Antioxidant Potential and Polyphenolic Composition of Acorn Flour from Different Mediterranean Oaks (Quercus spp.): A Comparative Study

Abstract

Acorn flours from the six Mediterranean Quercus species (Quercus cerris, Quercus petraea, Quercus robur, Quercus ilex, Quercus pubescens, and Quercus rotundifolia) were systematically fractionated for polyphenols using ultrasonic-assisted extraction with seven solvent systems varying in polarity and composition (water at 20 and 40 min; acetone, ethanol, and methanol at 20% and/or 70% v/v). The total polyphenol content (TPC), non-tannic phenolic content (NTPC), tannin content (TC), antioxidant potential (DPPH, ABTS, and FRAP), and individual phenolic profiles through high-performance liquid chromatography (HPLC) were determined. The results showed that the botanical species primarily determined the TPC and TC, while the solvent composition significantly influenced the NTPC yield. Q. cerris yielded the highest average TPC (105.1 ± 3.8 mg GAE/g) and TC, supported by a gallotannin-dominated profile. Conversely, Q. rotundifolia exhibited the lowest values but the highest NTPC/TPC ratio (32.0%). Q. ilex featured species-exclusive ellagitannins, while Q. pubescens showed the highest specific antioxidant activity. For the targeted recovery, 20% acetone is recommended for tannins and 70% ethanol for the non-tannic fractions. These findings establish a species-resolved framework for valorizing acorn flours as functional ingredients, identifying high-tannin species requiring detannification and “sweet” varieties suitable for direct food application.

1. Introduction

Acorns, the edible fruits of various species of the genus Quercus (Fagaceae), which are closely related to chestnuts, have a long history of human consumption across the Mediterranean, the Middle East, and Central Asia, where they were traditionally used as staple or emergency foods during periods of cereal scarcity [1,2,3]. Despite this extensive ethnobotanical and culinary heritage, acorns have been progressively marginalized from the human diet over the last century and are currently exploited mainly as animal feed or left underutilized in forest ecosystems [4,5]. In recent years, however, the growing interest in food security, sustainability, and non-cereal raw materials has renewed scientific and technological attention toward acorns as underutilized resources with potential applications in human nutrition, particularly in gluten-free formulations [1,6,7].

Acorn flour has been increasingly explored as an ingredient in bread, cakes, pasta, and cookies, owing to its gluten-free nature, high dietary fiber content, favorable lipid profile rich in unsaturated fatty acids, as well as the presence of phenolic compounds with antioxidant potential [3,4,8]. Nevertheless, previous studies also suggest that the technological performance and sensory acceptance of acorn-enriched products strongly depend on the level of incorporation, processing conditions, and product type, highlighting the need for formulation-specific optimization strategies [1,5,7].

Acorn flour is characterized by its high carbohydrate content, predominantly starch (approximately 50–60%), together with a significant proportion of lipids, dietary fiber, minerals, and bioactive compounds, which together define its nutritional and functional potential. Several studies report that acorn flour contains a considerable amount of unsaturated fatty acids, mainly oleic and linoleic acids, as well as minerals such as potassium, calcium, iron, copper, and manganese, contributing to its nutritional value and positioning it as a promising gluten-free ingredient [4,9,10]. In addition, acorn flour exhibits a notable phytochemical profile, with polyphenols representing one of its most significant bioactive fractions. Identified phenolic compounds include flavonoids and phenolic acids such as catechin, rutin, gallic acid, ellagic acid, and syringic acid, which have been associated with antioxidant capacity and potential health-promoting effects [4,10,11].

Among these compounds, tannins—water-soluble polymeric phenols—deserve particular attention due to their dual role in food systems. At moderate levels, tannins contribute to antioxidant, antimicrobial, and anti-inflammatory activities through their free radical scavenging capacity and interactions with biological macromolecules [12,13]. However, excessive tannin intake has been associated with reduced nutrient bioavailability and potential adverse effects, including hepatotoxicity or DNA damage, emphasizing the importance of controlled incorporation and appropriate processing strategies when using acorn-based ingredients in food formulations [14,15].

Previous studies have shown that acorns are a rich source of polyphenols with significant antioxidant potential; however, their experimental scope has generally been limited to single oak species and specific extraction parameters [16,17]. Recent research (2023–2026) has increasingly examined forest by-products and underutilized resources, such as acorn shells and flour, using modern green techniques like ultrasound-assisted extraction (UAE) to optimize the recovery of bioactives [18,19]. Despite these advancements, a significant gap remains: existing studies often focus on total phenolic content or a few individual compounds, frequently failing to distinguish between tannic and non-tannic phenolic fractions [11,20]. This distinction is critical because tannins represent a dominant and functionally distinct class of acorn polyphenols with specific nutritional and sensory impacts [13,21]. Furthermore, comparative evaluations of multiple Quercus species under standardized extraction conditions are scarce, making it difficult to assess true species-dependent differences in phytochemical recovery [3,10,16,17,22,23].

In this study, six Mediterranean oak species (Q. ilex, Q. pubescens, Q. cerris, Q. petraea, Q. robur, and Q. rotundifolia) were investigated using acorn flours and ultrasonic-assisted extraction under controlled and comparable conditions. The solvent systems, including water (20 and 40 min), acetone (20% and 70% v/v), ethanol (20% and 70% v/v), and methanol (20% v/v) were employed to evaluate their effects on the extraction of total polyphenols, individual phenolic compounds, and tannins. The study specifically tested the hypotheses that (i) polyphenol and tannin extraction efficiency differs significantly among oak species; (ii) solvent type, concentration, and extraction time exert a strong influence on phenolic recovery; and (iii) the choice of solvent under standardized UAE conditions can selectively modulate the recovery of tannin versus non-tannic phenolic fractions. By addressing these gaps, this work aims to support the targeted valorization of acorn flours as antioxidant-rich functional ingredients and to identify oak species with the highest potential for food and nutraceutical applications.

2. Materials and Methods

2.1. Plant Materials

Mature, healthy acorns from six Quercus species were harvested from distinct bioclimatic regions in Croatia and Portugal during peak physiological maturity (October–December). For each location, three independent biological batches (1 kg each) were collected, with each batch consisting of pooled acorns from three different individual trees. The specific collection sites and their respective geolocations were as follows: Quercus cerris (Borovik, Đakovo, Croatia; GPS: 45°13′05″ N, 18°20′13″ E), Quercus petraea (Mandičevac, Đakovo, Croatia; GPS: 45°18′40″ N, 18°14′20″ E), Quercus robur (Slavir, Otok, Croatia; GPS: 45°08′52″ N, 18°53′28″ E), Quercus pubescens (Brzac, Island of Krk, Croatia; GPS: 45°05′20″ N, 14°26′21″ E), Quercus ilex (Glavotok, Island of Krk, Croatia; GPS: 45°05′31″ N, 14°26′28″ E), and Quercus rotundifolia (Portugal: Bragança, 41°48′26″ N, 6°45′32″ W; Portalegre, 39°17′38″ N, 7°25′52″ W). After collection, acorns were visually inspected, and only mature, intact fruits were retained for flour production.

2.2. Chemicals and Reagents

High-purity analytical standards, including Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid, 97%), gallic acid (3,4,5-trihydroxybenzoic acid, 97.5–102.5%), and ellagic acid (4,4′,5,5′,6,6′-hexahydroxydiphenic acid 2,6,2′,6′-dilactone), were purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). Reagents for spectrophotometric assays, including Folin–Ciocalteu reagent, 2,2-diphenyl-1-picrylhydrazyl (DPPH), and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS), were also obtained from the same supplier. Reagents for the antioxidant capacity assays, specifically 2,4,6-tris(2-pyridyl)-s-triazine (TPTZ) and potassium persulfate, were sourced from Thermo Fisher Scientific Inc. (Waltham, MA, USA). Polyvinylpolypyrrolidone (PVPP), used for the selective removal of tannins, was obtained from Enologica Vason S.p.A. (Verona, Italy). HPLC-grade acetonitrile and formic acid were purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). All other solvents (methanol, ethanol, and acetone) and reagents (hydrochloric acid, ferric chloride hexahydrate, sodium carbonate, and glacial acetic acid) were of analytical grade and procured from Gram-Mol d.o.o. (Zagreb, Croatia), T.T.T. d.o.o. (Sveta Nedelja, Croatia), or LabExpert d.o.o. (Zagreb, Croatia). Demineralized water was used throughout all extraction and analytical procedures.

2.3. Methods

2.3.1. Preparation of Acorn Flour

After collection, the acorn material was submerged in water to remove impurities and discard damaged or rotten fruits. The clean acorns were then dried in a laboratory oven (Memmert GmbH + Co. KG, B40, Schwabach, Germany) at 40 °C for 24 h. Following the initial drying, the acorn pericarps were scored using an electric chestnut cutter (Xeoleo, Shenzhen Xeoleo Technology Co., Ltd., Guangdong, China) and dried again at 40 °C for another 24 h. The outer shells were then manually removed with knives, and the endosperms were subjected to an additional drying cycle at 40 °C for 24 h. In order to remove the remains of the test, the dried endosperms were sieved using a sieve with 1 mm openings on a vibratory sieve shaker (Fritsch GmbH—Mahlen und Messen, Analysette 3 PRO, Idar-Oberstein, Germany) set to an amplitude of 2.5 mm for 4 min. The material was subsequently milled using an ultracentrifugal laboratory mill (ZM 200, Retsch GmbH, Haan, Germany) and sieve with 250 µm openings. The prepared acorn flour was stored in a refrigerator until further analysis.

2.3.2. Ultrasonic-Assisted Extraction of Polyphenols

Ultrasound-assisted extraction (UAE) of polyphenols was performed using seven solvent systems to assess the effects of solvent type, organic solvent concentration, and extraction time on phenolic recovery. For each system, 10 mg of acorn flour were accurately weighed into a 15 mL centrifuge tube using an analytical balance (Mettler-Toledo International Inc., AB204-S, Greifensee, Switzerland), and 10 mL of the respective extraction solvent were added. The solvent systems used were: (1) demineralized water, 20 min (Water 20′); (2) demineralized water, 40 min (Water 40′); (3) aqueous acetone, 20% v/v, 20 min (Ace 20%, 20′); (4) aqueous acetone, 70% v/v, 20 min (Ace 70%, 20′); (5) aqueous ethanol, 20% v/v, 20 min (EtOH 20%, 20′); (6) aqueous ethanol, 70% v/v, 20 min (EtOH 70%, 20′); and (7) aqueous methanol, 20% v/v, 20 min (MeOH 20%, 20′). Preliminary kinetic experiments were conducted to evaluate the effect of extraction time (10–60 min) on total phenolic content. For aqueous extraction, the content of extracted compounds increased up to 40 min, after which no further increase was observed, and in some cases a slight decrease occurred at longer extraction times. In contrast, all other solvent systems reached maximum values at 20 min. Based on these results, 20 min was selected as the standard extraction time for all solvent systems to ensure consistency and comparability, while aqueous extraction was also performed at 40 min.

Organic solvent concentrations of 20% and 70% (v/v) were selected to assess phenolic recovery across a broad range of dielectric constants. The 20% concentration provides a high-dielectric environment that favors the recovery of hydrophilic hydrolysable tannins through efficient mass transfer. In contrast, the 70% concentration significantly reduces the dielectric constant, allowing investigation of solvent selectivity for non-tannic fractions and the disruption of specific phenolic–matrix interactions that are less accessible in purely aqueous systems.

