Attenuated Total Reflection Mid-Infrared (ATR-MIR) Spectroscopy and Chemometrics for the Identification and Classification of Commercial Tannins

Abstract

Attenuated total reflection Fourier transform infrared (FT-IR) spectroscopy was used to characterize 40 commercial tannins, including condensed and hydrolyzable chemical classes, provided as powder extracts from suppliers. Spectral data were processed to detect typical molecular vibrations of tannins bearing different chemical groups and of varying botanical origin (univariate qualitative analysis). The mid-infrared region between 4000 and 520 cm⁻¹ was analyzed, with a particular emphasis on the vibrational modes in the fingerprint region (1800-520 cm⁻¹), which provide detailed information about skeletal structures and specific substituents. The region 1800–1500 cm⁻¹ contained signals due to hydrolyzable structures, while bands due to condensed tannins appeared at 1300–900 cm⁻¹ and exhibited specific hydroxylation patterns useful to elucidate the structure of the flavonoid monomeric units. The spectra were investigated further using principal component analysis for discriminative purposes, to enhance the ability of infrared spectroscopy in the classification and quality control of commercial dried extracts and to enhance their industrial exploitation.

Keywords

Fourier transform infrared FT-IR Condensed tannins Hydrolyzable tannins Principal component analysis PCA

INTRODUCTION

Tannins are naturally occurring bioactive compounds, produced as secondary metabolites by plants. According to their botanical origin, tannins can be divided into two main classes: (1) hydrolyzable tannins, further divided in ellagitannins (lactones of ellagic acid) and gallotannins (natural polymers of gallic acid esters), which are mainly derived from woody part of plants and fruits, and (2) condensed tannins, commonly known as proanthocyanidins, composed of flavanol monomeric units (catechins) containing galloyl moieties (prodelphinidins) or catechols (procyanidins), which are mainly derived from grape, stone fruits, cocoa, coffee, mimosa, and quebracho tree species.^1,2 Figures 1 and 2 show the molecular structures of monomeric units and related polymers cited in this work.

Fig. 1.

Condensed tannins and their monomeric units.

Tannins find application in different industry settings, including wine making,³ water purification,⁴ ink manufacture, leather and dye industries,⁵ synthesis of metal nanoparticles,⁶ and manufacture of adhesives, surface coatings^7,8 and plastic resins.⁹ In particular, tannins act as natural preservatives, and blending them with adhesives and plastics enhances the technological properties of innovative polymers doped with these natural extracts.^10–12 For example, proanthocyanidins from the common grape (Vitis vinifera) were shown to be particularly effective in providing antioxidant activity while not affecting the mechanical properties of ethyl cellulose films.¹³

In addition to their multiple functional properties, a further advantage of tannins is their large availability due to the widespread use of traditional agricultural practices and an increasing interest in the recovery of agro-wastes for the production of “green” derivatives.^14,15 Despite the large availability of different commercial formulations, there is a need for (1) further information on their chemical composition and (2) rapid analytical techniques for effective discrimination between tannins derived from different botanical sources to further enhance and tailor industrial applications of tannins. Alternative analytical methods, both for hydrolyzable as well as for condensed tannins, such as matrix-assisted laser desorption-ionization time-of-flight mass spectrometer^16–19 (MALDI-TOF), ¹³C nuclear magnetic resonance²⁰ (NMR), and ³¹P NMR^21,22 are widely used to get further chemical information on the structure of tannins.

In this context, attenuated total reflectance (ATR) Fourier transform infrared (FT-IR) spectroscopy, being a simple and sensitive technique, has been widely applied in the field of quality control in industry. The versatility of this technique is also improved by its ability to operate with different matrices (both solid and liquid) without the need for sample pretreatments.²³ Furthermore, ATR infrared (IR) spectroscopy allows analysis of powders directly over the internal reflection element, without dilution in an IR-transparent medium (KBr, Nujol, etc.). The product as provided by suppliers can be directly analyzed in a few seconds.

The data elaboration of a single sample in molecular spectroscopy can be considered as a univariate approach to the study of chemical differences between tannins belonging to different botanical species. This is expensive in terms of time compared with fast and efficient methods required for industrial processes. On the other hand, FT-IR analysis is very effective for discriminative purposes when coupled with multivariate chemometric approaches.^24,25

The aim of this work is to identify IR spectral patterns that typify representative commercial tannins used in industrial applications (univariate approach) and to evaluate the performance of classification algorithms across the selected samples. The effectiveness of spectral and statistical analyses provides a valid contribution in evaluating the authenticity and the degree of purity of these compounds in a fast, reliable, and nondestructive way.

