The apples (Malus domestica) Red Delicious variety were provided by Kleppe company (Cipolletti, Rio Negro, Argentina).

The chemicals used were lactic acid (85%–90%. Pro-Analysis, ACS), citric acid (Pro-Analysis, ACS), anhydrous D (+)-glucose (Pro-Analysis, ACS), and phenol crystals (Pro-Analysis, ACS) were provided by Cicarelli Laboratories (Santa Fe, Argentina). D- (+)-galacturonic acid monohydrate, Protein Standard BSA, and 3-phenylphenol were provided by Sigma Aldrich (Missouri, USA). Methanol ACS was provided by Panreac AppliChem, (Castellar del Valles, Spain). Ácido sulfúrico (98%. ACS) and acetyl acetone (≥99%) were provided by Anedra (Buenos Aires, Argentina). Folin Ciocalteu reagent was provided by Biopack (Buenos Aires, Argentina). Deionized water was used.

The raw material was processed at the Food Pilot Plant of Departamento de Industrias (Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires). The apples were sanitized with an aqueous solution of sodium hypochlorite, peeled, cored, and cut into 12 pieces with a fruit peeling and corer machine (Zhengzhou Xinshijia Machinery Equipment Co., Model QPJ-APP10). They were subsequently immersed in citric acid aqueous solution (0.5 g/100 mL) and ascorbic acid aqueous solution (2.3 g/100 mL) in order to delay their browning during the juice extraction process by pressure (Metalúrgica VZ, Córdoba, Argentina). After obtaining the juice, the by-products (peel remaining from peeling and pulp remaining from pressing) were washed with water and ethanol 96% (v/v) to eliminate intrinsic sugars from the apple tissue that could interfere with the handling of the material during dehydration and affect the obtaining of fractions enriched in pectin.

To obtain the material enriched in cell wall components (CWM) from both the pulp (CWM-Pulp) and the peel (CWM-Peel), the washed material was dehydrated in a convection dryer at 70°C and with an air speed of 6 m/s until obtaining an aqueous activity between 0.3 and 0.4, thus ensuring its stability during storage (Clemenson et al., 2012). A 212–710 µm powder rich in the cell wall material (CWM) was obtained through grinding with a domestic appliance (Wemir E909, Buenos Aires, Argentina) and sieving with ASTM sieves 25 (standard 710 µm) and 70 (212 µm) (Zonytest, Buenos Aires, Argentina). The sieving was performed by means of a vibrating device (sieve shaker) manufactured by the technical staff of the Departamento de Industrias (Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires). The yield was calculated as g of CWM per 100 g of by-product.

To evaluate the cellulose and lignin content, acid hydrolyses were carried out under the action of sulfuric acid and the application of heat, according to Ng et al. (1998). Specifically, the CWM of each tissue was subjected to hydrolysis with sulfuric acid (72 g/100 g) for 3 hours at room temperature. This suspension was then brought to a sulfuric acid concentration of 1 mol/L by adding a sufficient amount of MilliQ water, and the samples were heated at 100°C for 2.5 h. Once cooled, they were centrifuged for 10 min at 10,000 rpm. The residue was carefully separated from the supernatant, washed with deionized water, centrifuged for 10 min at 10,000 rpm, and finally lyophilized. The dry residue obtained was weighed and reported as lignin. A second procedure applied to the CWMs consisted of their hydrolysis with a sulfuric acid solution with a concentration of 1 mol/L at 100°C for 2.5 h. Then, the same steps described above were followed, and the lyophilized residue obtained corresponded to cellulose + lignin. Both hydrolyses were performed in triplicate. Non-cellulosic carbohydrates, uronic acids, and protein content were determined in triplicate in the supernatants with the methods described by Dubois et al. (1956), Filisetti-Cozzi and Carpita (1991) and Lowry et al. (1951), respectively (Santo Domingo et al., 2015).

The moisture content of the powders was determined in triplicate by means of an infrared scale (Moisture Analyzer MB45Ohaus Corporation, New Jersey, USA), using a 0.5-g sample. Water activity was measured in duplicate at 25°C in a Decagon AquaLab (Series 3 Water Activity Meter, Pullman, WA, USA).

Yield was calculated as g of fraction isolated per 100 g of CWM used.

Non-linear fits and statistical evaluations were made using the Prism 5 software (GraphPad, California, USA). The statistical significance of the differences was carried out using ANOVA with a level, α, of 0.05. The Tukey test was used as an “a posteriori test” (Sokal and Rohlf, 2000).

