: A (+9.84), B (+23.10), C (+11.21), D (+9.66), BC (-19.21), and BD (-10.94), while the multiple regression equation for describing the relationship between protein and ingredients (coded form) after neglecting the non-significant terms is given below (Eq. 4)

Y1 (protein) = 9.84A + 23.101B + 11.21C + 9.661D - 14.59AB- 19.21BC - 10.94BD        (2)

R² = 98.11

The lack of fit for % crude protein was not significant (p≤0.05) with a p-value of 0.2842 which indicate that the model accommodated more observations by less than the variation between replicates, fit statistic standard deviation, C.V %, R², adjusted R², predicted R² and adeq precision were 0.2536, 10.99, 2.31, 0.9811, 0.9724, 0.9240 and 31.4735 respectively. The Model F-value of 112.43 implies the model is significant. The interactive effect of AB, BC and BD was significant with a p-value less than 0.0500, this is because the main effect B had a high amount of protein of about 36.405% (Figure 1).

R- squared value is an indication of the level of response that could be explained by the model (Okpalanma et al., 2020). The Predicted R² of 0.8829 is in reasonable agreement with the Adjusted R² of 0.9750. The adequate precision test measures the signal to noise ratio. It compares the range of the predicted values at the design points to the average prediction error. A ratio greater than 4 indicates adequate model discrimination and then the model can be used to navigate the design space. The ratio of 31.4735 indicates an adequate signal therefore the model can be used to navigate the design space.

There was a relative increase in protein content with an increase in soybean flour, this was seen in the positive coefficient of 23.19 for soybean flour in the predictive equation. Also, the positive correlation coefficient of 0.898 (Table 3) also indicates an increase in protein content or rice soy noodles with an increase in soybean flour. The 3D surface plot (Fig 2a) also shows that an increase in soybean settings resulted in a significant (p<0.05) increase in the overall protein content of the product, while an increase in rice flour and corn starch setting decreases the protein content of rice soy noodles. Emeka et al., (2020) also reported that increasing the quantity of cassava flour in the blend decreases the protein content and that increase in both soybean and wheat flour in the blend increased the protein content of the biscuit. However, soybean flour contributed more to the increase in protein than wheat flour.

Table 4: Constraints fixed for numerical optimization of independent variables and responses

Carbohydrate content of –rice soy noodles

The suggested quadratic model on the carbohydrate content is shown in (Table 5). Based on the ANOVA, the interactive effect AB, BC and BD were significant (p≤0.05) while AC, AD and CD were not significant to the model. The lack of fit for % carbohydrate was not significant (p≤0.05) with a p-value of 0.3066. also fit statistic standard deviation, C.V%, R², adjusted R², predicted R² and adeq precision were 0.6838, 80.17, 0.8529, 0.9602, 0.9244, 0.7570 and 19.2907 respectively. The coefficient estimate of carbohydrates for the components are: A (+82.63), B (+62.11), C (+79.86), C (+80.16), AB (-19.40), BC (+25.52), and BD (+22.30), while the multiple regression equation for describing the relationship between carbohydrate and ingredients (coded form) after neglecting the non-significant terms is given below (Eq. 4)

Table 5: Numerical optimization of protein, carbohydrate, texture, appearance, taste and overall acceptability of rice soy noodles.

Y2 (CHO) = 82.63A + 62.11B + 79.86C + 80.161D - 19.40AB + 25.52BC + 22.30BD                     (3)

R² = 96.02

The Model F-value of 26.81 implies the model is significant. The interactive effect of AB, BC and BD were all significant with a p-value less than 0.0500, this is because of main effects A, C and D (rice flour, corn and potato starch) had a considerable amount of carbohydrates of about 84.05, 87.92 and 90.23% respectively (Figure 1). A similar result was also reported by warda with carbohydrate content ranging from 76.55 to 82.48% in new rice for Africa flour.

The Lack of Fit F-value of 0.97 implies the lack of fit is not significant relative to the pure error. The Predicted R² is in reasonable agreement with the Adjusted R² Observed adeq Precision is desirable (Chang et al., 2006).