EXPERIMENTAL

Samples Description. Gallic acid, (+)-catechin mono-hydrate, and ellagic acid standards used for reference spectra were purchased from Sigma-Aldrich (Castle Hill, Australia). Forty commercial powder tannins (Table I) were analyzed as provided by suppliers (Enologica Vason, Verona, Italy; HTS Enologia, Marsala, Italy; Laffort, Bordeaux Cedex, France; AEB Group, Brescia, Italy) in order to ascertain the ability of infrared spectroscopy to classify and describe the different samples based on their chemical composition and declared botanical origin.

TABLE I.

Summary of commercial tannins classified according with chemical and botanical origin.

Sample cod	Botanical origin	Chemical description(C, condensed; H, hydrolyzable; M, mixture)
PR1	Green tea	C
PR2	Vitis vinifera leaves	C
PR3	Grape fruit	C
PR4	Grape fruit	C
PR5	Grape skin and seeds	C
PR6	White grape seeds	C
PR7	Grape seeds	C
PR8	Malbec red grape seeds	C
PR9	Quebracho	C
PR10	Not fermented grape skin	C
PR-HY1	Red fruits tree wood	M
PR-HY2	White fruits tree wood	M
PR-HY3	N/A^a	M
PR-HY4	N/A	M
PR-HY5	N/A	M
PR-HY6	N/A	M
PR-HY7	N/A	M
PR-HY8	Oak wood and grape	M
HY-OA1	Oak	H
HY-OA2	Oak	H
HY-OA3	Oak	H
HY-OA4	Oak	H
HY-OA5	Oak	H
HY-OA6	Oak	H
HY-OA7	Oak	H
HY-CH1	Chestnut	H
HY-CH2	Chestnut	H
HY-AD1	American Durmast	H
HY-AD2	American Durmast	H
HY-AD3	American Durmast	H
HY-FD1	French Durmast	H
HY-FD2	French Durmast	H
HY-FD3	French Durmast	H
HY-FD4	French Durmast	H
HY-FD5	French Durmast	H
HY-GEN1	N/A	H
HY-GEN2	N/A	H
HY-GEN3	Red fruits tree wood	H
HY-GEN4	N/A	H
HY-GALL1	Gallnut	H

N/A, information not available.

Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy. Attenuated total reflectance FT-IR analysis was carried out using a diamond ATR Smart Orbit accessory (Thermo Optec), equipped with a deuterated triglycine sulfate detector and a KBr window for measuring in the medium infrared (MIR) region; the incident beam had a 45° geometry with respect to the diamond surface, yielding 25 internal reflections. Tannins were analyzed without any pretreatment or purification; a few milligrams of the powder was placed over the macro ATR crystal, and a slight pressure was applied over the sample to maximize the surface of contact. MIR spectra range from 4000 to 520 cm⁻¹ and were averaged over 128 scans with a resolution of 4 cm⁻¹.

Data Processing and Multivariate Analyses. The ATR spectra in the MIR region were processed using principal component analysis (PCA) and a cluster analysis (CA) unsupervised method as classification tools based on the relationship between botanical sources and molecular structure of the selected tannins. The raw spectroscopic data were subjected to mathematical elaboration: all values reported are from triplicate spectral analyses, every single spectrum was processed performing a linear algorithm for baseline correction and a Savitzky-Golay smoothing filter (nine points). The results were mediated to obtain an average spectrum for peak assignments and subsequent multivariate elaboration. The collection and preprocessing of the spectral data was performed using the Omnic SPE 7.2 software (Thermo Fisher Scientific).

The averaged spectra were subjected to multivariate analysis using the The Unscrambler 9.7 software (Camo Software AS). The different spectral regions were tested for chemometric elaboration, and ultimately the whole spectral region was found to be the best option to maximize the information.

The validation method selected for PCA was the leverage correction on a full model size. The cluster analysis was performed using the correlation algorithm over the vector-normalized spectra. According with the similarity between data, the optimal number of clusters was six, with a satisfactory sum of distances value of 0.24.

