14
Flavan‐3‐ols and Condensed Tannin

14.1 Introduction

Amongst flavonoids, the class present at the largest quantity in grapes is the flavan‐3‐ols, with significant amounts in the seeds as well as in the berry skins. Flavanols are widespread in foods, with the major sources in the Western diet being chocolate, tea, and apples. The use of the number “3” in the name refers to the position of the alcohol (Figure 14.1). Flavanols with an alcohol group at position 4 do exist at trace levels, though these compounds are highly reactive and rarely reported. The flavan‐3‐ols are notable for the fact that both positions 2 and 3 can have isomers with cis and trans forms relative to the attached B ring, a situation unique to this subclass. This gives rise to the existence of variants of the C ring, with the cis form noted by the prefix “epi‐” when two natural forms exist. The natural forms have (2R) stereochemistry, but in the acid conditions of wine, C2 can equilibrate due to acid protonation on the C‐ring oxygen. Thus, in wine, the (2S) form exists as well, leading to partial racemization of the catechins [1].

Figure 14.1 Parent ring system for flavan‐3‐ols

On the B ring, there are only two variants observed among flavanols the “normal” 3ʹ,4ʹ‐dihydroxy substitution, and the “gallo” 3ʹ,4,ʹ5ʹ‐ version. Consequently, the flavan‐3‐ol with cis substitution on the C ring and trihydroxy substituents on the B ring is epigallocatechin. There is also substitution at the 3 position, and the flavan‐3‐ols have gallic acid esters depicted in some examples in Figure 14.2. Glycosides of flavan‐3‐ols have been detected but not quantified, and appear to be present at low levels [2]. There are five different monomeric flavanols found in grapes (Figure 14.2). The distribution of the flavanols in grape berries is not the same in all varieties, and will also vary between seed and skin [3] (Table 14.1).

Figure 14.2 Grape‐derived monomeric flavan‐3‐ols in wine

Table 14.1 Flavan‐3‐ol monomer wine composition

Winegrape variety	Catechin	Epicatechin	EGC	ECG	GC	Reference
Tempranillo	16	10	nr	nr	nr	[6]
Graciano	33	34	nr	nr	nr
Cabernet Sauvignon	42	20	nr	nr	nr
Merlot	27	19	nr	nr	nr
Cabernet Sauvignon	19	58	20	2	nr	[7]
Tannat	43	65	60	0	14	[8]
Nero D’Avola	25	32	nr	nr	nr	[9]
Cabernet Sauvignon	38	16	nr	nr	nr	[10]
Syrah	43	51	nr	nr	nr
Tempranillo	27	54	nr	nr	nr

EGC = epigallocatechin, ECG = epicatechin gallate, GC = gallocatechin, nr = not reported.

14.2 Monomeric catechins

The levels of catechin and epicatechin in Cabernet Sauvignon wine have been reported to be 37–80 mg/L, with the major proportion usually being catechin (Table 14.1, Figure 14.2) [4]. In contrast, epigallocatechin, gallocatechin, and epicatechin gallate have been reported in small amounts in wine [5]. While these are relatively minor components of wine, they are useful as markers for phenolic extraction, that is, from skins and seeds, since the analysis of these compounds is specific and can be precisely measured. This is in contrast with measurements of the oligomers and especially polymers of flavan‐3‐ols, see below. The analysis of these monomeric flavanols is typically via direct HPLC separation of wine samples [4].

While the catechins are minor components, they are also often used as models for the reactions expected of the abundant oligomers and polymers. Many of the reactions described for flavan‐3‐ols in general are based on reactions of the catechins. For instance, reactions with acetaldehyde were first demonstrated with catechin to form bridged dimers [11], and catechin was used to describe the oxidation of flavanols [12]. Catechin and epicatechin have been reported to isomerize via ring‐C opening at high temperatures to yield the enantiomer of the epimer. For instance, (+) catechin isomerizes to (+) epicatechin [13]; however, no enantiomers were observed in new wine pomace [14].

14.3 Oligomeric proanthocyanidins and polymeric condensed tannins

A significant portion, 25–50%, of phenolic compounds in a typical red wine exist as oligomers (proanthocyanidins with distinguishable components of different degrees of polymerization) and polymers (condensed tannins) of flavan‐3‐ols [15]. These are likely formed through the biochemical condensation of flavan‐3‐ol units, although the enzymes responsible for forming these linkages, if they exist, are still not elucidated [16], although the genes related to proanthocyanidin production are coming into focus [17]. The condensation forms covalent bonds between flavan‐3‐ol subunits, the most common linkages being 4 → 8 (i.e., procyanidin B1) and 4 → 6 positions (i.e., procyanidin B5) (Figure 14.3). Epicatechin is the predominant unit in condensed tannins from grapes and wine and catechin is the next most abundant (often found at the terminal position, that is, those units with no bond at the 4 position).