From equation 3, it can be seen that there is a relative increase in soybean flour resulting in a reduction in the carbohydrate content of rice soy noodles, also the negative correlation coefficient was -0.912 (Table 3) also agrees with the equation of coded factor. From this study, an increase in soybean addition resulted in a significant (p<0.05) reduction in the carbohydrate content of the gluten-free noodles and also an increase in rice flour and corn starch resulted in an increase in the carbohydrate content of the samples. Okpalanma et al. (2020) also reported that increasing soybean flour significantly decrease the carbohydrate content of biscuit. However, there was a need to control the carbohydrate level in gluten-free noodles as an increase in carbohydrate content can increase the cooking loss in noodles (Fu, 2008). The 3D plot also reveal that increase in the setting of rice flour, corn and potato starch increased the carbohydrate content of rice soy noodle, this is due to the high carbohydrate content of 84.95, 87.95 and 90.23% in rice flour, potato and corn starch respectively (Figure 1).

Texture of rice-soy noodles

The suggested special cubic model on the texture of rice soy noodle is shown in Table 4. Based on the analysis of variance (ANOVA), all the interactive effect AB, AC, AD, BC, BD, CD, ABC, ABD, ACD and BCD were significant (p≤0.05) to the model.

The lack of fit for texture was not significant (p≤0.05) with a p-value of 0.1751. Also, fit statistic standard deviation, C.V%, R², adjusted R², predicted R² and adeq precision were 0.071, 4.98, 1.42, 0.999, 0.99, 0.913 and 133.3235 respectively. The coefficient estimate of texture for the components are: A (+ 7.69), B (-28.96), C (+5.59), D(+15.42), AB (+23.76), AC (+5.81), AD (-39.74), BC (+83.02), BD (+37.46), CD (8.41), ABC (-180.21), ABD (+251.06), ACD (+125.71), and BCD (-165.64), While the Model for texture is presented as:

Y3 (Texture) =

7.69A - 28.96B + 5.59C + 15.42D + 23.76AB + 5.81AC - 39.74AD + 83.02BC + 37.46BD - 8.41CD - 180.21ABC + 251.06ABD + 125.71ACD - 165.64BCD               (4)

R² = 99.9

The ANOVA for texture indicates the interactive effect of the independent variable. The Model F-value of 3387.34 implies the model is significant. All the interactive effects were significant with a p-value less than 0.0500, this is because all the main effects played a major role in the texture of the gluten-free noodles.

The Lack of Fit F-value of 2.49 implies the Lack of Fit is not significant relative to the pure error. Non-significant lack of fit is required for the model to fit. R- squared value is an indication of the level of response that could be explained by the model (Okpalanma et al., 2020). The result shows that 99.99% of the response could be explained by the model at 0.0001 significant level. The Predicted R² of 0.9128 is in reasonable agreement with the Adjusted R² of 0.9996; due to the difference is less than 0.2.

The regression equation in terms of coded factors can be used to make predictions about the texture for given levels of each factor. Here, the levels should be specified in the original units for each factor.

The texture of rice soy noodles was affected by the addition of corn and potato starches. This may be due to the high content of amylose content of 26 and 32 %, respectively (Figure 1). Tungtrakul (1998 ) reported that amylose content greater than 27% is recommended for production of rice noodles. Generally, the best substrate for starch noodles is considered to be starch from legumes like mung beans, which normally have more than 30% amylose content (Hou, 2010). The alkaline salt composition has a significant effect on noodle color, whereas concentration has a significant impact on noodle texture (Hatcher & Anderson 2007). There was an inverse proportionality of soybean flour concentration to the texture of rice soy noodles with a correlation coefficient of -0.841. Wu & Corke (2005) also reported that due to the protein structure, related to the potential for crosslink formation, can affect noodle texture.

Appearance of rice-soy noodles

The suggested special cubic model on the analyzed appearance of gluten-free noodles shown in Table 4. All the interactive effect AB, AC, AD, BC, BD, CD, ABC, ABD, ACD and BCD were significant (p≤0.05) to the model.

