Vegetal burgers of cashew fiber and cowpea: formulation, characterization and stability during frozen storage 1

ABSCTRACT - This work aimed to obtain a vegetable burger from the cashew juice production residue (fiber) and cowpea that are two abundant materials in Northeast Brazil. Fiber was washed and pressed in an expeller for five times. Cowpea was soaked in water and, after skin removal, ground to obtain a paste. Hamburgers were made by mixing the cashew fiber, the cowpea paste, and the ingredients used in traditional burger recipes, then shaped and frozen. Four sensory tests were performed to define: (1) the need of cooking the cowpea paste with the other ingredients; (2) the fiber/paste proportion; (3) ingredients to enhance the sensory characteristics; (4) burger acceptance. The resultant formulation had 29.3% of cashew fiber, 29.3% of cowpea paste, 25.1% of tomato, 6.8% of onion, 5.3% of sweet pepper, 1.3% of garlic, 0.1% of black pepper, 0.2% of dehydrated parsley, 1.2% of salt and 1.4% of corn oil. The sensory acceptance was 7.8 in a 9-point hedonic scale. Proximate composition was 71.08% moisture, 2.07% ashes, 4.86% proteins, 1.19% lipids and 20.79% total carbohydrates. Stability was evaluated during frozen storage (-18 ºC) for 180 days. In this period, acidity increased and pH and ascorbic acid decreased. As to color, a* and b* values increased, indicating that the burgers became more orange/brown. The burger was considered microbiologically safe and shelf-stable for at least 6 months.


INTRODUCTION
The juice is the most important product obtained from the cashew apple (Anacardium occidentale L.) and its extraction generates a residue called cashew fiber, which represents 20 to 40% of the apple weight (ABREU et al., 2013;KUILA et al., 2011). The cashew apple fiber is usually used as animal feed (DANTAS FILHO et al., 2007;RAMOS et al., 2006) or even in ethanol production (CORREIA et al., 2013;LIMA et al., 2012;ROCHA et al., 2014), however, some authors have reported its use for human consumption, mainly because it is 12% dietary fibers (PINHO et al., 2011a). Good sensory results were obtained by adding 8 to 15% of dry cashew fiber in cookies (MATIAS et al., 2005). It was also reported the replacement of up to 10% of meat by dried cashew fiber in burgers without any sensory changes (PINHO et al., 2011b). Previous studies revealed that burgers made exclusively from cashew fiber had medium sensory acceptance (5.9 in a 9-point hedonic scale) and low protein value (5%) (LIMA, 2008).
On the other hand, cowpea is abundant in Brazil and could be used to enhance palatability and protein value of burgers made from cashew apple fiber. Cowpea (Vigna unguiculata L. Walp) is an annual legume that is commonly known as southern pea, black-eyed pea, alubia, caupi, tape or frijole (AVANZA et al., 2013). In Brazil, cowpea is frequently produced by small and medium size farmers for either personal consumption or trade. Cowpea protein content ranges from 20 to 30% (ADEGUNWA et al., 2012;AVANZA et al., 2013;CARVALHO et al., 2012;GIAMI, 2005;VASCONCELOS et al., 2010). Therefore, the objective of this work was to obtain a vegetable burger from the cashew juice production residue (fiber) mixed with cowpea (protein source) and evaluate its composition, sensory acceptance, and stability during 180 days of frozen storage (-18 ºC).

Ingredients
The residue from the cashew apple juice production (cashew apple fiber) was obtained at a factory located in Fortaleza city, Northeastern region of Brazil. The cashew fiber was washed and pressed in an expeller (Incomap 300, Fortaleza, Brazil) for five times in order to reduce size and acidity. At each time, water was added in the proportion of 1:1 (w:w). Cowpea was obtained from local suppliers (Fortaleza, CE, Brazil), soaked in water for 2 hours and, after skin removal, crushed to obtain a paste. The other ingredients were purchased at the local market.

Defining the vegetal burgers formulation
Four consecutive sensory tests were performed to define the burger formulation. Preliminary tests indicated that the mixture of cashew apple fiber and cowpea paste should constitute 55.2% of the burger formulation. The other ingredients were tomato (21.2%), onion (6.4%), sweet pepper (7.4%), garlic (1.3%), salt (0.9%), corn oil (1.3%), and water (6.3%). The ingredients were cooked all together, shaped (80 g per unit), packed in polyethylene bags and frozen.
Sensory judges were non-smokers and used to consuming burgers occasionally. Before the analysis, burgers were pan fried with 10 mL of corn oil for 5 minutes, alternating cooking sides. Burgers were cut in three pieces, each piece was placed in small plastic containers coded with random three-digit numbers and presented in a balanced order. Judges were instructed to consume the sample and rinse the mouth with water (room temperature) between sample evaluations. Tests were performed in individual acclimatized booths (24 ºC) (MEILGAARD; CIVILLE; CARR, 1999).
At first, a paired preference test applied to 40 judges was used to verify the need of cooking the cowpea paste together with the other ingredients. Burgers were prepared using cashew fiber and cowpea paste in the proportion of 3:1. Judges were asked to choose between burgers made with cooked and uncooked cowpea paste. Choice was compulsory, with a "no preference" option not being allowed. The results were analyzed using the critical number of correct responses in a two-sided directional difference test table (α=0.05) (MEILGAARD; CIVILLE; CARR, 1999).
Next, a ranking test applied to 48 judges was used to verify the preference among three formulations made with different cashew apple fiber/cowpea paste proportions (60/40, 50/50, 40/60). Judges were asked to rank the samples from the least (n° 1) to the most preferred (n° 3), and the results were analyzed by Friedman test, using Newell and MacFarlane (1987) tables.
Thirdly, a focus group meeting, involving eight professionals from the gastronomy course, was held to gather opinions on what ingredients could be changed in order to enhance the burgers' sensory characteristics.
Finally, an acceptance test was applied to 50 judges to verify the overall acceptability of burgers using a structured 9-point hedonic scale ranging from 1 (extremely dislike) to 9 (extremely like). Judges were also asked about their purchase intent using a structured 5-point scale ranging from 1 (I would certainly not buy it) to 5 (I would certainly buy it). Results were presented as frequency histograms.