Binary solvent mixtures were prepared volumetrically by combining the appropriate volume of organic solvent with demineralized water immediately before use. After solvent addition, all samples were briefly homogenized on a vortex mixer (Domel d.o.o., Tehtnica Vibromix 10, Železniki, Slovenia) for 30 s and then subjected to UAE in an ultrasonic water bath (Vevor, Digital Pro+, Ningbo, Zhejiang, China) operating at 40 kHz. UAE was performed at 30 °C to ensure controlled extraction conditions, minimize excessive heating from cavitation, and preserve the stability of phenolic compounds, which are susceptible to thermal degradation at elevated temperatures [24,25]. After extraction, the samples were centrifuged (Heraeus Holding GmbH, Multifuge 3L-R, Hanau, Germany) at 4000× g for 10 min at room temperature. The polyphenol-containing supernatants were collected and used immediately for further analyses.

2.3.3. Selective Precipitation of Tannins

Tannic phenolic compounds were removed from polyphenol extracts by selective adsorption onto polyvinylpolypyrrolidone (PVPP), which binds polymeric phenolics (tannins) through hydrogen bonding while leaving low-molecular-weight, non-tannic phenolic compounds in solution [26,27]. An aliquot of 5 mL of each polyphenol-containing supernatant (obtained in Section 2.3.2) was transferred to a 10 mL centrifuge tube containing 50 mg of PVPP. The suspension was allowed to stand for 15 min at room temperature with manual shaking every 5 min to ensure complete tannin adsorption. The samples were then centrifuged at 4000× g for 10 min. The resulting non-tannic polyphenolic supernatants were collected and used immediately for further analyses.

2.3.4. Determination of Total Polyphenol Content (TPC), Non-Tannic Phenolic Content (NTPC) and Tannin Content (TC)

Total polyphenol content (TPC) and the non-tannic phenolic content (NTPC) (polyphenol content of PVPP-treated extracts) were determined spectrophotometrically using the Folin–Ciocalteu (FC) method [28]. Aliquots of 100 µL of each extract were combined with 100 µL of FC reagent and allowed to react for 5 min at room temperature in the dark. Subsequently, 900 µL of 7.5% (w/v) sodium carbonate solution and 900 µL of demineralized water were added. After thorough mixing, the samples were incubated for 30 min in the dark, and the absorbance was measured at 765 nm against a reagent blank (solvent substituted for extract) using a Shimadzu UV-1280 spectrophotometer (Kyoto, Japan). Total polyphenol content was expressed as milligrams of gallic acid equivalents per gram of dry flour (mg GAE/g dw), based on a gallic acid calibration curve ranging from 0 to 150 mg/L (y = 167.36x − 10.947; R² = 0.998). The same procedure and expression were applied to total polyphenol and non-tannic polyphenol extracts.

Tannin content (TC) was calculated as the difference between the total polyphenol content (TPC) of the original extract and the non-tannic polyphenol content (NTPC) of PVPP-treated extracts:

$$\mathrm{T}\mathrm{C}\left(\mathrm{m}\mathrm{g}\mathrm{G}\mathrm{A}\mathrm{E}/\mathrm{g}\mathrm{d}\mathrm{w}\right)=\mathrm{T}\mathrm{P}\mathrm{C}\left(\mathrm{m}\mathrm{g}\mathrm{G}\mathrm{A}\mathrm{E}/\mathrm{g}\mathrm{d}\mathrm{w}\right)-\mathrm{N}\mathrm{T}\mathrm{P}\mathrm{C}\left(\mathrm{m}\mathrm{g}\mathrm{G}\mathrm{A}\mathrm{E}/\mathrm{g}\mathrm{d}\mathrm{w}\right),$$

(1)

2.3.5. Determination of Antioxidant Activity

Antioxidant activity was assessed using three mechanistically complementary assays: DPPH radical scavenging (hydrogen atom transfer, HAT), ABTS radical cation decolorization (combined HAT/electron transfer, ET), and ferric reducing antioxidant power (FRAP; ET mechanism). All assays were performed on both original polyphenol (TPC) and non-tannic polyphenolic (NTPC) extracts, enabling fractionated assessment of the individual contributions of total phenolics, and non-tannic compounds to antioxidant capacity, as well as the contribution of tannins, which was calculated from the difference in antioxidant activity between TPC and NTPC.

DPPH Assay

The DPPH radical scavenging activity was determined according to the method of Brand-Williams et al. (1995) [29], with minor modifications. A DPPH working solution was prepared by dissolving DPPH in methanol to a concentration of 0.1 mM, resulting in an initial absorbance of 1.000 ± 0.020 at 517 nm. Aliquots of 100 µL of each extract were mixed with 3.9 mL of DPPH solution and incubated in the dark for 30 min at room temperature. Absorbance was then measured at 517 nm against a methanol blank using a spectrophotometer. The percentage inhibition of the DPPH radical was calculated as:

$$\mathrm{\%}DPPHinhibition=\left[A_0-A_{sample}/A_0\right]\times 100,$$

(2)

where A₀ is the absorbance of the control (DPPH solution without extract) and A_sample is the absorbance of the test sample. Antioxidant activity was expressed as µmol Trolox equivalents per gram dry flour (µmol TE/g sample dw), using a Trolox calibration curve (0–1500 µM; y = 16.669x − 5.656; R² = 0.996). A calibration curve was constructed by plotting Trolox concentration against FRAP response, enabling direct conversion of absorbance values into Trolox equivalents. Here, x represents the percentage of DPPH inhibition, while y represents the Trolox equivalent concentration (µM).

ABTS Assay

The ABTS^•+ radical cation decolorization assay was performed according to Re et al. (1999) [30], with minor modifications. The ABTS^•+ radical cation was generated by reacting a 7 mM aqueous ABTS solution with aqueous 2.45 mM potassium persulfate (K₂S₂O₈) solution in the dark for 12–16 h at room temperature. Before use, the radical solution was diluted with demineralized water (1:59) to an absorbance of 0.700 ± 0.020 at 734 nm. Aliquots of 100 µL of each extract were mixed with 3.9 mL of the ABTS^•+ working solution and incubated in the dark for 6 min at room temperature. Absorbance was then measured at 734 nm against a reagent blank. The percentage inhibition was calculated as in the DPPH assay. Results were expressed as µmol TE/g sample dw using a Trolox calibration curve (0–1500 µM; y = 11.897x − 0.950; R² = 0.996). A calibration curve was constructed by plotting Trolox concentration against FRAP response, enabling direct conversion of absorbance values into Trolox equivalents. Here, x represents the percentage of ABTS inhibition, while y represents the Trolox equivalent concentration (µM).

FRAP Assay

The ferric reducing antioxidant power (FRAP) was determined according to Benzie & Strain (1996) [31], with minor modifications. The FRAP reagent was freshly prepared daily by mixing 300 mM acetate buffer (pH 3.6), 10 mM TPTZ acetic solution, and 20 mM FeCl₃·6H₂O in a 10:1:1 ratio (v/v/v), then equilibrated to 37 °C before use. Aliquots of 100 µL of each extract were added to 3.0 mL of FRAP reagent and incubated for 10 min at 37 °C in a water bath (Grant Instruments, JB Nova, Cambridge, UK). Absorbance was measured at 594 nm against a reagent blank. The FRAP value was calculated as:

$$FRAPvalue=A_{sample}-A_0,$$

(3)

where A₀ is the absorbance of the FRAP reagent without extract. Results were expressed as µmol TE/g sample dw using a Trolox calibration curve (0–1500 µM; y = 953.17x − 4.693; R² = 0.992). A calibration curve was constructed by plotting Trolox concentration against FRAP responsmine, enabling direct conversion of absorbance values into Trolox equivalents. Here, x represents the FRAP response (absorbance), while y represents the Trolox equivalent concentration (µM).

2.3.6. HPLC-DAD Analysis of Individual Phenolic Compounds

Prior to chromatographic analysis, plant material was subjected to hydroethanolic maceration (EtOH/H₂O, 60:40, v/v) using a 25 g/L biomass-to-solvent ratio. Maceration was performed at room temperature with continuous agitation for 60 min and re-peated once to ensure thorough recovery of target compounds. Chromatographic profiling was conducted on a Dionex UltiMate™ 3000 HPLC system (Thermo Fisher Scientific, San Jose, CA, USA) coupled to a diode array detector (DAD) and interfaced in series with an Orbitrap Exploris™ 120 high-resolution mass spectrometer (Thermo Fisher Scientific, San Jose, CA, USA). Analyte separation was achieved on a Spherisorb S3 ODS-2 C18 reversed-phase column (3 μm particle size, 4.6 mm × 150 mm; Waters, Milford, CT, USA) thermostated at 35 °C. The binary mobile phase consisted of 0.1% (v/v) formic acid in ultrapure water (phase A) and pure acetonitrile (phase B), delivered under the following gradient program: 85:15 (A:B) held for 5 min, linearly adjusted to 80:20 over 5 min, then to 75:25 over 10 min, 65:35 over 10 min, and 50:50 over 10 min, before returning to initial conditions over 10 min, with a final 10 min re-equilibration step. Chromatographic separation was run at a constant flow of 0.5 mL/min; sample volumes of 10 μL were injected per run. DAD acquisition covered the 180–700 nm wavelength range; extracted chromatograms at 280, 330, and 370 nm were used for quantitative profiling of distinct phenolic subclasses. Ionization was performed in negative electrospray ionization mode (ESI⁻) using an OptaMax NG ion source (Thermo Fisher Scientific, San Jose, CA, USA), with a spray voltage of 2.5 kV, an ion transfer tube temperature of 325 °C, and a vaporizer temperature of 350 °C. Nitrogen was used as the sheath (50 arb), auxiliary (10 arb), and sweep (1 arb) gas. Full-scan MS spectra were recorded across an m/z window of 110–1800 at a resolving power of 15,000, with the RF lens set to 70%. Data-dependent fragmentation (ddMS2) of the four most abundant precursor ions was triggered automatically, using stepped HCD with a base normalized collision energy of 30% (step values: 30, 50, and 150%). Redundant precursor selection was minimized via a dynamic exclusion window. All data acquisition and processing were performed using Xcalibur™ software version 4.3 (Thermo Fisher Scientific, San Jose, CA, USA). Tentative compound identification relied on a combination of chromatographic retention behavior on the C18 stationary phase, UV-Vis ab-sorption profiles, accurate precursor masses ([M–H]⁻), and diagnostic MS/MS fragmentation patterns, cross-referenced against authenticated reference standards, peer-reviewed literature, and spectral databases (NIST™, MZ Vault™, and MZCloud™) accessible through Freestyle™ 1.7 software (Thermo Fisher Scientific, San Jose, CA, USA). Quantification of identified compounds was performed against external calibration curves of reference standards: gallic acid (y = 23,857x – 13,524, R² = 0.9979, LOD = 0.41 μg/mL, LOQ = 1.05 μg/mL) served as the standard for compounds 1–13, 15–17, 19–21, 23, and 25–30, while ellagic acid (y = 14,803x – 151,526, R² = 0.9956, LOD = 0.48 μg/mL, LOQ = 1.30 μg/mL) was applied to compounds 14, 18, 22, and 24. Both calibration curves demonstrated adequate linearity across the tested concentration ranges (R² > 0.99). Since authentic standards are not commercially available for most complex tannins, individual gallotannins were quantified as gallic acid equivalents and ellagitannins as ellagic acid equivalents, following a class-calibration approach widely adopted in phenolic profiling of tannin-rich plant matrices; accordingly, the resulting values for individual tannins should be regarded as semi-quantitative estimates suitable for inter-species comparison rather than absolute concentrations. System repeatability was confirmed by repeated injections of calibration standards throughout each analytical sequence, with consistently stable peak areas and retention times observed across all runs. Formal recovery experiments were not performed in the present study; however, the extraction procedure was standardized and performed in duplicate, and the consistency between HPLC-quantified tannin fractions and independently determined tannin content obtained by the PVPP/Folin–Ciocalteu approach supports the reliability of the quantitative trends reported. All results were expressed as milligrams of analyte per gram of dry extract (mg/g extract).