The lack of fit for texture was not significant (p≤0.05) with a p-value of 0.7808. Also, fit statistic standard deviation, C.V%, R2, adjusted R2, predicted R2 and adeq precision were 0.1119, 5.13, 2.18, 0.997, 0.999, 0.916 and 84.2113 respectively. The coefficient estimate of appearance for the components are: A (+7.09), B (-37.68), C (+4.73), D(+20.03), AB (+38.51), AC (+11.99), AD (-47.12), BC (+110.45), BD (+51.96), CD (-13.96), ABC (-234.96), ABD (+281.66), ACD (+122.49), and BCD (-249.54), While the Model for protein is presented as:

Y5 (Apperance) = 7.09A - 37.68B + 4.73C + 20.03D + 38.51AB + 11.99AC - 47.12AD + 110.45BC + 51.96BD - 13.96CD - 234.96ABC + 281.66ABD +122.49ACD - 249.54BCD -                         (5)

R² = 99.7

The ANOVA for appearance noodles’ the interactive effect of the independent variable on the appearance of glutenfree noodles. All the interactive effects were significant (p<0.05), this is because all the main effects played a major role in the appearance of the noodles.

The Lack of Fit F-value of 0.7808 implies the Lack of Fit is not significant relative to the pure error. Non-significant lack of fit was observed. The Predicted R² is in reasonable agreement with the Adjusted R. adequate precision ratio of 84.211 indicates an adequate signal

The equation in terms of actual factors (Eq. 5) can be used to make predictions about the appearance for given levels of each factor.

With coefficient of -37.68 in the regression equation, it is evident from the result that a relative increase in % soybean resulted in a reduction in the degree of likeness of the appearance. Objective colour determination of rice soy noodle colour also indicated that soybean to L* resulted in a negative correlation coefficient of -0.719. The reduction of the degree of likeness of appearance is due to the mallard reaction that occurred during drying of rice soy noodle thereby leading to browning. This study reveals that an increase in rice flour, corn and potato starch resulted in a significant (p<0.05) increase in the degree of likeness of appearance of the rice soy noodles

Taste of rice-soy noodles

The suggested special cubic model on the taste of glutenfree noodles is shown in Table 4. Based on the analysis of variance (ANOVA), the interactive effect BC, ABC and ABD were significant (p≤0.05) to the model.

Model lack of fit was not significant (p≤0.05) with a p-value of 0.5104. also fit statistic standard deviation, C.V%, R2, adjusted R², predicted R² and adeq precision were 0.4452, 5.11, 8.71, 0.994, 0.9836, 0.8256 and 21.9650 respectively. The multiple regression equation for describing the relationship between taste and ingredients (coded form) after neglecting the non-significant terms is given below (Eq. 6)

Y5 (Taste) = 7.14A - 15.55B + 8.32C + 16.18D + 38.71BC - 123.36ABC + 199.17ABD         (6)

R² = 99.4

The Lack of Fit F-value of 0.7718 implies the Lack of Fit is not significant relative to the pure error. The Predicted R²of 0.8256 is in reasonable agreement with the Adjusted R² of 0.9836. Adeq Precision measures the signal-to-noise ratio. The result showed ratio of 84.211 indicating an adequate signal which implies that the model can be used to navigate the design space.

With a coefficient of -15.65 in equation 6, it is evident from the result that a relative increase in soybean flour resulted in a reduction in taste perception. Negative correlation coefficient of -0.844 (Table 7) indicates that an increase in soybean flour resulted in a reduction in the taste of rice soy noodles while increase in rice flour, corn starch and potato starch resulted in an increase in taste score with a correlation coefficient of 0.758, 0.430 and 0.259 respectively. 3D model (Figure 2a-f) also confirms the increase in soybean setting resulted in the reduction of taste score of rice soy noodles. This study reveals that an increase in rice flour, corn and potato starch resulted in a significant (p<0.05) increase in the degree of likeness of the appearance of the rice soy noodles ( Caulibaly et al., 2006).

Overall acceptability of rice-soy noodles

The suggested special cubic model on the overall acceptability of gluten-free noodles is shown in Table 5. Based on the ANOVA, all the interactive effect AB, AC, AD, BC, BD, ABC, ABD, ACD and BCD were significant (p≤0.05) to the model while CD (corn starch x potato starch) was insignificant to the model.

