Addition of Pachira aquatica oil and Platonia insignis almond in cookies: Physicochemical and sensorial aspects

Abstract This article investigated the use of Pachira aquatica (PA) fat and Platonia insignis (PI) nuts as ingredients in the preparation of cookies. Seven formulations containing PA fat and/or PI nuts were studied by changing the formulation proposed by AACC, and samples were evaluated considering physical–chemical, microbiological, and sensory characteristics. Formulations F1, F4, and F5 showed higher mass loss and lower expansion factor after cooking. Formulations F4, F6, and F7 presented a greater increase in diameter. In turn, formulations F5, F6, and F7 presented greater thickness. The content of fatty acids varied according to the composition of each biscuit, and formulations 2 and 3 presented the best lipid profile (oleic acid ~32%). In addition, it was observed that the addition of PI almond increased the fiber content (~7.15%). The sensory evaluation showed that formulation F5 obtained a score of 5, proving that the partial replacement of hydrogenated vegetable fat with PA fat and grated coconut with PI almond favored the panelists' purchasing decision. The results indicate that unconventional sources of lipids and nuts can be used without loss of quality in biscuits.

flavonoids, anthocyanins, and simple phenols. The PI oil contains fatty acids, including palmitic, linoleic, stearic, and oleic acids, and are, therefore, of particular interest to the food industry (Freitas de Lima et al., 2016).In human food, it is interesting to investigate the use of PA and PI as ingredients in the elaboration of food products that present a high nutritional value while adding value and favoring the preservation of the species. Thus, the insertion of PA fat or PI almond into products of great acceptability such as cookies may be an opportunity.
Cookies are widely accepted by people of all ages and social classes (Kurek & Wyrwisz, 2015). Formulations mainly include wheat flour, sugar, and hydrogenated fat. Because of their attractive sensory characteristics, the addition or substitution of certain ingredients in cookies is possible. In fact, the addition of other ingredients to cookies is widely performed, whether to incorporate functional constituents such as soluble fibers (Bick, Fogaça, & Storck, 2014;Vitali, Dragojević, & Šebečić, 2009), to replace wheat with flours obtained from unconventional fruits such as defatted sesame seed meal (Clerici, Oliveira, & Nabeshima, 2013), to add pequi, baru or buriti (Pineli et al., 2015;Santos et al., 2011;da Silva, Monteiro, & da Rosa, 2014), among others. The potential role of PA fat and PI almond in the formulation of cereal-based foods has not been addressed in the scientific literature yet. Therefore, evaluating the use of these functional products as a food ingredient would allow for the development of baked products with improved nutritional value, while valuing and favoring the preservation of the species. Given this promising application, this work aimed to study the effect of the addition of PA fat and PI almond in the physicochemical, nutritional, and sensory characteristics of cookies.

| Material
The PA and PI fruits were randomly collected in the municipality of Dourados-Mato Grosso do Sul (latitude 22°13′16″, longitude 54°48′20″ and altitude of 430 m). The fruits were hygienized and sanitized in sodium dichloroisocyanate solution dihydrate 0.66% (w/w) (active chlorine content 3% w/w) for 10 min. The almonds (endosperm) were removed for posterior use (PI) or even for oil extraction (PA). In both cases, the almonds were packed in inert plastic containers and then stored at −5°C until use.

| Oil extraction
The oil was extracted by cold pressing in an expeller-type press (Ecirtec -MPE-40P). For the extraction, the almonds were previously dehydrated at 45°C for 48 hr in an oven with air circulation at a speed of 0.5 m/s. The extracted oil was centrifuged at 4226g for 15 min at room temperature, transferred to amber containers and stored under refrigeration (7°C) until use.

| Cookie preparation
The cookies were prepared by modifying a method of the AACC (2000), in which wheat flour (60%) and corn starch (40%) accounted for 100%. Milk, eggs, baking powder, and flavoring were incorporated in the same amount in all formulations. The percentage of each ingredient was calculated using the total flour and starch mass, as follows: milk (9%), eggs (9%), baking powder (0.42%), vanilla flavor (0.33%), and salt (0.25%), in addition to the ingredients shown in Table 1.
The preparation of the cookies consisted of mixing the moist ingredients in a domestic shaker (300g) until a cream was obtained, and then, the dry ingredients were incorporated and mixed until a homogeneous mass was obtained. Chocolate pieces (66.66% w/w) were added in all formulations. The cookies were shaped in the form of 30 mm diameter and 5 mm thick disks and baked at 200°C for 15 min in a semi-industrial oven (Venâncio-Fc4emv). They were then cooled to room temperature (25°C) and conditioned in hermetically sealed containers until the time of analysis.

