The fruits were transported to the laboratory and their juices were extracted separately. First, the passion fruits were washed with clean water, then cut in two parts. The pulp inside was collected and kneaded with hot water. The juice obtained was filtered through a 0.5 mm mesh sieve. Before the cashew apple juice was prepared, the apples were detached from the nut. They were washed with clean water, then cut and crushed by a blender (Blender LB20E, Torrington, USA, 2002). The juice obtained by pressing the crushed material was filtered through a 0.5 mm mesh sieve. Finally, the different juices were stored for sixteen (16) days in the refrigerator (Fiocchetti, Mazzara, Italy) at 4˚C [19].

The pH was determined according to [20] using a pH meter (pH meter C861, Consort, bio block, Belgium). The calibration of the instrument was ensured by the use of two buffer solutions at pH 7.0 and 4.0 and this is done systematically before the pH measurements. The measurement was made by immersing the electrode in 5 mL of sample and the reading was repeated three times.

The determination of the titratable acidity of the juices was carried out according to the standard [21]. It was obtained by determination of 5 mL of the sample with a 0.1 N sodium hydroxide solution. The analysis was terminated by turning pink after the addition of 2 to 3 drops of phenolphthalein. The results obtained are the average of three tests. The titratable acidity expressed as a percentage of grams of lactic acid/L is calculated using the following formula:

% TitratableAcidity = N NaOH × V NaOH × 0.09 × 100 V test (1)

V_test: volume of the test sample;

N_NaOH: normality of NaOH poured (meq - g/L);

V_NaOH: volume of NaOH poured (mL);

0.09: milli-equivalent gram of lactic acid.

The rate of soluble solids was determined according to the [22] standard. The Brix (%) was calibrated according to the number of grams of cane sugar contained in a 100 g solution. The measurement was made at a temperature of 20˚C, of the refractive index of the prepared sample. The prisms of the refractometer were cleaned with distilled water in order to operate at the required temperature which must remain constant to within 0.5˚C during the determination. Then the test solution was brought to the measuring temperature. The Brix degree was measured by applying two to three drops of sample to the fixed prism of the refractometer and the reading was taken three times.

Dry matter was determined according to the method described by [23]. The crucible was previously dried in an oven at 110˚C for 3 hours, then cooled in a desiccator. This crucible was weighed (i.e. C₀ weight) and 5 g of juice sample was introduced into it. The juice was heated slowly on a hot plate until most of the organic constituents were burnt off, then the crucible was placed in the oven for 24 hours. At the end of this time the crucible was removed and cooled with a desiccator and weighed again to obtain a weight C₁. The analyses were performed in triplicate.

D M = C 1 − C 0 P e × 100 (2)

with P_e the initial weight of the sample;

C₀ the mass of the empty crucible (g);

C₁ the mass of the crucible after drying (g);

DM = dry matter (%).

The ash content has been determined according to [24]. A volume of 5 mL of juice sample is poured into a previously dried crucible and weighed. The whole is placed in a muffle furnace at 550˚C for 24 hours. The crucible is then removed and cooled in a desiccator and weighed again. The analyses are carried out in triplicate and the ash content is determined by the following formula:

%   C = C 1 − C 0 P e × 100 (3)

where:

C₀ is the mass of the empty crucible (g);

C₁ is the mass of the crucible after drying (g);

P_e is the test portion (g);

C is the percentage of ash (%).

The measurement was performed in a 250 mL foot cylinder. The sample was well homogenized and placed in the test tube. The density measurement was carried out by introducing the density meter into the sample at a temperature of 20˚C. The reading was taken three times.

The protein content was determined according to the Kjeldahl method [25]. In a matrass, 1 mL of juice was introduced and mineralized in the presence of 10 mL of concentrated sulfuric acid and 1 g of mineralization catalyst. Mineralization was carried out hot for one hour on a mineralization ramp. After mineralization and cooling of the samples, 300 mL of distilled water, as well as 50 mL of soda lye and two drops of phenolphthalein were added to each matrass containing the mineralization. The mineralization was placed in a distillation unit. During distillation, 200 mL of distillate was collected in an Erlenmeyer flask containing 25 mL of 4% boric acid and 3 drops of methyl red. The whole was titrated under stirring with a sulfuric acid solution (0.1 N) until the color changed from green to pink. The analyses were carried out in triplicate. The total protein content, expressed in g per 100 g of sample is obtained according to the following formula:

%   N = 0.14 × V P e (4)

N: Nitrogen present in the sample (%);

V: volume of sulfuric acid standard solution (mL);

M Nitrogen 0.14: molar mass of nitrogen (g∙mol⁻¹);

Pe: test sample (g).

% P = % N × 6.25 (5)

% P = protein content;

6.25 nitrogen to protein conversion factor.

