Characterization of tannin extracts derived from the bark of four tree species by HPLC and FTIR

Abstract The objective of this work was the extraction and quantification of tannins obtained from the bark of four tree species from the forests of Ixtlán de Juárez, Oaxaca, Mexico (Arbutus xalapensis Kunth., Prunus serotina Ehrh., Quercus crassifolia Humb. and Bonpl., and Quercus laurina Humb. and Bonpl.), In this extraction process, 90% ethanol was used as solvent at an extraction temperature of 70 °C for 2 h. The quantification of phenolic compounds, condensed tannins, and percentage of total extract, were obtained using the Stiasny number. The total extracts were 12.87, 16.38, 19.31, and 25.68% for P. serotina, Q. laurina, Q. crassifolia, and A. xalapensis, respectively. The condensed tannins were characterized by Fourier Transformed Infrared Spectroscopy (FTIR) where at 1626 cm−1 is observed an isomerization that implies a rotation between the C and B rings that corresponding at elongation of bonds in benzene structure of catechin. The purity of the extracted tannins was analyzed by High Performance Liquid Chromatography (HPLC). The effectiveness of the extraction process was proven; the importance of knowing the amount of tannins will promote the utilization of the bark of these species. The results of the condensed tannin content place Quercus crasifolia (10.29%) and Arbutus xalapensis (13.12%) as potential sources of tannins.


