Pineapple and papaya peels by-products generated from minimal fresh fruit processing were collected from two local small-scale food units (Las Exquisitas Food & Snacks, and Jugos y Chocos Tony). Bulk refined sugar, corn starch glucose (45 Globe/Gloria 2534, Ingredion México S.A. de C.V.), unflavored powder gelatin Bloom 290 (Coloidales Duche S.A. de C.V), citric acid (MABI), Capol 3073A (a release agent for gums and jellies with highly stable vegetable oils and natural waxes; Capol GmbH), chamoy and pineapple flavoring (Deiman S.A. de C.V.) and bulk cornstarch, were obtained from the local market.

The peels free of mold and visible degradation were selected, washed and disinfected by immersion for 10 min in a solution (2 mL/10L water) of Nicon-PQ (active component citric acid: Procesadores de Aditivos S.A. de C.V., Investa de México S.A. de C.V., Mexico), and subsequently submerged in water at 90˚C to inactivate enzymes. The excess of solution was removed in a vegetables centrifuge (Ilko, México), and the peels separated into two batches for each by-product. The first batch was dried with circulating hot air (Excalibur Products, 3500, USA) at 57˚C for 16 to 24 h for pineapple peels, and 8 h for papaya peels. The second batch, previously frozen to −78˚C (SO-LOW, U40-13, USA), was freeze-dried in a laboratory unit (Labconco, FreeZone^® Legacy, 7752020, USA) at 0.0099˚C and 610.5 Pa vacuum. The dehydrated peels were ground (Oster, 6859, MX) and sieved (Quimilab, Sieves, Flicc, S.A. de C.V.) for 30 min. The pineapple peel powder was divided into three particle size fractions: greater than 300 μm, 250 μm, and lower that 180 μm. That of papaya was divided into two fractions: greater than 300 μm, and 250 to less than 180 μm (because the yield of papaya peel powders with particle size of 250 was low). Six different pineapple peel powders, labeled as Pi-HA-1, Pi-HA-2, Pi-HA-3, Pi-LYO-1, Pi-LYO-2 and Pi-LYO-3, and four of papaya peels powders, identified as Pa-HA-1, Pa-HA-2, Pi-LYO-1 and Pi-LYO-2, were obtained, where HA stands for air dried and LYO for freeze-dried.

The properties of the powders were determined: Moisture [20]; Total dietary fiber [20]; Water activity; DPPH [21]; ABTS [22]; and Polyphenols [23] are shown in Table 1.

Control gummy candies, i.e. without fruit peel powders, were made according to an existing formulation [24]. This formulation was adapted to prepare the modified gels as follows; 50 mg/g sucrose was replaced by pineapple or papaya powder. Artificial colorant was not added, and flavoring was not added to

*mg eq. Trolox^®/100 g d.b.; **mg eq. gallic acid/100 g d.b.; d.b. = dry basis.

pineapple formulation. However, 1 mg/g of chamoy flavor was added to papaya formulation to counteract the strong taste of papaya. Citric acid was decreased by 2.8 mg/g because of the content of organic acids of the powders (Table 2).

To produce the gummy candies, gelatin was hydrated with 0.453 g/g of the total water of the formulation and left to stand for 30 min. Sucrose was dissolved with the remaining 0.546 g/g of the water at a temperature of 70˚C, then glucose was added the blend cooked until it reached 105˚C. The hydrated gelatin was added immediately to the control formulations and mixed to homogenize the ingredients. Then the citric acid previously dissolved in water, flavor (pineapple or chamoy) and yellow color were added for the pineapple control, and orange for the papaya one. For the modified gummy candies, after concentrating the syrup, the powders of pineapple or papaya peel were added and mixed. The hydrated gelatin was added and mixed to homogenize the ingredients, then the citric acid previously dissolved in water was added, and chamoy flavor only for papaya gels. The resulting gels were molded in starch chests to hemisphere molds, 1.5 cm in diameter, and gelled at 23˚C for 24 h. The gummy candies were demolded, brushed, waxed with Capol and stored in polypropylene bottles at 20˚C. The products were labeled as Control-Pi, G-Pi-HA-1, G-Pi-HA-2, G-Pi-HA-3, G-Pi-LYO-1, G-Pi, LYO-2, G-Pi-LYO-3 for pineapple, and Control-Pa, G-Pa-HA-1, Pa-HA-2, Pa-LYO-1, Pa-LYO-2 for papaya/chamoy.

The content of soluble solids (˚Bx) at 20˚C determined in a refractometer (Atago, H-93, USA), and pH at 25˚C measured with a potentiometer (Hanna Instruments, HI99163, USA) previously calibrated with buffer solutions pH = 7 and 4 were measured in triplicate. Water activity (a_w) was determined at 25˚C in a meter (Decagon Devices, Inc., AquaLab Series 3TE, USA) previously calibrated with aqueous standards of a_w = 0.250 (13.4 mol/kg LiCl), a_w = 0.5 (8.57 mol/kg LiCl), and a_w = 0.984 (0.5 mol/kg KCl). Instrumental color was determined in a colorimeter (Konica Minolta Sensing Americas, Inc, CR-400, Japan) and CIE-L*C*h*

coordinates (luminosity, L*, chroma, C*, and hue, h*) were obtained.

The texture profile analysis of one-day old hemispheric specimens 1.5 cm in diameter, and 1.5 cm in height was carried out at 20˚C in a texturometer (Stable Micro System, TA-XTplus, UK). Two different tests were run. The first one was done using an adaptation of the Gummy Confectionery Test available in the texturometer software. A cylindrical probe (compression platen P/75, 75 mm in diameter), a heavy-duty platform adjusted with a flat plate (HDP/90) and a trigger force of 50 mN were used. Before running the test, the probe was calibrated to recognize the analyzer’s test surface (100% voltage), and the distance (20 mm) back from the probe after compression was selected. Each gummy candy was compressed to 20% of his original height, at a speed of 1.0 mm/s. The cell was kept at this distance for 60 s and then removed to its initial position at a speed of 10.0 mm/s. The MACRO of the resulting force-distance curve was run, and the following texture properties recorded: firmness which corresponds to the total strength (N) of the specimen, and elasticity (%) which corresponds to the height that the sample recovers after being compressed.

