Effects of Different Heating Treatments on the Antioxidant Activity and Phenolic Compounds of Ecuadorian Red Dacca Banana

The banana is a tropical fruit characterized by its composition of healthy and nutritional compounds. This fruit is part of traditional Ecuadorian gastronomy, being consumed in a wide variety of ways. In this context, unripe Red Dacca banana samples and those submitted to different traditional Ecuadorian heating treatments (boiling, roasting, and baking) were evaluated to profile their phenolic content by ultra-high-performance liquid chromatography coupled to high-resolution mass spectrometry (UHPLC-HRMS) and the antioxidant activity by ORAC, ABTS, and DPPH assays. A total of sixty-eight phenolic compounds were identified or tentatively identified in raw banana and treated samples, highlighting the content in flavonoids (flavan-3-ols with 88.33% and flavonols with 3.24%) followed by the hydroxybenzoic acid family (5.44%) in raw banana samples. The total phenolic compound content significantly decreased for all the elaborations evaluated, specifically from 442.12 mg/100 g DW in fresh bananas to 338.60 mg/100 g DW in boiled (23.41%), 243.63 mg/100 g DW in roasted (44.90%), and 109.85 mg/100 g DW in baked samples (75.15%). Flavan-3-ols and flavonols were the phenolic groups most affected by the heating treatments, while flavanones and hydroxybenzoic acids showed higher stability against the heating treatments, especially the boiled and roasted samples. In general, the decrease in phenolic compounds corresponded with a decline in antioxidant activity, evaluated by different methods, especially in baked samples. The results obtained from PCA studies confirmed that the impact of heating on the composition of some phenolic compounds was different depending on the technique used. In general, the heating processes applied to the banana samples induced phytochemical modifications. Even so, they remain an important source of bioactive compounds for consumers.

Therefore, due to the limited information found in the literature regarding exhaustive phenolic compound content and profile in traditional cooking elaborations of bananas, the aim of this work was to study the impact in the Red Dacca banana pulp (Musa sp. AAA) of three traditional Ecuadorian heating treatments (boiling, roasting, and baking) on the antioxidant activity and phenolic compound profiles using ultra-high-performance liquid chromatography coupled to high-resolution mass spectrometry (UHPLC-HRMS).