Introduction
The name tannin comes from the ancient practice of using plant extracts to process animal skin into leather (Feng et al. 2013). Tannins are categorized into condensed and hydrolyzable (Fengel and Wegener 1989). Condensed tannins are polymers of flavonoid units linked by C-C bonds, which cannot be hydrolyzed but can be oxidized by a strong acid to yield anthocyanidins (Okuda 2005). Condensed tannins have been one of the leading research objects of wood adhesives and represent 90% of the world's tannin production. The annual industrial production of this tannin has reached up to 200,000 tons worldwide (Zhou and Du 2019), while the production of hydrolyzable tannins worldwide is about 10% (Arbenz and Av erous 2015). Hydrolyzable tannins are heterogeneous polymers containing phenolic acids, especially gallic acid, and simple sugars. Furthermore, they are smaller than condensed tannins, efficiently hydrolyzable, but less common in wood (Fengel and Wegener 1989).
Several researchers highlight the importance of tannins in the treatment of emerging pollutants, removal and recovery of pollutant ions (Sun et al. 2019), and urban water and wastewater treatment (Grehs et al. 2019;Bello et al. 2020). In addition to some applications for industrial purposes, such as adhesives for wood and corrugated cardboard (Pizzi and Mittal 2018;Pizzi 2019), metal corrosion inhibitors (Pizzi 2019), building insulation (Meikleham and Pizzi 1994) and in the substitution of some phenolic compounds (Santana et al. 1995). Also, due to their antimicrobial, antifungal, and antioxidant characteristics (Amarowicz et al. 2000;Sulaiman et al. 2011;Wei et al. 2011;Wong-Paz et al. 2015;Zhang et al. 2015), these chemical compounds have been valued by the pharmaceutical industry (Abu et al. 2016;Pizzi 2019).
The solvents commonly used in extraction processes have been water and various organic solvents or mixtures (Kemppainen et al. 2014;Abu et al. 2016;Ruiz-Aquino et al. 2021;Nisca et al. 2022). It is known that using water in extraction is more economical, and industrially hot water extraction is used (Kemppainen et al. 2014). However, the extract usually contains phenolic monomers, mineral substances, and carbohydrates (Bianchi et al. 2015).
Ethanol is an organic solvent and due to its polar nature, the extraction of tannins is more efficient, as mentioned by Duraisamy et al. (2020), who used water, methanol, and a mixture of water and methanol, highlighting the advantage of using polar solvents. In the same sense, Downey and Hanlin (2010) indicate that tannin extraction increases with increasing proportions of ethanol, up to a maximum of 50% ethanol in water, the selection of the solvent mixture is important and depends on the objective of the analysis to be performed.
Extraction using Soxhlet with ethanol is one of the best methods used in separating bioactive compounds from nature. This is because the heating that occurs during the extraction process can increase the ability to extract compounds that are not soluble at room temperature, in addition, ethanol is a solvent that produces the highest total tannin, and the lowest residue (Kusuma et al. 2022).
Some studies have implemented techniques to characterize total polyphenols and condensed tannins from extracts by Fourier transform infrared spectroscopy (FTIR) and high-pressure liquid chromatography (HPLC) (Chupin et al. 2013;Durgawale et al. 2016;Ruiz-Aquino et al. 2021). FTIR is one of the most widely used analytical methods for molecular analysis due to its relatively low cost, ease of use, and reliability of the results provided (Ricci et al. 2015). HPLC separates dissolved compounds in a sample and allows qualitative and quantitative analysis of each sample component (Moldoveanu and David 2016).
Tannins are found in most vascular plants around all their components (bark, wood, leaves, buds, stems, fruits, seeds, roots, and plant galls) (Haslam 1989;Pizzi 2019). Scientific reports indicate a high concentration of tannins in the bark of some trees (Paes et al. 2006;Sousa et al. 2019;Ruiz-Aquino et al. 2021). It is known that the barks from the Schinopsis, Acacia, and Pinus genera (Pizzi 2019) play an essential role in the industrial extraction of tannins. However, these chemical compounds have also been found in other tree species, such as Quercus spp., Eucalyptus spp., Arbutus spp. (Trugilho et al. 2003;Sartori et al. 2018;Ruiz-Aquino et al. 2021).
Barks are available as waste material and by-product of wood industry; the characterization of extractable substances can promote their use and add value. They have been reported to contain interesting molecules and show some bioactivity, such as antioxidant and antifungal, after wood, the bark is the second most important tissue of a trunk. It amounts to about 10-20% of a stem depending on the species and growing conditions (Arifiani et al. 2017). La corteza se considera un residuo del aprovechamiento forestall.
The genus Quercus consists of about 600 species worldwide, is one of the most imperative clades of woody angiosperms (Rasheed et al. 2017). In Mexico, oaks grow in all states of the republic (Mart ınez-Calder on et al. 2017). In the forests of Ixtl an de Ju arez, Oaxaca, Mexico, Quercus crassifolia Humb. and Bonpl. and Quercus laurina Humb. and Bonpl are the species with the greatest structural dominance (Ruiz-Aquino et al. 2015). The productivity of the oak forests of Ixtl an de Ju arez, Oaxaca, Mexico, has made it possible to analyze these species being used industrially due to their wood quality in terms of their physical and mechanical properties (Ruiz-Aquino et al. 2018). However, in the sawmilling process lignocellulosic residues are generated, including bark, although sometimes, during debarking in the field, it can be integrated into the soil (Gim enez et al. 2008).
In the forest of Ixtl an de Ju arez, Oaxaca, coniferous and broadleaf species cohabit. The wood of oaks and broadleaves, such as Arbutus xalapensis Kunth. and Prunus serotina Ehrh, does not have a stable market, its low utilization is attributed to the low cost per cubic meter (Santiago-Garc ıa et al. 2022), nevertheless, in a growing context of green and circular economy, gaining knowledge of the composition of every species is crucial, as this will allow for their full exploitation. Thus, the objective of the present work was to evaluate the tannin content of four tree species (Arbutus xalapensis Kunth., Prunus serotina Ehrh., Quercus crassifolia Humb. and Bonpl., and Quercus laurina Humb. and Bonpl.) from the Sierra of Oaxaca, Mexico using ethanol as extraction solvent and characterizing the extracts by FTIR and HPLC.

Sample collection and preparation
Bark samples were collected in the forest of Ixtl an de Ju arez, Oaxaca, Mexico; located within the geographic coordinates 17 18 0 16 00 -17 30 0 00 00 N and 96 21 0 29 00 -96 31 0 38 00 W (Santiago-Garc ıa et al. 2022). Samples of outer bark were removed from standing trees without damaging the entire circumference and dried in an oven at 103 C for 48 h (Paes et al. 2006). The dry samples were crushed, ground, and sieved, and the particle size used was 1 mm.