The lack of fit for overall acceptability was not significant (p≤0.05) with a p-value of 0.5869. Also fit statistic standard deviation, C.V%, R², adjusted R², predicted R² and adeq precision were 0.3216, 5.04, 6.38, 0.9974, 0.9917, 0.8600 and 29.9742 respectively. The coefficient estimate of overall acceptability for the components are: A (+7.28), B (-32.97), C (+3.68), D(+13.97), AB (+30.54), AC (+13.58), AD (-32.95), BC (+97.60), BD (+48.60), ABC (-225.27), ABD (+257.11), ACD (+81.07), and BCD (-193.55), While the Model for protein is presented as:

Y6 (Overall acceptability) =

7.28A - 32.97B + 3.68C + 13.97D + 30.54AB + 13.58AC - 32.95AD + 97.60BC + 48.60BD - 225.27ABC + 257.11ABD + 81.07ACD - 193.55BCD        (7)

R² =  99.69

The ANOVA for overall acceptability indicates the interactive effect of the independent variable on the overall acceptability of gluten-free noodles. The Model F-value of 187.47 in Table 4 implies the model is significant. It is evident from the result that all the interactive effects were all significant (p<0.05) except CD this is because the main effects A (Nerica rice) and B (Soybean flour) played a major role in the overall acceptability of the gluten-free noodles.

Acceptable lack of fit was observed. The result also showed that the Predicted R² is in reasonable agreement with the Adjusted R² of 0.9916; because the difference is less than 0.2. Adeq Precision measures the signal-to-noise ratio. This model provides adeq precision of 30.628 which indicates an adequate signal. This model can be used to navigate the design space. Pamplona & Roger GM (2005) the equation in terms of coded factors can be used to make predictions about the response (overall acceptability) for given levels of each factor. Here, the levels should be specified in the original units for each factor.

From the above equation, it is evident from the result that a relative increase in % soybean resulted in a reduction in the appearance of gluten-free noodles appearance, with a negative correlation coefficient of -0.829 (Table 7). The effect of soybean flour on the gluten-free noodle appearance in the special cubic model. This study reveals that an increase in soybean addition resulted in a significant (p<0.05) reduction in the appearance of the gluten-free noodles and also an increase in rice flour (Nerica 6, corn starch, and potato starch) resulted in an increase in the appearance of the gluten-free noodles. Savita et al., (2018) also reported that the overall acceptability of multigrain gluten enriched instant noodles reduced with increase in soy flour. Collins & Pangloli (1997) also reported that sensory acceptability of sweet potato and soy flour incorporated noodles decreased with increased level of soy flour. Similar findings were also reported by Singh, Chauhan & Bains (1989) for noodles prepared for the noodles prepared with incorporation of conventional soy flour

Optimized formulations

The protein and carbohydrate content of the gluten-free noodle was optimized by maximizing the value (10-13%) and 65-75% respectively, while the texture, appearance and overall acceptability of the gluten-free noodle were in the range (7-8.5). Stat-ease Inc (2021) the response optimizer of the software generated 28 solutions of the ingredient value with the desirability of 0.24 to 0.904 and their respective protein, carbohydrate, texture, appearance and overall acceptability values as shown in Table 5. Suggested solution (1) and three other solution (3, 5 and 9) were selected and coded sample A, B, C and D. this was due to the high desirability.

Design validation

The selected blend obtained through optimization were validated by the development of gluten-free noodle and by analyzing the actual values of the responses in triplicate. The results for all responses were compared by one sample t-test (Table 6). The values of all the responses for samples A, B, C and D had a residual value of <4. Sample A was not significantly different at 5% level of significance for protein, carbohydrate, appearance, taste and overall acceptability, sample B was not significantly different at 5% level of significance for protein, texture, taste and appearance, and sample C was not significantly different at 5% level of significance for texture and taste while sample D was not significantly different at 5% level of significance for appearance, taste and overall acceptability (Catling & Alam 1983).

Table 6: Predicted value and experimental quality attribute of rice soy noodles (One sample T-test)

Validation results showed that experimental values were in good agreement with the predicted response values, thereby confirming the adequacy of the selected model. This is due to the high desirability value 0.842, 0.891, 0.894 and 0.904 for selected solutions. A desirability of 1.00 means the goals were easy to reach and better results may be available (stat-ease, 2021). Savita et al., (2018) also reported high desirability of 0.89 when selecting the optimum solution during optimization. Control of the experiment was also monitored to avoid variation in the experimental result and predicted values (Table 7).