RESULTS AND DISCUSSION

Figure 3 shows the spectral profiles in the MIR region for the basic monomeric unit of tannins: (+)-catechin, gallic acid, and ellagic acid. Some differences due to the typical structure of these compounds were found.

Fig. 2.

Hydrolyzable tannins and their monomeric units.

Fig. 3.

ATR FT-IR spectra of the most common monomeric units of tannins, (a) (+)-catechin; (b) gallic acid; (c) ellagic acid.

The vibrational spectrum of catechin-based tannins showed a complex envelope of strong signals in the region 1250–900 cm⁻¹, where band occurrence and intensity is sensitive to the number and position of -OH aromatic substituents (Fig. 3a). Additional information on galloyl moieties and galloylation of monomers could be derived especially from this spectral region. In particular, flavan-3-ol-based structures have a strong peak around 1610–1620 cm⁻¹ as a doublet, due to in-plane bending of the benzene ring for the catechol moieties (ring B) and to the same combination in the resorcinol moieties.^26,27 The relative intensity of this spectral feature is sensitive to elongation of the C4-C8 or C4-C6 interflavonoid linkages, so that it could be used as an index of the degree of polymerization.²⁸

The MIR spectra of gallic acid (Fig. 3b) and ellagic acid (Fig. 3c) showed some similarities, because they belong to the same class of compounds. At the same time, there were some slight differences that can be used to discriminate the subclasses, as follows. A diagnostic region for hydrolyzable tannins is within 1750–1680 cm⁻¹, where the stretching of the C=0 group of the carbonyl moiety occurs, giving a strong peak.^29,30 This band is usually quite weak in condensed tannins, except if oxidation and rearrangement of OH group occurs as a consequence of extraction processes with proanthocyanidins.³¹ A specific peak around 1717 cm⁻¹ was assigned to the interaction of condensed tannins with some additives, usually cellulose.³²

A strong peak was present at around 1318 cm⁻¹ for the spectrum related to ellagic acid (Fig. 3c), and there is a strong doublet for gallic acid in the same spectral region. This signal could be considered a fingerprint band of the phenolic ring, and several authors have observed that it is a typical feature for hydrolyzable tannins.^33,34 This finding is also highlighted in Fig. 4, where the spectra of a mixture of ellagitannins (Fig. 4c), a gallnut extract (Fig. 4b), and a proanthocyanidin extract from grape seed (Fig. 4a) are compared. By contrast, there are few signals of interest in this region for the spectrum of flavonoids, with only weak to very weak bands for catechin (Fig. 3a).

Fig. 4.

ATR FT-IR spectra of three industrial tannins, (a) Proanthocyanidins from grape; (b) gallnut extract; (c) ellagitannins from red fruit tree wood.

The region 1250–900 cm⁻¹ showed diagnostic bands related to the hydroxylation patterns of aromatic rings present in tannins; the analysis of these spectral features confirms the information reported in the literature. A strong band around 1037 cm⁻¹ is diagnostic for the chestnut tannin extracts (C–O stretch of phenolic O–H groups) and is often present in oak tannins at a strong intensity. The 1145–1150 cm⁻¹ band is common to hydrolyzable compounds and is attributable to a combination of C–O stretching and -OH deformation motions of the carboxyl group. Quebracho tannin exhibits typical bands at 1160, 1114, and 1032 cm⁻¹, which allows it to be distinguished from the resorcinol group in grape tannins. Here the change in the monomeric unit, single OH substituent on the A ring in quebracho, produces a slight change in the hydroxylation pattern.^35,36 Condensed tannins also contain an oxygenated heterocyclic ring (ring C) that produces characteristic vibrations: a peak around 1280 cm⁻¹ due to the asymmetric stretching vibration of the pyran ring. When this signal is coupled with strong bands at 1162-1155, 1116-1110, 976, and 844–842 cm⁻¹ the extract can be unambiguously assigned to the condensed tannins group.^35,36

The univariate detection of diagnostic spectral features for industrial extracts was followed by a multivariate analysis of vibrational frequencies in the mid-infrared region. The PC1 component shows the highest explanation of differences between the chemical composition of the extract (78%), followed by PC2 (10%), and so these two variables were taken into account for clustering. Figure 5 shows the PCA-CA diagram with the recognition of the six obtained clusters. The first cluster contained six samples, including all tannins derived from American oak, together with one French oak extract and two hydrolyzable tannins derived from oak. The commercial denomination of European oak concerns two botanical species, Quercus petraea and Quercus robur, whereas American oak generally refers to the white oak (Quercus ssp. and Quercus alba). It is noticeable that from the botanical point of view, American and European oaks exhibit different MIR properties, which reflects their technological applications. American oak is richer in aromatic precursors such as the cis-trans isomers of 13-methyl-γ-octalactone, but the quantity of extractable polyphenols is very low compared with the European variety. Furthermore, the European variety Quercus robur is particularly high in ellagitannin content, and the quality could be considered higher according with the specific technological application.³⁷ From this perspective, the ability of the multivariate model to discriminate the two species could be useful to avoid adulteration or dilution in the extracts.