Physicochemical analysis of burgers
Burgers formulated according to the results obtained previously were analyzed for moisture, ash, lipid and protein contents (ASSOCIATION OF OFFICIAL ANALYTICAL CHEMISTS, 1997). Total carbohydrate content was estimated by the difference from other components: 100 -[weight in grams (water + proteins + lipids + ashes) in 100 g of food]. Analyses were performed with three repetitions and results were presented as means and standard deviations. The energy value was calculated using 4 kcal g -1 for carbohydrates and proteins and 9 kcal g -1 for lipids (BRASIL, 2003).

Stability of burgers
Burgers were prepared, packed in polyethylene bags (80 g), sealed and frozen (-18 ºC). The stability test (acidity, pH, ascorbic acid content, color, sensory acceptance and microbiological quality) was carried out under freezing temperature (-18 ºC), at about 40 day intervals, up to 180 days.
Total titratable acidity, pH and ascorbic acid (2,6dichloroindophenol titrimetric method) were determined according to the ASSOCIATION OF OFFICIAL ANALYTICAL CHEMISTS guidelines (1997). The color was assessed using a colorimeter (Chroma Meter CR-400, Konica Minolta Sensing Inc., Osaka, Japan) to determine the L*a*b* values. Analyses were performed in triplicates.
Sensory acceptability tests were carried out with the burger samples by an untrained 48-member panel consisting of employees and students who worked at Embrapa Tropical Agroindustry. Judges were not necessarily the same at the time of each evaluation; they were non-smokers and consumed burgers occasionally. Burgers were pan fried, as described above, and approximately 20 g of burger were placed in small plastic containers coded with random three-digit numbers. Judges used a 9-point hedonic structured scale, where 9 was extremely like and 1 was extremely dislike for the overall acceptability of samples (MEILGAARD; CIVILLE; CARR, 1999). The tests were performed in individual acclimatized booths (24 °C) with daylight like illumination.
Results from acidity, pH, ascorbic acid content, color and sensory acceptance were submitted to regression analysis using the SAS statistical program for Windows system (STATISTICAL ANALISYS SYSTEM, 2009) and presented as graphics.
The microbiological quality of samples was analyzed by the determination of fecal coliforms, Staphylococcus aureus and Salmonella sp., according to the American Public Health Association guidelines (DOWNES; ITO, 2001). Analyses were performed in triplicates.

Vegetal burger formulation
For the preference test used to choose between cooked and uncooked cowpea paste, the minimum number of agreeing judgments to establish significance (n=40, α=0.05) is 27. The number of judges that preferred the cooked sample was 23 and the number that preferred the uncooked was 17 (difference = 6), indicating that no preference was observed between the samples. Consequently, cooking the cowpea paste with the other ingredients was the procedure chosen for the next step, given that such procedure made burgers easier to mold. Furthermore, heating can improve the nutritive value of legumes by decreasing the levels of anti-nutrients and increasing protein digestibility (ADEGUNWA et al., 2012;AVANZA et al., 2013;CARVALHO et al., 2012;GIAMI, 2005;VASCONCELOS et al., 2010).
The ranking scores for the formulations with different cashew apple fiber and cowpea paste proportions are presented in Table 1. No preference was observed among the burgers, so the cashew fiber/cowpea paste proportion of 50/50 was chosen for the next test because it made molding burgers easier than the proportion of 60/40 and the former (50/50) used a higher amount of fiber than the latter (40/60).
Cashew fiber/cowpea proportion 60/40 50/50 40/60 Ranking scores 85 a 105 a 101 a As to the focus group, when presented to the whole fried burger and asked "What do you think about the burger's appearance, texture and smell?", it answered that the appearance was good, similar to those on the market; the texture was good, when cutting with a knife; and the smell was fine. The other question, when a piece of the burger was presented for tasting, was "What is your general opinion about the burger´s taste and your suggestions for its improvement?", and the answers were to increase the amounts of salt, onion, parsley and black pepper; and to decrease the amount of sweet pepper. As a result, the final formulation was: cashew fiber (29.3%), cowpea paste J. R. Lima et al. (29.3%), tomato (25.1%), onion (6.8%), sweet pepper (5.3%), garlic (1.3%), black pepper (0.1%), dehydrated parsley (0.2%), salt (1.2%) and corn oil (1.4%).
The mean of hedonic values for the sensory acceptance test was 7.8 in a 9-point scale. Ninety-two percent of the hedonic sensory scores (Figure 1) were within the acceptance range of the scale, which comprises categories from 6 (like slightly) to 9 (extremely like). Moreover, when asked about their purchase intent ( Figure  2), 86% of judges answered that they would buy it or certainly would buy it. Therefore, we can conclude that the burger was accepted by the consumers. Pinho et al. (2011b), using a 9-point hedonic scale, found lower averages (~ 5.0) for overall acceptance of burgers with 7 to 14% replacement of meat by cashew fiber. These authors suggested that the presence of tannins in the fiber could be responsible for the low acceptance of the burgers, an effect that was reduced in the burgers made in our study by the pressing step, which promoted fiber washing with the inclusion of water in each pressing cycle.