Figure 14.3 Condensed forms of flavan‐3‐ols

Proanthocyanidin is the overarching name given to the class that encapsulates the procyanidins and the prodelphinidins. These names arise from the fact that when these substances are treated with strong mineral acid, they break down into two specific anthocyanidins with the related substitution on the B ring, that is, the 3ʹ,4ʹ‐dihydroxy catechin and epicatechin yield cyanidin (procyanidins), while the 3ʹ,4ʹ,5ʹ‐trihydroxy subunits in gallocatechin yield delphinidin (prodelphinidins) (Chapter 16) (Figure 14.4) [18].¹ Only two products are found because this treatment hydrolyzes the gallate esters and eliminates stereochemical differences (normal versus epi) on the C ring (Figure 14.4). This type of method is commonly used to measure the amount of proanthocyanidin in supplements, to give “OPC” values that are calibrated based on cyanidin absorbance. A milder degradation, based on dilute acid‐catalyzed cleavage of the 4 → 8 and 4 → 6 interflavan bonds, followed by nucleophilic trapping of the reactive carbocation at position 4, yields all subunits with their original stereochemical configuration and C3 ester substitution retained [19] (see Table 14.2 for sample data). Apart from subunit composition, this method can also provide an estimate of the average degree of polymerization (DP) by comparing the amount of released terminal units that are not modified versus the released extension units that are modified at C4. For instance, a tetramer will yield three modified extension units and one terminal unit. Due to the greater amount of information, this method is preferred and widely used. Indeed, proanthocyanidin cleavage in the presence of a nucleophile (e.g., phloroglucinol = phloroglucinolysis or benzyl mercaptan = thiolysis) can be used to compare the ratio of seed‐derived gallate ester and skin‐derived epigallocatechin in order to assess seed versus skin extraction in winemaking [20]. A limitation of the method is seen with samples where the proanthocyanidin chain has a large amount of A‐type linkages or been modified by oxidation or anthocyanin reactions, and in these cases the treatment has low yields of the normal products (Chapter 24). The same chemical reaction pathway is important in the continuous rearrangement of proanthocyanidins in wine. See the later boxed section on electrophilic aromatic substitution.

Figure 14.4 Reactions to analyze the proanthocyanidin structure. Strong acid product with oxidation shown on top right, nucleophile (thiolysis or phloroglucinolysis) product on top left. R = H or gallic acid

Table 14.2 Proanthocyanidin composition of Cabernet Sauvignon wine, subunit composition of fractions by degree of polymerization [32]. Separation by polar LC followed by phloroglucinolysis of the fractions

DP	mg/La	Extension units				Terminal
		EGC	Cat	EC	ECG	Cat	EC
5	37.6	33.0	4.0	60.4	2.6	81.6	18.4
6	60.1	32.5	4.0	61.1	2.5	81.3	18.7
7	60.7	32.3	4.2	60.9	2.7	82.0	18.0
8	39.0	32.4	3.4	61.5	2.8	82.3	17.7
9	189.3	36.0	2.9	58.4	2.8	80.0	20.0
10	18.9	36.6	3.0	57.6	2.8	79.3	20.7
11	14.7	35.6	3.3	58.0	3.2	76.2	23.8
12	6.4	34.2	2.8	60.1	2.9	75.0	25.0
13	26.8	34.0	2.9	60.0	3.1	75.7	24.3
14	16.0	37.6	2.9	56.6	2.9	77.5	22.5
15	124.0	39.6	2.4	54.7	3.3	81.0	19.0
Total	593.5

^aBy HPLC.

Grape seeds can also contain A‐type proanthocyanidins, where there is an additional bond between the oxygen at the C‐7 position of one subunit and the C‐2 position on the unit “above” [21,22] (Figure 14.3). A‐type are also found as a result of oxidation under model conditions [23], so oxidative aging may contribute to additional A‐type proanthocyanidins.

