Improving the nutritional quality and maintaining consumption quality of akara using curdlan and composite flour

Abstract

Soybean flour and curdlan were incorporated into cowpea flour to determine their effect on lowering the fat content and on the physical properties of akara. At 20% substitution, soybean flour lowered the fat content of akara by 7.7% and increased the protein content by 28.7% without significantly affecting the firmness or the colour of akara. Addition of 1% curdlan decreased the fat content by 32.2% but significantly increased the firmness of akara and produced a darker-coloured product. The paste moisture content of akara containing 20% soybean flour and 1% curdlan was modified to obtain product characteristics comparable with the control (100% cowpea flour) while maintaining a lower fat content. Optimum results were obtained for paste with 63% moisture content. Firmness of this modified product was similar to the control and the fat content was lower (17%) compared with the control (26%).

Introduction

Deep-fat frying is a convenient method of food preparation; while it has accelerated US acceptance of ethnic foods, bolder flavours, and added a crunch to traditional meals (Sloan, 2000) it has also resulted in increased obesity, heart-related diseases, as well as some cancers. It is defined as the process of cooking foods by immersing them in edible oil or fat which is at a temperature above the boiling point of water, typically 150–200 °C (Farkas et al., 1996).

Many snacks and dishes all over the world use fried batter as a constituent, where fried batter is defined as a flour–water mixture which is deep-fried in hot oil (Mohamed et al., 1998). A fried cowpea batter product popular in West Africa is akara. The batter is made with imbibed dry seeds or cowpea flour and water, seasoned with bell or hot peppers, onions, and salt and deep-fat fried at 193 °C. Research has shown that akara contains about 20–25% fat, most of which is located in the crust of the product.

The consumer's preference for low-fat and fat-free products has been the driving force of the snack food industry to produce reduced fat products that still retain desirable texture and flavour (Garayo & Moreira, 2002). Research has identified several methods of lowering the oil content of foods such as the use of vacuum frying (Garayo & Moreira, 2002), edible films (Williams & Mittal, 1999) and modifiers (Patterson, 2002; Falade et al., 2003).

Soybean is an important source of high quality but inexpensive protein and oil. With an average protein content of 40% (of total dry matter) and oil content of 20%, soybean has the highest protein content of all food crops, and is second only to groundnut in terms of oil content among food legumes (International Institute of Tropical Agriculture: IITA, 2004). Studies have shown that addition of soybean flour to fried food formulations plays a role in reducing the oil uptake of the product during frying or on the overall fat content. Significant reductions in oil absorption with incorporation of soybean flour have been achieved for various food products (Johnson, 1970; Wolf & Cowan, 1971; Martin & Davis, 1986; Mohamed et al., 1995; Huse, 1996). Studies done by Wolf & Cowan (1971) and Mohamed et al. (1995) hypothesised that the decrease in oil absorption may be related to the rate of denaturation of the protein, and that the denatured proteins form a fat-resistant barrier.

Curdlan is a water-insoluble polysaccharide (β-1,3-glucan) (Fig. 1) produced by Alcaligenes faecalis var. myxogenes, and has a specific character to make an irreversible gel by heating of a water suspension (Harada et al., 1966). In its natural state, curdlan exists as a granule much like that of starch which is insoluble in distilled water, but dissolves easily in dilute alkali solution (Cheeseman & Brown, 1995). It is colourless, odourless and tasteless and is indigestible. Curdlan is used as a formulation aid, processing aid, stabiliser, and thickener or texturiser for use in food and was approved for these uses by the US Food and Drug Administration (FDA, 1996) in December 1996. Approval of curdlan was based on the fact that it consisted of a glucose polymer and a small amount of inorganic salts, mainly sodium chloride; it is non-toxic and the producing organism is also non-pathogenic; and there is a history of safe consumption of similar glucose polymers in food (Spicer et al., 1999). Unlike other gelling agents such as carrageenan, agar-agar, HM pectin, gellan and gelatin gel which gel after heating, then cooling, curdlan gels on heating alone (Deis, 1997) and its gelation mechanism has been reported by Zhang et al. (2002). Depending on the temperature, it forms two types of heat-induced gel: low-set (55–60 °C) and high-set (80 °C) (Harada et al., 1987). According to Funami et al. (1999), the thermal gelling properties of curdlan can be considered as a potential oil barrier-forming ingredient for fried foods. In their research on using curdlan to reduce the oil uptake of doughnuts during deep-fat frying, addition of 0.1–0.5% curdlan significantly decreased the final lipid content and oil uptake of doughnuts.

