TAPIOCA AND RICE FLOUR COOKIES : TECHNOLOGICAL , NUTRITIONAL AND SENSORY PROPERTIES

Tapioca flour is derived from the starch extracted from manioc and is a widely used food product in Brazil. Rice flour is produced from grains of rice and is used in the production of bread, porridge, cakes and cookies, which are recommended for people with celiac disease. The goal of this work was to add value to the aforementioned products by developing cookies based on tapioca and rice flours. Five formulations were prepared: A 100:0, B 75:25, C 50:50, D 25:75 and E 0:100 to tapioca and rice flour respectively, with the addition of brown sugar, and analyses its technological, nutritional and sensory properties. The following physical, physicochemical and nutritional properties were analyzed: dough texture profile, cookie weight, diameter and volume, acidity, water activity, carbohydrates, lipids, proteins, dietary fiber, ash content and moisture. A sensory evaluation was held using an affective test with 90 judges and a structured nine-point hedonic scale ranging from ‘like extremely’ to ‘dislike extremely’ for the attributes scent, color, texture, taste and overall impression in addition to purchase intent. The results indicate that cookies made of tapioca and rice flours with the addition of brown sugar have technological, physicochemical and nutritional profiles within legal standards; however, the fiber contents were below recommendations. The sensory evaluation showed good acceptance of the cookies, with average scores above 7.0. This study is part of an attempt to raise further discussions regarding the production of new low-cost bakery products that are nutritionally enriched, viable and easily accessible to all, including to people with celiac disease.


INTRODUCTION
In many countries, the names cassava starch, tapioca flour and tapioca starch are often mistaken as denoting tapioca flour while actually denote manioc starch (Manihot esculenta, Crantz), which is a different product (Milde et al., 2010;Poongodi, Vijayakumar;Boopathy, 2012;Silva et al. 2013).Tapioca flour is defined as a product obtained in granulated form manioc starch, and when subjected to the appropriate technological processing, it is found in the form of irregular, polyhedral or spherical granules (Brasil, 2005a).However, tapioca flour has several advantages for Tapioca and rice flour cookies...

515
Ciênc.Agrotec., Lavras, v. 39, n. 5, p. 514-522, set./out., 2015 consumption because it is produced throughout most of Brazil, has a relatively low-cost production chain, is quite palatable, has a long shelf life and is used in the preparation of many foods.Rice flour (Oryza sativa, L.) is obtained from milling polished or whole rice and contains large amounts of starch and dietary fiber, which are sources of energy and important for intestinal maintenance, respectively.Tapioca and rice flours do not contain gluten and, when used together, can boost the production of bakery products, such as glutenfree cookies, that are recommended for individuals with celiac disease.Cereda (2002) states that starch can be added to cookies to standardize the gluten content of the flour in a proportion of 15 to 20% of the weight of wheat flour.This process does not cause technological problems or alter the appearance or other fundamental characteristics of the cookies.In general, cookies made with mixed flour-such as the tapioca and rice flour mixture in this study are better accepted by the public because they taste better and are lighter than conventional cookies.The nutritional value of tapioca can be boosted by adding brown sugar and rice flour during cookie production, which may make the cookies nutritionally richer in addition to providing an easily accessible and longlasting product.At the same time, by boosting the nutritional value of these products, the identity, economy and culture of Brazilians are strengthened.In addition to the nutritional improvement, taking into consideration the addition of iron from the brown sugar, the present study aims to contribute to increase the availability of gluten-free products.People with gluten intolerance, called celiac disease, number 25 million worldwide which justifies research into alternatives to replace wheat in bakery products.

MATERIAL AND METHODS
The experimental design was random with a mixture of two factors (tapioca and rice flours), with five replicates for each analysis and three repetitions.The formulations and physical analysis of the cookies were performed in the Food Technology Laboratories of the State University of Feira de Santana (Universidade Estadual de Feira de Santana -UEFS), Bahia, Brazil.The protein analysis was conducted in the Bromatology Laboratory of the State University of Bahia (Universidade do Estado da Bahia -UNEB), Bahia, and a dietary fiber analysis was conducted at the Federal University of Viçosa (Universidade Federal de Viçosa -UFV), Minas Gerais, Brazil.

