New insight into combination tanning

Abstract The reactions between collagen and condensed vegetable tannin and oxazolidine have been examined using soluble collagen, hide powder and deaminated collagen as models for skin. The measurement of combined, crosslinked condensed tannins on collagen showed that 20-30% condensed tannin was immobilised through covalent bonding. By employing deaminated collagen, the reactive sites on collagen were found to be the e-amino groups of lysine and the guanidyl groups of arginine. Differential scanning calorimetry showed that the shrinkage temperature of the leather changed little after the non-combined vegetable tannin had been removed from the leather. This indicates that the high stability of the combination tanned leather comes from the covalently bound vegetable tannins and oxazolidine, analogous to observations in mineral tanning. The conclusion is supported by hydrothermal isometric tension measurements, when no relaxation process was observed for the combination tanned leather, following heating beyond the conventionally measured shrinkage temperature. Introduction In searching for a commercially viable organic tannage, condensed plant polyphenol and oxazolidine combination tannage is one of the most promising options. It is capable of achieving high hydrothermal stability and, since it is based on a renewable natural resource, it is perceived to be environmentally friendly^1-5. This tannage has been known from the 1980s⁶ but the first mechanistic study was reported in 1998 by Covington and Shi^3,4. They found that high stability leather (Ts>110°C) can be obtained from vegetable tanning followed by oxazolidine retanning, when mimosa is the preferred commercial tannin. Modelled in those studies by the condensed tannin monomer catechin, the reaction was described as the polymerisation of plant polyphenol molecules by the cross-linker oxazolidine, illustrated in Figure 1^3,4. However, this is not enough to provide a comprehensive definition of the mechanism of condensed tannin-oxazolidine combination tannage. Key questions remain: * From where does the high stability tannage effect originate? * If a multiply bound tanning is created within the collagen matrix, then how do the polyphenol-oxazolidine-collagen components interact? In this work, hydrothermal isometric tension (HIT) has been used to study the crosslinking mechanism. This technique has been used in studies of collagen materials, to investigate crosslinking density and properties^7-14. In this technique, a sample is held at its unloaded length while the surrounding medium is heated: when the collagen denatures, the forces exerted during the contraction and the subsequent relaxation processes are recorded¹⁵. In other work¹⁶, we have demonstrated that gallocatechin tannin (prodelphinidin, Figure 2), of all the types of plant polyphenol tannins, has the highest crosslinking activity with oxazolidine, because the reaction takes place at the flavanol α-ring and pyrogallol β-ring. Using model compounds, butyl amine or amino acids reacting with simple phenols and oxazolidine, the crosslinking occurs between amino groups and phenols. This indicates that the condensed tannins may be immobilised on the collagen fibres by covalent bonds, via ring-opened oxazolidine. In the present work, the combination characteristics of several vegetable tannins (pecan, myrica esculenta, mimosa, quebracho, neem, tea polyphenol) and oxazolidine on collagen have been studied. Experimental and materials Combination tanning of hide powder To 1g hide powder, soaked in 20ml of pH6 buffer solution for 12 hours at room temperature, 25 or 200% vegetable tannin was added, depending on the experiment; shake for 8 hours at room temperature, filter to remove non combined tannins. Resoak, then 10% oxazolidine was added and shaken for 30 minutes at 20°C, increased to 60°C for 4 hours. The powder was filtered and washed with 10ml buffer solution twice. The washing solution was collected, made to 500ml and the residue of oxazolidine in the solution was measured from the nitrogen content or by GC-MS (gas chromatography-mass spectrometry). Determination of tannin combination on leather Vegetable tanned or combination tanned hide powder was soaked in 100ml of 50% acetone aqueous solution or pH12 sodium carbonate solution for 20 minutes at 20°C and washed with the same solvent. The amount of tannin washed out from hide powder was analysed by the gravimetric