Tannin extraction
Tannin extraction was performed in triplicate using, in each case, 25.0 g of bark, 200 mL of ethanol 90% (JT Becker technical grade), and 200 mL of glacial acetic acid to avoid oxidation of the tannin extracts. The extraction process was refluxed at 70 C for 2 h, Popova et al. (2021), indicate that regardless of ethanol concentration and temperature, the maximum amount of tannins is extracted after the first 600 s interval. The solution was cooled to room temperature and filtered under vacuum using Whatman No. 54 paper. The tannin extract's recovery from alcohol was performed using a rotary evaporator, and the total extract was calculated by weight difference in the initial bark sample before and after extraction (Kusuma et al. 2022).
Stiasny number and condensed tannins 0.1 g of ethanolic extract, 10.0 mL of methanol, 1.0 mL of hydrochloric acid, and 2.0 mL of 37% formaldehyde were placed in an Erlenmeyer flask to determine the Stiasny number. The mixture was boiled for 30 min, then filtered under vacuum using Whatman paper No. 54. These experiments were carried out in triplicate. The dry residue corresponds to the amount of condensable tannin, according to the Stiasny number Lisperguer et al. 2016), calculated with the following expression: Where: NS ¼ Stiasny number, W 1 ¼ mass of the residue, W 2 ¼ dry mass of the initial sample.
With the Stiasny number and the amount of total extract, the number of condensed tannins was calculated according to the following expression (Paes et al. 2006):

HPLC-UV/VIS method for quantification and identification of compounds
The HPLC system used was a CC5-BAS liquid chromatograph equipped with a quaternary pump and UV-116 BAS detector. The HPLC column Thermo Scientific TM Hypersil BDS C18 reversed-phase 250 Â 4.6 mm, 5.0 mm was used a 30 C. Wavelengths of detection for (þ)-catechin 280 nm were purchased from Sigma-Aldrich. All reagents used were HPLC grade; catechin standard of 98% purity, methanol and phosphoric acid of Sigma Aldrich, and Milli-Q water from 18.0 mS. All reagents and standards were prepared using Milli Q deionized water were filtered and degassed before their use. The tannin samples were diluted in 1.0 mM of phosphoric acid this was done to avoid the decomposition of the tannin extracts.
A mobile phase of 60:40 (v/v%) nm methanol-water and a 1.0 mL/min flow rate were used under isocratic flow conditions with pressure of 13.65 MPa, at room temperature (30 C) and k ¼ 280 nm and 40 mL of the sample was injected into the chromatographic column. Concentrations of the analytes were calculated from chromatogram peak areas based on calibration curves. The method linearity was assessed by means of linear regression of the mass of analyte injected vs.

Analysis of tannin extracts by FTIR
For the study of surface chemistry and the identification of the functional groups, the material was analyzed by Fourier transform infrared spectroscopy (FTIR) on a Perkin Elmer Spectrum, Frontier model, in the wavenumber range 500-4500 cm À1 .
For this study, $0.015 g of tannin and 0.450 g of potassium bromide (KBr) were weighed and placed in the oven at 60 C for a week to ensure moisture removal. Then, the KBr was mixed with the sample, and by compression, the pellet was made for subsequent analysis with the FTIR spectrometer.

Tannin extraction
The total extract amount of A. xalapensis (25.68%) and Q. crassifolia (19.31%) (  Bianchi et al. (2015) for bark extracts of European softwood species. The amount of tannins depends on the species, geography, biological origin, age, type of tissue, e.g. sapwood, heartwood, and bark (Pizzi and Cameron 1986), also, the extraction method influences, which can be very variable (Das et al. 2020).