Technological development of the cookies
The ingredients for preparing the cookies were tapioca flour bought at an open-air market in the city of Salvador.Rice flour (Dular brand), brown sugar (Dular brand), eggs, vegetable shortening (margarine with 60% lipids, Deline brand) and sodium chloride were obtained from local markets of Feira de Santana city .The granulated variety of tapioca flour was chosen because it is easier to manipulate, is widely used by the population and provides the best sensory and visual qualities for cookie production.However, the influence of finer flours' grain granulometry is not a synonym of quality, and its mechanism is not fully understood (Borges et al., 2006).The tapioca flour and brown sugar were passed through a cloth sieve with 1.19 mm openings (Bertel brand).All the ingredients were weighed with a Shimadzu semi-analytical scale (except for the eggs).
The cookies formulations are show in Table 1.

Steps in the production of the tapioca and rice flour cookies
To produce the cookies, the 'creaming' method (Dendry; Dobraszczyk, 2001) was used, with the following steps: selecting and weighing of the ingredients, except eggs; mixing brown sugar and vegetable shortening to obtain a homogeneous cream; adding the tapioca and rice flours to obtain a homogenous dough; and then adding the eggs and kneading by hand for five minutes to obtain a firm dough.After the mixture was ready, it was stretched with a rolling pin and cut with a standard mold into 1 cm high and 4.4 cm diameter cookies.After being cut, the cookies were placed on rectangular aluminum trays and baked in an electric oven (Progas brand) at 150 ºC for 15 minutes.

Sample treatment
To conduct the technological, nutritional and sensory analyses, the baked samples were taken in significant and representative amounts, homogenized and, in some cases, crushed in a mortar with a porcelain pestle and then sieved.The samples were kept away from light and humidity and free from biological contamination.

Physical analyses
The cookie physical characteristics were assessed according to the 10-50D Method (American Association Of Cereal Chemists -AACC, 2000) cited by Moraes et al. (2010).The weight was determined using a Shimadzu semianalytic digital scale, before baking, and expressed in grams.The diameter was determined using a caliper after baking and expressed in centimeters.The specific volume was calculated as the ratio between the apparent volume (determined by the method of millet seed [Panicum miliaceum L.] displacement) and the cookie's weight after baking; the expansion factor was calculated as the ratio between the diameter of the unbaked dough and the cookie's diameter after baking.
The texture profiles of the unbaked dough and the baked cookies were analyzed with a TA.XT.plus texturometer and the Exponent software from Stable Micro Systems.The following dough parameters were measured: hardness, adhesiveness, elasticity, cohesiveness, gumminess and chewability, according to the methodology described by Assis et al. (2009) and Moraes et al. (2010).The following parameters were measured using the baked cookies: hardness and fracturability according to Assis et al. (2009) and Moraes et al. (2010).

Physicochemical and nutritional analyses
The moisture content of the cookies was determined in a drying oven at 105 °C until a constant weight was obtained, according to the methodology of the Adolfo Lutz Institute ( 2004).The water activity was assessed using an AquaLab device.The titratable acidity was assessed according to the methodological guidelines of the Adolfo Lutz Institute.The ash content analysis was performed according to the American Association of Cereal Chemists (AACC) 08-01 method (2000) by burning in a muffle furnace at 600 °C.The levels of lipids were analyzed with Soxhlet extraction by the Association of Official Analytical Chemists (AOAC) method #31.4.02 (2005); proteins were analyzed using the AOAC (2005) Kjeldahl method no.920.87, with a nitrogen-protein factor of 6.25; dietary fiber analysis was conducted using the Association Of Official Analytical -AOAC (2005) enzymatic-gravimetric method; and a carbohydrates were quantified by difference.

Sensory analysis
The sensory analysis of the cookies was conducted through affective testing using a nine-point structured hedonic scale varying between the terms 'like extremely' and 'dislike extremely' and an indication of purchase intention.The test was conducted in individual booths adequately lit for this purpose, with a set of 90 untrained judges varying between 18 and 50 years of age.Before beginning the test, the judges read and signed an informed consent form.Next, the judges were given samples in sequence, i.e., the judge evaluated and analyzed the first sample and only then was given the second sample.Both of the samples were identified by three Hindu-Arabic digits without repetition.The cookies were presented in balanced stacks, served on a disposable white plate with a glass of water for mouth rinsing before the next test.Two samples were tested, which were chosen according to the best technological profile, i.e., according to the physical and texture profile analyses.