The objectives of this study were to:

• 1

  Evaluate the effect of soybean flour and curdlan on the fat content of akara and determine their effect on the physical properties and proximate composition of the product;

• 2

  Modify the moisture content of paste containing soyflour and curdlan to obtain product characteristics similar to the control while maintaining improved nutritional quality.

Materials and methods

Cowpea flour production

Dry cowpea seeds (breeding line UCR 97-15-33) were obtained from Inland Empire Foods, Inc. (Riverside, CA, USA) and stored at 2 °C until used. These cream-type seeds were milled into flour without decortication using a hammer mill (Champion, Model no. 6X14; Champion Products Inc., Eden Prairie, MN, USA) equipped with a 2.54-mm screen.

Soybean flour and curdlan

Partially defatted soybean flour was provided by American Soy and Tofu Corporation, Macon, GA, USA. Curdlan was obtained from Sigma-Aldrich Corporation, St Louis, MO, USA, and was stored at 2 °C until used.

Preparation of cowpea paste and akara

Cowpea paste was prepared by adding sufficient water to a cowpea–soyflour mixture or cowpea flour–curdlan mixture to obtain a final paste moisture content of 61%. Paste containing soyflour was prepared by substituting 0%, 5%, 10%, 15% and 20% soyflour for cowpea flour and adding the appropriate amount of water. These levels of soyflour were chosen based on previous research by Huse (1996) where a 26% fat content reduction was obtained for up to 6% level of incorporation of soyflour. Paste containing curdlan was prepared by adding 0%, 0.3%, 0.5%, 0.7% and 1% (based on 230 g of cowpea paste) to 100 g of cowpea flour and adding the appropriate amount of water. These levels of curdlan were chosen based on a previous research by Funami et al. (1999) on reducing oil uptake in doughnuts. The mixture was then stirred gently with a rubber spatula and allowed to stand for 15 min. The resulting paste was whipped for 1.5 min in a household mixer (Model 2366; Sunbeam Corp., Delray Beach, FL, USA) at speed 12 (high). During whipping, a rubber spatula was used to continuously scrape off the paste that adhered to the sides of the mixing bowl to ensure effective mixing. Pastes not containing salt, pepper, and onion were used for physical property evaluations.

After whipping, 9.5% chopped bell peppers, 9.5% chopped onions, and 1.5% salt were manually folded into the paste with a rubber spatula. A #40 ice cream scoop (c. 20 mL) was used to dispense the paste into hot (193 °C) peanut oil in an atmospheric fryer (Kitchen Kettle^TM electric multi-cooker; National Presto Industries, Eau Claire, WI, USA). After 3 min frying (1.5 min on each side) akara balls were drained on absorbent paper towels, cooled to room temperature, counted, weighed (five samples per batch), and used for texture and colour measurements.

Effect of moisture content of composite flour paste on akara quality

Preliminary studies conducted to determine the effect of soybean and curdlan showed that incorporating these modifiers into cowpea flour decreased the foaming capacity of the paste. Composite flour containing both curdlan and soybean flour at a level that resulted in the greatest decrease in fat content and greatest increase in protein content of akara from the first part of the study was identified. This mixture was used for the study on moisture content, following the same steps as described above except the moisture content of the composite flour mixture was adjusted to achieve 61%, 63% and 65% paste moisture content. Akara was prepared as described above.

Apparent viscosity and specific gravity measurements

The apparent viscosity of whipped pastes was determined at 23 °C using a digital viscometer (Model HATD; Brookfield Engineering Laboratories, Inc., Stoughton, MA, USA) equipped with a T spindle [number C, cross bar length of 1.064 in (27.1 mm)] at 10 r.p.m. Whipped cowpea paste (250 mL) was poured into a 600-mL beaker and tapped ten times on the heel of the palm to remove any air pockets. An inbuilt leveller was used to level the instrument. Readings were recorded continuously by an attached recorder for the same sample as the spindle was immersed to a depth of 2.5 cm and returned to the surface (total distance = 5 cm). The readings were read off a graph and the average taken. The apparent viscosity (Pa s) was calculated by the following formula:

where spindle factor = 20 M/r.p.m., M = 1000, and r.p.m.=10.