The apple CWM-Pulp and CWM-Peel showed a peel yield of 17.8 g/100 g and a pulp yield of 6.6 g/100 g, with moisture contents of 5.5 g/100 g and 7.5 g/100 g and water activities (a_w) of 0.29 and 0.37, respectively (Table 1).

The CWM-Pulp showed a non-cellulosic carbohydrate content (57.8 g/100 g) higher than peel (40.4 g/100 g). Lignin and cellulose constitute the main structural components of the plant cell wall (George et al., 2020). In the CWM-Peel, a content of 30.3 g/100 g of lignin and 12.7 g/100 g of cellulose was determined. These data were significantly higher than those of the CWM-Pulp. This trend is consistent with the protective function of the fruit that the peel fulfills, lignin being primarily responsible for giving rigidity to the cell wall, making it resistant to impacts and bending (Escudero Álvarez and González Sánchez, 2006). The UA and protein contents were 11.4 g/100 g for the UA and 8.5 g/100 g for proteins in the CWM-Pulp, while for CWM-Peel, the values were 10.1 g/100 g and 3.2 g/100 g, respectively.

As seen in Table 2, pectic fractions with low moisture content (6.1–9.0 g/100 g) and a_w values between 0.24 and 0.45 were obtained, thus ensuring their stability against deterioration during storage (Clemenson et al., 2012). The fractions extracted with NADES were less luminous and showed an intensification of red colors and a decrease in yellows compared to those extracted with citric acid. This could affect their potential for application in food products.

Success in isolating a compound from a plant matrix depends on various factors such as the extraction method, the part of the plant treated, temperature, pressure, time, and external energy delivered (Al Ubeed et al., 2022). With respect to the solvent used, there are specific factors that influence the process, such as the affinity of the compound to be extracted with the solvent. The viscosity of the solvent must also be considered because a high viscosity produces a more difficult contact of the solvent with the tissue. In this investigation, those fractions obtained with CA had a higher yield (9.6–12.2 g/100 g) than those extracted with N1 and N2. The high viscosities of the two latter solvents, as well as the washings necessary to eliminate the remaining glucose in these fractions, might be at least a partial explanation for the trend. At 100 s⁻¹ at the working temperature (80°C), N1 presented a viscosity of 137 mPa s and N2 of 4.62 mPa s (Gómez Vargas et al., 2023), while the working temperature of the citric acid solution was approximately 1.1 mPa s (Personal communication). Likewise, it was observed that N2 gave rise to a higher yield than N1 in the fractions extracted from pulp (5.1 g/100 g, N2 vs. 2.5 g/100 g, N1) but not in peel where there were no significant differences (≈4.1 g/100 g).

All fractions had a high content of total non-cellulosic carbohydrates (76.1–84.0 g/100 g) and a protein content with values in the range of 3.2–11.1 g/100 g.)

UAs are an important component of pectin; therefore, their determination is a fundamental step in the analysis of its structure (Dranca et al., 2020). Although a notable increase in UA is observed in the fractions when compared to their CWM of origin, they must be identified as “fractions enriched in pectin” because they fail to reach the 65% (m/m) established by the International Pectin Producers Association (IPPA) to be defined as pectins. Those extracted with N2 allowed a significantly higher extraction of UA (58.1–62.8 g/100 g) than those obtained using N1 (50.1–50.9 g/100 g) or CA (50.7–51.0 g/100 g) as an extractant. In each case, no differences were observed in this content when the tissue used was pulp or peel for each extractant.

The degree of methoxylation (DM), or percentage ratio between the moles of methanol and the moles of GalA, is an important characteristic for the rheological behavior of the extracted pectin fraction (Gerschenson et al., 2021). In the present work, isolated pectins were for high methoxyl content with obtained values greater than 50%. It was observed that those obtained from N2 had a lower DM (53%–54%) than those obtained with N1 (59%–71%) or with CA (73%). The lower pH of the NADES probably determined the lower DM observed for the pectins isolated with them.

In the analysis of the effect of tissue source, significant differences were observed exclusively in those samples extracted with N1 that showed lower DM when they originated from peel. Previous works have reported that extracting pectin from apple gives rise to highly methoxylated polysaccharides using inorganic and organic acids, enzymes, and NADES. In particular, Riyamol and Jeevitha (2024) observed that 16 different NADES based on choline chloride, proline, or betaine as hydrogen bond acceptors produced higher methoxylation degrees for pectins isolated from onion peels than those obtained with conventional acid extraction, but this trend was not observed in the present investigation in which NADES of different composition were used.