Physicochemical analysis of burgers
The proximate composition of burgers made from cashew fiber and cowpea was 71.08% (± 0.07) moisture, 2.07% (± 0.09) ashes, 4.86% (± 0.05) proteins, 1.19% (± 0.12) lipids and 20.79 % (± 0.26) total carbohydrates. The moisture value obtained in this study is higher than that obtained in a previous study (LIMA, 2008) where burgers were made by using only cashew fibers (49.50%). This difference is probably due to the soaking cowpea step. In order to compare our results with this previous work, the values of both studies were converted to dry weight. The inclusion of cowpea in the formulation has promoted an Figur e 1 -Frequency histograms of sensory acceptance of burgers made from cashew fiber and cowpea (1=extremely dislike, 2=dislike very much, 3=dislike moderately, 4=dislike slightly, 5=neither like nor dislike, 6=like slightly, 7=like moderately, 8=like very much, 9=extremely like) Figure 2 -Frequency histograms of purchase intent of burgers made from cashew fiber and cowpea (1=I would certainly not buy it, 2=I would probably not buy it, 3= I'm not sure whether or not I would buy it, 4=I would probably buy it, 5=I would certainly buy it) increase in ashes (from 6% to 7%), proteins (from 11% to 17%) and carbohydrates (from 66% to 72%), and a decrease in lipids (from 16% to 4%).
The lipids (1.19%) and energy values (113 Kcal (100g) -1 ) obtained for the cashew fiber and cowpea burger were considerably lower than the values for the traditional meat burgers (10% lipids, 260 Kcal (100g) -1 ) (UNITED STATES DEPARTMENT OF AGRICULTURE, 2015). Hence, the burgers obtained in this study can be considered as a low-fat and low-calorie product.

Stability of burgers
Regression analysis for the stability data was significant (p<0.05) for acidity ( Figure 3) and pH ( Figure 4). Acidity increased from 0.29 to 0.34% and pH decreased from 5.77 to 5.30. Changes in pH and acidity are usually associated with microbial growth; however, this was not the case, considering the microbiological analyses. Throughout the 180 days of storage, all coliforms, including Escherichia coli, were lower than 3 MPN g -1 . Salmonella spp. was absent in 25 g and Staphylococcus aureus was lower than 10 CFU g -1 . Therefore, the burger was considered microbiologically safe for at least 6 months of frozen storage.
Ascorbic acid content was lower than 94 mg (100 g) -1 , value reported by Lima et al. (2013) for the cashew fiber, decreasing (p<0.05) from 34.03 to 18.11 mg (100 g) -1 ( Figure 5). Ascorbic acid reduction in foods can occur either chemically (non-enzymatic oxidation) or through enzymatic processes, as there are enzymes that can resist the processing stages (thermo-resistant) (PHILLIPS et al., 2016). Moreover, freezing is not enough to block all enzymatic reactions (MAZZEO et al., 2015). However, at the end of 180 days of storage, approximately 53% of the initial ascorbic acid content remained in the burger.    Regression analysis was not significant (p>0.05) for lightness (L*), with a mean value of 54.7 during the storage time, but was significant (p<0.05) for the other chromaticity coordinates, a* (green to red) and b* (blue to yellow) ( Figure 6). An increase in a* (from 2.8 to 4.9) and b* values (from 15.6 to 23.9) was observed, indicating that the burgers became more orange/brown, probably due to the ascorbic acid degradation (DAMASCENO et al., 2008). Moreover, industrial freezing and storage processes partially deprive vegetables of the initial antioxidant capacity (PACIULLI et al., 2015), meaning that the oxidation process may occur and that could account for the change in color (SHAHIDI; AMBIGAIPALAN, 2015).
Regression analysis was not significant (p>0.05) for sensory acceptance, so that the slight changes observed in the physicochemical characteristics during storage did not affect the sensory acceptance of burgers, which remained the same throughout the 180 days of storage (Figure 7), with a hedonic mean of 6.7, a value very close to 7.0, corresponding to "like moderately" in the hedonic scale.