The second test was an adaptation to the testing standard of the Gelatin Manufacturers Institute of America [25]. A cylindrical Derlin probe (P/0.5R, 1.27 cm in diameter), a heavy-duty platform adjusted with a flat plate (HDP/90) and a trigger force of 50 mN were used. The specimen was placed on the base aligned centrally under the probe and penetrated at a speed of 1.0 mm/s, and a depth of 4 mm (maximum force). The MACRO of the resulting force-distance curve was run and expressed as “gel strength/bloom value” (g) of the gummy candy gel. Water activity, color and texture were determined in ten candies per formulation.

Two tests were carried out with an untrained panel of fifty customary consumers of this type of product in the range of 18 - 25 years old, with the same number of females and males. The selection was based on a survey of habits to rule out people allergic to the product, as well as frequent consumers of alcohol and cigarettes. The first analysis was done to select the preferred particle size by means of an acceptance test of preference at a level of significance of 5%. Once the preferred particle size was selected, a second sensory analysis was carried out with the preferred samples. The test consisted of a five-point hedonic scale to express how close color intensity, firmness, pineapple or papaya/chamoy flavor, elasticity, sweetness and acidity were to those preferred by the panelists, i.e. 1 = too weak, 2 = a little weak, 3 = just as I like it, 4 = a bit strong, 5 = too strong [26]. The samples were presented as a 1.5 cm diameter hemisphere at room temperature (20˚C - 22˚C) in random order and were labeled with randomly generated three-digit codes.

Data of the physicochemical analyses and results of satisfaction tests using hedonic scales of gummy candies were expressed as mean ± standard deviation (n = 3) and subjected to normality and homoscedasticity analysis. One-way analysis of variance (ANOVA) was used applying Tukey’s analysis for means with significant differences (p < 0.05) in physicochemical analyses and Fisher’s LSD for satisfaction tests. Data were analyzed using the statistical package InfoStat for Windows. Results of preference test were expressed as frequency of the number of scores given by the panelists. Tables of significance were used at a confidence level of 5%.

There were no differences (p < 0.05) in the content of soluble solids (˚Bx) between the samples prepared with pineapple or papaya peel powders and their respective controls (Table 3). This result indicates an improvement in the quality of the product and serves to fix the final concentration of soluble solids necessary to maintain the physicochemical stability of the gelled candies [27].

The pH showed differences (p < 0.05) between the different formulations and their controls; pineapple gummy candies exhibited the lowest pH. This result is explained by the acidity of the pineapple powder. The pH of the products with papaya peel powder was significantly (p < 0.05) higher than that of their control because papaya is less acid than pineapple. However, the pH range of the products was 3.0 to 4.5, which is recommended for gummies made with 290˚ Bloom gelatin [24]. The pH of the product is critical; a very low pH is not desirable because the product will not be stable, or a gel might not be formed. If a hydrocolloid

*Control = powder-free, G = gummy, Pi = pineapple; Pa = papaya; HA = dehydrated with circulating hot air; LYO = freeze-dried. Pineapple: 1 = ≥300 mm, 2 = 250 mm, 3 ≤ 180 mm. Papaya: 1 ≥ 300 mm, 2 = 250 mm to <180 mm. **Different letters in each column indicate significant difference (p < 0.05).

is maintained at its isoelectric point, pH for which the net charge is zero, then gelation is likely to occur [27].

Water activities of the enriched gummy candies showed significant differences (p < 0.05) (Table 3). The a_w of the controls was less, however, all the gummy candies comply with the range of water activity recommended for this type of products (0.54 to 0.66) since a higher value would favor microbiological degradation of the product due to its sugar content [24]. Therefore, it was established that enrichment of the gelled candies with pineapple and papaya peel powders does not interfere with the physicochemical and microbiological stability of the products, since they had low pH, low moisture, high content of soluble solids and reduced a_w. All these factors are important to obtain good quality products [24] [27].

Color is an important indicator of quality and acceptability of foods [28]. Table 4 shows the results of color determination of pineapple and papaya gummy candies. Luminosity, hue and chroma of the controls and the modified gummy candies were significantly different (p < 0.05).

The gummy candies modified with pineapple powders were less luminous with a more intense saturation and a yellow-brown color. The control of papaya/chamoy had greater luminosity, but chroma and hue indicate a more saturated orange color regarding the modified gummy candies. However, pineapple G-Pi-LYO-3 and papaya G-Pa-LYO-2 samples are the closest to the controls. The insoluble dietary fiber of the powders interferes with color in both gummy candies [2] [29]. The fiber does not dissolve and therefore the particle size plays an important role in color. The smaller the particle, the more homogeneous is the distribution of the powders in the product and these results in a better color.

Table 5 also shows the results of the instrumental texture measurements of the gelled candies. The modified candies were firmer, more elastic, but with lower strength, compared to the control samples. These results are consistent with those reported by Cappa et al. (2015) [28], who developed gel-like fruit candies (GLFC) based on apple and cranberry modified with grape skin powder of different particle sizes, using pectin as a gelling agent. These authors demonstrated that the powders and their particle size modify the instrumental texture of the GLFC. The particle size of the powders plays an important role on the mechanical properties, i.e. hardness and energy required for penetration, since a larger particle size results in a smaller force to penetrate the sample.