Study by LC/ESI/MSn and ESI/HR/MS of SO2 interactions in flavanols-aldehydes nucleophilic reactions

Article abstract of DOI:10.1016/j.foodchem.2009.08.029

The study concerns the interactions between (+)-catechin and a representative oak wood aldehyde (5-(hydroxymethyl)furfuraldehyde for furanic aldehydes and vanillin for phenolic aldehydes) in a wine-like model solution in presence or not of SO₂.

Full text of DOI:10.1016/j.foodchem.2009.08.029

Food Chemistry 122 (2010) 488–494
Contents lists available at ScienceDirect
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
Study by LC/ESI/MSn and ESI/HR/MS of SO2 interactions in flavanols–aldehydes
nucleophilic reactions
a
a
b
b
M.F. Nonier a, , N. Vivas , N. Vivas de Gaulejac , C. Absalon , C. Vitry
*
a Tonnellerie Demptos Seconded to the CESAMO (Centre d’Etude Structurale et d’Analyse des Molécules Organiques), ISM, Université Bordeaux I,
351 Cours de la Libération, F-33405 Talence, France
b CESAMO, ISM, Université Bordeaux I, Talence, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The study concerns the interactions between (+)-catechin and a representative oak wood aldehyde
(5-(hydroxymethyl)furfuraldehyde for furanic aldehydes and vanillin for phenolic aldehydes) in a
wine-like model solution in presence or not of SO2. The formed condensation products were character-
ised by LC/MS, LC/ESI/MSn, and ESI/HR/MS and we studied the effect of SO2 on these condensation
reactions.
Received 28 April 2009
Received in revised form 19 June 2009
Accepted 11 August 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
(+)-Catechin
HMF
Vanillin
HPLC/UV–vis
LC/ESI–MS
MS/MS
ESI/HR/MS
1. Introduction
acid-catalysed condensation (Bendz, Marttensson, & Nilsson,
1967) giving access to a condensation product composed of
two flavanols (or one flavanol and one anthocyanidin) linked
by a chemical bridge depending on the aldehyde used for the
reaction. Such condensation products have been already detected
in wine where they are thought to intervene in red wine colour
and astringency evolution (Fulcrand et al., 1996a; Herjavec, Jero-
mel, Da Silva, Orlic, & Redzepovic, 2007; Mateus, Silva, Rivas-
Gonzolo, Santos Buelga, & De Freitas, 2002; Towel & Waterhouse,
1996). During wine ageing, condensation reactions are known to
occur with aldehydes released from oak barrels (De Freitas, Sou-
sa, Siva, Santos-Buelga, & Mateus, 2004; Es-Safi, Cheynier, &
Moutounet, 2000; Sousa, Mateus, Perez-Alonso, Santos-Buelga,
& De Freitas, 2005).
The aim of the present work was to study, by LC/MS techniques,
the condensation products of flavanols with the different oak wood
aldehydes in model solution systems in presence of SO2, and to
study the effect of SO2 on these reactions.
We know that SO2 is very reactive with aldehydes and particu-
larly acetaldehyde to form sulfite–aldehydes link. We studied SO2
effect on condensation reactions between flavan-3-ols and oak
wood aldehydes. We compared SO2 effect on the kinetic reaction
in the case of furanic aldehyde (HMF) and phenolic aldehyde (van-
illin) with (+)-catechin during 80 days.
Procyanidins extracted from grapes contribute to young red
wines’ astringency. These procyanidins are polymers of flavanol
units with C4–C6 and/or C4–C8 linkages. Ageing wine in oak bar-
rels for several months is a procedure commonly employed by
winemakers in order to improve wines’ quality. During this period,
the wood transfers phenolic (vanillin and syringaldehyde) and fur-
anic (furfuraldehyde, 5-(hydroxymethyl)furfuraldehyde, 5-methyl-
furfuraldehyde) aldehydes as well as ellagitannins to the wine,
directly contributing to wine aroma and taste. Lignins and polysac-
charides are the source of the aldehydes in the toasted oak wood
(Nonier et al., 2007a).