14.3.1 Role of proanthocyanidins in wine

The proanthocyadins are quality indicators in grape and wine samples and have several key roles:

• They may react with anthocyanins to form some of the stable wine pigments found in aged red wine (Chapters 16 and 25).
• As high‐concentration phenolic compounds, they can both accelerate the rate of oxygen consumption and react with products of oxidation (Chapter 24). Oxidation yields o‐quinone products from the catechol groups on the B ring, which then react with other phenols, primarily via the nucleophilic A ring to create new bonds.
• With the nucleophilicity of the A ring, these compounds will react with numerous oxidation products, such as quinones and aldehydes, essentially scavenging oxidation products and avoiding their accumulation.
• They are highly correlated with the perception of astringency in red wines (i.e., the perceived loss of lubrication in the mouth), as described in Section 14.4.

14.3.2 Proanthocyanidin concentrations and measurements in grapes and wine

Concentrations of proanthocyanidins in grapes are typically reported to be in the range of 0.5–1.5 g/L, with the caveat that the amount is method dependent – see below. Proanthocyanidins are incompletely extracted during fermentation (Chapter 21), and red wine proanthocyanidin concentrations are typically <50% of concentrations in grapes. As discussed later, red wine proanthocyanidin concentrations are often poorly correlated with their corresponding grape concentrations, due to differences in maceration approach and tannin extractability across grapes. White wine proanthocyanidin concentrations are substantially lower, being in the range of 10–50 mg/L, with higher concentrations usually observed in harder press fractions.

As a caveat, comparison of proanthocyanidin values across the literature is complicated because several methods for their measurement exist [24]. This problem arises not only because of the complexity of proanthocyanidins, but also because there is no standard reference material available or even a standardized definition. Although proanthocyanidins absorb strongly at 280 nm, direct measurement using a single UV wavelength is not appropriate because of the presence of other phenolics. Published methods usually specify an isolation or separation step [25], followed by one of several types of procedures:

• Classic tannin analyses rely on condensation reactions with aldehydes such as vanillin to yield a colored product [26]. These approaches will detect monomers as well, and are not often reported for wine analysis.
• Methods that precipitate and isolate proanthocyanidins using protein [27] or polysaccharides [28] prior to UV‐vis measurement. These methods correlate well to perceived astringency, described in more detail below.
• Direct HPLC measurement using polar or gel permeation columns [29, 30], both which separate based on molecular size.
• Approaches that degrade the proanthocyanidin into modified monomers prior to HPLC analysis, for example, phloroglucinolysis as described above.
• Finally, secondary methods that use UV‐vis spectral measurements combined with multivariate statistics, and are calibrated against the aforementioned methods (Chapter 32) [31].

The correlation between any two of these analytical methods among a group of wines varies – for example, normal phase HPLC and protein precipitation show a strong correlation (r = 0.93), but much lower correlations (r < 0.7) are observed with an aldehyde‐based reagent [33]. Additionally, even when methods show strong correlations, they can still differ by a constant factor, for example, using methylcellulose in place of a protein as a precipitant to isolate proanthocyanidins results in 3‐fold higher measured values [34]. Correlations among methods are even weaker in grapes [35], presumably because of greater variation in proanthocyanidin extraction in the presence of grape solids. As a result, we recommend caution in comparing absolute proanthocyanidin concentrations across the numerous studies regarding the effects of winemaking or grapegrowing conditions on these compounds, for example, Reference [36], especially when methods differ.

14.4 Sensory effects

The monomers are bitter and astringent, and as the DP increases to dimers and trimers, bitterness decreases and astringency increases [37, 38]. LC‐Taste studies of red wine have identified the polymeric fraction, termed “procyanidins,” as responsible for the majority of wine’s astringency, but also noted different sensory responses in the smaller and larger molecular weight subfractions [39]. As mentioned above, some analytical methods for tannins are based on the interaction between protein [27] or modified carbohydrates [28] – these methods typically show excellent correlation (r² > 0.8) between perceived astringency and measured tannin [34]. Aging results in the hydrolysis of tannins and their reaction with other wine components (Chapter 25). The resulting modified tannins are more hydrophobic, yield low DP values by phloroglucinolysis, and are less astringent [40]. Other challenges in tannin sensory are discussed later (Chapter 33).

---

Chemical principles: electrophilic aromatic substitution

One of the most important reactions for aromatic rings is electrophilic aromatic substitution  (EAS). In this reaction an electrophile, usually possessing a positive charge, reacts at an unsubstituted position on a benzene ring to displace a proton. The reverse can also occur, with a proton displacing an electrophile.