Phenolic Compounds Profile
A total of sixty-eight phenolic compounds were identified and quantified in raw and heat treatment banana samples, including nine flavan-3-ols, nine flavanones, thirteen flavonols, sixteen hydroxybenzoic acids, sixteen hydroxycinnamic acids, and five phenylacetic acids. Table S1 (Supplementary Information) shows the characteristics for the identification of compounds, such as retention time (RT), the experimental accurate mass, and the error (ppm) between the exact accurate mass and the mass found of the detected compounds. The family of flavan-3-ols showed the highest concentration of phenolic compounds in the raw banana samples, reaching 390.51 mg/100 g DW (88%). The composition of this group consisted mainly of dimers and trimers of procyanidin and (−)-epicatechin (Table 1). In general, most of the literature has focused the research of flavan-3-ols on monomers, such as (−)-epicatechin, (+)-catechin, or (−)-epigallocatechin [8,14,17,18], neglecting the contribution of flavan-3-ols polymers in the banana pulp. In this sense, some publications have previously described (−)-epicatechin as the main flavonoid in green banana flour [17], whilst procyanidins were determined in high concentrations in banana peels [19]. Concerning our data, the content and profile of the flavan-3-ols group found in the banana pulp are not in agreement with most of the literature since we note a higher contribution of flavan-3-ols polymers. Other flavonoids were identified in bananas samples, including flavanones, eriodyctiol hexosides being the most abundant, and flavonols, where quercetin dihexoside and myricetin rutinoside were the compounds with the highest concentrations. Regarding the hydroxybenzoic acids group, 4 hydroxy-3methoxybenzoic acid and 3-hydroxy-4-methoxybenzoic acid compounds showed the highest contents, whilst cinnamic acid, 3-(4 -hydroxyphenyl)propanoic acid), and 4 -hydroxy-3methoxycinnamic acid hexoside were the main compounds of the hydroxycinnamic acids group. Concerning the phenylacetic acids family, 4 -hydroxy-3 -methoxyphenylacetic acid was the main compound, followed by 3,4-dihydroxyphenylacetic acid and 3 -hydroxy-4methoxyphenylacetic acid. Generally, the total content and profile of phenolic compounds described in the literature are very variable. This evidence could be attributed to different considerations such as pre-harvest (varieties, species, cropping areas, etc.) and post-harvest factors (ripening, preservation, and cooking, among others). Moreover, most of the studies in the literature were focused on estimating the total phenolic content of banana samples using spectrophotometry techniques, estimating the total content with respect to gallic acid (3,4,5-Trihydroxybenzoic acid) [7,20].
Regarding the impact of the heating treatments used in our study, the total content of phenolic compounds significantly decreased for all the elaborations evaluated, from 442.12 mg/100 g DW in raw banana to 338.60, 243.63, and 109.85 mg/100 g DW in the boiled (23.41%), roasted (44.90%), and baked (75.15%) samples, respectively (Table 1). The literature reports a high variability of the effects of cooking on the content of phenolic compounds in fruits and vegetables, sometimes being positive [15,21] and others not [6,21]. These results show that the different heating treatments tested significantly affect the content of phenolic compounds in bananas. Additionally, the decrease in phenolic compounds in boiled samples can be explained by an extraction process that takes place during boiling. In the literature, it is demonstrated that the previously mentioned impact increases when the banana is boiled without the peel [16]. Mean values with different letters in the same column present significant differences. Significant level: ns = not significant. * = p < 0.05. ** = p < 0.01. *** = p < 0.001. <LOD: less than limit of detection.
The content of flavan-3-ols significantly decreased for all preparations, being lower in the boiled samples (286.63 mg/100 g DW) and more pronounced in the baked ones (74.45 mg/100 g DW). This decrease was mainly related to the degradation of dimers and trimers of procyanidin (Table 1). Additional decreases were observed for (+)-catechin, (−)-epigallocatechin, or (+)-gallocatechin. Borges et al. [14] observed that (+)-catechin decreased in Pelipita and D'Angola banana varieties after boiling. It was observed that (−)-epicatechin significantly increased in boiled samples, while significantly decreased for the other treatments (roasted and baked bananas). Hence, these results could suggest that procyanidins could be degraded into (−)-epicatechin to some extent ( Figure 1). The formation of flavan-3-ols monomers from procyanidins has been confirmed previously in cooking research [22]. Therefore, the cooking temperature can be directly related to the chain cleavage of procyanidins [23].
Plants 2023, 12, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/plants epicatechin significantly increased in boiled samples, while significantly decreased for the other treatments (roasted and baked bananas). Hence, these results could suggest that procyanidins could be degraded into (−)-epicatechin to some extent ( Figure 1). The formation of flavan-3-ols monomers from procyanidins has been confirmed previously in cooking research [22]. Therefore, the cooking temperature can be directly related to the chain cleavage of procyanidins [23]. Total flavanones represented 0.31% of the total phenolics of fresh bananas. No significant changes were observed in the concentration of this family of compounds between the raw samples and the samples submitted to heat treatments, while a slight impact was observed among the boiling and roasting treatments and baking ones (Table 1). In addition, changes in individual flavanone content were observed in different elaborations, mainly due to changes in the concentration of naringenin and eriodyctiol hexosides. Hexoside isomer II increased significantly in the boiled and roasted samples, remaining stable in the baked samples. On the other hand, it was observed that eriodyctiol aglycone increased in the roasted and baked samples, probably due to the fragmentation of the monosaccharide from eriodyctiol hexoside isomers during the heating treatment ( Figure 2). Although a similar thermohydrolysis phenomenon seems to occur for the naringenin hexosides, such as naringenin-7-glucoside or naringenin hexoside I and III, naringenin aglycone were not detected after the heating processes. This could be due to additional cleavage reactions of the core compounds. The thermohydrolysis process is a chemical reaction widely described in the literature [6,15,19]. Total flavanones represented 0.31% of the total phenolics of fresh bananas. No significant changes were observed in the concentration of this family of compounds between the raw samples and the samples submitted to heat treatments, while a slight impact was observed among the boiling and roasting treatments and baking ones (Table 1). In addition, changes in individual flavanone content were observed in different elaborations, mainly due to changes in the concentration of naringenin and eriodyctiol hexosides. Hexoside isomer II increased significantly in the boiled and roasted samples, remaining stable in the baked samples. On the other hand, it was observed that eriodyctiol aglycone increased in the roasted and baked samples, probably due to the fragmentation of the monosaccharide from eriodyctiol hexoside isomers during the heating treatment ( Figure 2). Although a similar thermohydrolysis phenomenon seems to occur for the naringenin hexosides, such as naringenin-7-glucoside or naringenin hexoside I and III, naringenin aglycone were not detected after the heating processes. This could be due to additional cleavage reactions of the core compounds. The thermohydrolysis process is a chemical reaction widely described in the literature [6,15,19]. The total concentration of flavonols represented 3.24% of the pulp of the raw banana, and it was affected by all the heat treatments (Table 1), passing from 14.33 mg/100 g DW in the raw banana to 12.47 mg/100 g DW in the boiled samples (13%), to 6.67 mg/100 g DW in the roasted samples (53%) and, finally, to 2.90 mg/100 g DW in the baked samples (80%). Although most of the flavonols significantly decreased during the transformation treatments, this was most noticeable for quercetin dihexoside and myricetin rutinoside. Similar observations were highlighted by some authors regarding quercetin derivatives (rutin and quercetin deoxyhexose-hexoside) in boiled banana peels [14]. The myricetin rutinose content decreased from 6.32 mg/100 g DW to 5.54, 2.60, and 0.99 mg/100 g DW in the boiled, roasted, and baked samples, respectively. In contrast, Passo-Tsamo et al. [14] obtained different results for myricetin deoxyhexose-hexoside, where no significant differences were observed in boiled banana pulps and peels. Other flavonol hexosides, such as quercetin hexosides or kaempferol 3-rutinoside, showed a similar trend, decreasing their content after heat treatments (Table 1).
The total content of the hydroxybenzoic acids group represented 5.44% of the raw banana pulp and this group of compounds showed higher stability against the heating treatments tested. In our study, statistically significant differences only being observed The total concentration of flavonols represented 3.24% of the pulp of the raw banana, and it was affected by all the heat treatments (Table 1), passing from 14.33 mg/100 g DW in the raw banana to 12.47 mg/100 g DW in the boiled samples (13%), to 6.67 mg/100 g DW in the roasted samples (53%) and, finally, to 2.90 mg/100 g DW in the baked samples (80%). Although most of the flavonols significantly decreased during the transformation treatments, this was most noticeable for quercetin dihexoside and myricetin rutinoside. Similar observations were highlighted by some authors regarding quercetin derivatives (rutin and quercetin deoxyhexose-hexoside) in boiled banana peels [14]. The myricetin rutinose content decreased from 6.32 mg/100 g DW to 5.54, 2.60, and 0.99 mg/100 g DW in the boiled, roasted, and baked samples, respectively. In contrast, Passo-Tsamo et al. [14] obtained different results for myricetin deoxyhexose-hexoside, where no significant differences were observed in boiled banana pulps and peels. Other flavonol hexosides, such as quercetin hexosides or kaempferol 3-rutinoside, showed a similar trend, decreasing their content after heat treatments (Table 1).
The total content of the hydroxybenzoic acids group represented 5.44% of the raw banana pulp and this group of compounds showed higher stability against the heating treatments tested. In our study, statistically significant differences only being observed between the roasted and baked samples (24.97 vs. 20.78 mg/100 g DW, respectively) ( Table 1). In this context, 4-hydroxy-3-methoxybenzoic and 3-hydroxy-4-methoxybenzoic acids were the main compounds of the group, representing 70 and 24.2% of the raw banana pulp, respectively. 4-hydroxy-3-methoxybenzoic acid maintained a stable concentration during all the heat treatments. On the contrary, the 3-hydroxy-4-methoxybenzoic acid content was heavily affected by the treatments, with an important decrease for the baked samples and, to a lesser degree, in the roasted samples. On the other hand, it was observed that 3,4,5-Trihydroxybenzoic acid, 3,4,5-Trihydroxybenzoic acid hexoside, and 4-hydroxybenzoic acid increased during the preparations, especially in the roasted and baked samples. The increase of 3,4,5-Trihydroxybenzoic acid and hydroxybenzoic acids after cooking elaborations has been observed in previous studies [16].
Differences in the effects of heating treatments were observed in the total content of hydroxycinnamic acids (representing 1.69% of the fresh banana pulp). The concentration of this group increased in the boiled (9.19 mg/100 g DW) and roasted (9.09 mg/100 g DW) samples, while the baked (7.40 mg/100 g DW) samples remained unchanged (Table 1). Considering the individual profile, the compounds of this group had a different evolution depending on the treatment. The boiled samples showed a significant increase in 3 ,4dihydroxycinnamic acid, caffeoyl hexoside isomer I, and 4 -hydroxy-3 -methoxycinnamic acid hexoside. Similarly, cinnamic acid, 4 -hydroxy-3 -methoxycinnamic acid, 3 -hydroxy-4 -methoxycinnamic acid, and 4 -hydroxy-3 -methoxycinnamic acid hexoside increased their concentrations in the roasted samples. Differences in profile were also observed between the fresh and baked bananas, although the total content did not vary. In this case, cinnamic and 3 -hydroxy-4 -methoxycinnamic acid compounds increased in the baked samples, whilst caffeoyl hexoside isomers, 4 -hydroxy-3 -methoxycinnamic acid conjugate, and syringin were higher in the raw banana. In the literature, some authors have reported changes in the content of hydroxycinnamic acids in banana pulp after cooking, mainly 4 -Hydroxy-3 -methoxycinnamic acid hexoside in Moto Ebanga cultivar [14]. They observed an increase in 4 -Hydroxy-3 -methoxycinnamic acid from the cleavage of their hexoside, similar to that observed in roasted samples (Table 1). Additionally, the significant increase in some hydroxycinnamic acids, such as cinnamic acids in roasted and boiled samples or 3 ,4 -dihydroxycinnamic acid in boiled samples, may be due to the release of bound phenolics [24].