Data analysis
The experiment was carried out three times in a completely randomized design and data were analyzed using analyses of variance (ANOVA), and the mean values were compared with Tukey's test at 5% significance using the SAS software version 9.0.

Physical analyses
The physical analyses should measure weight, weigtht loss, expansion factor and specific volume.Table 2 present the results of the physical analyses and show which samples had the most significant results.
Ciênc.Agrotec., Lavras, v. 39, n. 5, p. 514-522, set./out., 2015 Regarding specific volume, the value of the control sample was smaller and very close to 1, meaning there was a balance between weight and volume.The cookies made using tapioca and rice flours had higher values than the control, indicating a smaller weight and a larger increase in volume after baking (Table 2).Assis et al. (2009), using oat flour and rice flour parboiled, found specific volumes between 0.92 and 1.76 mL.g -1 , similar to what was found in this study.This increase is due to the presence of more rice flour and consequently greater water absorption and expansion during the starch gelatinization that takes place in the oven.On the other hand, weight loss did not differ significantly among the formulations (p > 0,05).Regarding the expansion factor, sugar and vegetable shortening exert great influence because, in addition to conferring pleasant taste and texture, these ingredients favor spread and viscosity and reduce the cookies' thickness (Moraes et al. 2010.).Sample E had the lowest expansion factor and differed significantly from sample A (p < 0.05).This difference due to the increased quantity of fiber in rice bran, which reduced the cookie expansion.It is possible that the greater water absorption by the fibers causes a reduction in absorption by the starch, and, as a consequence, less gelatinization.This process was also reported by Feddern et al. (2011) for cookies made from wheat bran and rice flour: the expansion factors reported in that study were between 2.11 and 3.09, respectively, and differed from the values found in this work (0.93 and 1.04, respectively).Additionally, the behavior of tapioca flour compared to rice flour, present in a larger proportion in sample A, tended to reduce the cookie expansion factor.This decrease did not occur in sample E.

Raw dough texture profile
The texture profile of raw dough was assessed using a TA.XT Plus texturometer and the profiles for hardness, adhesiveness, cohesiveness and chewability for the raw dough are shown in Table 3 .The elasticity and gumminess profiles did not differ significantly among the formulations.
The elasticity values for cookies produced by Assis et al. (2009) also did not display significant difference (p > 0.05) among samples.Increasing the amount of rice flour from reduced the hardness.This pattern may be explained by the amount of fiber present in rice flour, which absorbs more water.This pattern was also observed by Assis et al. (2009), who observed that raw cookie dough containing rice flour displayed lower hardness.For adhesiveness, samples D was significantly different from CF. Regarding cohesiveness, the value for the CF was higher than those of all the other formulations because the CF contained 100% wheat flour, suggesting that the viscoelastic characteristics of gluten's protein chains may resist a second deformation relative to the behavior of a previous deformation, i.e., tensile strength is greater due to the presence of gliadin or glutenin molecules.The CF formulation was significantly different in chewability from all the other formulations exception sample A. The rice flour has a smaller amount of fiber, leaving the cookies more chewable.

Cookie texture profile
The cookies from the A, B, C, D and E formulations displayed very similar hardness values (Table 4).Whereas the hardness of the CF formulation was higher, which may be attributed to gluten's strong proteic network and the fact that the CF cookies contained only wheat flour.Alternatively, the reduction in hardness for tapioca and rice flour cookies might be associated with the product's lipid content, which grants more softness and palatability (Moraes et al., 2010).Ciênc.Agrotec., Lavras, v. 39, n. 5, p. 514-522, set./out., 2015 In regard to fracturability, there were significant differences among formulations: formulations A, B and CF did not significantly differ in fracturability, nor did formulations C, D and E differ from each other.However, the values of the latter formulations were higher than those for A, B and CF because, as the proportion of tapioca flour was reduced and that of rice flour was increased to 50%, 75% and 100% in C, D and E, respectively, fracturability increased.Another important aspect in regard to fracturability is the lipid content.A fat content between 30 and 38% leads to less fracturable cookies (Moraes et al., 2010).In this study, the lipid content was 35%, producing softer cookies.Lower hardness values were observed in comparison to CF as well as increased cookie fracturability as the proportion of rice flour in the cookie formulations exceeded 50%.The control formulation (100% wheat flour) displayed the greatest hardness, but its fracturability was the same as those of samples A and B.