The determination of reducing sugars was carried out according to the method described by [26]. Prior to the determination, the sample was prepared. Indeed, 10 mL of 80˚C ethanol was added to 1 mL of juice sample and then centrifuged at 2500 rpm for 15 min. The supernatant was recovered and then 2 mL lead acetate was added and centrifuged at 2500 rpm for 15 min. The supernatant was collected again and 2 mL oxalic acid was added. Everything is centrifuged at 2500 rpm for 15 minutes. In addition, the supernatant is collected and brought to the water bath (80˚C). After cooling, distilled water was added to the sample to the previous level. For the determination of reducing sugars, a volume of 0.2 mL of juice sample is introduced into test tubes. To this is added 1.8 mL distilled water and 0.6 mL of dinitrosalicylic acid (DNS). The whole was brought to the bath—marinate at 80˚C/5min and then leave for cooling. An orange coloration appears. 4 mL of distilled water was added and then the reading was taken with the spectrophotometer at 546 nm. The concentration of reducing sugars was determined with reference to a standard glucose range.

The determination of total sugars was carried out by the phenol-sulfuric method [26]. Prior to the determination, the sample was prepared. Indeed, 10 mL of 80˚C ethanol was added to 1 mL of juice sample and then centrifuged at 2500 rpm for 15 min. The supernatant was recovered and then 2 mL lead acetate was added and centrifuged at 2500 rpm for 15 min. The supernatant was collected again and 2 mL oxalic acid was added. Everything is centrifuged at 2500 rpm for 15 minutes. In addition, the supernatant is collected and brought to the water bath (80˚C). After cooling, distilled water was added to the sample to the previous level. For the determination of total sugars, 2 mL phenol (5%) was added to 2 mL of the solution taken from a test tube. Then 8 mL concentrated sulfuric acid (H₂SO₄) (96%) was added rapidly without running down the walls and the mixture was stirred immediately. Everything is put in the dark for 30 min. A yellow coloration develops, stable for several hours. Absorbance is measured at 490 nm. The concentration of total sugars is determined with reference to a standard glucose range.

The passion fruit juice was added to the cashew apple juice in proportions (Table 1). Thus, the mixtures are juice A (50 mL of passion fruit juice/50mL of cashew apple juice), juice B (60 mL of passion fruit juice/40mL of cashew apple juice ), C juice (70 mL passion fruit juice/30mL cashew apple juice), D juice (80 mL passion fruit juice/20mL cashew apple juice), juice E (90 mL of passion fruit juice/10mL of cashew apple juice), F juice (40 mL of passion fruit juice/60mL of cashew apple juice) and G juice (30 mL of passion fruit juice/70mL of cashew apple juice).

The sensory evaluation of the organoleptic characteristics and formulations of the mixed juice was carried out at the National Public Health Laboratory (LNSP) in Abidjan, Côte d’Ivoire. In Côte d’Ivoire sensory analysis tests do not require formal written consent. Simply agreeing to participate in the tests is considered consent. During the test, participants simply fill out a form. Also, sensory tests do not require ethical approval. A hedonic test on a 9-point linear scale (Figure 2) was used to assess the degree of appreciation of the organoleptic characteristics. This study followed the method described by [27] and applied more specifically to cashew nut juice by [28]. Before the tests were conducted, subjects were instructed not to smoke or consume anything except water at least one hour before the assessment to avoid biasing the results. Fruit juices were served in transparent, plastic tasting glasses and labelled for correct evaluation. They were presented one after the other to each evaluator, taking care to remove each sample after tasting and before putting the next sample back in. Water at room temperature was given to the tasters so that they could rinse their mouths before and between each of the samples they tasted. In addition, they did not communicate with each other during the sensory evaluation. Three (03) tasting sessions were held on Thursdays over three (03) consecutive weeks under the same working and blending conditions. These sessions, lasting 2 hours for the nine (09) samples, were organized in such a way as not to cause weariness and saturation of the subjects.

A panel of 30 people chosen from among the trainees (in Superior Technician’s Certificate and agrifood engineering) and some employees of the LNSP was formed. This panel was made up of people not trained on the characteristics

chosen. However, astringency was briefly explained to them. The panel therefore proceeded to test the acceptability of the different samples of cashew apple juice, passion fruit juice and the formulations of the mixed juice prepared.

The assessment of cashew apple juice samples was based on color, flavour (taste), odour, decanting and overall appreciation using a linear hedonic scale. This nine-point scale, ranging from “extremely poor” (point 1) to “extremely good” (point 9) was used [28]. Thus, for color, the scale varied from “extremely faded” to “extremely colored”. For taste, the range varied from “extremely bitter” to “extremely sweet”. For the smell the range varied from “extremely unpleasant” to “extremely pleasant”. For decanting the range was from “extremely unpleasant” to “extremely pleasant” and finally the overall assessment was from “extremely unpleasant” to “extremely pleasant” (Figure 2).