Multivariate Analyses: Principal Component Analysis (PCA)
A multivariate analysis, principal component analysis (PCA), was carried out to evaluate the effect of the heating treatments on the concentration and profile of phenolic compounds and antioxidant activities in raw bananas and three elaborations. We sampled the raw banana before the heating treatments, and these samples were analyzed by duplicate. Every heating treatment was performed by triplicate and each sample obtained was analyzed by duplicate (for a total of six data per treatment). Figure 3 shows the corresponding scores (A) and loadings (B) plotted using the first two principal components, explaining 76.76% of the cumulative variance. PC1 (54.41% of the data variability) shows that the sam-ples were separated according to a higher content of phenolic compounds. Thus, the raw banana samples were placed on the right side of the scores plot, highly related to most of the phenolic compounds found, followed by boiled and roasted banana samples ( Figure 3B). Most of the flavan-3-ols (mainly dimers and trimers) and flavonols were highly related to the raw samples. These flavonoid families were sensible to heat treatments, decreasing in the three types of cooking. Boiling and roasting had a lower impact on the phenolic content than baking. The antioxidant activity methods (ABTS, DPPH, and ORAC) were placed on the right side, and highly correlated with most phenolic compounds. Additionally, the compounds more related to baking (middle-left part) were 3,4,5-Trihydroxybenzoic acid, 3,5-dihydroxy-4-methoxybenzoic acid, 3,4,5-Trihydroxybenzoic acid hexoside, and 4 hydroxybenzoic acid, belonging to the family of hydroxybenzoic acids, as well as eriodyctiol, which increased significantly after the heating treatment. On the other hand, PC2 tends to separate the boiled samples (bottom right part) from the rest ( Figure 3A). The boiled samples were highly related to dihydroxybenzoic acid isomers, 3 ,4 -dihydroxycinnamic acid, caffeoyl hexoside isomer I, quercetin trihexoside isomers, and (−)-epicatechin. These compounds significantly increased in the banana samples after boiling. The fresh and roasted samples are in the positive area of PC2 (upper middle-right part), correlated with 3-galloyl-gallic acid, 4 -hydroxy-3 -methoxycinnamic acid conjugate, eriodyctiol hexoside isomer III, quercetin 3-rutinoside-7-rhamnoside, kaempferol 3-rutinoside-7-rhamnoside, and eriodyctiol rutinoside. These compounds were quantified in the raw samples and most remained stable after roasting. Regarding antioxidant activity, ORAC was tending toward the center line in PC2, since it does not present significant differences between raw, baked, and roasted samples. However, ABTS and DPPH were placed in the upper middle-right part and were correlated with raw and roasted samples. In general, these results indicate that the impact on the composition of some phenolic compounds differed for each heating treatment, allowing the different groups on the PCA to be separated.