Stiasny number and condensed tannins
The Stiasny number indicates the reactivity of the extracts to formaldehyde. This information can help us to determine if the extracts can be used as adhesives. The values obtained range from 36% (P. serotina) to 53% (Q. crassifolia) ( Table 1) and are in the range reported (22-73%) for different hardwood barks (Paes et al. 2006). Compared with other tree species, the maximum value of condensed tannins obtained here is lower than the 43.44% (Alnus incana) and 34.27% (Alnus glutinosa), extracted with ethanol-water by Janceva et al. (2011). This study obtained a Stiasny value of 53% for Q. crassifolia and 51% for A. xalapensis species,  produced suitable quality adhesives obtaining a Stiasny number of 46%. The concentration of condensed tannins is directly proportional to the yield in Stiasny number and total extract, and the highest values of condensed tannins corresponded to the bark of Q. crassifolia (10%) and A. xalapensis (13%), similar values were reported by Bianchi et al. (2015) for Scots pine (13%). Figure 1 shows the calibration curve was performed at concentrations of 10, 25,40,80,130,180,250, and 300 mg L À1 of HPLC-grade catechin standard, and good linearity proportional to the concentration analyzed can be observed. Also, in Figure 1 the chromatogram peak corresponding to 1.0 mg L À1 of catechin standard, were peak chromatogram elution of the catechin is an elution time of 2.54 min. In this way, it is possible to confirm that the method developed by Ruiz-Aquino et al. (2021) is feasibility. These conditions of the analysis by chromatography were used to analyze the tannin extracts obtained with ethanol from the different tree species, the results are shown in Figure 2. Figure 2 shows the elution time of the tannin at t ¼ 2.78 min. Impurities or phenolic compounds, such as (À)-epicatechin or (À)-epigallocatechin, were also observed in lower concentrations at t ¼ 3.20. The chromatographic signals of the species Q. laurina and A. xalapensis show that the catechin obtained is of high purity and presents an absorbance of 0.95 and 0.97 for 250 mg L À1 of the tannin obtained. The species Q. crassifolia and P. serotina present an absorbance of 0.71 and 0.92, respectively, and the purity and efficiency of the extraction can be observed when ethanol is used. This extraction efficiency depends on the polarity and the dielectric constant of the solvent (D), which is what determines the ability to dissolve solute molecules through a polarization effect. That is, the ability of the solvent to function as an insulator of electrical charges. Phenolic compounds, such as tannins are soluble in polar solvents due to the presence of hydroxyl groups and can be extracted using acidified solvents with small amounts of hydrochloric or formic acid, where acidification of the extraction solution can prevent oxidation of the phenolic compounds (Ruenroengklin et al. 2008). The effects of different extraction conditions on the polyphenol, flavonoids components, and antioxidant activity of Polyscias fruticosa roots was reported by (Turkmen et al. 2006) where they used 90% ethanol as a polar solvent, acetone as a moderately polar solvent, and hexane as a non-polar solvent. where they used 90% ethanol as a polar solvent, acetone as a moderately polar solvent, and hexane as a non-polar solvent, obtaining high concentrations of polyphenols with 90% ethanol.

Analysis of the tannin extracts by HPLC
With this extraction method in ethanol, we can observe the selectivity of ethanol toward catechin, which is the base compound of condensed tannins, whose structure is based on the union of catechin molecules in the formation of dimers, trimers, and tetramers. These Tannins are of great interest for industrial and biological properties, as shown in Figure 3.