The tendency for this reaction to occur is based on the reactivity of the electrophile, but also on the nucleophilic properties of the aromatic ring. The latter is based on the other substituents on the ring, and whether they contribute to and activate or withdraw from and deactivate the electron density of the benzene ring. Electrophiles will react more quickly with molecules that have a higher electron density. Oxygen as a phenol or alkylated substituent is a strong activator, with anisole reacting 100 times faster than benzene.

On the A‐ring of flavonoids, there are three oxygens, making this very reactive towards electrophiles. In addition, their alternating positions enhance the effect, which can be rationalized by the resonance structures of the transition state of the reaction. At the transition state, the electrophile has bonded to the ring and the positive charge is now distributed around the aromatic ring. Since the charge is concentrated at positions 2, 4, and 6 relative to the point of reaction, substituents that can stabilize the positive charge at those positions have the most impact. Oxygen can provide stabilization from its lone pair electrons into the aromatic pi orbitals. Thus, the alternating oxygens on the A ring create one of the most reactive aromatic rings to electrophilic aromatic substitution.

Resonance structures of electrophile addition to a phloroglucinal moiety, showing the charge delocalization that stabilizes the intermediate

Two important examples of EAS in wine chemistry involve the flavan‐3‐ols, the most reactive to electrophiles. When protonated by the acid in wine, acetaldehyde becomes a potent electrophile and reacts at the 6 and 8 positions of flavan‐3‐ols, creating a new covalent bond. The benzylic alcohol product will be protonated at wine pH, resulting in a benzylic carbon. This carbon is a very good electrophile, so a second EAS occurs, resulting in a bridge between two flavanol systems. Nearly any aldehyde can react in a similar manner, and this EAS reaction is the basis of the vanillin method for quantifying condensed tannin (although the vanillin does not create bridged structures). A second example is the cleavage and re‐formation of proanthocyanidin chains [41]. The bond from the 4 position on one catechin unit to the 8 position on another can be cleaved by EAS by a proton from acid. The released benzylic cation can then react with the A ring of a different catechin unit elsewhere, resulting in rearrangement of the proanthocyanidin chains. When this reaction occurs with a large amount of a monomer, the monomer will dominate the terminal position in the equilibrated sample [42], and the mean degree of polymerization of the proanthocyanidins will decrease. If other nucleophiles are present, additional products can be formed [43]. One example would be anthocyanin (see Chapter 16).

Example of acid‐catalyzed re‐formation of a proanthocyanidin oligomer, a process that is happening constantly in wine

