Obtaining and pre-processing raw material

Grape (merlot variety) and pomegranate (wonderful variety) residues were obtained in the regions of Parras and Cuatro Ciénegas, Coahuila, respectively. Since these residues were perishable, a dehydration pretreatment was performed on them in a convection drying stove (Biobase Biodustry BOV-T70C) at 60 °C for 24 hours. Grape and pomegranate dry residues were ground and classified by means of Montinox sieves (Mexico) of three sizes, #8 mesh (2.36 mm), #12 mesh (1.40 mm) and #20 mesh (0.85 mm). One gram of each vegetable sample was weighted, placed in 40 mL of distilled water and stirred for 20 min. Then the mixture was filtered and collected in glasses with ice, which was used for further analysis, considered as the filtered sample.

Determining total sugars

For quantifying total sugars (^{Dubois et al., 1956}) in the extract, 500 μL of the filtered sample was placed in a tube with 500 μL of 5 % phenol (Fermont, Mexico) and it was placed in an ice water bath for 5 minutes. With the tubes inside of the ice water bath, 1000 μL sulfuric acid (Jalmek, Mexico) were slowly added on the walls of the tubes and the mixture was let rest for 15 minutes. To stop the reaction, tubes were softly stirred and brought in a hot water bath at 50 °C for 5 minutes. Then, they were let cool at room temperature for 5 minutes. Absorbances were read in a BIOBASE EL-10 A ELISA microplate reader at 470 nm. Calibration curve was performed by using 0.1 % glucose (Jalmek, Mexico).

Quantifying reducing sugars

For determining reducing sugars (^{Miller, 1959}), 1000 μL of the filtered mixture were added to 1000 μL of the DNS (3,5-Dinitrosalicylic acid) prepared and the mixture was placed in a water bath (Thermo Scientific), at 50 °C for 5 minutes. 5 mL of distilled water were added, it was let rest for 5 minutes at room temperature. Absorbances were read in a BIOBASE EL-10 A ELISA microplate reader at 520 nm. Calibration curve was performed by using 0.1 % glucose (Jalmek, Mexico). For preparing DNS reactive, 10 mL distilled water were heated at 50 °C in a water bath (Thermo Scientific), stirring and 3,5-Dinitrosalicylic acid (Sigma-Aldrich, U.S.A.), sodium hydroxide (Jalmek, Mexico), sodium and potassium tartrate (Jalmek, Mexico), phenol (Fermont, Mexico) and sodium sulphate (Fermont, Mexico) were added in this order, it was diluted to 100 mL with distilled water, for conservation it was kept in dark and refrigerated.

Analyzing antioxidant compounds

For determining antioxidant compounds (^{Molyneux, 2004}), 100 mg of sample were weighted and 1.5 mL methanol (Hycel, Mexico) were added in a 2 mL conic tube. The mixture was stirred in vortex for 20 seconds, sonicated for 5 minutes and then centrifuged at 12,000 rpm for 10 minutes. The supernatant was extracted and filtered with nylon syringe filter with pore size of 0.45 μm for further analysis. 2,2-Diphenyl-1-picrylhydrazyl (DPPH; Sigma-Aldrich, U.S.A.) was prepared at a concentration of 60 % w/v. The vegetable material sample and the DPPH radical were placed in a 50:50 ratio. Absorbances were read in a BIOBASE EL-10 A ELISA microplate reader at 540 nm.

Optimizing the thermal hydrolysis process

Optimization process was performed according to a fractional factorial design with orthogonal arrays, by using Minitab 17.1.0 software. Three factors were selected (time, pressure and particles size) and an orthogonal matrix was established with three levels of variation per factor (Table 1).

Thermal hydrolysis process

For the hydrolysis, 2 g of sample were used of each size of vegetable material and were placed in fiber filter bags (Ankom Technology F57), with pore size of 0.25 μm, with 500 mL distilled water. Then, the bags were placed inside a Molded Aluminum Pressure Reactor. Samples were classified as “slurry” referring to grape bagasse after the hydrolysis treatment; and the liquid fraction derived from the hydrolysis treatment was called hydrolyzed broth. Total sugars were determined in these two fractions. After obtaining results of each of the samples, they were compared and a factorial design analysis was performed using Minitab (17.1.0 version) statistical software to be able to select the best values.

Statistical analysis

For processing data, Minitab (17.1.0 version) statistical software package was used. An analysis of variance (ANOVA) and a Tukey’s mean comparison test were performed (p ≤ 0.05). Results were represented as mean ± standard deviation.

Experimental analysis of the thermal hydrolysis treatment

The levels experimentally evaluated to be able to study the thermal hydrolysis process, presented heating and cooling profiles with times of 13±2 min and 22±2 min, respectively. Table 3 shows the analysis of variance performed on reducing sugars data obtained from the treatments derived from the L-9 orthogonal design. For grape liquid hydrolysate, time, pressure and particles size factors in the individual levels evaluated did not showed significant effects (p ≤ 0.05). However, the model estimated by Taguchi’s analysis achieved to determine that the best effect for releasing reducing sugars for the hydrolysate was given by the particles size, followed by the duration time of the process (Figure 1).