Specific gravity of unwhipped and whipped pastes was determined in triplicate according to the modified method of Kethireddipalli et al. (2002).

Instrumental colour measurement

Five akara balls were randomly selected from each batch; the Hunter colour values (L*, a*, b*) of the exterior surface were measured using a Minolta colorimeter (Model CR-200; Osaka, Japan) calibrated with a brown reference tile (L* = 69.82, a* = 19.17 and b* = 31.75). Each ball was measured twice, one on either side. Hue angle was calculated as tan⁻¹(b*/a*), chroma as (a*² + b*²)^1/2 and total colour difference (ΔE) as [(L* − L* reference)² + (a* − a* reference)² + (b* − b* reference)²]^1/2.

Instrumental texture measurement

An Instron universal testing machine (Model 5544; Instron, Inc., Canton, MA, USA) fitted with a 2000 N load cell was used to determine the texture of akara balls. The same akara balls used for the colour measurements were used. A cube of 1 cm was cut from the crumb portion of the akara ball and compressed twice in reciprocating motion each time to 25% of its original height at a crosshead speed of 50 mm/min (Kethireddipalli et al., 2002). Peak heights were measured and the force required to shear was reported as Newtons (N). Firmness, cohesiveness, chewiness, and springiness were calculated from the force–deformation curve as follows: firmness = maximum force from the first compression (Newtons); cohesiveness = the strength of the internal bonds making up the body of the product (ratio of positive force areas under first and second compressions); springiness = distance over which the sample recovers its height between the end of the first compression and the start of the second compression; and chewiness = firmness × cohesiveness × springiness (Szczesniak, 2002).

Proximate analysis

The moisture content of akara samples was determined by grinding and vacuum drying at 70 °C for 24 h (American Association of Cereal Chemists: AACC, 1976, method 44-40). The fat content was determined by extraction with petroleum ether for 24 h in a Goldfisch apparatus (Labconco, Kansas City, MO, USA) (AACC, 1976, method 30-26). Ash content was determined by using the moisture-free and fat-free samples (AACC, 1976, method 08-01). Protein content was determined by the nitrogen combustion method (LECO, FP-2000, Warrendale, PA, USA), using moisture-free and fat-free samples. A factor of 6.25 for cowpea and 5.71 for soybean was used to convert nitrogen to protein content (Food and Agriculture Organization of the United Nations: FAO, 1970). The carbohydrate content was determined by subtracting the sum of ash, fat and protein content from 100%.

Statistical analysis

All data were analysed using analysis of variance (Anova) procedures from the Statistical Analysis System (SAS). Mean comparisons were performed using Duncan's Multiple Range Test (SAS, 2000).

Results and discussion

Specific gravity and apparent viscosity measurements

Foaming properties are important to the textural character of akara (Kethireddipalli, 1999). Whipping of the paste is essential for formation of foam (Patterson et al., 2002), and it incorporates air into the paste and helps to evenly distribute the air bubbles. Specific gravity measurements of the paste can be used to determine the foaming capacity. The higher the foaming capacity of a paste or mixture, the greater the amount of air incorporated into it during whipping and the lower will be its specific gravity (Campbell et al., 1979).

Specific gravity values of soy-substituted cowpea flour pastes (Table 1) show that specific gravity of the whipped pastes was significantly affected by the addition of soyflour. The addition of soyflour significantly increased the specific gravity of the pastes after whipping. The lowest specific gravity after whipping was obtained for the control (0.65) which contained no soyflour and the highest was obtained for samples containing 15% and 20% soyflour (0.74). Soyflour used in the formulation contained twice as much protein (48%) as cowpea flour (24%) thus the formulations containing soyflour will have higher protein content than the control. According to Wolf & Cowan (1971), soy proteins contain numerous polar side chains along their peptide backbones, thus making the protein hydrophilic. As a result, addition of soy proteins such as flours to food products increases water absorption as the proteins absorb water and tend to retain it in finished food products. Aeration of food products produces foams which are defined by Niranjan (1999) as gas bubbles separated by thin films and these thin films are liquid layers. Thus due to increased water absorption of the soy-containing pastes, less liquid was made available for the formation of bubbles to produce a foam, resulting in the reduced foaming capacity of these pastes as the amount of soyflour was increased.