2.4. Statistical Analysis

All experimental analyses were performed in triplicate (n = 3), and results are expressed as means ± standard deviation (SD). To confirm and quantify the effects of botanical species and extraction solvent on the measured parameters (TPC, NTPC, TC, and AOA), data were subjected to two-way analysis of variance (ANOVA) without interaction, with species and solvent treated as fixed factors. Prior to inferential analysis, the normality of residual distributions was assessed using the Shapiro–Wilk test, and homogeneity of variance was verified using Levene’s test. Statistically significant main effects were resolved using Fisher’s least significant difference (LSD) post hoc test for pairwise comparisons between group means. For individual phenolic compounds obtained via single-factor design, one-way ANOVA followed by Fisher’s LSD test was applied.

The degree of linear association between different extraction systems, and between phenolic fraction concentrations and antioxidant activity values, was quantified using Pearson’s linear correlation coefficient (r). Statistical significance was defined as p < 0.05 (*) and p < 0.01 (**). All statistical analyses were performed using Statistica^® (version 14.0.0.15, TIBCO Software Inc., Palo Alto, CA, USA). Data processing and graphical visualization were performed using Microsoft Excel^® LTSC MSO (version 16.0.14334.20570, Microsoft Corp., Redmond, WA, USA).

3. Results and Discussion

3.1. Total Phenolic Content and Extraction Efficiency

As shown in Figure 1, interspecific variation in TPC substantially exceeds intraspecific variation due to solvent selection across all studied species, visually confirming the dominant statistical effect of the species factor.

The most notable pattern is the consistently high TPC of Q. cerris across all extraction systems (97.3–109.1 mg GAE/g), with minimal variation between solvents, indicating that this species’ exceptional phenolic richness is reliably recovered regardless of the extraction medium. In contrast, Q. robur exhibits a pronounced solvent-dependent profile: while most systems yield values between 88 and 96 mg GAE/g, 70% acetone drops to 58.0 mg GAE/g—the largest absolute solvent-induced reduction observed in any species—suggesting a phenolic matrix in this species that is particularly sensitive to solvent polarity. Q. rotundifolia displays the lowest TPC profile across all solvents (25.9–40.0 mg GAE/g), consistent with its reduced phytochemical richness. Notably, in Q. ilex, Q. robur and Q. pubescens, 70% acetone consistently yielded the lowest recovery (35.7–58.0 mg GAE/g, respectively), reinforcing the general pattern that higher organic fractions reduce extraction efficiency for the predominantly hydrophilic tannin pool characteristic of Quercus acorns [28,32].

The two-way ANOVA results (

Table 1. Two-way ANOVA—Total phenolic content.

Source	DF	SS	MS	F	p
Model	11	28,822.5	2620.2	52.6	<0.0001 *
Species	5	27,871.5	5574.3	111.8	<0.0001 *
Solvent	6	951.0	158.5	3.2	0.016 *
Error	30	1495.3	49.8
Total	41	30,317.8
Post hoc analysis (Fisher’s LSD test)
Species	TPC (mg GAE/g)		Solvent	TPC (mg GAE/g)
Q. cerris	105.1 ± 3.8 a		Water 20′	62.7 ± 27.3 bc
Q. ilex	49.9 ± 5.8 c		Water 40′	72.6 ± 27.9 a
Q. petraea	88.0 ± 9.0 b		Ace 20% v/v 20′	74.7 ± 30.1 a
Q. robur	85.3 ± 13.0 b		Ace 70% v/v 20′	60.8 ± 32.4 c
Q. pubescens	51.2 ± 9.1 c		EtOH 20% v/v 20′	70.0 ± 30.5 ab
Q. rotundifolia	32.0 ± 5.2 d		EtOH 70% v/v 20′	71.4 ± 26.1 a
			MeOH 20% v/v 20′	67.9 ± 27.8 abc

* p < 0.05. Values are presented as mean ± SD. Different letters within the same column indicate statistically significant differences between acorn flour samples according to Fisher’s LSD test (p < 0.05).

) confirm that both botanical species and solvent type are statistically significant determinants of total phenolic content (TPC) in acorn extracts, with the species factor exerting a markedly dominant influence (F = 111.8; p < 0.0001) over the solvent system (F = 3.2; p = 0.016). A comprehensive comparison of solvent performance across all six species for TPC, TC, and NTPC is provided in the Supplementary Materials, Table S1. This demonstrates that the intrinsic genetic architecture of Quercus species is the primary driver of phytochemical variability, consistent with other authors findings [10,16].

The magnitude of this disproportion between F-values (approximately 35-fold) requires further mechanistic consideration. The species factor reflects the cumulative outcome of genotype-determined biosynthetic capacity, including the expression of galloyl transferases, laccase-type oxidases, and glycoside hydrolases responsible for assembling hydrolysable tannin scaffolds, whereas the solvent factor operates exclusively at the level of physicochemical extraction selectivity [33,34]. The practical consequence is that interspecific selection is a much more powerful lever for modulating polyphenol yield than solvent optimization, implying that species identity should be the primary decision criterion in the industrial sourcing of acorn-based ingredients, with solvent choice reserved for fine-tuning fractional composition rather than maximizing overall yield.

Post hoc analysis revealed a distinct and statistically resolved interspecific hierarchy: Q. cerris > Q. petraea > Q. robur > Q. pubescens > Q. ilex > Q. rotundifolia. Q. cerris exhibited the highest average TPC (105.1 ± 3.8 mg GAE/g), approximately three times greater than Q. rotundifolia (32.0 ± 5.2 mg GAE/g). These findings are consistent with earlier studies identifying Q. cerris as exceptionally rich in secondary metabolites, with tannin concentrations documented as high as 11.69% on a dry weight basis [20,35]. Conversely, the comparatively low pheminnolic content of Q. rotundifolia reflects its designation as a “sweet” acorn variety, characterized by naturally reduced tannin levels that facilitate direct human consumption without extensive processing [4,36]. The high values recorded for Q. petraea (88.0 mg GAE/g) and Q. robur (85.3 mg GAE/g) confirm their substantial bioactive potential and agree with recent HPLC-ESI-MS/MS characterizations reporting TPC values for Q. robur in the range of 103–105 mg GAE/g [16,17].

The placement of Q. ilex and Q. pubescens in the lower-middle tier (49.9 and 51.2 mg GAE/g, respectively) requires contextual interpretation. Both are Mediterranean sclerophyllous species adapted to summer drought and high UV irradiance—conditions that, paradoxically, often induce upregulation of secondary metabolite biosynthesis as a photoprotective response [10]. The relatively modest absolute TPC values recorded here may partly reflect the specific provenance of the sampled material (island populations of Krk, Croatia) and inter-population genetic variation, rather than being invariant species-level traits. Future studies incorporating multi-provenance sampling across the full geographic range of these species would help distinguish genotypic from environmental contributions to phenolic variability.

Regarding solvent influence, ultrasound-assisted aqueous extraction at 40 min (Water 40′; 72.6 mg GAE/g) and 20% acetone (74.7 mg GAE/g) were the most effective systems for total phenolic recovery, while 70% acetone yielded the lowest results (60.8 mg GAE/g). The superior performance of the extended aqueous extraction suggests that many acorn phenolics exist in bound or complexed forms within the matrix, the release of which is enhanced by prolonged ultrasound-assisted extraction (UAE), facilitating improved mass transfer and the disruption of phenolic–matrix interactions [32,37]. The effectiveness of 20% acetone aligns with the established principle that aqueous–organic binary mixtures of moderate polarity provide optimal solvation for both hydrophilic phenolic acids and higher-molecular-weight tannins [20,26].

The consistently higher phenolic recovery observed in this study compared to many reports using conventional maceration can be attributed to the specific physical and chemical effects of acoustic cavitation [32,37]. Unlike traditional maceration, which relies on passive solvent diffusion, UAE facilitates the mechanical disruption of acorn cell walls, effectively targeting cell wall-bound phenolics complexed with polysaccharides and structural proteins [32,37]. These compounds are held within the matrix by strong non-covalent interactions, such as hydrogen bonding, which require the localized high-pressure microenvironments generated by ultrasonic waves to be effectively disrupted [26,28]. Furthermore, the phenomenon of ‘partial acoustic hydrolysis’ observed in our results suggests that UAE does not merely solubilize existing free phenolics but actively promotes the release of monomeric gallic and ellagic acid units from gallotannin and ellagitannin scaffolds [38,39]. This active liberation of bound and complexed forms explains why UAE provides a more comprehensive representation of the acorn’s bioactive potential than conventional stirring or soaking methods, which often fail to penetrate the recalcitrant lignocellulosic matrix of the acorn flour [32,37].

The species-specific solvent response patterns described above reveal biologically meaningful differences in matrix composition. The smallest standard deviation (SD) for Q. cerris across all seven extraction systems (SD = 3.8) contrasts sharply with the high variability of Q. robur (SD = 13.0), suggesting that the former has a phenolic matrix whose constituents are thermodynamically accessible across a wide range of dielectric constants—probably as a result of its overwhelming dominance by uniformly hydrophilic pentagalloylglucose and tetra-O-galloyl-β-D-glucose isomers. In Q. robur, the coexistence of gallotannins and structurally more heterogeneous ellagitannins creates a phenolic pool with a broader polarity range, explaining its greater sensitivity to solvent dielectric properties [33]. From a practical standpoint, the solvent insensitivity of Q. cerris implies that routine quality control of this species’ acorn flour can be performed with any green, food-safe extraction medium without compromising analytical accuracy—a significant advantage for industrial implementation, where ethanol- or water-based systems are preferred for regulatory and safety reasons [19].

The Pearson correlation matrix (

Table 2. Pearson’s correlation matrix—Total phenolic content.

	Water 20′	Water 40′	Ace 20% v/v 20′	Ace 70% v/v 20′	EtOH 20% v/v 20′	EtOH 70% v/v 20′
Water 40′	0.993 **	-	-	-	-	-
Ace 20% v/v 20′	0.977 **	0.956 **	-	-	-	-
Ace 70% v/v 20′	0.896 *	0.922 **	0.865 *	-	-	-
EtOH 20% v/v 20′	0.997 **	0.995 **	0.977 **	0.922 **	-	-
EtOH 70% v/v 20′	0.972 **	0.957 **	0.925 **	0.789	0.954 **	-
MeOH 20% v/v 20′	0.995 **	0.996 **	0.953 **	0.902 *	0.993 **	0.974 **

* p < 0.05; ** p < 0.01.

) demonstrates exceptional methodological congruence, with coefficients exceeding r = 0.95 in most pairwise comparisons. The near-perfect correlations between the 20 min aqueous extraction and low-concentration binary mixtures—specifically 20% ethanol (r = 0.997) and 20% methanol (r = 0.995)—confirm the predominantly hydrophilic nature of the Quercus phenolic profile and indicate that these systems are functionally interchangeable for rapid screening purposes. A notable deviation was observed for 70% acetone, which showed the lowest overall correlation coefficients and a non-significant relationship with 70% ethanol (r = 0.789). This divergence likely reflects a shift in solvent selectivity as the organic fraction increases, decreasing the dielectric constant and favoring the extraction of mid-polarity compounds or higher-molecular-weight condensed tannins less accessible to aqueous systems [20,28].

3.2. Non-Tannic Phenolic Content and Solvent Selectivity

Figure 2 shows a markedly different extraction pattern of NTPC compared to TPC. The most prominent visual feature is the consistent superiority of 70% ethanol across all six species, which in every case yields the highest NTPC value—peaking at 31.9 mg GAE/g in Q. petraea and 28.2 mg GAE/g in Q. robur. This solvent-driven dominance provides a direct graphical counterpart to the reversed ANOVA factor hierarchy observed for NTPC (F = 95.0 for solvent vs. F = 60.5 for species;

Table 3. Two-way ANOVA—Non-tannic phenolic content.

