Quality attributes of formulated snacks from blends of cassava starch and soy protein isolate deep fried in palmolein oil

Cassava starches from three varieties (TMS‐30572, TMS‐950289 and TME‐419) and soy protein isolate (SPI) were used to prepare dough at different proportions of 85:15 and 95:5 ratio of cassava starch to soy protein isolate. Dough were sheeted, cut and fried in palmolein oil at deep frying conditions of 170°C and 180°C for 5 and 15 min for temperature and time, respectively. Texture, chemical and colour properties were evaluated using a 2‐level full factorial experimental design (23) for three factors. The data obtained were analysed using analysis of variance (ANOVA) and numerical optimization in design expert 6.0.8. There were significant (p < .05) effects on expansion, colour and appearance. The optimized SPI level, frying temperature and frying time of 15%, 170°C and 2 min were obtained for TMS‐950289; 15%, 180°C and 4 min for TME‐419; and 15%, 170°C and 4 min for TMS‐30572, respectively.


| INTRODUCTION
Snack contributes an important part of many consumers' daily nutrient and calorie intake according to Bhattacharyya, Chakraborty, Chattory and Makherjee (1997). Gradually, snacks are taking over the main portion of the meal due to the convenience and ready-to-eat status. They are produced mostly from locally available agricultural raw materials and other ingredients as needed to make it palatable, appealing and most importantly, nutritious. Consumer awareness has increased on the importance of eating products that are wholesome and healthy, with specific characteristics though with minimal oil content. Snacks have been produced from whole flour from different crops like wheat, cassava, soybean and maize, which contains the nutrients in all forms, some of which are needed in larger quantities than the other. Building of controlled structures in food products requires understanding the behaviour of the individual constituent prior to or during processing, and product formulation provides alternative method for new products with controlled quality attributes (Aguilera, 2006). 'Formulated products based on cassava starch and soy protein isolate which is gluten-free could make it possible to investigate the effects of different product formulations on oil absorption, contributing to the understanding of the mechanisms that may be involved and its use in the production of fried products' (Gazmuri & Bouchon, 2009). These structures will be controlled to enhance uniformity in food product and helps to have a better understanding of how each element functions before and during processing.
Frying is an ancient preparation technique that impacts unique flavour and texture to food products that transcend from generation to generation. This makes fried snacks to be in high demand and appreciated by consumers, hence the reason why it is being extensively used in food processing. Tropical Manihot esculenta (TME) 419 also called NICASS 20 is a high dry matter (25%), high CMD resistance, high yielding (>25 t/ha), poundable and high starch cassava variety. Tropical Manioc Selection (TMS) 30,572 also called NICASS 1 is a high dry matter (25%), moderate CMD resistance, early bulking, high starch and high yielding (>25 t/ha), and TMS-950289 is an International Institute of Tropical Agriculture (IITA) clone, part of genetic gain collection as reported by IITA. Cassava starch has been reported to have low pasting temperature, low amylose content and a low tendency for syneresis when compared with other commercial starches. Soybean is known to be a dry powder food ingredient that has 90-95% protein and nearly carbohydrate and fat-free. The use of palmolein oil in frying snacks has been identified as a good oil as it does not emit undesirable odour, is highly resistant to oxidation, does not contain linolenic acid and has a favourable nutritional composition for being free of trans fatty acids and presenting tocopherols in its composition (Basiron, 2005). Omidiran et al. (2019) reported some findings on some characteristics of these formulated snacks deep fried in soybean oil. Hence, the objective of the study is to assess and optimize the quality attributes of fried snacks from cassava starch of different varieties and soy protein isolate (SPI) blends using palmolein oil.

| Cassava starch preparation
Fresh cassava roots were harvested, and starch was extracted from the roots. Cassava starch was produced according to the method of Aseidu (1989). The cassava root was weighed immediately after harvesting from farm. The peeled cassava root was washed in water, the washed root was grated and the screened starch was allowed to settle and decanted. The final product was packaged in high density polyethylene (Ziploc).

| Dough preparation
Preparation of the dough was carried out according to the modified method of Gazmuri and Bouchon (2009), by adding 152 ml of hot water (90-100 C) to 85:15 of cassava starch to SPI for 200 g of the mixture. For 95:5, 140 ml of the hot water was added. The mixture was kneaded until a smooth dough was obtained, sheeted and cut with a cutter to get accurate shape (diameter 10 cm and thickness 0.25 cm).

| Experimental design
A 2 3 (two-level) full factorial in the design expert software was adopted to study the interactive effect of frying temperature (170 C and 180 C), frying time (2 and 4 min) and SPI level (5% and 15%) at two levels stated, on some quality attributes of the fried snacks and the experimental runs as presented in Table 1.