Antioxidant Activity
The antioxidant activity of the raw, boiled, roasted, and baked banana samples was evaluated using several assays, such as ABTS, DPPH, and ORAC ( Figure 4). Regarding the ABTS test, the values from samples of fresh and roasted samples did not show differences, whilst the boiled banana samples were significantly lower. In particular, the antioxidant activity of the baked samples drastically decreased in ABTS assays. Similar results were observed in the DPPH test, where the antioxidant activity of the roasted samples remained stable, while that of the boiled and baked samples declined. Results from the ORAC method were slightly different since the antioxidant activity of the baked samples significantly decreased, but no differences were observed between the fresh, boiled, and roasted samples. These differences in the results of antioxidant activity methods are due to different chemical principles, namely ABTS and DPPH assays which are based on the antioxidant activity of sample extracts to scavenge the free-radical cation, while the ORAC method evaluates peroxyl-radical scavenging [17,[25][26][27]. In general, all the methods found that the antioxidant activity of the baked samples was highly affected. These results could be explained due to this product presenting a significant drop in phenolic compounds (from 442.12 to 109.85 mg/100 g DW), which play an important role in the antioxidant activity of these products. In general, thermal treatment can cause the degradation of phenolic compounds and other bioactive compounds, such as organic acids, vitamins, etc., affecting antioxidant activity [15,28].     oside_I; Caf_H_II: Caffeoyl-hexoside_II; Sin_H_I: Sinapic_acid_hexoside_I; Sin_H_II: Sin-apic_acid_hexoside_II; Fer: 4′-Hydroxy-3′-methoxycinnamic acid; Isofer: 3′-hydroxy-4′-methoxycinnamic acid; Cin: Cinnamic_acid; Dhcaf: 3-(3′,4′-dihydroxyphenyl)propanoic acid; 3_HPP: 3-(4′-Hydroxyphenyl)propanoic_acid; Syringin: Syringin; PA: Phenylacetic_acid; 3_HPA: 3-Hydroxy-phenylacetic_acid; HV: 4′-hydroxy-3′-methoxyphenylacetic acid; IHV: 3′-hydroxy-4′-methoxyphenylacetic acid; 3,4_DHPA: 3,4-Dihydroxyphenylacetic acid.