Source	DF	SS	MS	F	p
Model	11	1947.8	177.1	79.3	<0.0001 *
Species	5	675.5	135.1	60.5	<0.0001 *
Solvent	6	1272.3	212.1	95.0	<0.0001 *
Error	30	67.0	2.2
Total	41	2014.8
Post hoc analysis (Fisher’s LSD test)
Species	NTPC (mg GAE/g)		Solvent	NTPC (mg GAE/g)
Q. cerris	18.3 ± 6.2 a		Water 20′	10.9 ± 4.5 d
Q. ilex	8.4 ± 5.5 d		Water 40′	17.1 ± 5.2 b
Q. petraea	18.8 ± 7.0 a		Ace 20% v/v 20′	7.6 ± 2.5 e
Q. robur	15.7 ± 6.4 b		Ace 70% v/v 20′	10.1 ± 4.5 d
Q. pubescens	11.0 ± 6.2 c		EtOH 20% v/v 20′	14.8 ± 5.1 c
Q. rotundifolia	10.5 ± 5.1 c		EtOH 70% v/v 20′	25.2 ± 5.0 a
			MeOH 20% v/v 20′	10.6 ± 4.9 d

* p < 0.05. Values are presented as mean ± SD. Different letters within the same column indicate statistically significant differences between acorn flour samples according to Fisher’s LSD test (p < 0.05).

). Equally noteworthy is the consistently low performance of 20% acetone, which produced the lowest or near-lowest NTPC in five of the six species, with values as low as 3.7–4.0 mg GAE/g in Q. ilex. This inverse behavior—where acetone-rich systems excel at total and tannin-fraction recovery but underperform for non-tannic phenolics—underscores the opposite thermodynamic preferences of high-molecular-weight polymeric tannins and low-molecular-weight phenolic acids [26,32].

The time effect in aqueous extraction is also graphically evident: Water 40′ consistently outperforms Water 20′ across all species, with the largest absolute gain observed in Q. cerris (14.7 vs. 24.7 mg GAE/g) and Q. petraea (17.2 vs. 21.5 mg GAE/g), consistent with the proposed mechanism of partial tannin hydrolysis under extended ultrasound-assisted aqueous conditions [1,17].

The ANOVA results for non-tannic phenolic content (NTPC; Table 3) reveal a significant shift in the relative importance of the two factors compared to TPC analysis: the solvent effect (F = 95.0; p < 0.0001) now surpasses the species effect (F = 60.5; p < 0.0001) in statistical magnitude. This inversion suggests that while genetic factors set the overall ceiling of bioactive potential, the selective extraction of low-molecular-weight phenolic compounds is largely governed by the thermodynamic properties of the solvent system [17,32].

The mechanistic basis for this factor inversion requires explicit discussion, as it reflects a fundamental difference in the physical chemistry of tannin versus non-tannic phenolic extraction. High-molecular-weight hydrolysable tannins are primarily recovered through solvent-mediated disruption of non-covalent tannin–matrix complexes (hydrogen bonding with cell wall polysaccharides and structural proteins), a process governed mainly by the hydrogen-bond-accepting capacity of the solvent—a property intrinsically linked to the plant matrix composition and therefore dominated by the species factor [26,28]. In contrast, low-molecular-weight non-tannic phenolics—gallic acid, ellagic acid, catechin, and their glycosides—are more freely distributed within the vacuolar compartment and intercellular spaces of the flour matrix, and their recovery is principally governed by solvent polarity and the ability to disrupt lipophilic barriers surrounding intracellular pools [32,37]. This structural distinction explains why the solvent factor becomes the dominant determinant of NTPC yield, and why the F-ratio for solvent increases nearly 30-fold relative to its TPC counterpart—from 3.2 to 95.0—while the species F-ratio declines from 111.8 to 60.5.

Post hoc analysis identified Q. cerris and Q. petraea as yielding the highest absolute NTPC values (18.3 and 18.8 mg GAE/g, respectively), while Q. rotundifolia exhibited the significantly highest NTPC/TPC ratio (32.0%;

Table 4. Two-way ANOVA—Non-tannic phenolic content (%NTPC/TPC).

Source	DF	SS	MS	F	p
Model	11	3792.9	344.8	24.3	<0.0001 *
Species	5	1132.1	226.4	16.0	<0.0001 *
Solvent	6	2660.8	443.5	31.3	<0.0001 *
Error	30	424.9	14.2
Total	41	4217.8
Post hoc analysis (Fisher’s LSD test)
Species	NTPC/TPC (%)		Solvent	NTPC/TPC (%)
Q. cerris	17.4 ± 5.6 cd		Water 20′	18.9 ± 8.0 cd
Q. ilex	16.5 ± 9.5 d		Water 40′	24.5 ± 4.1 b
Q. petraea	22.0 ± 10.2 b		Ace 20% v/v 20′	11.1 ± 4.1 e
Q. robur	18.4 ± 6.4 bcd		Ace 70% v/v 20′	17.9 ± 6.3 d
Q. pubescens	21.3 ± 10.1 bc		EtOH 20% v/v 20′	22.6 ± 5.9 bc
Q. rotundifolia	32.0 ± 12.0 a		EtOH 70% v/v 20′	38.0 ± 10.8 a
			MeOH 20% v/v 20′	15.8 ± 4.7 d

* p < 0.05. Values are presented as mean ± SD. Different letters within the same column indicate statistically significant differences between acorn flour samples according to Fisher’s LSD test (p < 0.05).

). This finding is of considerable practical relevance: it confirms that in “sweet” acorn varieties, a larger proportion of the phenolic pool consists of low-molecular-weight, potentially more bioavailable compounds rather than astringent polymeric tannins [4,40]. This phytochemical profile positions Q. rotundifolia favorably for functional food applications where antioxidant activity is desired without the antinutritional effects associated with high tannin loading [20,36].

The clear superiority of 70% ethanol for non-tannic phenolic recovery (25.2 mg GAE/g; NTPC/TPC ratio of 38.0%) is consistent with this solvent’s capacity to solubilize free phenolic acids—including gallic, ellagic, syringic, and protocatechuic acids—as well as flavonoids such as catechin and quercetin, as confirmed by HPLC profiling [11,16]. The higher efficiency of 70% relative to 20% ethanol indicates that an elevated organic fraction is required to disrupt the vegetal matrix and stabilize monomeric phenolics during extraction [12,41]. Notably, extending the aqueous extraction duration from 20 to 40 min also significantly increased NTPC recovery (10.9 to 17.1 mg GAE/g), suggesting that partial acoustic hydrolysis of complex tannins into simpler, measurable phenolic units occurs under prolonged acoustic exposure—a phenomenon previously documented in tannic acid processing [39,42].

This time-dependent NTPC increase in purely aqueous media requires additional investigation in order to draw a conclusion about this behavior. According to some previous research, ultrasonic cavitation creates localized, transient microenvironments with elevated temperature and pressure that can accelerate the cleavage of ester bonds within gallotannin and ellagitannin scaffolds, releasing free gallic and ellagic acid moieties into solution [39,42]. The result is a measurable transfer of phenolic mass from the tannin to the non-tannic fraction—evidenced by the simultaneous increase in NTPC (10.9 → 17.1 mg GAE/g, Table 3) and decrease in the TC/TPC ratio under Water 40′ compared to Water 20′—without a proportional increase in absolute TPC. This trend is also confirmed when expressed as the relative proportion of NTPC to total phenolics, which increased from 18.9% to 24.5% with extended extraction time (Table 4). This partial acoustic hydrolysis effect is functionally relevant from a food processing perspective. Our results showed that increasing extraction time causes a measurable shift from complex tannins to simpler phenolic units. It suggests that conventional aqueous cooking or blanching of acorn flour (which involves higher temperatures and longer durations than the present UAE conditions) may substantially alter the tannin-to-non-tannin balance of the product. This provides a powerful tool to strategically modulate sensory properties (reduced astringency) and nutritional profile (shift toward more bioavailable phenolic forms). These factors should be considered as a critical framework in developing processing protocols for acorn-based functional ingredients.

The practical significance of the NTPC/TPC ratio goes beyond analytical characterization, as it indicates the proportion of the phenolic pool most likely to survive gastrointestinal transit and reach systemic circulation in a bioavailable form. High-molecular-weight hydrolysable tannins (MW > 3 kDa) are generally poorly absorbed in the small intestine and instead undergo extensive microbial biotransformation in the colon, producing bioactive metabolites such as urolithins (from ellagitannins) and pyrogallol derivatives (from gallotannins), whose systemic bioavailability is further influenced by individual microbiome composition [34,40]. In contrast, free gallic acid, ellagic acid, and their low-molecular-weight glycosides show substantially higher and more predictable small intestinal absorption. Accordingly, the markedly elevated NTPC/TPC ratio of Q. rotundifolia (32.0%, compared to 16.5–22.0% for the other species) suggests not only reduced astringency but also a more directly bioavailable phenolic profile—a distinction relevant for functional food formulations targeting acute antioxidant responses rather than the delayed and microbiome-dependent effects of tannin fermentation products [4,40]. Conversely, for applications where slow-release colonic metabolites are the desired bioactive output—for example, in gut microbiota modulation or sustained anti-inflammatory effects—the high-tannin species such as Q. cerris and Q. ilex would be the more appropriate substrates.

The Pearson correlation matrix for NTPC (

Table 5. Pearson’s correlation matrix—Non-tannic phenolic content.

	Water 20′	Water 40′	Ace 20% v/v 20′	Ace 70% v/v 20′	EtOH 20% v/v 20′	EtOH 70% v/v 20′
Water 40′	0.878 *	-	-	-	-	-
Ace 20% v/v 20′	0.851 *	0.928 **	-	-	-	-
Ace 70% v/v 20′	0.921 **	0.946 **	0.952 **	-	-	-
EtOH 20% v/v 20′	0.956 **	0.922 **	0.922 **	0.942 **	-	-
EtOH 70% v/v 20′	0.922 **	0.867 *	0.958 **	0.920 **	0.950 **	-
MeOH 20% v/v 20′	0.975 **	0.954 **	0.926 **	0.979 **	0.969 **	0.935 **

* p < 0.05; ** p < 0.01.

) supports these findings. The exceptionally high correlation of 20% methanol with all other systems (r = 0.975–0.979) highlights its value as a robust reference solvent for the phytochemical screening of low-molecular-weight metabolites.

The strong correlation between 70% ethanol and 20% acetone (r = 0.958) indicates that both systems effectively penetrate the plant matrix to recover non-tannic phenolics, with solvent polarity, rather than specific chemical affinity, being the primary governing factor [32,37]. The relatively lower correlation between the two water extraction times (r = 0.878) further supports a time-dependent release mechanism for simple phenolics in purely aqueous media.

From a methodological perspective, the high inter-solvent correlations for NTPC (r > 0.85 across all system pairs, except for Water 20′ vs. Ace 20%) confirm that the ranking of species by non-tannic phenolic content is largely solvent-independent, similar to the pattern observed for TPC. This indicates that the analytical conclusions regarding the relative NTPC richness of the six species are methodologically robust and not an artifact of solvent choice. However, the substantially lower absolute NTPC values obtained from acetone-based systems compared to ethanolic ones (7.6 vs. 25.2 mg GAE/g for Ace 20% and EtOH 70%, respectively) confirm that, while the species ranking is preserved, absolute quantification is strongly solvent-dependent. This distinction directly affects interlaboratory comparability: studies reporting NTPC values for Quercus species without specifying solvent composition may not be directly comparable, highlighting the need for standardized extraction protocols in comparative phytochemical research on oak acorns. The detailed breakdown of individual solvent efficiencies for non-tannic fractions confirms the consistent superiority of ethanol-based systems (Table S1).

3.3. Tannin Content and Interspecific Variability

As shown in Figure 3, the graphical representation of tannin content provides important visual context for the statistical findings in

Table 6. Two-way ANOVA—Tannin content.