| Atmospheric frying
Palmolein oil was preheated for 30 min; samples were then placed inside the frying basket of deep fat fryer (SAISHO, Model S-616, China) that was covered with a lid in order to ensure samples do not float in the frying oil during frying. Frying baskets were immersed into the oil at 170 C and 180 C for 2 and 4 min, and the fried snacks were removed and cooled after every batch. The fried snacks were then packaged in air-tight containers until further analyses. AOAC (2000) methodology was used to determine the proximate composition (total moisture content, crude fat, ash content) of the snacks. Protein content was analysed using Kjedahl method (AACC, 2000, 46-12.01). The carbohydrate content was obtained by calculating the difference from the sum total.

| Sensory evaluation
Acceptance testing method of Meilgaard, Civille and Carr (1999) was used to assess the sensory attributes (sweetness, expansion, crispness, oiliness, colour, appearance and overall acceptability) of the fried snacks prepared using the optimized frying conditions. A total of 50 untrained panelists, students of Federal University of Agriculture, Abeokuta that love eating snacks, participated in evaluating the snacks. The panelists were both male and female, age ranged between 20 and 30. The samples were served at room temperature (28 ± 2 C) and evaluated under a red lighting. Each sample attribute was rated using a 9-point Hedonic scale. The 9 points on the Hedonic Scale were dislike extremely = 1, dislike very much = 2, dislike moderately = 3, dislike slightly = 4, neither like nor dislike = 5, like slightly = 6, like moderately = 7, like very much = 8 and like extremely = 9. The average and mean values of scores for each of attributes were computed and analysed statistically.

| Expansion analysis
The procedures of Maeda and Cereda (2001) were used in the deter-

| Colour analysis
The colour parameters of the fried snacks were measured using a

| Data analysis
Data obtained were analysed using analysis of variance (ANOVA) to determine significant effect on the responses at 5% level, and models were generated showing the relationship between the independent variable and the responses using Design Expert version 6.0.8. expansion, hence its maximum value at a reduced SPI level. Gazmuri and Bouchon (2009) reported that gluten content was the most significant factor determining product expansion in predried discs of wheat gluten and starch matrices during deep-fat frying.

| Chewiness and hardness
Expansion was not significantly affected by the processing when fried in palmolein oil.

| Colour parameters
The mean value for Lightness of sample (Tables 2-4)

| Protein content
Protein content of the sample ranged from 3.28% to 6.12%, 1.31% to 6.12% and 3.06% to 6.36% for snacks from TMS-950289, TME-  to an increased frying temperature, frying time and decreased SPI level. Bouchon and Pyle (2004) reported that stronger and more elastic network can result in a less permeable outer layer that may act as an effective barrier against oil absorption. After the food has been dried completely and removed from the fryer, the product begins to cool, and water vapour condenses with a consequent decrease in internal pressure. Dana and Saguy (2006) reported that the mechanism of oil absorption occurs during the cooling phase, and this is known as cooling effect. Decrease in oil content observed could be due to the fact that the snacks were formulated from starch and protein constituents, and the surface oil was removed during cooling to prevent further uptake.

| Moisture content
The mean values of moisture content of samples ranged from 12.49% to 16.79%, 11.10% to 17.03% and 11.84% to 17.31% for snacks from  Akubo (1997) reported that the lower the moisture content of a product, the better the storage stability of the product. Ngadi, Watts and Correia (1997)

| Sensory evaluation
The result of sensory evaluation is presented in Figure 1 showing the relationship of the optimized fried snacks, based on their attributes.
All the samples were found to be moderately acceptable by the panellists because their scores were above average with overall acceptability rating ranged from 6.17 to 7.53. The sample (TME-419) was rated highest in terms of sweetness (6.67), crispness (7.47), oiliness (7.43), colour (7.27) and overall acceptability (7.53). There were no significant (p < .05) differences observed for expansion, colour and appearance. However, significant (p < .05) differences were observed in terms of sweetness, crispness and oiliness and overall acceptability of the samples. had the highest overall acceptability. TME-419 had the lowest oil content, highest expansion and chewiness. Fried snacks produced from TMS-950289 had the most preferable quality attributes at frying temperature of 170 C, frying time of 2 min and SPI of 5%.

ACKNOWLEDGEMENTS
The effort of Cassava Breeding Unit, IITA, Nigeria is acknowledged and appreciated for the provision of cassava roots used and laboratory space for processing into starches. Department of Food Science and Technology, Federal University of Agriculture, Abeokuta, Nigeria is also acknowledged for their support.

CONFLICT OF INTEREST
No conflict of interest was declared by the authors.

ETHICS STATEMENT
I hereby declare that this manuscript has not been published elsewhere, there is no conflict of interest and all co-authors met the criteria for authorship. The study does not require any ethical approval. If any error is subsequently found in the manuscript, I will inform the journal.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.

FUNDING INFORMATION
The study was self-funded.