Fig. 5.

PC1/PC2 score plot using PCA analysis and sample grouping based on cluster analysis.

The second cluster (Fig. 5) grouped the gallnut tannins coupled with a hydrolyzable tannin of undeclared origin (sample HY-GEN2), a vegetable extract with high antioxidant activity, most probably with gallic acid monomers arranged around a simple sugar (tannic acid–like structure). Figure 6 shows how the ATR FT-IR spectra of these two extracts overlapped in the fingertip region (1820-520 cm⁻¹), confirming a perfect match for the two spectral profiles.

Fig. 6.

Comparison between gallnut extract (gray line) and sample HY-GEN2, a hydrolyzable tannin (black line).

The CA approach was particularly effective for the discrimination of condensed tannins. All procyanidins derived from Vitis vinifera (fruit pulp, seed, skin) were grouped in cluster 3, together with a sample described as a mixture of ellagic and catechin tannins (Fig. 5), of unknown purity and ratio of the two components. The robustness of the CA model was further confirmed by its ability to distinguish in a group, namely, cluster 6, the condensed tannins derived from exotic botanical sources, such as green tea (PR1) and quebracho (PR9). It is well known that the extent of polymerization and spatial arrangement of condensed tannins are strongly sensitive to the botanical source.³⁸ The most common condensed tannins are procyanidins and prodelphinidins from Vitis vinifera, and profisetinidins and prorobinetinidins for quebracho. The content and relative ratios of these compounds provide useful information about the botanical origin of commercial tannins. Procyanidins are found in grape skins and seeds, whereas prodelphinidins are present in grape skins only.³⁹ In particular, the fisetinidol and robinetinidol subunits of quebracho polymers differ from grape monomers, since they are missing a phenolic group in the C5 position of the resorcinol ring.⁴⁰

Quebracho tannin is mainly composed of profisetinidin, a linear polymer resulting from a combination of catechins and gallocatechins, namely, a proanthocyanidin blended with hydrolyzable tannins, as observed through MALDI-TOF analyses.⁴¹ Therefore, quebracho also contains some gallic acid, which could explain the inclusion in cluster 6 of the samples PR-HY1 and PR-HY7.

Grape tannins are random polymers mainly composed of procyanidins and galloylated procyanidins, having an average degree of polymerization of up to five for seeds and higher for skin extracts.^42,43 Polymerization usually follows a linear C4-8 sequence (interflavonoid linkage), but branching of monomeric units through the lateral moieties (usually C4-6) can occur.⁴⁴

Green tea is mainly composed of monomeric units of epigallocatechin gallate (EGCG), followed by epicatechin-3-gallate, epigallocatechin, epicatechin, and catechin.^45,46 Based on the chemical structures, the grouping of tea tannins with quebracho tannins was unexpected and supports the idea that gallic acid, galloyl, and gallate structures were present in both quebracho and tea. The spectral similarities between the quebracho and the green tea tannins over the region 1300–1000 cm⁻¹ could then be related to the hydroxylation patterns of aromatic molecules. At the same time, the absence of tannic acidlike structures in green teas makes the above hypothesis somewhat controversial.⁴⁶

Cluster 4 included two odd tannins: a catechin-based extract from Vitis vinifera leaves coupled with an ellagitannin from toasted oak (Fig. 5). Since there is no clear link between clustering and available information, the two tannins of cluster 4 should presently be considered as outliers.

Cluster 5 was the most abundant, including eight ellagitannins, three French oak tannins (HY-FD2, 3, 4), two chestnut tannins (HY-CH1 and 2), and six mixtures of condensed and hydrolyzable tannins. Grouping of these samples indicates that hydrolyzable are the prominent tannins.