The addition of curdlan to the cowpea flour resulted in a significant increase in the specific gravity of the paste after whipping (Table 1). According to Damodaran (1994), the basic requirements for a protein to be a good foaming agent include the ability to rapidly adsorb at the air–water interface during whipping. Curdlan absorbs and retains moisture thus making the moisture unavailable to the proteins of the cowpea flour. As the proteins were not sufficiently hydrated, poor foaming capacity resulted, which caused a reduction in the specific gravity of the paste. There was an increase in the specific gravity of the pastes after whipping as the amount of curdlan was increased from 0.3% to 1%. The lowest specific gravity was obtained for the control (0.65) and the highest was obtained for the paste with the highest quantity of curdlan (1% curdlan) (0.86). Comparison between soy-containing akara and curdlan-containing akara showed that addition of 1% curdlan to akara formulation resulted in the highest decrease in foaming capacity of the paste. Effect on the foaming capacity of the paste by adding 0.5% curdlan was similar to that obtained by incorporating 15 g and 20 g soyflour to akara formulation.

Increasing the soy content of the paste resulted in a significant increase in the apparent viscosity (Table 1) with the control paste having the lowest apparent viscosity (49.75 Pa s) and paste containing the highest amount of soyflour (20%) having the highest apparent viscosity (74.00 Pa s) for soy-containing pastes. Apparent viscosity is an indication of resistance to flow. Increasing the amount of air incorporated into the paste due to an increase in foaming capacity will result in pastes with less resistance to flow. As expected, increasing the soyflour content of the paste resulted in decreased foaming capacity, thus, the higher the soyflour content of the paste, the higher the resistance to flow (apparent viscosity).

As shown in Table 1, the addition of curdlan to cowpea flour resulted in an increase in the apparent viscosity of the pastes. Increasing the amount of curdlan increased the resistance to flow of the paste resulting in an increase in the apparent viscosity. This is also because curdlan absorbs moisture making it unavailable to the paste, thus making the paste thicker and more resistant to flow. The highest resistance to flow for curdlan-containing pastes was observed in the paste containing 1% curdlan which had the highest apparent viscosity (148.00 Pa s) and the lowest resistance to flow was observed in the control paste (49.75 Pa s). Apparent viscosity of paste containing 1% curdlan was twice that obtained for paste containing 20 g soyflour. This is an indication that incorporating 1% curdlan caused a higher increase in moisture absorption compared with incorporating 20 g soyflour in akara formulation. These results are confirmed in values obtained for reduction in specific gravity (Table 1) where a lower reduction in specific gravity is an indication of a higher degree of reduction of foaming capacity of the paste due to the treatment. The results show a decrease in reduction of specific gravity with increasing amounts of soyflour and increasing amounts of curdlan, with paste containing 20 g soyflour exhibiting almost twice the reduction in specific gravity (29.56%) compared with paste containing 1% curdlan (17.13%).

Texture

The textural quality of the soy-containing akara samples (Table 2) was not significantly affected by the addition of soyflour; however, curdlan-containing akara (1%) showed a significant increase in firmness (Table 2). The firmness of akara containing 0.3%, 0.5% and 0.7% curdlan was not significantly different from the control but on addition of 1% curdlan, the firmness of the akara increased (9.10 N, compared with 7.42 N for the control). Cohesiveness is defined as the extent to which a material can be deformed before it ruptures (physical definition) or the degree to which a substance is compressed between the teeth before it breaks (sensory definition) (Szczesniak, 2002). There was a significant decrease in cohesiveness as the amount of curdlan increased with akara containing 1% curdlan being the least cohesive (0.14). Curdlan-containing akara balls were generally less cohesive and less springy than soy-containing akara.