The contribution of aldehydes, in general, furanic and phenolic
in particular, is now well characterised: they are able to react
with flavanols (Fulcrand, Docco, Es Safi, Cheynier, & Moutounet,
1996a; Nonier, Vivas, Vivas de Gaulejac, Pianet,
& Fouquet,
2007b; Saucier, Bourgeois, Vitry, Roux, & Glories, 1997) and/or
anthocyanidins (Escribano-Baillon, Dangles, & Brouillard, 1996;
Fulcrand, Cameira dos Santos, Sarni-Manchado, Cheynier,
&
Favre-Bonvin, 1996b; Timberlake & Bridle, 1976) through a Bayer
* Corresponding author. Tel.: +33(0) 5 40 00 25 77; fax: +33(0) 5 40 00 26 23.
E-mail address: mf.nonier@ism.u-bordeaux1.fr (M.F. Nonier).
0308-8146/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2009.08.029

M.F. Nonier et al. / Food Chemistry 122 (2010) 488–494
489
H5
H4
H1
(+)-catechin
HMF
A
Without SO2
H2 H3
H7
H6
40
30
Time (min)
50
60
0
10
20
(+)-catechin
Abs 280 nm
B
HMF
Products formed with
SO2
H1
With SO2
H7
H2 H3
H6
40
30
Time (min)
50
60
0
10
20
Fig. 1. HPLC chromatogram at 280 nm during the HMF/(+)-catechin reaction performed in hydroalcoholic solution (12% EtOH, 5 g/L tartaric acid, 1 N NaOH for pH
adjustment) at pH = 3.5 without and with SO2.
Fig. 2. HPLC chromatogram at 280 nm during the HMF/(+)-catechin reaction at 20 days – mass spectra of compound adducts of catechin–HMF (H1, H2, H3, H6 and H7)
formed without SO2 and the corresponding structures.

490
M.F. Nonier et al. / Food Chemistry 122 (2010) 488–494
the basis of peak areas at 280 nm. Identification of the formed
2. Material and methods
products was achieved on the basis of their UV–vis and MS
spectrum.
2.1. Reagents
The same experiments were carried out with SO2 to study its
influence on the solutions’ reactivity: 30 mg/l of NaHSO3 (37.5%)
were added to each solution.
The samples were prepared and subsequently analysed in rep-
licate (data not shown).
Deionized water was purified with a Milli-Q system (Millipore,
Bedford, MA) prior to use. Acetonitrile, ethanol, methanol and for-
mic acid were obtained from Prolabo (Pessac, France). (+)-Catechin,
tartaric acid, sodium hydroxide, 5-(hydroxymethyl)furfuraldehyde
(HMF), and vanillin were purchased from Aldrich (Saint-Quentin,
France).
2.3. HPLC/UV–vis analysis
HPLC/UV–vis analyses were performed with a Waters separa-
tion module system, a Waters UV–vis detector, and Millenium32
chromatography manager software. UV–vis spectra were recorded
at 280 nm and 440 nm. The column was a reverse-phase Interchim
2.2. Reactions
To investigate the reaction between flavanols and oak wood
aldehydes, (+)-catechin was separately incubated with 5-(hydroxy-
methyl)furfuraldehyde, and vanillin during 85 days at pH 3.5. The
formed compounds were monitored by HPLC with UV–vis and
diode array detection, and characterised by LC/ESI/MS.
The model wine solution used was 12% EtOH buffered to pH 3.5
(5 g/l tartaric acid, 1 N NaOH to pH = 3.5). (+)-Catechin (1.65 mM)
was prepared in each resulting solution (50 ml). The aldehydic
derivatives (5-(hydroxymethyl)-furfuraldehyde, vanillin) were
added at a concentration of 3.14 mM.
The prepared solutions were incubated at 20 °C and in the ab-
sence of light. They were stored with a little headspace and under
nitrogen in order to avoid oxidation. Reactions were periodically
monitored by liquid chromatography (LC) coupled with an UV
detector and with an electrospray mass spectrometry (ESI/MSn)
detector during a period varying according to the nature of the
aldehyde. Quantification of residual (+)-catechin was achieved on
C18 (10
lm packing, 250 Â 4.7 mm i.d.) protected with a guard
column of the same material; solvent A, water/formic acid (98:2,
v/v); solvent B, acetonitrile/water/formic acid (80:18:2, v/v). The
column was placed at ambient temperature (T = 21 °C). The elution
programme was performed at a constant flow of 1 ml/min, passing
from 5% to 30% of B in 40 min, and then rising to 40% of B in 10 min,
and finally to 100% of B in 5 min, followed by washing and re-
equilibrating the column during 15 min. The injection volume
was 20 ll.