Table 7: Effect of ingredient ratio on the proximate, sensory attribute and cooking time of rice soy noodles .

Conclusion

Gluten-free noodles developed from composite blends of rice flour (X1), soybean flour (X2), corn starch (X3) and potato starch (X4) using I-optimal mixture design were evaluated for their quality and sensory attributes. All the statistically significant terms such as R2, F value and lack of fit show the adequacy of the model. It is also reported that soybean flour had a significant impact on the protein, carbohydrate, appearance, texture, taste and overall acceptability of ricesoy noodles. Current research suggests the effect of corn and potato starch had a significant (P<0.05) effect on the textural attribute of gluten-free noodles. Furthermore, due to their high protein level, these noodles can be used as part of mid-day meals to overcome the problem of protein-energy malnutrition. Suggested solutions with better desirability have provided varied options to choose from by manufacturers that would satisfy targeted consumers. Hence, saving time and cost. Four optimum formulae were selected, for independent variables for rice flour for the selected formula were 70.00%, 73.297%, 72.995% and 72.198% for samples A, B, C and D respectively. Soybean flour predicted optimum levels were 13.133%, 11.664%, 11.585 and 11.283% while corn starch and potato starch were 11.749% and 5.117%, 5.082% and 9.956%, 5.199% and 10.222% and 11.020% and 12.528% for sample A, B, C and D respectively. The Successful utilization of these local commodities for the development of high protein-rich gluten-free noodles will definitely spread new ideas to the food sectors.

References

Adekoyeni OO, Malomo O, Ogunmoyela OAB,  Jimoh O, Oluwajoba SO, et al (2012). Rheological & functional properties of soy-poundo yam flour. Int J Food Sci Nutr. 2(6): 101-107.

Indexed at, Google Scholar, Cross Ref

Ampofo V (2009). Production & sensory analysis of soybean & wheat flour composite cake, HND Dissertation, Cape Coast Polytechnic, Cape Coast, Ghana. 1: 5-7.

Google Scholar

Baik BK & Lee MR (2003). Effects of starch amylose content of wheat on textural properties of white salted noodles. Cereal Chemistry. 80: 304–309.

Indexed at, Google Scholar, Cross Ref

Buddenhagen IW, Persley GJ (1978). Rice in Africa. In: Proceedings of a conference held at the International Institute of Tropical Agriculture Ibadan, Nigeria. 7-11 March 1977.

Indexed at, Google Scholar

Catling HD, Alam PR (1983). Agronomic practices & yield assessment of rice in Bangladesh. Field Crops Res.  6: 109 – 132.

Indexed at, Google Scholar, Cross Ref

Caulibaly M, Akator K, Ata T (2006). Existence of two pathotypes of rice yellow virus, genus sabemovirus in mali. Plant Pathology J. 5(3): 368- 372.

Indexed at, Google Scholar, Cross Ref

Chang YH, Lin CL, Chen JC (2006). Characteristics of mung bean starch isolated by using lactic acid fermentation solution as the steeping liquor. Food Chem. 99: 794-802.

Indexed at, Google Scholar, Cross Ref

Crosbie GB, Ross AS, Moro & Chiu PC (1999). Starch and protein quality requirements of Japanese alkaline noodles (Ramen). Cereal Chem. 76: 328–334.

Indexed at, Google Scholar, Cross Ref

Hatcher DW & Gerson MJ (2007). Influence of alkaline formulation on oriental noodle color and texture. Cereal Chem. 84: 253–259.

Indexed at, Google Scholar, Cross Ref

Hoseney RC (1990). Pasta and noodles. In: Principles of science & technology. American Association of Cereal Chemists, St. Paul, MN, USA. 277–291.

Hou GG (2010). Asian noodles: Science, technology, and processing. John Wiley & Sons.

Indexed at, Google Scholar

Jones B & Goos P (2012). I-optimal versus D-optimal split-plot response surface designs. J Qual Technol. 44: 85-101.

