Concerning acidified tannins (path 1), because of the low solubility
in water of the fractions and the low reactivity of these
tannins toward paraformaldehyde, it was not possible to establish
gel times. This behavior is in agreement with the catechinic
rearrangement previously proposed which should affect the more
reactive A aromatic ring of the catechin moiety and consequently
should decrease the water solubility (as previously noticed) and the
reactivity toward paraformaldehyde. Concerning lyophilized tannins
(path 2) the gel time results of tannins formaldehyde-based
glues are given in table accompanied with the pH of the tannins
solutions in water (30%, w/w tannin/water). It appears that the gel
times are directly correlated with the Stiasny number of the tannins
fractions and a dramatic increase in the gel timewas observed
when high sodium hydroxide concentrations were employed. Fig. 5
gives the gel times of 3 lyophilized tannins extracts (tests 1, 2 and
3) as a function of pH of the tannins solution. The pH of each tannins
solution was adjusted by addition of sodium hydroxide or HCl
solution (see Section 2). The best result was observed for test 2using a concentration of NaOH of 10% (w/w) during the extraction
process: in these conditions, extracts bearing short gel times were
isolated (72 s at pH 8 and 32 s at pH 10). Compared to previously
described results, this tannins fraction cannot reach the gel times
yet obtained at pH 8 for pine tannins and pecan nut tannins, which
are equal to 26 and 22 s, respectively (Pizzi, 1994). However, despite
their lower purity (about 50% of active phenolic ingredients), this
grape pomace extract displays gel time values lower than the ones
of mimosa and quabracho tannins, demonstrating the good reactivity
of these crude tannins toward paraformaldehyde. Compared to
acidified tannins, the higher reactivity of the lyophilized extract is
probably due to the low catechinic acid content observed by NMR.