Analysis of extracts by FTIR
Fourier transform infrared (FTIR) spectroscopy provides information on functionalized simple polymers' surface chemistry and properties.
Tannin compounds have a wide application in different technological fields, including food science (Pizzi 2019). Figure 4 shows the infrared spectrum of high-purity catechin, which is used as a reference to analyze and compare the tannin extracts obtained in this study. The FTIR spectra show the main signals related to the different phenols and polymers in the catechin extract.
Generally, the FTIR spectrum shows a strong absorption around 3500 and 3200 cm À1 with a strong broad band centered at 3317.90 cm À1 . These bans are assigned to the stretching vibrations of the hydroxyl (OH) groups attributed to the wide variety of hydrogen bonds between the OH groups. Figure 5 shows a strong signal at 2929.43 and 2852.97 cm À1 associated with the symmetric and antisymmetric -C-Hstretching vibrations of the CH 2 groups, respectively.
The strain vibration of the carbon-carbon bonds in the phenolic groups is absorbed in the 1605-1496 cm À1 . In the range of 1284.81 and 1183.55 cm À1 , the C-O bond tension and the bonds of carboxylic groups are found. At 729-696 cm À1 , it shows the distortional vibration resulting from C¼C in the benzene rings. Under these reference conditions of the catechin standard, the FTIR spectra corresponding to the barks of (a) Q. laurina, (b) A. xalapensis, (c) Q. crassifolia, (d) P. serotina, which are shown in Figure 5, were analyzed.
The FTIR spectra for each tree species show that at 3350 cm À1 , broadband corresponds to the stretching vibration of the hydroxyl groups (OH). However, it can be observed that the P. serotina specie presents a higher transmittance peak than the other tree species because it has a greater amount of OH ions. For tannins characterized by a high degree of polymerization and a low percentage of monomeric units, the maximum is located at 3400 cm À1 (Sartori et al. 2018). The region 1800-1680 cm À1 is vital for qualitative analysis because it exhibits a peak at 1737 cm À1 for carbonyl stretching of hydrolyzable tannins. This weak signal of 1737 cm À1 usually indicates the presence of flavonols in the mixture, or it may also be the oxidation of some OH groups in the flavonol molecules because of the extraction process (Jensen et al. 2008). Some authors assign a specific peak around 1717 cm À1 to a C¼O-H bond interaction of catechin immersed in a polymeric matrix (Zhu et al. 2004), which provides a diagnostic tool for studying interactions that occur in polymers that include condensed tannins. Khedkar et al. (2010) performed a theoretical analysis providing two specific catechin frequencies, showing an isomerization at 1626 cm À1 , which involves a rotation between the C and B rings and a local rearrangement and consequent elongation of some bonds in the benzene structure of catechin. In the FTIR spectrum of each tree species' barks, a welldefined peak at 1626 cm À1 can be observed for each extract analyzed. The region from 1650 to 1400 cm À1 is mainly occupied by vibrational movements of C¼C groups in the aromatic rings. The intensity of this peak is also affected by the C 4 -C 8 interflavonoid bond elongation during the proanthocyanidins condensation process. In this way, a higher intensity can be considered evidence of a prolonged polymerization degree. Thus at 1626 cm À1 , we have an isomerization signal involving a rotation between the C and B rings, a local rearrangement, and an elongation of the aromatic ring.
The region from 1450 to 900 cm À1 is a more complex structure, can be considered the most significant for the description of substituent rings, and is characterized by an envelopment of strong bands that result from the combination of aromatic C-H bending, stretching C-O, and C-OH deformations.
Condensed tannins are the result of a mixture of flavanol monomers, galloflavanol, or flavan-3-ols polymers that are a subgroup of flavonoids (Figure 6)   catechin, and the movement of the resulting C-OH occurs in the catechin with a pronounced peak around 1280 cm À1 , Gallocatechin also exhibits an additional peak near 740 cm À1 , which is not observed for catechin, attributed to bending out of the -CH plane of the B ring. The catechin and gallocatechin molecular groups have the trans configuration at C-2 and C-3 in the heterocyclic ring, where the only significant modification is the variation in the number of OH substituents (Ricci et al. 2015). Furthermore, proanthocyanidin released an anthocyanidin by breaking interflavanic bonds due to acid hydrolysis in alcohol at high temperatures. This is the result of acid hydrolysis in the presence of alcohol at high temperature, the proanthocyanidin releases an anthocyanidin by breaking the interflavanic bonds.

Conclusions
FTIR spectroscopy was a useful technique to determine the signal corresponding to catechin and the characteristic functional groups of polyphenols, and the effectiveness of the extraction process was confirmed. In ascending order, the total extracts were 12.87, 16.38, 19.31, and 25.68% for P. serotina, Q. laurina, Q. crassifolia, and A. xalapensis, respectively. The chromatographic signals of the species Q. crassifolia and P. serotina present an absorbance of 0.71 and 0.92, respectively. Therefore, in these tannin extracts, the purity and efficiency of the extraction can be observed when ethanol is used. Furthermore, the percentage of condensed tannins in the extracts shows Q. crasifolia (10.29%) and A. xalapensis (13.12%). Consequently, these compounds obtained from the bark of woody plants could be exploited on an industrial scale.

Disclosure statement
No potential conflict of interest was reported by the author(s).