CONCLUSIONS

Forty commercial powder tannins were analyzed with ATR FT-IR spectroscopy, and the resulting MIR spectra provided valuable chemical information on tannin composition. This analysis was effective as a univariate approach, with the detection of typical vibrational bands related to specific molecular features, and multivariate analysis, for classification purposes. The univariate approach focused mainly on the fingertip region (1800-520 cm⁻¹), where diagnostic peaks were identified for discriminative purposes. The spectral region 1750–1680 cm⁻¹ contains strong bands that are unambiguously attributable to stretching vibrations of carbonyl moieties that are typical of hydrolyzable tannins. Condensed tannins do not show significant bands in this region, while they exhibit a typical doublet around 1610–1620 cm⁻¹, related to catechol and resorcinol moieties. The region around 1250 to 900 cm⁻¹ provides information about hydroxylation patterns, so that it could be used both to discriminate between hydrolyzable and condensed tannins and to obtain further discrimination between catechins and gallate catechins. principal component and clustering analyses were found to be effective tools for the classification of tannins based on chemical and botanical sources, especially in the case of proanthocyanidins from Vitis vinifera, profisetinidin from quebracho, and ellagitannins from white oak. The minor concerns about the nature of green tea tannins and the presence of a few outliers that were not properly described by the model would require additional samples of known origin to further improve this preliminary dataset.

Footnotes

ACKNOWLEDGMENTS

We gratefully acknowledge Enologica Vason, HTS enologia, Laffort and AEB group for providing tannins for this research. Authors Arianna Ricci, Giuseppina Parpinello, and Andrea Versari also acknowledge the School of Chemical Sciences of the University of Auckland, where they were guest scientists, and Arianna Ricci for the support of a Ph.D. fellowship from University of Bologna.

References

Haslam

. “Vegetable Tannins”. In: Swain

Harbone

J.B.

Van Sumere

C.F.

, editors. Biochemistry of Plant Phenolics. New York: Springer, 1979.

Haslam

. “Plant Polyphenols: Vegetable Tannins Revisited”. In: Phillipson

J.D.

Ayres

D.C.

Baxter

, editors. Chemistry and Pharmacology of Natural Products. Cambridge, UK: Cambridge University Press, 1989.

Versari

du Toit

Parpinello

G.P.

. “Oenological Tannins: A Review”. Aust. J. Grape Wine Res. 2013. 19(1): 1–10.

Thakur

S.S.

Choubey

. “Use of Tannin Based Natural Coagulants for Water Treatment: An Alternative to Inorganic Chemicals”. Int. J. ChemTech Res. 2014. 6(7): 3628–3634.

Ramakrishnan

Krishnan

M.R.V.

. “Tannin—Classification, Analysis and Applications”. Ancient Sci. Life. 1994. 13(3-4):232–238.

Ahmad

. “Reviewing the Tannic Acid Mediated Synthesis of Metal Nanoparticles”. J. Nanotechnol. 2014. 3: 1–11.

Matsuo

Kamigama

Saitoh

. “Adsorption Characteristics of Immobilized Kaki-Tannin and Industrial Application”. Acta Hortic. (ISHS). 1995. 398: 285–296.

Bisanda

E.T.N.

Ogola

W.O.

Tesha

J.V.

. “Characterization of Tannin Resin Blends for Particle Board Applications”. Cement Concrete Compos. 2003. 25(6): 593–598.

Zhu

Abhyankar

Nassiopoulos

Njuguna

. “Tannin-Based Flax Fibre Reinforced Composites for Structural Applications in Vehicles”. IOP Conference Series: Mater. Sci. Eng. 2012. 40: 1–8.

10.

Ballerini

Despres

Pizzi

. “Non-Toxic, Zero Emission Tannin-Glyoxal Adhesives for Wood Panels”. Holz Roh- Werkst. 2005. 63(6): 477–478.

11.

Moubarik

Pizzi

Allal

Charrier

. “Cornstarch and Tannin in Phenol-Formaldehyde Resins for Plywood Production”. Ind. Crop. Prod. 2009. 30(2): 188–193.

12.

Rivero

Garcia

Pinotti

. “Crosslinking Capacity of Tannic Acid in Plasticized Chitosan Films”. Carbohydr. Polym. 2010. 82(2): 270–276.

13.

Olejar

K.J.

Ray

Ricci

Kilmartin

P.A.