2.4. LC/ESI/MS analysis
LC/ESI/MS, LC/ESI/MS/MS and ESI/HR/MS were performed on a
Q-StarTM instrument using an electrospray ionisation source in
negative-ion mode. Ion spray voltage was selected at 4.5 kV, the
A
Vanillin
(+)-catechin
V1
V4
Without SO2
V5
V6
V3
V2
V7
0
10
20
30
40
50
60
Time (min)
Abs 280 nm
B
Vanillin
(+)-catechin
Products formed
with SO2
With SO2
V1
V4
V5
0
10
20
30
40
50
60
Time (min)
Fig. 3. HPLC chromatogram at 280 nm during the vanillin/(+)-catechin reaction performed in hydroalcoholic solution (12% EtOH, 5 g/L tartaric acid, 1 N NaOH for pH
adjustment) at pH = 3.5 without and with SO2.

M.F. Nonier et al. / Food Chemistry 122 (2010) 488–494
491
capillary temperature was selected at 275 °C, and the source tem-
3. Results and discussion
perature was selected at 400 °C. The values of sheath gas flow
rate were: nebulisation gas = 2.85 l/min, turbo gas = 4.8 l/min,
curtain gas = 1.48 l/min. For MS/MS measurements, the energy
collision and the gas collision were respectively À30 eV and
5 eV. Isolation window was 1 uma. For HR/MS analysis, we used
an internal calibration using PPG. The resolution obtained was
12,000.
3.1. Kinetic study of the reaction
When (+)-catechin was incubated in the presence of HMF or
vanillin, a reaction between these two reactants occurred with
the formation of analogous oligomeric compounds (Nonier, Vivas,
Vivas de Gaulejac, Pianet, & Fouquet, 2007b; Nonier et al., 2008).
HPLC technique permitted us to follow the reaction between (+)-
catechin/hydroxymethylfurfuraldehyde (Figs. 1A and 2) and (+)-
catechin/vanillin (Figs. 3A and 4). We studied in previous works
(Nonier et al., 2007b), the kinetic study of the reaction of (+)-cate-
chin with aldehydes derived from toasted wood.
At pH 3.5, a general decrease in the concentration of (+)-cate-
chin (Fig. 5) and the appearance of numerous products initially ab-
sent in the mixture were observed in the case of reaction with and
without SO2.
The kinetic study showed that in presence of SO2, the consump-
tion of (+)-catechin is slower than in the case of reactions without
SO2. So SO2 decreased the kinetic of the reaction. Fig. 5 summarised
this result. The bisulfite is generally reactive with aldehydes in par-
ticular acetaldehyde, forming a sulfite–aldehyde combination. In
the case of furanic and phenolic aldehydes, a similar reaction can
occur to limit the formation of catechin/aldehyde adducts. So, a
part of SO2 combine with aldehydes and the quantity of free alde-
hydes is inferior in front of reactions without SO2, so the polymer-
isation reaction is slow down.
The column was a reverse-phase Interchim C18 (10 lm packing,
250 Â 4.7 mm i.d.) protected with a guard column of the same
material; solvent A, water/acetic acid (98:2, v/v); solvent B, aceto-
nitrile/water/acetic acid (80:18:2, v/v). The column was placed at
ambient temperature (T = 21 °C). The elution programme was per-
formed at a constant flow of 1 ml/min using the LC inlet and split-
ter outlet connection permitting to reduce the flow at 0.5 ml/min,
passing from 5% to 30% of B in 40 min, and then rising to 40% of B in
10 min, and finally to 100% of B in 5 min, followed by washing and
re-equilibrating the column during 15 min. The injection volume
was 20 ll.
2.5. Absorption spectra
Spectrophotometric measurements were performed using an
Anthelie SecomamTM spectrophotometer and UV–vis spectra were
recorded with a spectrophotometer fitted with a quartz cell
(1 cm). The wine-like samples were diluted (1/20) and then their
spectra measurements were taken.
Fig. 4. HPLC chromatogram at 280 nm during the vanillin/(+)-catechin reaction at 31 days – mass spectra of compound adducts of catechin–vanillin (V1, V2, V3, V4, V5, V6
and V7) formed without SO2 and the corresponding structures.

492
M.F. Nonier et al. / Food Chemistry 122 (2010) 488–494
In the case of wine-like solutions, SO2 at 30 mg/l (free SO2, as a
(Fig. 2). Their molecular weight (688 amu) corresponds exactly to
a structure in which two (+)-catechin units are linked by a hydrox-
ymethylfurfuryl. They are dimer adducts.