Figure 1 shows the FT-IR spectra of the pectic fractions obtained. A change in intensity was observed in the bands depending on the residue (pulp or peel) and the extractant used (CA, N1, or N2).

The band observed at ≈3,400 cm⁻¹ corresponds to the stretching vibration of the -OH groups of galacturonic acid (Zheng et al., 2021). In the different fractions, this band was observed between 3,434 cm⁻¹ and 3,437 cm⁻¹. The CH stretching vibrations of the CH2 groups are responsible for generating the small band at ≈2,900 cm⁻¹ (Xu et al., 2022). For fractions isolated from CWM-Pulp, bands were found between 2,927 cm⁻¹ and 2,936 cm⁻¹. In the fractions obtained from CWM-Peel, the bands were observed at 2,917–2,919 cm⁻¹. Two bands were found for all the fractions obtained from the peel for all the extractants used, one at approximately 2,920 cm⁻¹ and the other at 2,850 cm⁻¹. The band at approximately 2,850 cm⁻¹ is due to remnants of the plant cuticle, such as cutin, waxes, and cutan, and is determined by the symmetrical stretching vibrations of their CH2 groups (Heredia-Guerrero et al., 2014).

The band at ≈1730 cm⁻¹ can be ascribed to the vibration of the carboxyl groups esterified with methyl (COO-R). In this band, a shift was observed according to the extracting solvent used. In the case of CA, the wavenumber was 1750 cm⁻¹. In the case of N1, it occurred at approximately 1715–1716 cm⁻¹, and for N2, it occurred at 1723–1727 cm⁻¹. The band observed between 1,632 cm⁻¹ and 1,638 cm⁻¹ is caused by the asymmetric stretching of the non-esterified carboxylate ions (COO-) to which the band that occurs at approximately ≈1,400 cm⁻¹ (carboxylate symmetric stretching band) is also attributed (Koziol et al., 2022). This last band was only detected in PF4 and PF5 with values of 1,448 cm⁻¹ and 1,451 cm⁻¹, respectively.

In the isolated fractions, the weak band observed at approximately 1,230 cm⁻¹ is attributed to the vibrations of the pectin side chains. Bands with wave numbers at approximately 1,105 cm⁻¹ and 1,014 cm⁻¹ were detected, allowing the identification of pectic polysaccharides rich in uronic acids (Alonso-Simón et al., 2004). The bands of medium or low intensity observed below 920 cm⁻¹ in the “fingerprint” region are fundamentally attributed to vibrations of the C–O–C bridges typical of polysaccharides (Koziol et al., 2022).

The FT-IR spectra recorded were characteristic of the typical polygalacturonic acid backbone of pectins.

Knowledge of rheological properties is essential for the adequate formulation of food products as well as for the design of their preservation processes. Therefore, the rheological properties of pectin-enriched fractions were investigated to evaluate their functionality. For this, aqueous solutions of 1 g/100 mL and 2 g/100 mL were evaluated with and without sucrose and acidification. Note that, for the formation of a gel, high methoxyl pectins require the addition of sucrose and acid that favor the establishment of hydrophobic interactions and hydrogen bonds (Zhao et al., 2020). Sucrose induces the heterogeneous distribution of water around the pectin macromolecules, producing self-assembly of the polymer chains, which contributes to the increase in viscosity and could result in macroscopic gelation in an acidic medium.

The results obtained can be observed in Figure 3 and Tables 6, 7. From the flow curves, it can be concluded that all fractions formed viscous solutions that behaved like pseudoplastic fluids because their viscosity decreased as the shear rate increased. This behavior can be attributed to structural changes in the macromolecular solutions due to the increase in the shear stress applied (Dranca et al., 2020). Analogous results were reported by Morales-Contreras et al. (2020) for high methoxyl apple pectin isolated with hydrochloric acid. Fissore et al. (2009) studied the flow behavior of pectins with a high degree of methoxylation, isolated by enzymatic treatment (hemicellulase), from Cucurbita moschata Duch ex Poiret, observing that their aqueous solutions presented pseudoplastic behavior that was attributed to the structure generated by the high density of interactions established between hydrated macromolecules.