| Physicochemical analysis
The mass of the cookies was determined in a semianalytical balance (BEL Engineering) before and after baking. The diameter and TA B L E 1 Formulation of cookies containing or not fats of PA and/or PI almond thickness were determined with a digital caliper (Digimess). The expansion factor was calculated by the ratio between the diameter and the thickness of the cookies after baking (AACC method 10-50D, 2000). The specific volume was determined by the kernel seed displacement method (Moraes, Zavareze, Miranda, Salas-Mellado, 2010), and the yield was calculated from the difference between the final (after baking) and the initial (before baking) mass of the cookies, expressed as a percentage.
The color analysis was performed using a Konica Minolta (Model CR-400/Cr-410) colorimeter, with a D65 illuminator, a 10° Angle observation and a CIE L* a* b* system (Commission Internationale de L'Eclairage, 1986). The parameters evaluated were as follows: Luminosity (L*) ranging from 0 (black) to 100 (white), chromaticity of red-green (a*) and blue-yellow (b*), hue angle (°h), and color saturation (C). All physical analyses were performed randomly for each batch of the different formulations made.

| Proximate composition
Moisture determinations were carried out in a greenhouse at 105°C (AOAC, 1995), lipids were determined by the cold method (Bligh & Dyer, 1959), fixed mineral residue was determined by muffle incineration at 550°C, protein was determined by the micro-Kjeldahl method (Labtec DT2008), and dietary fiber was determined according to the method proposed by AOAC (1995). Carbohydrates were calculated by the difference of the other components. The total energy value (TEV) was calculated using the following Atwater conversion factors: 4 kcal/g proteins, 9 kcal/g lipids, and 4 kcal/g carbohydrates (Merrill & Watt, 1955).

| Fatty acid composition
Total lipids were extracted from the samples using the Bligh and Dyer method (1959). For the transesterification of the triglycerides, approximately 50 mg of the sample was transferred to 15 ml falcon-type tubes, in which 2 ml of heptane was added. The mixture was subsequently stirred until complete dissolution of the grease material, and then, 2 ml of KOH (2 M) in methanol was added. The mixture was stirred for approximately 5 min, and after phase separation, 1 ml of the higher phase (heptane and fatty acid methyl esters) was transferred to 1.5 ml Eppendorf flasks. These flasks were hermetically sealed and stored in a freezer at −18°C for further chromatographic analysis.
The fatty acid composition was determined by gas chromatography (Shimadzu, GC-2010 Plus Capillary GC) using gas chromatography with a flame ionization detector. For elution, a 100 m × 0.25 mm × 0.20 μm molten silica capillary column was employed. The oven temperature was programmed to start at 100°C and was held for 1 min and then raised to 170°C at 6.5°C/min.
Subsequently, another elevation of 170-215°C was performed at 2.75°C/min, and the temperature was maintained for 12 min. Finally, a final elevation was performed from 215 to 230°C at 40°C/min. The injector and detector temperatures were 270 and 280°C, respectively. The 0.5 μl samples were injected in split mode, using nitrogen as a carrier gas with a drag speed of 1 ml/min. The identification of the fatty acid methyl esters was performed by comparison to the retention times of the sample compound with the standards (Sigma) eluted under the same conditions as the sample.

| Sensory analysis
A nine-point hedonic scale was used for sensory analysis, anchored at extremes 1 (dislikes extremely) and 9 (likes extremely). The attributes analyzed were aroma, color, taste, sweetness, crispness, and global acceptance. The purchase intention was assessed using a fivepoint scale (1 = certainly would not buy and 5 = certainly would buy).
Ninety-six untrained judges of both sexes, aged between 17 and 58, participated in the analysis. The panelists signed the informed consent form before the test, according to Resolution 466/12 of the Ministry of Health (Brasil, 2013). The analysis was performed on different days.
In the first step, the samples of four formulations (F1, F2, F3, and F4) were presented to the panelists. In the second step, the samples of formulations F5, F6, and F7 were evaluated, as well as F4, which presented higher acceptance in the first stage of the test. The samples were presented in containers labeled using three-digit codes and were served monadically with water, in order to clean the taste buds.
The sample presentation order was distributed randomly, balanced and randomized in complete blocks. This study was approved by the

| Microbiological analysis
The Salmonella sp., Coliform 45°C and Coagulase-positive Staphylococcus counts were performed on the cookies with the best formulations, according to the methodology described by Downes, Ito, and American Public Health Association (2001).

| Statistical analysis
The StatSoft program (Statistica 8.0) was used for statistical analysis.
The data referring to the physicochemical analyses and sensory acceptance tests were subjected to the analysis of variance (ANOVA) by the F test and to the comparison of means by the Tukey test (p < .05).
For the sensory preference ranking test, the difference between treatments was verified by the Newell and MacFarlane table (p < .05).