Instrumental colour

The Hunter system of colour measurements defines L* as the darkness of a product with L* = 100 as white and L* = 0 as black. It also defines a* as the redness and b* as the yellowness of a product. There was no significant change in the darkness of akara containing soyflour for all samples except at the 15% soyflour level where a slight darkening of the colour (L* = 59.47) occurred (Table 3). There was also no significant change in the redness or yellowness of the soy-containing samples except at the 20% soyflour level where decreased redness and yellowness were observed. As a result, the total colour change (ΔE) due to treatment was not significantly different for all of the samples except for akara containing 15% soy (ΔE = 13.47). According to McWatters et al. (2001), for akara, hue angles between 40° and 75° represent brown colours with a lower hue angle indicating more brown colour than a higher hue angle. The results show that there was no significant change in the hue angles of the soy-containing akara except at the15% soyflour level where more brown colour (hue angle = 71.52) developed. Chroma indicates how saturated or intense a colour is; the results show that colour saturation for all of the samples was not significantly different except for akara containing 20% soyflour which exhibited lower colour saturation (33.06). This could be due to the fact that this sample was less yellow (lower b*) and less red (lower a*) than the rest of the samples.

The results show that increasing the amount of curdlan resulted in a significant increase in the darkness (lower L*) of the akara. The control samples were lighter (L* = 62.46) than akara containing curdlan, with akara containing 1% curdlan having the darkest colour (L* = 55.69). Curdlan remains colourless after heating; therefore, the change in colour of akara could be due to the fact that curdlan absorbed moisture from the paste, thus the lower water activity may have favoured an increase in the Maillard reaction, causing the crust of the akara to brown faster. Addition of curdlan also increased the redness (higher a*) and yellowness (higher b*) of the akara, compared with the control. Consequently, hue angles decreased as the amount of curdlan increased. The control had the highest hue angle (73.99) depicting less brown colour than curdlan-containing akara (68.42–71.35). Akara containing 0.5–1% curdlan had similar hue angles, which were lower than that of akara containing 0.3% curdlan. Akara containing curdlan also had more saturated, intense (higher chroma values) colour than the control. Thus there was a significant change in total colour due to addition of curdlan as shown by ΔE values. The control showed the least change (12.14) while akara containing 1% curdlan showed the most colour change due to treatment (15.15). A general comparison between soy- and curdlan-containing akara showed that curdlan-containing akara was darker than soy-containing akara, as depicted by the lower L* values. Also, curdlan-containing akara exhibited a higher degree of redness (higher a* values) and yellowness (higher b* values), thus having lower hue angle values (more brown colour). Colour saturation was also higher for curdlan-containing akara and as a result, the total colour change due to treatment was higher for curdlan-containing akara than for soy-containing akara.

Weight, number and proximate composition of akara

The lower the foaming capacity of the paste containing soyflour, the higher the weight of the finished product (Table 4). Increasing the amount of soyflour in the sample resulted in a significant increased weight of akara balls with the control weighing the least (17.48 g) and the samples containing 15% and 20% soy weighing the most (19.83 and 19.91 g, respectively) for soy-containing akara. This is because of the decreased foaming capacity of the pastes resulting in more solids per volume and less air incorporated as the concentration of soyflour increased. Thus there was a corresponding decrease in the number of balls obtained per batch (100 g composite flour), however, this decrease was not statistically significant.

Increasing the amount of curdlan added to cowpea flour resulted in a progressively significant increase in the weight of akara balls obtained (Table 4). This is due to absorption of moisture by the curdlan, resulting in poorer foaming capacity of the paste. There was also a decrease in the number of akara balls obtained per 100 g batch of cowpea flour compared with the control. Akara prepared with curdlan generally weighed more than that prepared with soyflour.

Table 4 also shows the proximate composition of the soy-containing samples. There was a significant decrease in the fat content of the samples with the control exhibiting the highest fat content (25.40%). With the addition of 5% soyflour, the fat content decreased to 23.23%, then to 22.38% and 22.69% for samples containing 10% and 15% soyflour, respectively. Addition of 20% soyflour then resulted in a slight increase in the fat content to 23.44%. Increasing the soyflour content of the akara resulted in a significant increase in the protein content (Table 4). The control contained the lowest protein content (16.87%) while samples made with 15% and 20% soyflour contained the highest (21.21% and 21.72%, respectively). Substituting cowpea flour with 20% soyflour resulted in a 28.7% increase in the protein content. Substituting cowpea flour with soyflour to make akara was not only beneficial in lowering the fat content of the product but also in increasing the protein content. Cowpeas and soybeans are low-cost sources of protein; their usage in foods will not only benefit the developing countries where meat and eggs are expensive and sometimes unavailable to the poorer regions but also the developed countries where a low-carbohydrate, high-protein diet has been shown to be healthy.