Antioxidant Activity
The antioxidant activity of the raw, boiled, roasted, and baked banana samples was evaluated using several assays, such as ABTS, DPPH, and ORAC ( Figure 4). Regarding the ABTS test, the values from samples of fresh and roasted samples did not show differences, whilst the boiled banana samples were significantly lower. In particular, the antioxidant activity of the baked samples drastically decreased in ABTS assays. Similar results were observed in the DPPH test, where the antioxidant activity of the roasted samples remained stable, while that of the boiled and baked samples declined. Results from the ORAC method were slightly different since the antioxidant activity of the baked samples significantly decreased, but no differences were observed between the fresh, boiled, and roasted samples. These differences in the results of antioxidant activity methods are due to different chemical principles, namely ABTS and DPPH assays which are based on the antioxidant activity of sample extracts to scavenge the free-radical cation, while the ORAC method evaluates peroxyl-radical scavenging [17,[25][26][27]. In general, all the methods found that the antioxidant activity of the baked samples was highly affected. These results could be explained due to this product presenting a significant drop in phenolic compounds (from 442.12 to 109.85 mg/100 g DW), which play an important role in the antioxidant activity of these products. In general, thermal treatment can cause the degradation of phenolic compounds and other bioactive compounds, such as organic acids, vitamins, etc., affecting antioxidant activity [15,28].

Materials and Samples Preparation
Red Dacca banana (Musa sp. AAA) samples were provided from a field of bananas in "el Cantón la Maná" (Cotopaxi, Ecuador) planted in 2019. Banana bunches were harvested in February 2022, and the harvesting criterion was selecting fruits with a diameter of 39-46 mm around 8 weeks after the inflorescence. The samples complied with all the postharvest quality controls, and they were packed in cardboard boxes of approximately 6.9 kg. All samples were stored at room temperature (25 ± 2 • C) and 70 ± 5% relative humidity until processing. The samples were stored for 15 days from the harvest until the elaborations. Additionally, the maturity of the bananas (unripe bananas) was the same for the raw samples and subsequent heat treatments.
In order to evaluate the impact of the heating treatments on the bioactive compounds, the banana samples were cooked with different preparation methods, including boiling, roasting, and baking ( Figure 5). All the analyses were conducted on peeled banana samples.

Extraction of Phenolic Compounds
The extraction of polyphenols from banana samples was adapted from Ordóñez-Díaz et al. [27] with some modifications. Briefly, 200 mg of the sample was mixed with 1 mL of a methanol/acidified water mixture (80:20, v/v) with 1% formic acid. The samples were sonicated using an ultrasonic bath (Selecta, Barcelona, Spain) for 10 min and centrifuged at 15,000 rpm for 15 min at 4 °C using a microcentrifuge (Eppendorf, Hamburg, Germany), and finally, supernatants were collected. The pellet was re-extracted with 1 mL of the same solvent as described above. All the supernatants were pooled and made up to a final volume of 2 mL and stored at −80 °C until analysis.