Mass spectrometry analysis revealed also the compounds H6
and H7 all had molecular weights of 1086 amu ([MÀH]À ion signals
at m/z 1085 amu) (Fig. 2) and corresponded to a structure in which
three (+)-catechin are linked by two hydroxymethylfurfuryl bridge.
They are trimer adducts.
standard level for wine ageing) affects the kinetic of the reaction,
but only in the range of 5–10%, in comparison with the control
sample without SO2, so its effect was limited.
3.2. Characterisation of the compounds by LC/ESI/MSn and ESI/HR/MS
HPLC/DAD technique (Figs. 1A and 2) allows us to follow the
reaction between (+)-catechin and HMF in hydroalcoholic solu-
tion (Nonier et al., 2007b). Among the formed compounds, the
UV traces recorded at 280 nm at 20 days of reaction, revealed
the presence of seven major peaks in addition to those of (+)-cat-
echin and HMF (Figs. 1A and 2). The first two peaks on the chro-
matogram correspond to HMF and (+)-catechin respectively, and
the seven others are condensed products of (+)-catechin and
HMF, characterised by LC/MS in previous work (Nonier et al.,
2008).
We observed at 20 days of reaction with and without SO2
(Fig. 1), the presence of compounds H1, H2, H3, H6 and H7.
Mass spectrometric analysis, conduced in the negative-ion
mode, revealed that the compounds H1, H2, H3 gave all [MÀH]À
ion signals at m/z 687 as shown on the obtained mass spectrum
But, in presence of SO2, at 20 days of reaction, four new peaks
appeared: X1, X2, X3 and X4 (Fig. 1B).
LC/MS analysis, performed in the negative-ion mode, allowed
molecular weight determination of these new products formed
through this reaction. Mass spectrometric analysis revealed that
they gave all [MÀH]À ion signals at m/z = 751 as shown on the ob-
tained mass spectrum (Fig. 6).
The m/z values of compounds X1, X2, X3, X4 detected at 751
(Fig. 6) did not establish the structure of these compounds: MSn
analysis are conduced. Fig. 4 presented the MS2 spectrum obtained.
Their MS/MS fragmentation gave three signals at m/z = 669, 599
and 517. The m/z 669 and 517 ions correspond respectively to
the loss of H2SO3 and a specific break presented Fig. 4. Mass exact
was performed on a Q-Tof instrument. The molecular formula ob-
tained was [C36H31O16S]À, with
a Dm (error) = 0.44 ppm (or
0.33 mDa). A molecular structure was presented Fig. 6. X1, X3,
X3 and X4 were four positional isomers. One could hypothesise
that there were other three positional isomers arising from a differ-
ent position of the H2SO3 moiety, acting as a bridge between close
OH groups, not necessarily linked to the same benzene ring: the
two phenolic OH groups on the catechin unit at the left side of
Fig. 6 structure, one of these groups and the close OH group on
the other catechin unit, a catechin OH group and the CH2OH group
belonging to hydroxymethyl furfuraldehyde. This hypothesis was
clearly confirmed by the presence, in the MS/MS spectrum, of a
m/z 599 ion, corresponding to the m/z 517 ion plus H2SO3 moiety.
HPLC/DAD technique (Figs. 3 and 4) permits us to follow the
reaction between (+)-catechin and vanillin in hydroalcoholic solu-
tion (Nonier et al., 2007b). At 40 days of reaction, the UV traces re-
corded at 280 nm, revealed the presence of seven peaks in addition
to those of (+)-catechin and vanillin (Figs. 3A and 4). They are con-
densed products of (+)-catechin and vanillin characterised by LC/
MS in previous work (Nonier et al., 2008).
100
90
Vanillin with SO2
80
70
Vanillin
60
50
40
30
HMF with SO2
20
HMF
10
0
0
10
20
30
40
50
60
70
80
Time (days)
Fig. 5. Evolution of the reaction rate for catechin in the case of vanillin and HMF
without and with SO2. The reactions were performed in hydroalcoholic solution
(12% EtOH, 5 g/L tartaric acid, 1 N NaOH for pH adjustment) at pH = 3.5. The
reaction was followed by reverse-phase HPLC.
Fig. 6. Mass spectrum of compound adducts of catechin–HMF (X1, X2, X3 and X4) formed in presence of SO2 and MS2 mass spectrum obtained from ion m/z: 751 and the
corresponding structure.