The experimental data were analyzed using the Ostwald model and the Herschel–Bulkley model. The results can be seen in Tables 6, 7. In general, the polymeric solutions of the fractions presented a flow behavior that significantly adjusted to the Ostwald model, with the exception of PF4 (1 g/100 mL) and PF6 (2 g/100 mL), both with sucrose and acidification, whose flow behavior significantly adjusted to the Herschel–Bulkley model, presenting a minimum shear stress (τ₀) of ≈5.9 Pa. This indicates they are interesting hydrocolloids for the formulation of foods due to the possibility of using them as thickeners that also allow controlling the rate of discharge from their containers.

Although all fractions gave rise to aqueous solutions with the behavior of pseudoplastic fluids, significant differences were found in the flow behavior index (n) and the consistency coefficient (K) depending on the technique applied for obtaining the fraction, its concentration, and the presence of sucrose and acidification. In all cases, K increased, and n decreased i) when incorporating sucrose, both at 1 g/100 mL and at 2 g/100 mL, or ii) when the fraction concentration was increased from 1 g/100 mL to 2 g/100 mL in the presence or absence of sucrose. Those fractions isolated with CA and N2 showed, in general, higher K and lower n in the 1 g/100 mL or 2 g/100 mL solutions than those isolated with N1 (Tables 6, 7). This might be linked to the composition of the different isolated fractions and, more specifically, to the characteristics of the macromolecules present in them. Davitadze (2023) reported that for pectins, the content of neutral sugar and, therefore, the shielding of the carboxyl groups by those sugars contributes to the twisting and formation of globular structures by the macromolecules, thus contributing to its rheological behavior. Therefore, prolonged thermal treatments of pectic extracts, which destroy arabinose, hydrogen bonds, and arabinan, galacturonan, and galactan bonds, can affect the rheological behavior.

The fractions PF1 and PF4 at 2 g/100 mL in aqueous solution containing sucrose presented a significantly higher K than the rest (32.8 Pa sⁿ and 25.8 Pa sⁿ), thus evidencing interesting properties due to their important potential contribution to the texture of foods. Their low n (0.29) showed an important decrease in viscosity with the increase in shear rate.

The viscosity and flow characteristics of the systems indicate that they can be used as potential food thickeners and stabilizers.

The objective of ultrasound pretreatment in the extraction of pectin using NADES was to evaluate whether cavitation increased the yield of the pectin fractions through its action on apple tissues. This evaluation was performed with N2 extraction because the characterization of fractions obtained with this solvent without US pretreatment showed a greater content of UA, higher purity, larger ramifications on Rha, and higher Tg. For all this, and although it is known that US can alter the chemical structure and functionality of pectin, N2 seemed to be the most interesting NADES to be explored. The results obtained are reported below.

The yield when the US pretreatment amplitude was 20% and the solvent treatment duration was 30 min were 13.9 g/100 g CWM-Pulp (USPFA) and 6.4 g/100 g CWM-Peel. Under the same conditions with a US amplitude of 50%, they were 15.1 g/100 g CWM-Pulp (USPFB) and 6.3 g/100 g CWM-Peel.

The yield when the US amplitude in pretreatment was 20% and the treatment duration was 60 min were 8.14 g/100 g CWM-Pulp and 17.9 g/100 g CWM-Peel (USPFC). Under the same conditions, but with a US amplitude of 50%, they were 7.4 g/100 g CWM-Pulp and 13 g/100 g CWM-Peel (USPFD). This indicates that in the case of pulp, at both amplitudes, shorter times (30 min) produced higher yields, but in the case of peel, longer treatment times (60 min) obtained higher yields. Accordingly, work continued with the fractions obtained with the treatments that gave greater yield for each tissue, and their composition was characterized.

The results obtained can be seen in Table 8. The yields even exceeded the values reported in Table 2 for the treatments with CA, which were the highest in the absence of US assistance. It can be seen in Table 8 that in the proposed technique with US, a greater enrichment in UA was obtained (USPFA, USPFB, and USPFD). In particular, the fractions obtained from pulp with values of 70.7 g/100 g for USPFA and 70.6 g/100 g for USPFB could be called pectins as established by the IPPA (International Pectin Producers Association, 2024) because they exceed the value of 65 g/100 g. In contrast, the fraction obtained from peel with US pretreatment of 50% amplitude, although it showed higher UA content (63.6 g/100 g in USPFD) than the fraction obtained with N2 and without the assistance of US (Table 2, Fraction PF6, 58.1 g/100 g), did not reach the minimum established to be considered pectin under the IPPA standard. The values for DM ranged between 68% and 70%, being higher than those observed for PF3 and PF6 (53%–54%) (Table 2).