Meats and produce are a major part of a low-carbohydrate diet which according to a recent article by Golden (2004) can put a strain on pocket books. This is because according to this article, the Food Marketing Institutes 2004 Trends Report states that a one-person household spends on average $59 a week on groceries, but to follow a low-carbohydrate meal plan, that cost jumps to $99.89 for the Atkins Diet and $91.28 for the South Beach Diet. This almost doubles the grocery cost. As a result, the national health interview study shows that 26% of those with income less than $17 000 are overweight compared with 18% for those making over $67 000 per year. So for those on a limited income, a diet consisting of animal proteins causes a financial strain. Supplementing cowpea with soybeans improves the protein content of a product due to the high protein content of both cowpeas and soybeans. In the US cowpeas are mainly boiled and eaten as-is and soybeans are processed for oil. Akara made from these two legumes will provide a more nutritious and innovative alternative use and encourage increased production of these two legumes in the country.

Table 4 shows a slight increase in the protein content of the curdlan-containing akara, compared with the control. There was a slight increase in the ash content of akara containing 0.7% and 1.0% curdlan. Addition of curdlan resulted in a significant decrease in the fat content of akara. Addition of 1% curdlan to akara formulation resulted in a 32.2% decrease in the fat content of the product. The control sample had the highest fat content (25.40%) while akara with 1% curdlan had the lowest (17.23%) (Table 4). This could be due to the fact that curdlan absorbs moisture and retains it in the structure of the product, thus preventing loss of moisture during frying. As a result, fewer voids are left in the product to be filled by the frying oil. This is confirmed by the results obtained for the moisture content of akara which showed a significant increase with increase in the amount of curdlan added.

Comparing all samples, addition of 0.7% and 1% curdlan resulted in the greatest decrease in fat content. Generally, soy-containing akara had higher protein content than curdlan-containing akara; this is because soyflour contains protein and curdlan does not. Addition of 20 g soyflour resulted in the greatest increase in the protein content of akara while addition of 1% curdlan resulted in the greatest decrease in the fat content of akara.

Effect of moisture content of composite flour paste on akara quality

The thickening and foaming property of proteins is affected by the extent of interaction with solvent water (Kinsella & Damodaran, 1981). Table 5 shows the specific gravity and apparent viscosity of cowpea pastes with specified formulations. Increasing the moisture content of the paste resulted in a decrease in the specific gravity after whipping. The reduction in paste specific gravity after whipping is a good measure of foaming capacity (Kethireddipalli et al., 2002) because according to Campbell et al. (1979), the greater the amount of air incorporated into the paste the lower will be its specific gravity. There was a significant difference in the specific gravity of the pastes either at 61%, 63% or 65% moisture content. The lowest specific gravity after whipping was obtained for the control paste (0.65), followed by the paste with the highest moisture content (65%) with a specific gravity of 0.77. The highest specific gravity (0.97) was obtained for the paste containing 20% soyflour, 1% curdlan, and 61% (lowest) moisture content. Curdlan tends to absorb moisture so less moisture is available to hydrate the soy and cowpea proteins. Soyflour has twice as much protein (48%) as cowpea flour and these proteins are hydrophilic. An increase in the protein content thus led to an increase in the absorption of moisture and due to the increase in water absorption, less moisture was made available for the formation of bubbles to make foam. Thus, increasing the moisture content of the paste led to an increase in its foaming capacity.

Increasing the moisture content of the paste also resulted in a corresponding decrease in the apparent viscosity. As more water was made available in the paste due to the increase in moisture content, more bubbles were formed leading to incorporation of air into the paste. Paste with 63% moisture content was similar in apparent viscosity (47.17 Pa s) to the control (48.30 Pa s). The lowest apparent viscosity, however, was obtained for paste with 65% moisture content (28.67 Pa s).