Source	DF	SS	MS	F	p
Model	11	22,195.6	2017.8	37.6	<0.0001 *
Species	5	20,645.9	4129.2	76.9	<0.0001 *
Solvent	6	1549.7	258.3	4.8	0.002 *
Error	30	1611.1	53.7
Total	41	23,806.7
Post hoc analysis (Fisher’s LSD test)
Species	TC (mg GAE/g)		Solvent	TC (mg GAE/g)
Q. cerris	86.8 ± 5.4 a		Water 20′	51.8 ± 23.9 bc
Q. ilex	41.5 ± 5.7 c		Water 40′	55.5 ± 23.0 b
Q. petraea	69.3 ± 14.7 b		Ace 20% v/v 20′	67.1 ± 28.0 a
Q. robur	69.6 ± 12.2 b		Ace 70% v/v 20′	50.7 ± 28.6 bc
Q. pubescens	40.2 ± 9.0 c		EtOH 20% v/v 20′	55.1 ± 25.8 b
Q. rotundifolia	21.5 ± 4.3 d		EtOH 70% v/v 20′	46.2 ± 22.8 c
			MeOH 20% v/v 20′	57.3 ± 23.7 b

* p < 0.05. Values are presented as mean ± SD. Different letters within the same column indicate statistically significant differences between acorn flour samples according to Fisher’s LSD test (p < 0.05).

and

Table 7. Two-way ANOVA—Tannin content (%TC/TPC).

Source	DF	SS	MS	F	p
Model	11	3792.9	344.8	24.3	<0.0001 *
Species	5	1132.1	226.4	16.0	<0.0001 *
Solvent	6	2660.8	443.5	31.3	<0.0001 *
Error	30	424.9	14.2
Total	41	4217.8
Post hoc analysis (Fisher’s LSD test)
Species	TC/TPC (%)		Solvent	TC/TPC (%)
Q. cerris	82.6 ± 5.6 ab		Water 20′	81.1 ± 8.0 bc
Q. ilex	83.5 ± 9.5 a		Water 40′	75.5 ± 4.1 d
Q. petraea	78.0 ± 10.2 c		Ace 20% v/v 20′	88.9 ± 4.1 a
Q. robur	81.6 ± 6.4 abc		Ace 70% v/v 20′	82.1 ± 6.3 b
Q. pubescens	78.7 ± 10.1 bc		EtOH 20% v/v 20′	77.4 ± 5.9 cd
Q. rotundifolia	68.0 ± 12.0 d		EtOH 70% v/v 20′	62.0 ± 10.8 e
			MeOH 20% v/v 20′	84.2 ± 4.7 b

* p < 0.05. Values are presented as mean ± SD. Different letters within the same column indicate statistically significant differences between acorn flour samples according to Fisher’s LSD test (p < 0.05).

. The dominant interspecific gradient is immediately clear: Q. cerris consistently shows high tannin recovery across all solvents (78.1–90.0 mg GAE/g), while Q. rotundifolia remains uniformly low (16.9–27.7 mg GAE/g), with no overlap between these extremes, regardless of the extraction medium used. This visual pattern reinforces the primacy of the species factor over solvent selection, as subsequently confirmed by ANOVA.

A particularly informative feature is the contrasting intra-solvent behavior of the acetone series across species. In Q. petraea and Q. robur, 20% acetone yields the highest tannin recovery among all solvents (69.5 and 86.0 mg GAE/g, respectively), whereas 70% acetone drops sharply—to 43.4 mg GAE/g in Q. petraea and 45.7 mg GAE/g in Q. robur. This steep intra-series contrast directly supports the thermodynamic principle that moderate water content in acetone–water mixtures is essential for tannin solvation: as the organic fraction increases to 70%, the dielectric constant falls below the threshold required for effective solubilization of the predominantly hydrophilic tannin pool characteristic of Quercus acorns [26,28]. Conversely, in contrast to the TPC and NTPC figures where 70% ethanol was consistently among the top performers, Figure 3 shows that 70% ethanol produces the lowest or near-lowest tannin recovery in four of the six species—most notably in Q. cerris (78.1 mg GAE/g, the lowest of all solvents for this species) and Q. pubescens (21.6 mg GAE/g). This graphical inversion between Figure 2 and Figure 3 for the 70% ethanol bars is the clearest visual demonstration of the opposite thermodynamic selectivity of the non-tannic and tannic phenolic fractions, and it provides direct graphical support for the practical recommendation to use 20% acetone for tannin-targeted extraction and 70% ethanol for non-tannic phenolic recovery. Water extraction kinetics are also evident: Water 40′ consistently exceeds Water 20′ across all species, though the absolute gain is modest relative to the solvent polarity effect, which is consistent with the interpretation that tannin release in aqueous media is primarily limited by matrix binding rather than solubility [32,37].

The ANOVA results for tannin content (TC; Table 6) confirmed the dominant role of the species (F = 76.9; p < 0.0001) over solvent selection (F = 4.8; p = 0.002) in determining both the absolute tannin concentration and its relative proportion within the total phenolic pool (TC/TPC ratio). Q. cerris exhibited the highest TC (86.8 ± 5.4 mg GAE/g), representing 82.6% of its total phenolic content (Table 7), which is consistent with its status as an exceptionally tannin-rich species [20,35]. At the opposite extreme, Q. rotundifolia showed the lowest tannin content (21.5 ± 4.3 mg GAE/g) and the lowest TC/TPC ratio (68.0%), confirming its classification as a low-bitterness variety that requires minimal leaching or thermal detannification for human consumption [4,40].

These values also provide an important reference point in the context of acorn processing and detannification strategies reported in the literature. Previous studies on thermal leaching, boiling, fermentation, and prolonged soaking treatments have consistently shown substantial reductions in tannin concentration, often ranging from 40% to over 80%, depending on species and process intensity [3]. In this context, the present untreated flour values establish a baseline for defining species-specific processing thresholds. For highly tannin-rich species such as Q. cerris and Q. robur, more intensive detannification may be required to reach acceptable sensory and nutritional targets, whereas Q. rotundifolia and Q. pubescens may require only mild processing to reduce astringency while preserving antioxidant functionality. This interspecific baseline is particularly valuable for optimizing processing severity according to the intended food application.

Q. robur and Q. petraea exhibited statistically similar intermediate values (approximately 69 mg GAE/g), which is consistent with their characterization as major sources of hydrolysable tannins, including gallotannins and ellagitannins [16,17].

Regarding the solvent effects (Table 6 and Table 7), 20% acetone was the most effective medium for tannin recovery (67.1 mg GAE/g; TC/TPC = 88.9%), which is consistent with the established principle that acetone–water mixtures are superior for disrupting strong hydrogen bonds between tannins and the plant matrix or complexed proteins [38,40]. Conversely, 70% ethanol yielded the lowest tannin concentration (46.2 mg GAE/g) and relative proportion (TC/TPC = 62.0%), indicating its selectivity for lower-molecular-weight phenolics rather than polymeric tannins [32,43]. Extending the water extraction from 20 to 40 min resulted in only a marginal, statistically insignificant (p > 0.05) increase in absolute TC (51.8 to 55.5 mg GAE/g) but a significant decrease in the TC/TPC ratio (81.1% to 75.5%), suggesting that prolonged acoustic exposure may facilitate partial depolymerization or hydrolysis of complex tannins into smaller phenolic units, thereby increasing the measurable non-tannic fraction [39,42]. The extraction-optimized framework, organized by species and solvent polarity, is summarized in Table S1.

From a food safety and regulatory perspective, the finding that TC/TPC ratios exceeded 75% in four of the six studied species (Q. cerris, Q. ilex, Q. robur, and Q. petraea) is practically significant. At high dietary concentrations, tannins reduce protein digestibility and mineral bioavailability and have been associated with potential hepatotoxic effects [13,15]. Therefore, using high-tannin species such as Q. cerris and Q. ilex in food formulations requires prior detannification—by aqueous leaching, alkaline treatment, or thermal processing—or limiting their incorporation to levels consistent with safe intake thresholds [14]. In contrast, the comparatively lower tannin content of Q. rotundifolia (TC/TPC = 68.0%; TC = 21.5 mg GAE/g) and Q. pubescens (TC/TPC = 78.7%; TC = 40.2 mg GAE/g) makes these species more suitable for direct incorporation into functional foods, provided that consumption remains within established safety parameters [4,40].

A particularly noteworthy finding from the TC/TPC ratio analysis (Table 7) is that Q. ilex exhibited the highest relative tannin proportion (83.5 ± 9.5%), slightly exceeding Q. cerris (82.6 ± 5.6%), despite its substantially lower absolute TC (41.5 vs. 86.8 mg GAE/g). This apparent paradox reflects the qualitatively distinct phenolic composition of Q. ilex, whose profile is dominated by structurally complex ellagitannins—including trigalloyl-HHDP (hexahydroxydiphenoyl) glucoside derivatives and O-galloyl-castalagin—rather than the simpler galloyl ester monomers prevalent in Q. cerris. The Folin–Ciocalteu reagent, while broadly responsive to all reducing phenolics, interacts differently with HHDP-bearing ellagitannins compared to galloyl esters, likely generating a higher reducing signal per unit of total phenolics in Q. ilex [28,34]. From a nutritional and technological perspective, the high TC/TPC ratio of Q. ilex indicates that its phenolic bioactivity is strongly tannin-mediated, with a comparatively limited contribution from free phenolic acids—a consideration directly relevant when targeting selective antioxidant fractions for nutraceutical applications [10].

The correlation matrix for tannin content (

Table 8. Pearson’s correlation matrix—Tannin content.

	Water 20′	Water 40′	Ace 20% v/v 20′	Ace 70% v/v 20′	EtOH 20% v/v 20′	EtOH 70% v/v 20′
Water 40′	0.980 **	-	-	-	-	-
Ace 20% v/v 20′	0.973 **	0.954 **	-	-	-	-
Ace 70% v/v 20′	0.841 *	0.882 *	0.850 *	-	-	-
EtOH 20% v/v 20′	0.991 **	0.991 **	0.976 **	0.901 *	-	-
EtOH 70% v/v 20′	0.958 **	0.926 **	0.877 *	0.709	0.924 **	-
MeOH 20% v/v 20′	0.992 **	0.990 **	0.950 **	0.855 *	0.989 **	0.965 **

* p < 0.05; ** p < 0.01.

) shows a high degree of methodological congruence across solvent systems, with most coefficients exceeding r = 0.95 among high-polarity systems. The near-perfect correlations between Water 20′, Water 40′, 20% ethanol (r = 0.991), and 20% methanol (r = 0.992) confirm the highly hydrophilic nature of the predominant tannins in acorn flours, which is consistent with their classification as hydrolysable gallotannins and ellagitannins rather than condensed proanthocyanidins [20,44,45]. The high linearity between the two water extraction times (r = 0.980) further suggests that tannin recovery is governed by rapid initial mass transfer, with extended extraction yielding diminishing returns in metabolic profile diversity [32,37]. From a practical perspective, this statistical redundancy confirms that pure water, as green and food-safe solvent, is fully adequate for representative tannin quantification in acorn flours, eliminating the need for high-concentration organic systems in routine analytical workflows [18,36].

The lowest correlation in the matrix (70% acetone vs. 70% ethanol; r = 0.709, non-significant) highlights a critical divergence in solvent selectivity at elevated organic fractions. As the dielectric constant decreases with increasing organic content, the two solvents diverge thermodynamically: 70% acetone preferentially disrupts hydrogen-bonding interactions between the condensed tannins and the plant matrix, favoring their solubilization [20,28,46], while 70% ethanol shows a greater affinity for lower-molecular-weight flavonoids and phenolic acids, as evidenced by its superior NTPC recovery (Table 3). This complementary selectivity underscores the fact that, while high-polarity aqueous systems provide a reliable general assessment of total polyphenol load, acetone–water mixtures remain indispensable for targeted investigations into complex tannin fractions—a distinction that is directly relevant when designing extraction protocols for specific industrial applications [20,28].