. “Superior Antioxidant Polymer Films Created Through the Incorporation of Grape Tannins in Ethyl Cellulose”. Cellulose. 2014. 21(6): 4545–4556.

14.

Gandini

. “Polymers from Renewable Resources: A Challenge for the Future of Macromolecular Materials”. Macromolecules. 2008. 41(24): 9491–9504.

15.

C.S.

Mun

S.P.

. “Antioxidant Activities of Ethanol Extracts from Seeds in Fresh Bokbunja (Rubus coreanus Miq.) and Wine Processing Waste”. Bioresour. Technol. 2008. 99(10): 4503–4509.

16.

Pizzi

Pasch

Rode

Giovando

. “Polymer Structure of Commercial Hydrolyzable Tannins by MALDI-TOF Mass Spectrometry”. J. Appl. Polym. Sci. 2008. 113(6): 3847–3859.

17.

Radebe

Rode

Pizzi

Pasch

. “Microstructure Elucidation of Polyflavonoid Tannins by MALDI-TOF-CID”. J. Appl. Polym. Sci. 2013. 127(3): 1937–1950.

18.

Radebe

Rode

Pizzi

Giovando

Pasch

. “MALDI-TOF-CID for the Microstructure Elucidation of Polymeric Hydrolysable Tannins”. J. Appl. Polym. Sci. 2013. 128(1): 97–107.

19.

Giovando

Pizzi

Pasch

Pretorius

. “Structure and Oligomers Distribution of Commercial Tara (Caesalpinia spinosa) Hydrolysable Tannin”. Pro Ligno. 2013. 9(1): 22–31.

20.

Navarrete

Pizzi

Pasch

Rode

Delmotte

. “MALDI-TOF and ¹³C NMR Characterisation of Maritime Pine Industrial Tannin Extract”. Ind. Crops Prod. 2010. 32(2): 105–110.

21.

Melone

Saladino

Lange

Crestini

. “Tannin Structural Elucidation and Quantitative ³¹P NMR Analysis. 1. Model Compounds”. J. Agric. Food Chem. 2013. 61(39): 9307–9315.

22.

Melone

Saladino

Lange

Crestini

. “Tannin Structural Elucidation and Quantitative ³¹P NMR Analysis. 2. Hydrolyzable Tannins and Proanthocyanidins”. J. Agric. Food Chem. 2013. 61 (39): 9316–9324.

23.

Tondi

Petutschnigg

. “Middle Infrared (ATR FT-MIR) Characterization of Industrial Tannin Extracts”. Ind. Crop. Prod. 2015. 65: 422–428. doi:10.1016/j.indcrop.2014.11.005.

24.

Regmi

Palma

Barroso

C.G.

. “Direct Determination of Organic Acids in Wine and Wine-Derived Products by Fourier Transform Infrared (FT-IR) Spectroscopy and Chemometric Techniques”. Anal. Chim. Acta. 2012. 732: 137–144.

25.

Koch

Posch

A.E.

Goicoechea

H.C.

Herwig

Lendl

. “Multi-Analyte Quantification in Bioprocesses by Fourier-Transform-Infrared Spectroscopy by Partial Least Squares Regression and Multivariate Curve Resolution”. Anal. Chim. Acta. 2014. 807: 103–110.

26.

Murugananthan

Bhaskar Raju

Prabhakar

. “Removal of Tannins and Polyhydroxy Phenols by Electro-Chemical Techniques”. J. Chem. Technol. Biotechnol. 2005. 80(10): 1188–1197.

27.

Socrates

. Infrared and Raman Characteristic Group Frequencies: Tables and Charts. New York: John Wiley and Sons, 2004. 3rd ed.

28.

Kim

H.-J.

. “Curing Behavior and Viscoelastic Properties of Pine and Wattle Tannin-Based Adhesives Studied by Dynamic Mechanical Thermal Analysis and FT-IR-ATR Spectroscopy”. J. Adhes. Sci. Technol. 2003. 17(10): 1369–1383.

29.

Jensen

J.S.

Egebo

Meyer

A.S.

. “Identification of Spectral Regions for the Quantification of Red Wine Tannins with Fourier Transform Mid-Infrared Spectroscopy”. J. Agr. Food Chem. 2008. 56(10): 3493–3499.

30.

Ping

Gambier

Pizzi

Guo

Z.D.

Brosse

. “Wood Adhesives from Agricultural By-Products: Lignins and Tannins for the Elaboration of Particleboards”. Cellul. Chem. Technol. 2012. 46(7-8):457–462.