Physicochemical and nutritional analyses
The formulations B and C displayed similar results to those found in the scientific literature in terms of percentage composition (Table 5).The values for moisture (2.97 and 2.96), titratable acidity (0.18 and 0.19) and water activity (0.30 and 0.30) for samples B and C, respectively, did not vary significantly (p<0,05) from each other but are in accordance with the parameters recommended in the legislation (Brasil, 2005b).The low moisture contents and water activity indicate a product that inhibits microbial growth in conjunction with containing significant carbohydrate concentrations, which also make the medium inhospitable for microbial reproduction.The following results were obtained for formulations B and C, respectively: ash content (0.77 and 0.83); proteins (2.56 and 3.52); lipids (19.07 and 15.64); carbohydrates (73.77 and 76.14); and fiber (0.86 and 0.91).Regarding nutritional value, similar values for ash, proteins, lipids,  Despite their low fiber content, these cookies are still nutritious because they are complete in their nutritional composition and they contain macroproteins and mineral residue (ash) in addition to the caloric load provided by the carbohydrates and lipids.

Sensory analysis
Results for the acceptability index per cookie attribute can be seen in Figure 1.The sensory evaluation was conducted only for samples B and C because these formulations showed better technological profiles during the raw and baked dough texture tests.Moreover, were the samples showed better performance in the cookies preparation.Samples B and C were significantly different (p < 0.05) from one another in all the studied sensory attributes.Sample B had higher acceptability means than sample C, including the flavor attribute, in which it scored 7.79.All the mean values obtained in the sensory analysis, for both the B and C samples, were between the hedonic values of 6 and 8, corresponding to 'like slightly' and 'like very much'.
For the scent, color, texture, flavor attributes and overall evaluation, the mean values of formulation B were significantly higher than those for formulation C.This predilection may be related to the sample B have more tapioca flour (75%) than sample C (50% and presentation more flavorful.Regarding purchase intention, 45.56% of the judges responded that they would most likely buy cookie B, and 43.34% indicated similar purchase intent for cookie C.However, 33.34% of the judges said they would certainly buy cookie B, and 14.45% said they would certainly buy cookie C. In addition, 20% of the judges reported that they had doubts regarding whether they would buy sample B, in comparison to 32.23% for sample C (Figure 2).

CONCLUSIONS
The results obtained in this study indicate viability in producing cookies from tapioca and rice flours from technological, nutritional and sensory standpoints because the cookies' acceptance was satisfactory given a considerable sample (n=90).The technology involved in cookie production and subsequent commercialization does not generate a high demand for resources, time or labor.Therefore, it is noteworthy that mixed flours are a good option for manufacturing new bakery products and similar products because mixed flours do not negatively affect the partial or complete substitution of wheat flour.However, other factors may interfere in the sensory quality of the products, such as the quality of the flours, moisture contents, excess and/or scarcity or inadequate choice of the type of sugar and fat or any structural ingredient of the dough.Another evident benefit of mixed flours is that they allow people with food restrictions to eat gluten-free foods that have good sensory and nutritional quality.

Figure 1 :
Figure 1: Sensory evaluation of samples B and C. Means of the ratings given to the sensory aspects scent, texture, color, flavor and overall impression.Mean values followed by the same letter in each column are not different from each other based on Tukey's test (p > 0.05).

Figure 2 :
Figure 2: Purchase intent for cookies B and C. Percentage data for purchase intent based on a sample of 90 judges.Legend: DP (Definitely Purchase); PP (Probably Purchase); MP (Maybe Purchase); PNP (Probably Not Purchase); DNP (Definitely Not Purchase).

Table 1 :
Amount of each ingredient, in baker´s percentage, used to prepare the formulations.

Table 2 :
Technological assessment of tapioca and rice flour cookies.Expansion factor (EF) and specific volume (SV).Mean values followed by the same letter in each column do not differ from one another based on Tukey's test (p > 0.05).

Table 3 :
Parameter values for hardness, adhesiveness, cohesiveness and chewability for the raw dough texture profile of tapioca and rice flour cookies.