Antioxidant Activity Analysis
The antioxidant activity of the banana samples was analyzed using three different assays: ABTS (2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt), DPPH (2,2-diphenyl-1-picrylhydrazyl), and ORAC (Oxygen Radical Absorbance Capacity) methods. The antioxidant activity analyses were performed using a Synergy HTX Multi-Mode Microplate Reader (Biotek Instruments, Winooski, VT, USA). Boiled preparations were performed by boiling the bananas in 500 mL of drinking water for 15 min in a covered stainless-steel pot heated on a laboratory hotplate RH digital KT/C (IKA ® , Staufen, Germany). Previously, the temperature was controlled to reach 100 • C before introducing the samples into the pot. For the roasted elaboration, a covered grill was used to cook the banana samples for 20 min at 120 • C. The samples were turned for a homogeneous elaboration. For the baking process, the banana samples were sliced and dried in an oven (Selecta Digitheat 2001243, Barcelona, Spain) for 2 h at 100 • C, turning the samples for a homogeneous elaboration.
All the elaborations were performed by triplicate, and each repetition was formed by two peeled bananas. The final samples were cut into small pieces, lyophilized in a freeze dryer ECO EVO (Tred Technology S.R.L., Ripalimosani, Italy), ground, and stored at −80 • C until analysis.

Extraction of Phenolic Compounds
The extraction of polyphenols from banana samples was adapted from Ordóñez-Díaz et al. [27] with some modifications. Briefly, 200 mg of the sample was mixed with 1 mL of a methanol/acidified water mixture (80:20, v/v) with 1% formic acid. The samples were sonicated using an ultrasonic bath (Selecta, Barcelona, Spain) for 10 min and centrifuged at 15,000 rpm for 15 min at 4 • C using a microcentrifuge (Eppendorf, Hamburg, Germany), and finally, supernatants were collected. The pellet was re-extracted with 1 mL of the same solvent as described above. All the supernatants were pooled and made up to a final volume of 2 mL and stored at −80 • C until analysis.

ABTS Method
Free radical ABTS scavenging activity was assessed in the phenolic extract following the methods previously described by Ordóñez-Díaz et al. [29]. The antioxidant activity results were expressed as mmol of Trolox equivalents per 100 g of dry weight (mmol TE/100 g DW). The ABTS analyses were performed in triplicate.

DPPH Method
The free radical scavenging activity was measured using the DPPH method adapted for the microplate reader [29]. The antioxidant activity results were expressed as mmol of Trolox equivalents per 100 g of dry weight (mmol TE/100 g DW). The DPPH analyses were performed in triplicate.

ORAC Method
The oxygen radical scavenging activity in the phenolic extractions was analyzed by the ORAC assay according to the method previously published by Madrona et al. [30] and modified by Pereira-Caro et al. [31]. Briefly, 25 µL of banana samples or standard solutions (Trolox) are mixed with 150 µL of fluorescein solution (8.5 × 10 −5 mM) prepared in 75 mM of phosphate buffer (pH 7.4) and 30 µL of AAPH (153 mM) as a peroxyl radical generator. The fluorescence was measured every 90 s for 90 min at 485 and 528 nm excitation and emission wavelengths, respectively. The ORAC values were expressed as mmol Trolox equivalents per gram of dry sample (mmol TE/g DW). The ORAC analyses were performed in triplicate.

Phenolic Compound Analysis
The identification and quantification of the phenolic compounds in the fresh bananas and their products were performed by using a UHPLC-HRMS mass spectrometer system (Thermo Scientific, San José, CA, USA) comprised of a UHPLC pump and an autosampler operating at 10 • C (ThermoFisher Scientific, San Jose, CA, USA). The column was a Zorbax SB-C18 RRHD column (100 × 2.1 mm i.d., 1.8 µm (Agilent, Santa Clara, CA, USA)), and a guard pre-column of the same stationary phase was used. The injection volume was 5 µL and the separation was obtained at a flow rate of 0.2 mL/min at 40 • C with the gradient program employed by Ordóñez-Díaz et al. [27].
The column eluate went directly to an Exactive Orbitrap mass spectrometer (Thermo Scientific, San José, CA, USA) fitted with a heated electrospray ionization probe (HESI) operating in negative ionization mode for the determination of phenolic compounds. Full scans were recorded in m/z range from 100 to 1200 with a resolution of 50,000 Hz and with a full AGC target of 100,000 charges, using 2 microscans. Analyses were also based