3.4. Antioxidant Potential and Correlations with Phenolic Fractions

The antioxidant potential of the acorn flour samples (

Table 9. Antioxidant potential of acorn flour total phenols (TP), non-tannic phenols (NTP) and tannins (T) (µmol TE/g sample dw).

Species	Total Phenols (TP) (µmol TE/g) *	Non-Tannic Phenols (NTP) (µmol TE/g)	Tannins (T) (µmol TE/g)	NTP/TP (%)	T/TP (%)
DPPH
Q. cerris	1183.6 ± 23.5 a	210.0 ± 9.7 a	973.6 ± 13.8 a	17.7 ± 0.5 ab	82.3 ± 0.5 ab
Q. ilex	511.8 ± 13.5 d	58.5 ± 20.7 c	453.2 ± 34.2 c	11.5 ± 4.3 b	88.5 ± 4.3 a
Q. petraea	960.6 ± 21.1 c	147.5 ± 23.4 b	813.2 ± 2.3 b	15.3 ± 2.1 b	84.7 ± 2.1 a
Q. robur	1022.1 ± 0.7 b	150.0 ± 15.0 b	872.1 ± 15.8 b	14.7 ± 1.5 b	85.3 ± 1.5 a
Q. pubescens	524.1 ± 16.3 d	64.7 ± 10.7 c	459.4 ± 27.0 c	12.4 ± 2.4 b	87.6 ± 2.4 a
Q. rotundifolia	267.8 ± 26.7 e	73.7 ± 15.7 c	194.1 ± 42.4 d	27.9 ± 8.6 a	72.1 ± 8.6 b
ABTS
Q. cerris	1184.6 ± 19.4 a	281.8 ± 37.3 a	902.7 ± 17.9 a	23.8 ± 2.8 b	76.2 ± 2.8 c
Q. ilex	616.1 ± 16.5 c	107.0 ± 0.2 b	509.1 ± 16.4 b	17.4 ± 0.4 cd	82.6 ± 0.4 ab
Q. petraea	1107.0 ± 21.1 b	253.5 ± 24.6 a	853.5 ± 3.5 a	22.9 ± 1.8 b	77.1 ± 1.8 c
Q. robur	1136.2 ± 37.3 ab	234.3 ± 10.8 a	901.9 ± 48.1 a	20.7 ± 1.6 bc	79.3 ± 1.6 bc
Q. pubescens	640.1 ± 19.0 c	101.3 ± 14.3 b	538.7 ± 4.7 b	15.8 ± 1.8 d	84.2 ± 1.8 a
Q. rotundifolia	329.5 ± 7.6 d	116.7 ± 0.3 b	212.7 ± 7.3 c	35.4 ± 0.7 a	64.6 ± 0.7 d
FRAP
Q. cerris	1228.4 ± 65.8 a	238.2 ± 16.1 a	990.3 ± 49.7 a	19.4 ± 0.3 c	80.6 ± 0.3 b
Q. ilex	579.2 ± 16.1 d	83.9 ± 5.2 c	495.3 ± 11.0 c	14.5 ± 0.5 d	85.5 ± 0.5 a
Q. petraea	1137.2 ± 2.5 b	253.2 ± 4.4 a	884.0 ± 6.9 b	22.3 ± 0.4 b	77.7 ± 0.4 c
Q. robur	1053.1 ± 1.6 c	195.9 ± 5.5 b	857.2 ± 7.1 b	18.6 ± 0.6 c	81.4 ± 0.6 b
Q. pubescens	626.2 ± 17.2 d	100.7 ± 1.5 c	525.6 ± 18.6 c	16.1 ± 0.7 d	83.9 ± 0.7 a
Q. rotundifolia	295.7 ± 13.2 e	99.6 ± 9.0 c	196.1 ± 4.2 d	33.6 ± 1.5 a	66.4 ± 1.5 d

* Values are mean ± SD. Different letters in the same column indicate statistically significant difference between acorn flour samples according to Fisher’s LSD test (p < 0.05).

) was assessed using the Water 40′ extract. This system was selected based on three criteria: (i) among purely aqueous systems, it provided the most representative capture of the total phenolic pool, including both soluble hydrophilic phenolic acids and the hydrolysable tannins; (ii) water is the closest food-grade analog to gastrointestinal extraction conditions, making the results physiologically relevant for potential in vivo bioavailability; and (iii) the high Pearson correlation coefficients of Water 40′ with all other solvent systems confirm that it reflects each species’ intrinsic bioactive character rather than solvent-specific extraction artifacts.

In all three assays (DPPH, ABTS, FRAP), absolute antioxidant potential followed the species hierarchy established for TPC, with Q. cerris consistently showing the highest total phenol-based activity (1183.6–1228.4 µmol TE/g) and Q. rotundifolia the lowest (267.8–329.5 µmol TE/g). This ranking is unsurprising given the strong linear relationship between phenolic content and radical scavenging capacity documented across plant matrices [14,21].

However, the fractionated data (NTP/TP and T/TP ratios; Table 9) reveal important qualitative differences that absolute values alone obscure. The implementation of PVPP-based fractionation was a cornerstone of our analytical approach, offering a more nuanced view of antioxidant capacity than conventional total phenolic assays. Q. rotundifolia showed markedly elevated NTP/TP ratios across all assays (27.9% DPPH, 35.4% ABTS, 33.6% FRAP), indicating that a disproportionately large share of its antioxidant activity originates from low-molecular-weight non-tannic compounds. This is consistent with its characteristically simple phenolic profile and low tannin content (Table 6) and supports its suitability for food applications where milder astringency is required [4,40]. Conversely, Q. ilex exhibited the lowest NTP/TP ratios (11.5% DPPH, 17.4% ABTS, 14.5% FRAP), confirming that the great majority of its antioxidant capacity is tannin-mediated—consistent with its ellagitannin-dominated HPLC profile, and the highest TC/TPC ratio recorded among the six species (Table 7). A notable discordance was observed between assays for Q. petraea and Q. robur: in the DPPH assay, Q. robur (1022.1 µmol TE/g) exceeded Q. petraea (960.6 µmol TE/g), while this ranking was reversed in both ABTS and FRAP assays (Q. petraea: 1107.0 and 1137.2 µmol TE/g, respectively; Q. robur: 1136.2 and 1053.1 µmol TE/g). This inversion most likely reflects mechanistic differences between the assays: DPPH measures predominantly hydrogen atom transfer (HAT) activity, whereas ABTS and FRAP are more sensitive to electron transfer (ET) [47,48]. The divergence in ranking therefore implies that the phenolic profiles of these two species differ in the relative proportions of HAT-active and ET-active structural motifs—a distinction consistent with the HPLC data showing that Q. robur is richer in pentagalloylglucose isomers (which are efficient HAT donors), while Q. petraea uniquely contains pedunculagin (peak 4), an ellagitannin whose HHDP moiety is particularly active in ET-based mechanisms [34,49].

Pearson correlation analysis was conducted to quantify the relationship between phenolic fractions and antioxidant potential across all six species (

Table 10. Pearson’s correlation coefficients between radical scavenging activity (DPPH and ABTS), antioxidant capacity (FRAP) and total phenolic, non-tannic phenolic and tannin content in acorn flour samples.

Antioxidant Potential (µmol TE/g)	TPC (mg GAE/g)	NTPC (mg GAE/g)	TC (mg GAE/g)
Total phenols
DPPH	0.993 *	-	-
ABTS	0.974 *	-	-
FRAP	0.982 *	-	-
Non-tannic phenols
DPPH	-	0.948 *	-
ABTS	-	0.937 *	-
FRAP	-	0.944 *	-
Tannins
DPPH	-	-	0.989 *
ABTS	-	-	0.957 *
FRAP	-	-	0.974 *

* p < 0.01.

). The results showed strong linear associations, with all correlation coefficients exceeding r = 0.937. The strongest associations were found for total phenolic content (TPC), with coefficients of r = 0.993 for DPPH, r = 0.974 for ABTS, and r = 0.982 for FRAP, all statistically significant at p < 0.01. Tannin content (TC) also showed a near-perfect correlation with radical scavenging capacity (r = 0.957 to 0.989, p < 0.01), highlighting its dominant role in the bioactive pool. Even the non-tannic phenolic fraction (NTPC), though a minor component, contributed significantly to the antioxidant response, with correlation coefficients ranging from 0.937 to 0.948 (p < 0.01). These findings indicate that the antioxidant capacity of acorn flours is directly and predictably determined by their polyphenolic content, particularly the hydrolysable tannin fraction.

Similarly strong correlations were observed between TC and the antioxidant activity (r = 0.989, 0.957, and 0.974 for DPPH, ABTS, and FRAP, respectively), reflecting the predominance of tannins within the phenolic pool and their well-established radical scavenging and metal-chelating properties [12,21]. NTPC also showed significant correlations with all three assays (r ≈ 0.94–0.95), underscoring the independent and meaningful contribution of free phenolic acids and flavonoids to the overall antioxidant response, even when they represent a minor fraction of the total phenolics [50].

The analysis of the specific antioxidant potential (

Table 11. Specific antioxidant potential of acorn flour samples (µmol TE/mg of total phenols (TP), non-tannic phenols (NTP) or tannins (T)).

Species	Total Phenols (µmol TE/mg) *	Non-Tannic Phenols (µmol TE/mg)	Tannins (µmol TE/mg)
DPPH
Q. cerris	10.8 ± 0.2 a	8.5 ± 0.4 a	11.5 ± 0.2 a
Q. ilex	9.9 ± 0.3 b	5.0 ± 1.8 b	11.4 ± 0.9 a
Q. petraea	10.5 ± 0.2 ab	6.8 ± 1.1 ab	11.7 ± 0 a
Q. robur	11.3 ± 0.0 a	8.4 ± 0.8 a	12.0 ± 0.2 a
Q. pubescens	9.9 ± 0.3 b	4.4 ± 0.7 b	12.0 ± 0.7 a
Q. rotundifolia	6.7 ± 0.7 c	6.0 ± 1.3 ab	7.0 ± 1.5 b
ABTS
Q. cerris	10.9 ± 0.2 b	11.4 ± 1.5 ab	10.7 ± 0.2 c
Q. ilex	12 ± 0.3 a	9.2 ± 0.0 c	12.8 ± 0.4 b
Q. petraea	12.1 ± 0.2 a	11.8 ± 1.1 a	12.2 ± 0.1 b
Q. robur	12.5 ± 0.4 a	13.2 ± 0.6 a	12.4 ± 0.7 b
Q. pubescens	12.0 ± 0.4 a	6.8 ± 1.0 d	14.0 ± 0.1 a
Q. rotundifolia	8.2 ± 0.2 c	9.5 ± 0.0 bc	7.7 ± 0.3 d
FRAP
Q. cerris	11.3 ± 0.6 b	9.7 ± 0.7 b	11.7 ± 0.6 c
Q. ilex	11.2 ± 0.3 b	7.2 ± 0.4 cd	12.4 ± 0.3 bc
Q. petraea	12.5 ± 0.0 a	11.8 ± 0.2 a	12.7 ± 0.1 b
Q. robur	11.6 ± 0.0 b	11.0 ± 0.3 a	11.8 ± 0.1 c
Q. pubescens	11.8 ± 0.3 ab	6.8 ± 0.1 d	13.7 ± 0.5 a
Q. rotundifolia	7.4 ± 0.3 c	8.1 ± 0.7 c	7.1 ± 0.2 d

* Values are mean ± SD. Different letters in the same column indicate statistically significant difference between acorn flour samples according to Fisher’s LSD test (p < 0.05).