31.

Ogata

Nakano

. “Mechanisms of Gold Recovery from Aqueous Solutions Using a Novel Tannin Gel Adsorbent Synthesized from Natural Condensed Tannin”. Water Res. 2005. 39(18): 4281–4286.

32.

Zhu

Yamane

Kimura

Nishida

Inoue

. “Effect of Steric Hindrance on Hydrogen-Bonding Interaction Between Polyesters and Natural Polyphenol Catechin”. J. Appl. Polym. Sci. 2004. 91(6): 3565–3573.

33.

Laghi

Parpinello

G.P.

Del Rio

Calani

Mattioli

A.U.

Versari

. “Fingerprint of Enological Tannins by Multiple Techniques Approach”. Food Chem. 2010. 121(3): 783–788.

34.

Özacar

Şengil

I.A.

Türkmenler

. “Equilibrium and Kinetic Data, and Adsorption Mechanism for Adsorption of Lead onto Valonia Tannin Resin”. Chem. Eng. J. 2008. 143(1): 32–42.

35.

Falcão

Araújo

M.E.M.

. “Tannins Characterization in Historic Leathers by Complementary Analytical Techniques ATR-FTIR, UV-Vis and Chemical Tests”. J. Cult. Herit. 2013. 14(6): 499–508.

36.

Falcão

Araújo

M.E.M.

. “Application of ATR-FTIR Spectroscopy to the Analysis of Tannins in Historic Leathers: The Case Study of the Upholstery from the 19th Century Portuguese Royal Train”. Vib. Spectrosc. 2014. 74: 98–103.

37.

Chatonnet

Dubourdieu

. “Comparative Study of the Characteristics of American White Oak (Quercus alba) and European Oak (Quercus petraea and Q. robur) for Production of Barrels Used in Barrel Aging of Wines”. Am. J. Enol. Vitic. 1998. 49(1): 79–85.

38.

Marles

M.A.S.

Ray

Gruber

M.Y.

. “New Perspectives on Proanthocyanidin Biochemistry and Molecular Regulation”. Phyto-chemistry. 2003. 64(2): 367–383.

39.

Vivas

Nonier

M.-F.

de Gaulejac

N.V.

Absalon

Bertrand

Mirabel

. “Differentiation of Proanthocyanidin Tannins from Seeds, Skins and Stems of Grapes (Vitis vinifera) and Heartwood of Quebracho (Schinopsis balansae) by Matrix-Assisted Laser Desorption/lonization Time-of-Flight Mass Spectrometry and Thioacidolysis/Liquid Chromatography/Electrospray Ionization Mass Spectrometry”. Anal. Chim. Acta. 2004. 513(1): 247–256.

40.

Pizzi

. Advanced Wood Adhesives Technology. Boca Raton, FL: CRC Press, 1994.

41.

Pizzi

Pasch

Rode

Giovando

. “Polymer Structure of Commercial Hydrolyzable Tannins by Matrix-Assisted Laser Desorption/lonization-Time-of-Flight Mass Spectrometry”. J. Appl. Polym. Sci. 2009. 113(6): 3847–3859.

42.

Herderich

M.J.

Smith

P.A.

. “Analysis of Grape and Wine Tannins: Methods, Applications and Challenges”. Aust. J. Grape Wine Res. 2005. 11(2): 205–214.

43.

Perret

Pezet

Tabacchi

. “Fractionation of Grape Tannins and Analysis by Matrix-Assisted Laser Desorption/lonisation Time-of-Flight Mass Spectrometry”. Phytochem. Anal. 2003. 14(4): 202–208.

44.

Hanlin

Hrmova

Harbertson

Downey

. “Review: Condensed Tannin and Grape Cell Wall Interactions and Their Impact on Tannin Extractability into Wine”. Aust. J. Grape Wine Res. 2010. 16(1): 173–188.

45.

Frazier

R.A.

Deaville

E.R.

Green

R.J.

Stringano

Willoughby

Plant

Mueller-Harvey

. “Interactions of Tea Tannins and Condensed Tannins with Proteins”. J. Pharm. Biomed. Anal. 2010. 51(2): 490–495.

46.

Graham

H.N.

. “Green Tea Composition, Consumption, and Polyphenol Chemistry”. Prev. Med. 1992. 21(3): 334–350.