Indexed at, Google Scholar, Cross Ref

Lu S & Nip WK (2006). Manufacture of Asian (Oriental) noodles. Hand book of Food Science, Technology, and Engineering. 149.

Indexed at, Google Scholar, Cross Ref

Mestres C, Colonna P, Buleon A (1988). Characteristics of starch networks within rice flour noodles and mung  bean starch vermicelli. J Food Sci. 53: 1809–1812.

Indexed at, Google Scholar, Cross Ref

Mohammed I Ahmed AR, Senge B (2012). Dough rheology and bread quality of wheat–chickpea flour blends. Ind Crops Prod. 36:196-202.

Indexed at, Google Scholar, Cross Ref

Ocheme OB, Adedeji OE, Chinma CE, Yakubu CM & Ajibo UH (2018). Proximate composition, functional, and pasting properties of wheat and groundnut protein concentrate flour blends. Food Sci Nutr. 6: 1173-1178.

Indexed at, Google Scholar, Cross Ref

Okpalanma EF, Ishiwu CN, Chukwu C (2020). Development of biscuits made with wheat, soybean and cassava flour blends using mixture design. Int J Food Sci Nutr. 5(2): 73-79.

Indexed at, Google Scholar

Pamplona & Roger GM (2005). Encyclopedia of foods and their healing power: A guide to food science and diet therapy. Education and Health library, Hang Tai Printing Co. Ltd. Hong Kong. 24-25.

Indexed at

Pereira LJ, Van der Bilt A (2016). The influence of oral processing, food perception and social aspects on food consumption: A review. J Oral Rehabil. 43: 630-648.

Indexed at, Google Scholar, Cross Ref

Plahar WA, Okezie BO, Annan NT (2003). Nutritional quality and storage stability of extruded weaning foods based on peanut, maize and soybean. Plant Foods Hum Nutr. 58:1-16.

Indexed at, Google Scholar, Cross Ref

Porter MA & Jones MA (2003). Variability in soy flour composition.  J Am Oil Chem Soc. 80: 557-562.

Indexed at, Google Scholar, Cross Ref

Sabanis D, Tzia C (2009). Effect of rice, corn and soy flour addition on the characteristic of bread produced from different wheat cultivars. Food Sci Biotechnol. 2: 68-79.

Indexed at, Google Scholar, Cross Ref

Savita R, Rakhi S, Barjinder PK, Ashutosh U, Dinkar BK (2018). Optimization and evaluation of multigrain gluten-enriched instant noodles. Appl Biol Chem. 61: 531–541.

Indexed at, Google Scholar, Cross Ref

Sharma HR & Chauhan GS (2000). Physical, sensory and chemical characteristics of wheat breads supplemented with fenugreek. J Food Sci Technol. 37: 91-94.

Indexed at, Google Scholar

Singh N, Chauhan GS, Bains GS (1989). Effect of soyflour supplementation on the quality of cooked noodles. Int. J. Food Sci. Technol. 24: 111-114.

Indexed at, Google Scholar, Cross Ref

Stat-ease Inc (2021). Numerical optimization report - design-expert documentation.

Sun QJ (2006). Theory and technology of rice noodle processing. China Light Industry Press. 71-129.

Google Scholar

Tungtrakul P (1998). Quality and physicochemical properties of rice related to rice noodle. NFRI. 0:19.

Google Scholar

Wang Ch, Kovacs MIP, Fowler DB, Holley R (2004). Effects of protein content and composition on white noodle making quality: Color. Cereal Chem. 81: 777–784.

Indexed at, Google Scholar, Cross Ref

Wu J & Corke H (2005). Quality of dried white salted noodles affected by microbial transglutaminase. J Sci Food Agric. 85: 2587-2594.

Indexed at, Google Scholar, Cross Ref

Young SK, Dennis PW, James H, Patricia B (1996). Suitability of edible bean & potato starches for starch noodles. AACC. 73: 302-308.

Indexed at, Google Scholar

Citation: Ntukidem, James, Barber, et al (2023). Optimization of ingredients for the production of noodles made from rice flour, soybean flour, corn and potato starch -I- optimal mixture design. AJFST: 031.