The data generated by this study were statistically processed using SPSS 11.19 statistical software. The analysis of variance (ANOVA) was used to process the data from the sensory analysis of the mixed juices produced. Whenever a significant difference (α < 0.05) was found, the ANOVA test was supplemented by

Tukey’s post-ANOVA test to identify the variable(s) with highly significant differences from the control values.

The physico-chemical and biochemical composition of the juices analyzed is presented in Table 2. Cashew apple juice has a pH of 4.2 ± 0.00 and a refractive index of (1.35 ± 0.00) RI which are higher than the pH of passion fruit juice which is 3.14 ± 0.02 and the refractive index which is (1.34 ± 0.00) RI. Nevertheless, in passion fruit juice the titratable acidity is (1.17 ± 0.01)g of lactic acid/L, the density is (1.60 ± 0.00) and the Brix degree is (14.17 ± 0.00), 01)˚B are higher than in cashew apple juice with a titratable acidity of (0.79 ± 0.00) g of lactic acid/L, a density of (1.05 ± 0.00) and a Brix degree of (13.7 ± 0.0)˚B. In terms of biochemical composition, cashew apple juice has a dry matter content of (19.9 ± 0.1)%, ash content of (2.5 ± 0.7)%, total sugars of 85.3 g/L and reducing sugars of 9, 174 g/L higher than (13.43 ± 0.06)% dry matter, (0.68 ± 0.10)% ash, 43.6 g/L total sugars and 16.714 g/L reducing sugars in passion fruit juice. However, the protein content of (1.34 ± 0.05)% in passion fruit juice is higher than in cashew apple juice which is (0.22 ± 0.04)%.

Figure 3 gives the appreciation of the juices according to the attribute of the color. The color of the Passion Control juice (T passion) is more appreciated with a score of 6.94 than the color of the Cashew Control juice (T cashew) which has a score of 5.75. In addition, the color of C, D, E and F juices, which score

*Values in the same line with the same letter are not significantly different from each other according to Duncan’s multiple comparison test at the 5% threshold (P > 0.05). Values are expressed as Mean ± Standard Deviation (n = 3 determinations).

6.66, 6.66, 6.90 and 6.52 respectively, is close to that of T passion juice. While the color of A, B and G juices, whose scores are 6.32, 6.23 and 4.47 respectively, is similar to that of T cashew juice. The color of juice E is more appreciated as it has the highest score which is 6.90.

Figure 4 shows the evaluation of juices according to the flavour attribute. Cashew T Juice with a score of 5.19 and Passion T Juice with a score of 5.04 have virtually the same flavors; however, it should be noted that the flavor of Cashew T Juice is more appreciated. In addition, the flavor of E and F juices with a score of 5.47 and 5.66 respectively is close to that of Cashew T juice. While the flavor of A, B, C, D and G juices whose respective scores are 5.04; 5.09; 5.04 and 4.95 are close to that of T passion juice. Juice F is more appreciated because it obtained the majority score corresponding to 5.66.

Figure 5 gives the appreciation of the juices according to the attribute of decantation. Passion T juice with a score of 5.33 and cashew T juice with a score of 6.14 have virtually the same decantation; however, it should be noted that the decantation of Passion T juice is more appreciated. On the other hand, the decantation of D, E and F juices with respective scores of 5.52, 6.28 and 5.04 is close to that of cashew T juice. While the decantation of A, B, C and G juices,

which have scores of 4.33, 4.71, 4.80 and 4.71 respectively, is similar to that of passion fruit T juice. The decanting of juice A is more appreciated because it settles less. It obtained a score of 4.33.

Figure 6 shows the evaluation of juices according to the odour attribute. The odor of Passion T juice with a score of 7.17 is more appreciated than the odor of Cashew T juice which has a score of 5.52. In addition, the smell of B, C, D, E and F juices, whose scores are 5.90; 6.09; 6.09; 6.42 and 5.80 respectively, is similar to that of T passion juice. While the odor of juices A and G, whose scores are 5.38 and 4.80 respectively, is similar to that of cashew T juice. The smell of E juice is highly appreciated because the majority score corresponds to 6.42.

Figure 7 shows the assessment of juices according to the attribute of global acceptance. Passion T juice with a score of 6.00 is more appreciated than cashew T juice with a score of 5.72. In addition, the overall acceptance of C, E and F juices, with scores of 5.96, 6.53 and 5.96 respectively, is close to that of passion T juice. While the overall acceptance of juices A, B, D and G, which have scores of 5.72; 5.53; 5.67 and 5.19 respectively, is close to that of juice T cashew. Juice E is the most popular with a score of 6.53.