M.F. Nonier et al. / Food Chemistry 122 (2010) 488–494
493
Fig. 7. Mass spectrum of compound adducts of catechin–vanillin (Y1 and Y2) formed in presence of SO2 and MS2 mass spectrum obtained from ion m/z: 505 and the
corresponding structure.
At 40 days of reaction, we observed with and without SO2
(Fig. 2), the presence of dimer adducts: V1, V4, V5.
the flow rate at 3 ml/min. The four and the three eluted fractions
respectively in the case of HMF and vanillin were frozen and lyo-
philisated. Each fraction was immediately analysed by LC/DAD
technique in order to control the purity of the compound and then
was analysed by NMR. As in the case of the precedent purification,
the interpretation of the obtained spectra was impossible, because
each compound was modified at ambient temperature giving a
complex mixture.
Mass spectrometric analysis, conduced in the negative-ion
mode, revealed that the compounds V1, V4, V5 gave all [MÀH]À
ion signals at m/z 713 amu as shown on the obtained mass spec-
trum (Fig. 4). Their molecular weight (714 amu) corresponds ex-
actly to a structure in which two (+)-catechin units are linked by
a vanillyl. They are dimer adducts.
But in presence of SO2, two new peaks appeared: Y1 and Y2.
LC/MS analysis, performed in the negative-ion mode, allowed
molecular weight determination of these new products formed
through this reaction. Mass spectrometric analysis revealed that
they gave all [MÀH]À ion signals at m/z = 505 as shown on the ob-
tained mass spectrum (Fig. 7).
Each fraction was acetylated with pyridine–acetic anhydride in
order to protect hydroxyl groups and then analysed by ESI/MS. Re-
sults showed that the reaction of acetylation was not complete and
the obtained compound was not pure.
It appears very difficult to characterise the different compounds
by NMR. They are sensible products. So, we proposed in each case a
structure thanks to the interpretation of MS/MS fragmentations
and mass exact.
The m/z values of compounds Y1 and Y2 detected at m/z
505 amu (Fig. 7) did not permit to establish the structure of these
compounds: MSn analysis are conduced. Fig. 5 presented the MS2
spectrum obtained. Their MS/MS fragmentation gave two signals
at m/z = 423 and 271 which correspond respectively to the loss of
H2SO3 and a specific break presented Fig. 7. Mass exact was per-
formed on a Q-Tof instrument. The molecular formula obtained
A
0.12
70 days without SO2
0.10
was [C23H21O11S]À, with a
Dm (error) = 2.1 ppm (or 1.1 mDa). A
70 days with SO2
0.08
molecular structure was presented Fig. 7. There was only another
possible bridging between close OH groups, i.e. the one arising
from vanillin condensation with catechin and a catechin OH group,
and H2SO3, then two positional isomers were expected and con-
firmed by the presence of Y1 and Y2 products.
0.06
0.04
0.02
T0 with and without SO2
0
We tried to isolate the different compounds: X1, X2, X3, X4 and
Y1, Y2 by different techniques.
500
Wavelength (nm)
550
600
360
400
450
After a twelve-day reaction period in the case of HMF or a
thirty-day reaction period in the case of vanillin, the solvent was
concentrated from the model solution using a rotary evaporator
at 25 °C, to a volume of 5 ml. The solution was applied to a
300 Â 25 mm TSK Toyopearl gel HW-40 (S) column (Tosoh, Japan),
and the compounds were eluted with methanol, at a flow rate of
0.8 ml/min. The four and the three eluted fractions respectively
in the case of HMF and vanillin were dried using a rotary evapora-
tor at 20 °C. Each fraction was immediately analysed by LC/DAD
technique in order to control the purity of the compound and then
was analysed by NMR. The interpretation of the obtained spectra
was impossible, because each compound was modified at ambient
temperature giving a complex mixture.
B
0.02
0.01
70 days without SO2
70 days without SO2
0
375
425
475
Wavelength (nm)
525
575
T0 with and without SO2
So, we isolated these compounds using an other technique:
semipreparative HPLC purification. The column was a reversed-
Fig. 8. Visible spectra (350–600 nm) of the solution during the reaction between
(A) catechin–HMF and (B) catechin–vanillin under acidic conditions, without and
with SO2. The reactions were performed in hydroalcoholic solution (12% EtOH, 5 g/L
tartaric acid, 1 N NaOH for pH adjustment) at pH = 3.5.
phase C18 (5
l
m packing, 250 Â 10 mm i.d.). Conditions of HPLC
were identical to those described in Section 2.3., we only increased

494
M.F. Nonier et al. / Food Chemistry 122 (2010) 488–494
Fulcrand, H., Cameira dos Santos, P.-J., Sarni-Manchado, P., Cheynier, V., & Favre-
3.3. Evolution of the colour
Bonvin, J. (1996b). Structure of new anthocyanin-derived wine pigments.