)—activity normalized per milligram of phenolic compound—showed that quantitative dominance does not equate to per-unit efficiency. While Q. cerris recorded the highest absolute antioxidant activity, Q. robur and Q. petraea exhibited superior or statistically equivalent specific potentials across most assays. In the ABTS assay, Q. robur reached 12.5 µmol TE/mg and Q. petraea 12.1 µmol TE/mg, both of which were significantly higher than that of Q. cerris (10.9 µmol TE/mg). In the FRAP assay, Q. petraea recorded the highest specific potential among all species (12.5 µmol TE/mg). These findings suggest that Q. cerris achieves high absolute activity primarily through phenolic mass rather than structural efficiency, while Q. robur and Q. petraea contain structurally more efficient radical scavengers—most likely monomeric ellagitannins with dense vicinal hydroxyl groups enabling superior radical stabilization [51,52].

A particularly unexpected finding was the exceptionally high specific tannin antioxidant activity of Q. pubescens in the ABTS (14.0 µmol TE/mg) and FRAP (13.7 µmol TE/mg) assays—the highest among all species in these assays—despite its relatively modest absolute TC (40.2 mg GAE/g). This discordance between the bulk concentration and per-unit activity points to the presence of structurally specific tannins with exceptional per-molecule efficiency. As discussed in Section 3.5, Q. pubescens uniquely exhibits the highest ellagic acid hexoside concentration among all species (1.94 mg/g; peak 14), a compound whose glucosylated ellagic acid scaffold may confer particularly favorable redox properties in ET-based assays [34,49]. Q. rotundifolia consistently showed the lowest specific antioxidant activity across all assays and fractions; notably, its tannin fraction (7.0–7.7 µmol TE/mg) exhibited a lower per-unit activity than its non-tannic fraction in several assays—a pattern that is inverse to those of all other species, where tannins consistently outperform NTPC on a per-mg basis—further confirming the structural simplicity and low functional density of its tannin pool.

3.5. Individual Phenolic Profile by HPLC-DAD Analysis

The HPLC-DAD analysis (

Table 12. Tentative identified phenolic compounds and their quantification (mg/g) * in acorn flour extracts.

Peak	Phenolic Compound **	Rt (min)	[M-H]- m/z	Q. cerris	Q. ilex	Q. petraea	Q. robur	Q. pubescens	Q. rotundifolia
1	Gallic acid 1	4.44	169	6.67 ± 0.07 a	2.9 ± 0.02 c	5.29 ± 0.08 b	5.25 ± 0.06 b	2.94 ± 0.03 c	1.92 ± 0.05 d
2	Galloyl-HHDP-glucose isomer I 2	4.69	633	0.49 ± 0.01 a	0.38 ± 0.01 c	0.35 ± 0.00 d	0.42 ± 0.01 b	0.18 ± 0.00 e	n.d.
3	Digalloyl-glucose isomer I 3	4.84	483	1.45 ± 0.00 a	0.21 ± 0.00 d	0.33 ± 0.01 b	0.29 ± 0.01 c	0.20 ± 0.00 d	0.31 ± 0.01 bc
4	Pedunculagin 2	5.05	783	n.d.	n.d.	0.13 ± 0.00 a	n.d.	n.d.	n.d.
5	Galloyl-HHDP-DHHDP-hexoside 2	5.47	951	n.d.	n.d.	0.23 ± 0.00 a	n.d.	n.d.	n.d.
6	Digalloyl-glucose isomer II 3	5.85	483	n.d.	0.35 ± 0.01 d	0.59 ± 0.01 b	0.71 ± 0.00 a	0.53 ± 0.00 c	n.d.
7	Galloyl-HHDP-glucose isomer II 2	6.84	633	0.6 ± 0.01 a	0.24 ± 0.00 d	0.29 ± 0.00 c	0.35 ± 0.00 b	0.29 ± 0.00 c	n.d.
8	Digalloyl HHDP glucoside isomer I 2	7.15	785	0.76 ± 0.00 a	n.d.	n.d.	0.52 ± 0.00 b	0.37 ± 0.01 c	0.36 ± 0.00 d
9	Digalloyl HHDP glucoside isomer II 2	7.58	785	0.62 ± 0.00 a	n.d.	0.29 ± 0.00 c	0.32 ± 0.01 b	n.d.	n.d.
10	tri-O-galloyl-β-D-glucose isomer I 3	8.25	635	0.18 ± 0.00 b	0.14 ± 0.00 d	0.11 ± 0.00 e	0.20 ± 0.00 a	0.17 ± 0.00 c	0.09 ± 0.00 f
11	Digalloyl HHDP glucoside isomer III 2	9.39	785	1.03 ± 0.00 b	n.d.	0.48 ± 0.00 d	1.15 ± 0.02 a	0.68 ± 0.02 c	n.d.
12	tri-O-galloyl-β-D-glucose isomer II 3	10.69	635	1.82 ± 0.06 a	n.d.	n.d.	0.36 ± 0.00 b	0.36 ± 0.01 b	0.14 ± 0.00 c
13	tri-O-galloyl-β-D-glucose isomer III 3	11.38	635	1.86 ± 0.01 a	n.d.	n.d.	0.70 ± 0.02 b	0.37 ± 0.00 c	0.28 ± 0.00 d
14	Ellagic acid hexoside 1	11.94	463	1.18 ± 0.00 e	1.28 ± 0.00 c	1.37 ± 0.01 b	1.26 ± 0.01 d	1.94 ± 0.01 a	1.1 ± 0.00 f
15	Digalloyl HHDP glucoside isomer IV 2	12.86	785	0.34 ± 0.00 b	0.2 ± 0.00 c	0.37 ± 0.00 a	0.21 ± 0.01 c	0.10 ± 0.00 d	0.11 ± 0.00 d
16	1,2,3,6-tetra-O-galloyl-β-D-glucose isomer I 3	14.13	787	1.34 ± 0.02 a	n.d.	0.32 ± 0.00 d	0.72 ± 0.01 b	0.46 ± 0.00 c	n.d.
17	Trigalloyl-HHDP-glucoside isomer I 2	15.87	937	1.76 ± 0.02 b	0.45 ± 0.00 d	n.d.	1.85 ± 0.02 a	1.14 ± 0.02 c	n.d.
18	Ellagic acid pentoside 1	15.99	433	n.d.	n.d.	2.00 ± 0.02 b	2.11 ± 0.04 a	n.d.	n.d.
19	Trigalloyl-HHDP-glucoside isomer II 2	15.99	433	0.58 ± 0.00 a	0.52 ± 0.02 b	n.d.	n.d.	0.11 ± 0.00 c	0.13 ± 0.00 c
20	O-galloyl-castalagin 2	16.37	1085	n.d.	0.44 ± 0.00 a	n.d.	n.d.	n.d.	n.d.
21	1,2,3,6-tetra-O-galloyl-β-D-glucose isomer II 3	16.76	787	14.11 ± 0.14 a	0.27 ± 0.00 e	0.26 ± 0.00 e	3.85 ± 0.10 b	1.74 ± 0.04 c	0.63 ± 0.00 d
22	Ellagic acid deoxyhexoside 1	17.19	447	n.d.	n.d.	1.27 ± 0.01 b	n.d.	1.13 ± 0.00 c	1.48 ± 0.02 a
23	Trigalloyl-HHDP-glucoside isomer III 2	17.98	937	n.d.	0.82 ± 0.00 a	n.d.	n.d.	n.d.	n.d.
24	Ellagic acid 1	18.56	301	5.58 ± 0.05 a	4.42 ± 0.11 b	5.42 ± 0.10 a	3.97 ± 0.00 c	3.38 ± 0.01 d	1.59 ± 0.01 e
25	Trigalloyl-HHDP-glucoside isomer IV 2	18.74	937	n.d.	4.14 ± 0.07 a	n.d.	n.d.	n.d.	2.01 ± 0.07 b
26	Pentagalloyl glucose isomer I 3	19.26	939	2.16 ± 0.04 a	n.d.	n.d.	1.70 ± 0.04 b	0.75 ± 0.02 c	0.45 ± 0.01 d
27	Pentagalloyl glucose isomer II 3	20.12	939	8.86 ± 0.29 a	n.d.	n.d.	4.54 ± 0.14 b	1.72 ± 0.01 c	0.96 ± 0.02 d
28	Pentagalloyl glucose isomer III 3	20.95	939	3.45 ± 0.11 a	n.d.	n.d.	0.87 ± 0.01 b	n.d.	n.d.
29	Pentagalloyl glucose isomer IV 3	21.62	939	3.62 ± 0.02 a	n.d.	n.d.	n.d.	n.d.	n.d.
30	Hexagalloyl-glucoside 3	25.14	1091	0.22 ± 0.00 b	0.45 ± 0.01 a	n.d.	n.d.	n.d.	n.d.
	Total phenolic content (TPC) 1,2,3			58.7 ± 0.09 a	17.22 ± 0.01 e	19.11 ± 0.01 c	31.33 ± 0.23 b	18.58 ± 0.04 d	11.55 ± 0.14 f
	Non-tannic phenolic content (NTPC) 1			13.43 ± 0.12 b	8.61 ± 0.09 e	15.35 ± 0.02 a	12.59 ± 0.03 c	9.39 ± 0.02 d	6.08 ± 0.06 f
	Tannin content (TC) 2,3			45.27 ± 0.21 a	8.61 ± 0.08 c	3.76 ± 0.01 e	18.74 ± 0.26 b	9.19 ± 0.06 c	5.46 ± 0.08 d
	Ellagitannins 2			6.18 ± 0.02 b	7.18 ± 0.06 a	2.15 ± 0.01 f	4.81 ± 0.01 c	2.88 ± 0.03 d	2.60 ± 0.06 e
	Gallotannins 3			39.09 ± 0.23 a	1.43 ± 0.02 e	1.61 ± 0.02 e	13.93 ± 0.27 b	6.31 ± 0.02 c	2.87 ± 0.01 d

* Values are mean ± SD. Different letters in the same row indicate statistically significant difference between acorn flour samples according to Fisher’s LSD test (p < 0.05). ** Detailed MS² fragmentation patterns and λ_max are provided in Supplementary Table S2. ¹ Non-tannic phenolic compound; ² Ellagitannins; ³ Gallotannins.

) identified and tentatively assigned 30 phenolic compounds across the six Quercus species, including hydrolysable tannins from both gallotannin and ellagitannin subclasses, as well as several non-tannic phenolic acids. The results demonstrated a highly heterogeneous distribution of constituents among the investigated Quercus species, confirming that differences in phenolic composition are strongly species-dependent. The profile was predominantly represented by hydrolysable tannins, which is consistent with previous findings indicating that acorns are rich in ellagitannins and gallotannins, while condensed tannins are generally negligible or absent in the cotyledon fraction of acorns [3]. The distribution of individual compounds was highly species-specific in both composition and relative abundance, reflecting genetic predispositions and environmental adaptations that govern secondary metabolism. Some quantitative discrepancies between the HPLC results and the Folin–Ciocalteu phenolic content were identified. These discrepancies are well documented in the literature and can be explained by the presence of a wide range of non-phenolic reducing compounds with which the Folin reagent reacts with, and the influence of large polymerized tannins that may not be fully detected by standard HPLC methods [52]. These limitations are well recognized in the field and do not compromise the comparative value of either approach; however, they should be considered when interpreting absolute values across methodologies.

Among the identified compounds, gallic acid and ellagic acid were the most abundant non-tannic phenolics, with gallic acid reaching its highest concentration in Q. cerris (6.67 mg/g), while ellagic acid was particularly abundant in Q. cerris and Q. petraea. These compounds are well known as the hydrolysis products of gallotannins and ellagitannins, respectively, and their presence may reflect their natural metabolic turnover [3].