The textural quality of akara obtained from the modified cowpea pastes was significantly affected by the change in moisture content. Table 6 shows that the firmness of the control was similar (7.63 N) to akara made from pastes with 61% (7.09 N) and 63% (6.87 N) moisture content; however, akara made from paste with 65% moisture content was less firm (6.17 N). Proper hydration of the cowpea proteins in the paste led to an increase in the foaming capacity and a decrease in the apparent viscosity, as shown in Table 5. According to McWatters et al. (1988), the foaming capacity of paste is significant in determining the textural quality of akara. Cohesiveness and springiness of akara made from paste with 65% moisture content were similar to the control. The control sample was more chewy (0.76 N mm) than the samples containing soyflour and curdlan. This may be due to the greater strength of internal bonds making up the body of the product.

Akara prepared from the control paste was significantly lighter (higher L*) than akara made from the other pastes (Table 7). The hue angles show that akara prepared from the control paste and from paste with 65% moisture content was less brown (73.68° and 73.45°, respectively) than akara prepared from paste with 61% and 63% moisture content. The lower chroma values (lower colour saturation) of these products (control and akara prepared from paste with 65% moisture content) can be attributed to their lower a* (less redness) and lower b* (less yellowness) values.

Akara balls prepared from the control paste weighed less than the rest of the samples (17.57 g) (Table 8) due to higher foaming capacity of the paste (Table 5) which resulted in a more porous structure. Akara balls prepared from paste containing 20% soyflour, 1% curdlan, and 61% moisture content weighed more (23.6 g) than the rest of the samples. Increasing the moisture content of the paste resulted in a decrease in the weight of the akara balls as well as an increase in the number of balls obtained from each batch of paste. However, more akara balls (14) were obtained from the control than from the soy/curdlan-supplemented samples, and these weighed less because the control had the highest foaming capacity.

Akara prepared from the control paste had the lowest moisture content (51.7%) (Table 8). The control paste and paste containing 20% soyflour, 1% curdlan, and 61% moisture content were both prepared with enough water to obtain the same initial paste moisture content, but the final moisture content of the cooked product prepared from the control had lower moisture content. This is because the presence of the soyflour and curdlan in the rest of the samples retained moisture in the product during frying. During frying, as the temperature increases, moisture evaporates from the product leaving voids that are filled by the frying fat. Akara from the control paste thus had more fat than the rest of the samples. The fat content of akara increased with increasing moisture content of paste due to an increase in the foaming capacity of the pastes; increasing the foaming capacity of the paste provided a product structure that was open and porous, resulting in greater oil absorption during frying. Akara prepared from pastes containing soyflour and curdlan contained more protein compared with the control due to the presence of the soyflour. Ash values for all samples were similar.

Conclusions

This study has shown that soybean flour and curdlan were successfully used individually to reduce the fat content of akara. While they each reduced the foaming capacity of the paste, soybean flour did not affect the texture of akara. The addition of soybean flour to the akara formulation had the added benefit of increasing the protein content of the end product. Although curdlan reduced oil absorption during frying, it increased the firmness of the end product compared with the control.

Incorporating both soybean flour and curdlan into the akara formulation further decreased oil absorption; however, this resulted in a further decrease in the foaming capacity of the paste and increased firmness of the end product. Proper hydration of cowpea composite paste was essential for producing akara that had similar physical/functional properties to traditionally made, good quality akara. Increasing the moisture content of the paste produced pastes with foaming and flow properties comparable with the control as well as less firm end products. However, the best qualities were obtained for paste with 63% moisture content because in addition to having similar flow properties as the control, the firmness of the product was similar, while the fat content (17.06%) was significantly less than the control (26.3%). As a result, proper hydration of the paste resulted in a product with similar physical characteristics as the control but with improved nutritional (fat and protein content) qualities.

Acknowledgments

This study was supported by the Bean/Cowpea Collaborative Research Support Program (Grant no. DAN-1310-G-SS-6008-00), U.S. Agency for International Development, and by state and Hatch funds allocated to the University of Georgia Agricultural Experiment Station-Griffin Campus. We are grateful to Glenn Farrell and Lary Hitchcock for their technical support.

References