| Proximate composition
The cookies were prepared by modifying the AACC method (AOAC, 1995), and the results of the proximal composition are shown in Table 2.
The moisture content varied from 4.38 to 6.19 g/100 g, which is in the range of commercial cookies that typically present a moisture content that varies from 1 to 6 g/100 g. In general, cookies are considered verylow moisture content products and it is important to highlight that the moisture content can contribute to the texture, sensorial and microbial to claim that a food is a "source of dietary fiber," it should contain at least 3 g/100 g of serving of total dietary fiber, whereas the claim "high in dietary fiber" is assigned to food categories with at least 6 g/100 g (Fuller, Beck, Salman, & Tapsell, 2016).
The importance of dietary fiber intake in human consumption is due to the benefits already evidenced by several studies in the prevention of hypertension, diabetes, obesity, gastrointestinal diseases, and among others (Westenbrink, Brunt, & van der Kamp, 2013).

| Technological characteristics
The physicochemical analysis was used to characterize the cookies, and the results are presented in Table 3. In all formulations, the cookies presented a lower mass after baking, which was already expected due to the loss of water. The F1, F4, and F5 cookies presented higher mass loss, but there was no significant difference (p ≥ .05). The F5 cookies presented higher diameter and expansion factor after baking (5.29 mm and 2.98%, respectively); however, there was no significant change in thickness when compared to the other cookies (p ≥ .05). This can be justified by the presence of fat added to the dough, which interferes with the gluten expansion during the baking process, forming a porous texture, which in turn favors the evaporation of liquids, thus guaranteeing the quality of the texture of the cookies (Jacob & Leelavathi, 2007).
The largest diameter increase was found in the formulations with brown sugar (F4, F6, and F7). The F5, F6, and F7 formulations presented greater thickness. The substitution of the coarse sugar by the brown sugar and the grated coconut by the PI almond (F6) favored the increase of the expansion factor. The expansion factor is related to the ability of the ingredients to absorb water.
Generally, cookies with a high content of insoluble fibers present a lower expansion factor (Fradinho, Nunes, & Raymundo, 2015). The PA fat influenced the reduction of the expansion factor as can be observed in the F1, F4, and F5 formulations. According to Gaines (1993), the diameter and the expansion factor in cookies may be used to predict the quality of the products, since these factors influence the size of the sample presenting variability, which hinders the standardization of the product.
The cookies with the addition of PA and PI almond (F5) presented a lower volume. As reported by Armbrister and Setser (1994), the interaction of sugars with fat substitutes can produce cookies with less expansion and, consequently, with lower specific volume and greater softness. However, the specific volume variation is lower, which favors product standardization.
Color is a fundamental attribute to evaluate food quality and visual analysis is the first of the senses to be used in the choice and acceptance of the product, considering the color parameters (Table 4). In the texture evaluation, the cookies presented differ-

| Fatty acid profile
Fats and oils are available to the body as fatty acids that play a significant role in human health. Table 5

| Microbiological and sensory analysis
All formulations were microbiologically evaluated, and results are presented in Table 6. The absence of Salmonella sp. in 25 g was observed in all formulations. The presence of thermotolerant coliforms was lower than 0.3 NMP/g and the values of Staphylococcus coagulase were <10 CFU/g showing that the conditions of handling and processing were adequate, making these foods safe for sensory analysis.
The panelists' scores for the sensory attributes of the cookies are presented in Table 7. Cookies of all formulations presented scores >6 (slightly liked) for all attributes, all of which were sensorially accepted.
However, the F5 formulation (50% of the PA fat and 50% of the PI almond) presented superior acceptability and sensorial preference, obtaining significantly (p ≤ .05) higher scores for all attributes evaluated in comparison with the other formulations. Concerning the intention to purchase, the average scores represented the consumer's intention to "maybe buy/maybe not buy" (note 3), possibly buy (note 4), and certainly buy (note 5). The F5 formulation cookies obtained a score of 5, proving that the partial replacement of hydrogenated vegetable fat with PA fat and grated coconut with PI almond favored the panelists' purchasing decision.
Cookies with total fat replacement with PA fat (F1) and coconut with PI almond (F3) presented a lower percentage of purchase intention, probably due to the fat perception. The flavor and overall acceptance mean values (Table 5) confirm these results. Therefore, the F5 formulation cookies elaborated with 50% of PA fat and 50% of PI almond presented better results in all evaluations.

| CON CLUS IONS
Cookies made with PA fat and PI almond presented high levels of protein, lipids, carbohydrates, and energy. The saturated fatty acids, especially palmitic and lauric acids, were predominant in the lipid profile of the cookies formulated with PA. Among the unsaturated fatty acids present, oleic acid had a higher percentage.
Cookies formulated with 50% PA fat and 50% PI almonds (F5) had acceptable sensory and microbiological characteristics. The high content of dietary fiber indicates that this product may be a good source of fibers. The results presented throughout this study indicate that unconventional sources of lipids such as PA fat and palm kernel such as PI can be used without loss of quality in cookies.
The use of these fruits in the increment of new ingredients enables their use in the development of food products while valuing natural resources.

ACK N OWLED G M ENTS
The authors are thankful to the Coordination for Improvement of Higher Education Personnel (CAPES) and the Foundation for Support to the Development of Education, Science and Technology of the State of Mato Grosso do Sul (FUNDECT) for the financial support of this research.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.