Q. cerris was characterized by a pronounced dominance of high-molecular-weight gallotannins, with 1,2,3,6-tetra-O-galloyl-β-D-glucose isomer II as the single most abundant compound across all species (14.11 mg/g; peak 21), and multiple pentagalloylglucose isomers collectively reaching approximately 18.1 mg/g (peaks 26–29). Along with the highest gallic acid concentration recorded (6.67 mg/g; peak 1), this gallotannin-dominated profile reflects the species’ exceptional capacity for sequential galloylation of glucose through the central tannin biosynthetic pathway [33,34,35]. The accumulation of pentagalloylglucose (PGG) is of particular biosynthetic significance: PGG is the obligate branch-point precursor from which ellagitannins are formed via oxidative C–C coupling of adjacent galloyl groups into HHDP units [34]. The fact that Q. cerris accumulates PGG at high levels without proceeding substantially to ellagitannins suggests either a low flux through the oxidative coupling step in this species or a high rate of PGG turnover in other pathways.

From an antioxidant perspective, the dense array of free hydroxyl groups on galloyl residues—five per PGG molecule—provides a structurally favorable scaffold for hydrogen atom transfer, which is mechanistically consistent with the high absolute DPPH activity recorded for this species (Table 9) [51].

A similar, although less pronounced, pattern was observed in Q. robur, which also exhibited substantial levels of tetra- and pentagalloyl glucose derivatives. This is in agreement with earlier studies showing that gallotannins can accumulate extensively in certain Quercus species and represent key intermediates in tannin biosynthesis. The high abundance of these compounds explains the elevated total tannin content and suggests a metabolic shift toward galloylation processes [53,54]. Furthermore, Q. robur displayed a mixed gallotannin–ellagitannin profile, with gallotannins totaling 13.93 mg/g and a substantial ellagitannin complement (4.81 mg/g), including multiple digalloyl-HHDP glucoside isomers (peaks 8, 9, 11, 15) and trigalloyl-HHDP glucoside isomer I (1.85 mg/g; peak 17). The coexistence of both biosynthetic arms—galloylation and oxidative HHDP coupling—is consistent with the characterization of Q. robur as a species expressing both gallotannin and ellagitannin pathways simultaneously [16,17]. Notably, Q. robur was one of only two species (alongside Q. petraea) in which ellagic acid pentoside was detected (2.11 mg/g; peak 18), a compound absents in Q. cerris and Q. ilex. Ellagic acid pentoside represents a hydrolysis or glycosylation product of ellagitannins and has been associated with sustained antioxidant activity and distinct bioavailability profiles relative to free ellagic acid [33]. Its exclusive presence in Q. robur and Q. petraea may in part explain the higher specific DPPH activity of Q. robur relative to that of Q. cerris (11.3 vs. 10.8 µmol TE/mg; Table 11), despite the latter carrying greater absolute phenolic mass.

Q. ilex exhibited a distinctly ellagitannin-enriched signature that sets it apart from all the other studied species. Its profile was dominated by trigalloyl-HHDP-glucoside isomer IV (4.14 mg/g; peak 25)—by far the highest concentration of this compound across any species—alongside trigalloyl-HHDP-glucoside isomers II and III (peaks 19 and 23). The high concentration of these compounds suggests the enhanced oxidative coupling of galloyl groups into HHDP units, which is a key step in ellagitannin formation [34]. Although Q. ilex, together with Q. rotundifolia, has the lowest polyphenolic content, the high content of ellagitannins, and especially Trigalloyl-HHDP-glucoside isomer IV, means that the astringency potential of Q. ilex is significant. Namely, ellagitannins can trigger a feeling of astringency faster and for longer than other tannins because their binding kinetics with saliva proteins are faster, and their dissociation is slower than that observed in gallotannins [55]. Furthermore, O-galloyl-castalagin (0.44 mg/g; peak 20) is a structural derivative of castalagin, one of the most complex naturally occurring ellagitannins, and it was detected exclusively in Q. ilex among all six species, constituting a species-specific chemical marker [34]. Hexagalloyl-glucoside was detected at its highest concentration in Q. ilex as well (0.45 mg/g vs. 0.22 mg/g in Q. cerris; peak 30), indicating a biosynthetic preference for extensively oxidized and coupled polyphenolic scaffolds. This shift toward structurally complex ellagitannins, involving the multi-step oxidative coupling of galloyl moieties into HHDP units, is consistent with the known biochemistry of ellagitannin biosynthesis and has been associated with antioxidant mechanisms distinct from those of simple galloyl esters, including metal chelation and redox cycling [34,49,56]. This structural complexity provides a mechanistic basis for the high relative tannin efficiency of Q. ilex in ET-based assays (ABTS specific tannin activity: 12.8 µmol TE/mg; FRAP: 12.4 µmol TE/mg; Table 11) despite its comparatively low absolute TC.

Q. petraea exhibited the most chemically diverse profile among the six species. Its unique features include the exclusive presence of pedunculagin (0.13 mg/g; peak 4) and galloyl-HHDP-DHHDP-hexoside (0.23 mg/g; peak 5), both of which were detected only in Q. petraea. Pedunculagin is a dimeric ellagitannin containing two HHDP groups, and its presence confirms that the oxidative coupling pathway in this species extends beyond the initial HHDP formation step [34]. Galloyl-HHDP-DHHDP-hexoside provides further evidence of the advanced oxidative modification, involving a dehydro-HHDP (DHHDP) unit formed through additional C–C coupling. The simultaneous detection of ellagic acid pentoside (2.00 mg/g; peak 18) and ellagic acid deoxyhexoside (1.27 mg/g; peak 22) further supports a diversified ellagitannin hydrolysis and secondary modification metabolism. Structurally, the presence of multiple HHDP-bearing scaffolds in Q. petraea aligns with the superior per-unit electron-transfer activity recorded for this species in both the ABTS (12.1 µmol TE/mg) and the FRAP assays (12.5 µmol TE/mg; Table 11), as HHDP-bearing ellagitannins are particularly active due to their extended aromatic conjugation and favorable redox potential [49,51].

Species such as Q. rotundifolia and Q. pubescens displayed comparatively lower concentrations of both gallotannins and ellagitannins, alongside a relatively higher contribution of non-tannic phenolics (>50%). This pattern may indicate either reduced tannin biosynthesis or increased degradation, as suggested in previous studies on acorn composition and variability [3,57]. The lower tannin content of acorns from the Q. rotudifolia and Q. pubescens species was confirmed by other authors [4,58], which could be a crucial factor in the selection of less bitter species for use in food production.

Q. pubescens displayed a moderate overall phenolic content but was notable for one structurally significant observation: it had the highest concentration of ellagic acid hexoside among all six species (1.94 mg/g; peak 14), a compound present in all species but for which Q. pubescens ranked distinctly first (compared to 1.37 mg/g in Q. petraea, the next highest). Ellagic acid hexoside is a glycosylated form of free ellagic acid whose glucosyl substituent confers improved aqueous solubility and may modulate both the bioavailability and the intrinsic redox activity relative to the aglycone [33,49]. The prominence of this compound in Q. pubescens is directly relevant to the antioxidant results discussed in Section 3.4: this species recorded the highest specific tannin antioxidant activity in both the ABTS (14.0 µmol TE/mg) and the FRAP (13.7 µmol TE/mg) assays, despite a modest absolute TC (40.2 mg GAE/g). The selective enrichment of an ellagic acid conjugate with favorable electron-transfer properties in a species otherwise unremarkable for its bulk phenolic concentration suggests a qualitative rather than quantitative basis for its exceptional per-unit activity and requires further investigation by targeted MS² fragmentation and in vitro bioavailability assays.

Q. rotundifolia exhibited the simplest and least diverse phenolic profile among the six species, which was consistent with its low absolute TPC and TC values. Most of its phenolic content was accounted for by a small number of compounds distributed across classes: gallic acid (1.92 mg/g; peak 1), free ellagic acid (1.59 mg/g; peak 24), and minor contributions from the digalloyl-glucose isomers and the pentagalloylglucose isomers, which were present only at trace levels compared to those in Q. cerris and Q. robur. However, one structurally noteworthy observation emerged from the quantitative data: Q. rotundifolia had the highest concentration of ellagic acid deoxyhexoside among all the species (1.48 mg/g; peak 22)—the only compound in Table 12 for which this species ranked first, being absent or below the detection limit in Q. cerris, Q. ilex, and Q. robur. Ellagic acid deoxyhexoside is a deoxy-sugar conjugate of ellagic acid whose biosynthetic origin is not fully established but may involve salvage pathways from ellagitannin hydrolysis coupled with non-standard glycosylation [33]. Its selective accumulation in an otherwise tannin-poor species suggests species-specific regulation of ellagitannin catabolic or secondary modification pathways and may partially explain the comparatively higher specific antioxidant activity of its non-tannic fraction, which exceeded that of its tannin fraction in per-unit activity in several assays—a pattern that is unique to Q. rotundifolia across the dataset (Table 11).

Overall, the HPLC-DAD profiling results confirm that the phenolic composition of acorn flour is highly species-specific, with distinct metabolic patterns ranging from gallotannin-dominated (Q. cerris, Q. robur) to ellagitannin-rich (Q. ilex) and low-tannin profiles (Q. rotundifolia). These differences can also be attributed to the harvest timing and the fruit’s stage of maturity. The literature indicates that oxidation plays a key role in the breakdown of phenolic compounds. Therefore, factors such as the maturity stage, growing conditions, handling practices, and storage conditions can also influence variations in the phenolic composition of samples. These differences are expected to significantly influence the antioxidant properties and functional behavior of the flour in food systems [57,58].

The results presented in this study, along with the converging structure–activity relationships established, validate the fractionation strategy employed and underscore the complementary and irreplaceable role of HPLC-DAD profiling alongside bulk spectrophotometric assays in the full characterization of Quercus-derived phenolic ingredients.

3.6. Study Limitations and Future Research Directions

Although this study provides a comprehensive comparative framework for Mediterranean acorn flours, several limitations should be acknowledged. While UAE was efficient for recovering bound phenolics, further comparative studies with conventional maceration are needed to fully characterize how ultrasonic cavitation specifically targets cell wall-bound compounds in the acorn matrix. Additionally, discrepancies between HPLC-DAD and spectrophotometric results suggest that highly polymerized tannins or non-phenolic reducing agents may not have been fully detected by standard chromatographic methods. Future investigations should employ advanced mass spectrometry or specialized assays to further resolve these complex tannin fractions. Finally, although this study establishes a robust baseline for antioxidant potential, it lacks in vivo bioavailability and safety assessments. Subsequent research should focus on the metabolic fate of acorn tannins, particularly their microbial biotransformation in the colon, and verify the safety thresholds for direct human consumption of high-tannin varieties. Such studies are essential to confirm the functional efficacy and long-term safety of acorn-based ingredients in human nutrition.

4. Conclusions

This study demonstrated that the botanical species is the primary determinant of phenolic composition in acorn flours, while the solvent selection mainly governs the distribution between tannin and non-tannin fractions. Among the investigated species, Q. cerris and Q. robur showed the highest phenolic and tannin contents, driven by a strong gallotannin accumulation, which also explains their high antioxidant capacity. In contrast, Q. rotundifolia exhibited the lowest total phenolic content but the highest relative contribution and efficiency of non-tannic compounds, making it the most suitable species for direct food application without prior detannification. Q. ilex was characterized by an ellagitannin-enriched profile, while Q. petraea showed the greatest compositional diversity. Notably, Q. pubescens exhibited high specific antioxidant activity despite its moderate tannin levels.

Extraction conditions significantly affected the phenolic partitioning, with 20% aqueous acetone favoring tannin recovery and 70% ethanol enhancing the recovery of non-tannic phenolics. Water-based ultrasound-assisted extraction proved to be a suitable green alternative for general screening. The high tannin dominance in Q. cerris, Q. ilex, Q. robur, and Q. petraea highlights the need for the detannification of acorns if they are used in large quantities in food production formulations, whereas Q. rotundifolia and Q. pubescens are viable candidates for direct use.

Overall, these findings establish a framework for the valorization of acorn flours based on species selection and targeted extraction, emphasizing the importance of detailed chromatographic profiling alongside spectrophotometric methods for an accurate functional characterization.