Journal of Chemical Society, Perkin Transactions, 1, 735–739.
The colour of the solution was followed by spectrophotometry.
Results are summarised in Fig. 8.
We can note, in the case of reaction without SO2, a significant
increase in solution absorbance, with time with a maximum at
440 nm (yellow).
The absorbance of the solution in presence of SO2 evaluated dif-
ferently. At 70 days of reaction, solutions with SO2 absorbed at
440 nm. At the same time, without SO2 in the solution, this absor-
bance has a superior value.
Herjavec, S., Jeromel, A., Da Silva, A., Orlic, S., & Redzepovic, S. (2007). The quality of
white wines fermented in Croatian oak barrels. Food Chemistry, 100, 24–128.
Mateus, N., Silva, A. M. S., Rivas-Gonzolo, J. C., Santos Buelga, B., & De Freitas, V.
(2002). Identification of anthocyanin–flavanol pigments in red wines by NMR
and mass spectrometry. Journal of Agricultural and Food Chemistry, 50,
2110–2116.
Nonier, M. F., Vivas, N., Vivas de Gaulejac, N., Absalon, C., Soulié, P., & Fouquet, E.
(2007a). Pyrolysis–gas chromatography/mass spectrometry of Quercus sp.
wood. Application to structural elucidation of macromolecules and aromatic
profiles of different species. Journal of Analytical and Applied Pyrolysis, 75,
181–193.
Nonier, M. F., Vivas, N., Vivas de Gaulejac, N., Pianet, I., & Fouquet, E. (2007b). A
kinetic study of the reaction of (+)-catechin with aldehydes derived from
toasted oak. Journal of the Science and Food and Agriculture, 87, 2081–2091.
Nonier, M. F., Vivas, N., Absalon, C., Vitry, C., Fouquet, E., & Vivas de Gaulejac, N.
(2008). A kinetic study of the reaction of (+)-catechin with aldehydes derived
from toasted oak. Food Chemistry, 107, 1494–1505.
References
Bendz, G., Marttensson, O., & Nilsson, E. (1967). Studies of flavylium compounds and
their properties. Arkiv foer Kemi, 27, 67–77.
De Freitas, V., Sousa, C., Siva, M. S. A., Santos-Buelga, C., & Mateus, N. (2004).
Synthesis of a new catechin–pyrylium derived pigment. Tetrahedron Letters, 45,
9349.
Saucier, C., Bourgeois, G., Vitry, C., Roux, D., & Glories, Y. (1997). Characterisation of
(+)-catechin acetaldehyde polymers:
A model of colloidal state of wine
polyphenols. Journal of Agricultural and Food Chemistry, 454, 1045–1049.
Sousa, C., Mateus, N., Perez-Alonso, J., Santos-Buelga, C., & De Freitas, V. (2005).
Preliminary study of oaklins,
a new class of brick-red catechinpyrylium
Escribano-Baillon, T., Dangles, O.,
& Brouillard, R. (1996). Coupling reactions
pigments resulting from the reaction between catechin and wood aldehydes.
Journal of Agricultural and Food Chemistry, 53, 9249–9256.
Timberlake, C. F., & Bridle, P. (1976). Interactions between anthocyanins phenolic
compounds and acetaldehyde. American Journal of Enology and Viticulture, 27,
97–105.
between flavylium ions and catechin. Phytochemistry, 41, 1583–1592.
Es-Safi, N., Cheynier, V., & Moutounet, M. (2000). Study of the reactions between
(+)-catechin and furfural. Derivatives in the presence or absence of
anthocyanins and their implication in food color change. Journal of
Agricultural and Food Chemistry, 48, 5946–5954.
Towel, J. P., & Waterhouse, A. L. (1996). Barrel to barrel variation of volatile oak
extractives in barrel-fermented chardonnay. American Journal of Enology and
Viticulture, 47, 17–20.
Fulcrand, H., Docco, T., Es Safi, N., Cheynier, V., & Moutounet, M. (1996a). Study of
the acetaldehyde induced polymerisation of flavan-3-ols by liquid
chromatography. Journal of Chromatography A, 752, 85–91.