---

References

1. 1. Nay, B., Monti, J.P., Nuhrich, A., et al. (2000) Methods in synthesis of flavonoids. Part 2: High yield access to both enantiomers of catechin. Tetrahedron Letters, 41 (47), 9049–9051.
2. 2. Delcambre, A. and Saucier, C. (2012) Identification of new flavan‐3‐ol monoglycosides by UHPLC‐ESI‐Q‐TOF in grapes and wine. Journal of Mass Spectrometry, 47 (6), 727–736.
3. 3. Mattivi, F., Vrhovsek, U., Masuero, D., Trainotti, D. (2009) Differences in the amount and structure of extractable skin and seed tannins amongst red grape varieties. Australian Journal of Grape and Wine Research, 15 (1), 27–35.
4. 4. Ritchey, J.G. and Waterhouse, A.L. (1999) A standard red wine: monomeric phenolic analysis of commercial Cabernet Sauvignon wines. American Journal of Enology and Viticulture, 50 (1), 91–100.
5. 5. Ricardo da Silva, J.M., Rosec, J.P., Bourzeix, M., Heredia, N. (1990) Separation and quantitative determination of grape and wine procyanidins by high‐performance reversed phase liquid chromatography. Journal of the Science of Food and Agriculture, 53 (1), 85–92.
6. 6. Monagas, M., Suarez, R., Gomez‐Cordoves, C., Bartolome, B. (2005) Simultaneous determination of nonanthocyanin phenolic compounds in red wines by HPLC‐DAD/ESI‐MS. American Journal of Enology and Viticulture, 56 (2), 139–147.
7. 7. Canals, R., Llaudy, M.D.C., Canals, J.M., Zamora, F. (2008) Influence of the elimination and addition of seeds on the colour, phenolic composition and astringency of red wine. European Food Research and Technology, 226 (5), 1183–1190.
8. 8. Boido, E., Garcia‐Marino, M., Dellacassa, E., et al. (2011) Characterisation and evolution of grape polyphenol profiles of Vitis vinifera L. cv. Tannat during ripening and vinification. Australian Journal of Grape and Wine Research, 17 (3), 383–393.
9. 9. La Torre, G.L., Saitta, M., Vilasi, F., et al. (2006) Direct determination of phenolic compounds in Sicilian wines by liquid chromatography with PDA and MS detection. Food Chemistry, 94 (4), 640–650.
10. 10. Gutierrez, I.H., Lorenzo, E.S.P., Espinosa, A.V. (2005) Phenolic composition and magnitude of copigmentation in young and shortly aged red wines made from the cultivars, Cabernet Sauvignon, Cencibel, and Syrah. Food Chemistry, 92 (2), 269–283.
11. 11. Saucier, C., Little, D., Golries, Y. (1997) First evidence of acetaldehyde‐flavanol condensation products in red wine. American Journal of Enology and Viticulture, 48 (3), 370–373.
12. 12. Guyot, S., Cheynier, V., Souquet, J.‐M., Moutounet, M. (1995) Influence of pH on the enzymatic oxidation of (+)‐catechin in model systems. Journal of Agricultural and Food Chemistry, 43 (9), 2458–2462.
13. 13. Wang, H.F. and Helliwell, K. (2000) Epimerisation of catechins in green tea infusions. Food Chemistry, 70 (3), 337–344.
14. 14. Rockenbach, I.I., Jungfer, E., Ritter, C., et al. (2012) Characterization of flavan‐3‐ols in seeds of grape pomace by CE, HPLC‐DAD‐MSⁿ and LC‐ESI‐FTICR‐MS. Food Research International, 48 (2), 848–855.
15. 15. Singleton, V.L. (1992) Tannins and the qualities of wines, in Plant Polyphenols (eds Hemingway, R.W. and Laks, P.E.), Plenum Press, New York, pp. 859–880.
16. 16. Dixon, R.A., Xie, D.Y., Sharma, S.B. (2005) Proanthocyanidins – a final frontier in flavonoid research? New Phytologist, 165 (1), 9–28.
17. 17. Hancock, K.R., Collette, V., Fraser, K., et al. (2012) Expression of the R2R3‐MYB transcription factor TaMYB14 from Trifolium arvense activates proanthocyanidin biosynthesis in the legumes Trifolium repens and Medicago sativa. Plant Physiology, 159 (3), 1204–1220.
18. 18. Porter, L.J., Hrstich, L.N., Chan, B.G. (1986) The conversion of procyanidins and prodelphinidins to cyanidin and delphinidin. Phytochemistry, 25 (1), 223–230.
19. 19. Kennedy, J.A. and Jones, G.P. (2001) Analysis of proanthocyanidin cleavage products following acid‐catalysis in the presence of excess phloroglucinol. Journal of Agricultural and Food Chemistry, 49 (4), 1740–1746.
20. 20. Des Gachons, C.P. and Kennedy, J.A. (2003) Direct method for determining seed and skin proanthocyanidin extraction into red wine. Journal of Agricultural and Food Chemistry, 51 (20), 5877–5881.
21. 21. Passos, C.P., Cardoso, S.M., Domingues, M.R.M., et al. (2007) Evidence for galloylated type‐A procyanidins in grape seeds. Food Chemistry, 105 (4), 1457–1467.
22. 22. De Marchi, F., Seraglia, R., Molin, L., et al. (2014) Study of isobaric grape seed proanthocyanidins by MALDI‐TOF MS. Journal of Mass Spectrometry, 49 (9), 826–830.
23. 23. He, F., Pan, Q.H., Shi, Y., Duan, C.Q. (2008) Chemical synthesis of proanthocyanidins in vitro and their reactions in aging wines. Molecules, 13 (12), 3007–3032.