Figure 1 Main effects for the means of reducing sugars in the grape hydrolysate. 

For the slurry of grape residues, the three factors evaluated showed a significant effect for the release of reducing sugars (p ≤ 0.05), being the time the factor which most benefits the process, followed by the pressure in the system (Figure 2).

Figure 2 Main effects for the means of reducing sugars in the grape slurry. 

In the case of the liquid hydrolyzed and the slurry of pomegranate residues, time, pressure and particles size factors in the individual levels evaluated, did not show significant effects (p ≤ 0.05) in either treatment.

The greatest effect for the release of reducing sugars for the hydrolysate was given by the pressure during the process, followed by the duration time of the process (Figure 3); and the greatest effect for the release of sugars in the slurry was the particles size, followed by the time of the process (Figure 4).

Figure 3 Main effects for the means of reducing sugars in the pomegranate hydrolysate. 

Figure 4 Main effects for the means of reducing sugars in the pomegranate slurry. 

The difference detected in ANOVAs and the impact of the factors evaluated in the release of monomeric sugars can be due to the characteristics of the vegetable material. When using residues of industrial processes where sugars are required as raw material, they lack of hydro-soluble available sugars for their aqueous extraction at low time and temperature conditions.

Analyzing and optimizing thermal hydrolysis conditions for grape residues showed that time-size and pressure-size interactions presented the highest severity indexes (27.41 % and 20.76 % respectively). For pomegranate residues, the most critical interaction with 34.29 % of severity was the pressure-size one. The particles size makes the hydrolysis easier, due to the extension of the contact area, in addition to improve solubility for the smaller particles sizes (^{Grajales Martínez et al., 2018}).

Therefore, the conditions that most favored the release of reducing sugars of grape residues hydrolysates can be determined, such as: time of 10 minutes, pressure of 10 psi and size of 1.40-1.70 mm; while for slurry, they were: time of 15 minutes, pressure of 15 psi and size of 2.36-2.80 mm. The conditions for pomegranate residues hydrolysates were time of 15 minutes, pressure of 20 psi and size of 1.40-1.70 mm; while for slurry, they were: time of 15 minutes, pressure of 15 psi and size of 2.36-2.80 mm.

The conditions achieved a better extraction of reducing sugars content (Table 5), allowing to generate an added value to these residues, and being able to use them as a substrate in diverse processes for obtaining biomolecules or biofuels.

In both vegetable residues, the hydrolysis process achieved to increase reducing sugars extraction present in vegetables, compared to initial values of the respective grape and pomegranate values. The quantity of water absorbed in the particles allows to give access to water molecules, helping swell the vegetable material, generating amorphous regions and favoring the hydrolysis process (^{Kapoor et al., 2019}).

Hydrolysis treatments achieved to convert recalcitrant vegetable wall, mainly made up by lignin, into hemicellulose compounds of higher availability for monomeric sugars. ^{Zietsman et al. (2017)} identified that the thermal treatment to which their grape residues samples were submitted managed to eliminate pectin and to increase released carbohydrates (xylan, xyloglucan, cellulose and mannan polymers) in the supernatants obtained. ^{Dos Santos Rocha et al. (2017)}, when performing hydro-thermal treatments studies on sugarcane bagasse, reported that thermal treatments can remove hemicellulose up to 85 %, favoring the conversion of complex carbohydrates into oligomers and monomers.

Pressure (temperature) conditions and particles size evaluated achieved to remove lignin from the liquid phase (hydrolysate), being more evident in pomegranate residues than in grape residues (Table 4). The low quantity of glucose in hydrolysate grape can be due to the conversion of initial monomeric sugars into hydroxymethylfurfural and furfural (^{Dos Santos Rocha et al., 2017}).

The analysis of the antioxidant capacity of materials of grape bagasse and pomegranate residues were observed to be related to times of residence and temperature, due to the characteristics presented by proanthocyanidins, anthocyanins and flavonoids when being submitted to hydrolysis and when increasing hydrolysis times (^{Petrović et al., 2016}), which represents a decrease in its antioxidant activity when being submitted to higher times and temperatures of the thermal process. The antioxidant activity is related to phenolic compounds present in the vegetable material and to the derivatives of lignin degradation by thermal treatment. In the case of grape and pomegranate hydrolysates, reductions of the antioxidant effect were shown in samples treated with thermal hydrolysis. The concentration and the type of phenolic compounds present in the hydrolysates depend on operating conditions (temperature and residence times), as well as on the treated lignocellulosic material and its initial phenolic compounds content (^{Moure et al., 2017}).