24. 24. Herderich, M.J. and Smith, P.A. (2005) Analysis of grape and wine tannins: methods, applications and challenges. Australian Journal of Grape and Wine Research, 11 (2), 205–214.
25. 25. Jeffery, D.W., Mercurio, M.D., Herderich, M.J., et al. (2008) Rapid isolation of red wine polymeric polyphenols by solid‐phase extraction. Journal of Agricultural and Food Chemistry, 56 (8), 2571–2580.
26. 26. Caceres‐Mella, A., Pena‐Neira, A., Narvaez‐Bastias, J., et al. (2013) Comparison of analytical methods for measuring proanthocyanidins in wines and their relationship with perceived astringency. International Journal of Food Science and Technology, 48 (12), 2588–2594.
27. 27. Harbertson, J.F., Picciotto, E.A., Adams, D.O. (2003) Measurement of polymeric pigments in grape berry extracts and wines using a protein precipitation assay combined with bisulfite bleaching. American Journal of Enology and Viticulture, 54 (4), 301–306.
28. 28. Sarneckis, C.J., Dambergs, R.G., Jones, P., et al. (2006) Quantification of condensed tannins by precipitation with methyl cellulose: development and validation of an optimised tool for grape and wine analysis. Australian Journal of Grape and Wine Research, 12 (1), 39–49.
29. 29. Waterhouse, A.L., Ignelzi, S., Shirley, J.R. (2000) A comparison of methods for quantifying oligomeric proanthocyanidins from grape seed extracts. American Journal of Enology and Viticulture, 51 (4), 383–389.
30. 30. Kennedy, J.A. and Taylor, A.W. (2003) Analysis of proanthocyanidins by high‐performance gel permeation chromatography. Journal of Chromatography A, 995 (1–2), 99–107.
31. 31. Dambergs, R.G., Mercurio, M.D., Kassara, S., et al. (2012) Rapid measurement of methyl cellulose precipitable tannins using ultraviolet spectroscopy with chemometrics: application to red wine and inter‐laboratory calibration transfer. Applied Spectroscopy, 66 (6), 656–664.
32. 32. Hanlin, R.L., Kelm, M.A., Wilkinson, K.L., Downey, M.O. (2011) Detailed characterization of proanthocyanidins in skin, seeds, and wine of Shiraz and Cabernet Sauvignon wine grapes (Vitis vinifera). Journal of Agricultural and Food Chemistry, 59 (24), 13265–13276.
33. 33. De Beer, D., Harbertson, J.F., Kilmartin, P.A., et al. (2004) Phenolics: a comparison of diverse analytical methods. American Journal of Enology and Viticulture, 55 (4), 389–400.
34. 34. Mercurio, M.D. and Smith, P.A. (2008) Tannin quantification in red grapes and wine: comparison of polysaccharide‐ and protein‐based tannin precipitation techniques and their ability to model wine astringency. Journal of Agricultural and Food Chemistry, 56 (14), 5528–5537.
35. 35. Seddon, T.J. and Downey, M.O. (2008) Comparison of analytical methods for the determination of condensed tannins in grape skin. Australian Journal of Grape and Wine Research, 14 (1), 54–61.
36. 36. Downey, M., Mazza, M., Seddon, T.J., et al. (2012) Variation in condensed tannin content, composition and polymer length distribution in the skin of 36 grape cultivars. Current Bioactive Compounds, 8 (3), 200–217.
37. 37. Peleg, H., Gacon, K., Schlich, P., Noble, A.C. (1999) Bitterness and astringency of flavan‐3‐ol monomers, dimers and trimers. Journal of the Science of Food and Agriculture, 79 (8), 1123–1128.
38. 38. Robichaud, J.L. and Noble, A.C. (1990) Astringency and bitterness of selected phenolics in wine. Journal of the Science of Food and Agriculture, 53 (3), 343–353.
39. 39. Hufnagel, J.C. and Hofmann, T. (2008) Orosensory‐directed identification of astringent mouthfeel and bitter‐tasting compounds in red wine. Journal of Agricultural and Food Chemistry, 56 (4), 1376–1386.
40. 40. McRae, J.M., Schulkin, A., Kassara, S., et al. (2013) Sensory properties of wine tannin fractions: implications for in‐mouth sensory properties. Journal of Agricultural and Food Chemistry, 61 (3), 719–727.
41. 41. Lea, A.G.H. (1978) The phenolics of ciders: oligomeric and polymeric procyanidins. Journal of the Science of Food and Agriculture, 29 (5), 471–477.
42. 42. Vidal, S., Cartalade, D., Souquet, J.‐M., et al. (2002) Changes in proanthocyanidin chain length in winelike model solutions. Journal of Agricultural and Food Chemistry, 50 (8), 2261–2266.
43. 43. Foo, L.Y., McGraw, G.W., Hemingway, R.W. (1983) Condensed tannins – preferential substitution at the interflavanoid bond by sulfite ion. Journal of the Chemical Society – Chemical Communications, (12), 672–673.

Note

1 A dramatic culinary example of this phenomenon is the production of Spanish membrillo, where colorless quince fruit is transformed into a red paste following cooking due to proanthocyanidin hydrolysis. However, oxidation is necessary for anthocyanidin formation, so a trace of iron is essential, and the use of purified water can defeat color formation.