method. The Ts of the hide powder was measured by DSC. Differential scanning calorimetry (DSC)¹⁴ The hide powder, tanned or untanned, was adjusted to water content 80-90% and 10-20mg samples were sealed in standard aluminium pans. DSC runs were carried out from 30 to 125°C at 5°C/min. The enthalpy was calculated on dry weight. Combination tanning of pickled pelt³ The pH of the pelt was adjusted to 4.5, then 10% vegetable tannin was added, followed by another 10% after 1 hour; running continued for 24h. After discarding the float, 100% water and 10% oxazolidine were added, then the tannage ran for 4 hours. Hydrothermal isometric tension (HIT) of leather^8-13 The leather was shaved to 0.8mm and soaked in tris-HCl buffer solution at pH7.4 for 48 hours. The specimens 30x7mm were stamped out with a press knife and clamped between two steel jaws; samples were conditioned in liquid paraffin for 30 minutes before starting the heating process. The tension before heating was adjusted so that there was nearly zero tension on the samples. The temperature was increased steady from 20 to 125°C at a rate of 2°C/min. The tension and temperature data were recorded by computer. Results and discussion Crosslinking in combination tanned leather Vegetable-oxazolidine combination tannage is analogous to hydrolysable tannin + Al(III) combination tannage in achieving high stability synergistically^17-21. In vegetable + oxazolidine combination, crosslinks may coexist between vegetable tannin and collagen, vegetable tannin and oxazolidine, oxazolidine and collagen. In which case, which is the determining factor for high stability? In the case of vegetable tanned collagen, since the normal aldehyde reaction site is the amino groups of lysine and the peptide links are believed to be effectively masked by tannins, then it might be deduced that oxazolidine retanning only crosslinks the polyphenol tannin^22,23. In this way, a multiply bound tanning matrix is created within the collagen matrix and the concerted effects of the phenol condensation products are the cause of high stability. However, from this present work, that view must be modified. Fixation of oxazolidine in tanning system From Table 1, it can be seen that the amount of oxazolidine reacted with tanned hide powder, Ox_hptan, is always less than the sum of the fixation by vegetable tannin plus fixation by hide powder, (Ox_tan + Ox_hp), by about 5%. If the polyphenolic tannin had masked the collagen completely, the fixation of oxazolidine by the vegetable tanned hide powder would be the same as the amount fixed by tannin alone. The results show that the total consumption of oxazolidine was not completely adversely affected by prior reaction between vegetable tannin and collagen. This result is in agreement with the literature, in which it has been shown that the vegetable tanning process has little effect on the ability of the collagen to combine with acid or formaldehyde^20,21. The relatively small crosslinking molecules can manage to find the reactive sites on both collagen or tannin molecules. From Table 1, it is also shown that different vegetable tanned collagens exhibit different fixation capabilities towards oxazolidine: gallocatechin (prodelphinidin) tannins such as pecan and myrica esculenta are more reactive, hence the tanned collagen can absorb more oxazolidine. Combination of tannins in the tanning system a. Intact collagen Lyotropic agents such as 50% acetone, 6M urea or concentrated Na₂CO₃ solution can break down hydrogen bonds and release the tannin molecules from tanned pelt or hide powder²⁴. Up to 95% of the bound tannins can be washed out from leather by 50% acetone, leaving 5% tannin that is irreversibly combined with collagen²⁵. From Table 2, less tannin is washed out after the crosslinking process, showing that 20-30% was irreversibly fixed in combination tanning of collagen. There are three possible reasons for this phenomenon. Some of the tannin molecules might be fixed on collagen chains by covalent bonding. Hydrogen bonded tannins or the condensation product with oxazolidine should be displaced from the collagen by sodium carbonate solution at pH12 but not if there is covalent crosslinking between the collagen and the tannin molecules. This is the most likely mechanism; it means there are about 20g of tannin fixed per 100g collagen via covalent oxazolidine bonds. The amount of tannin covalently fixed depends on the type of condensed tannin; gallocatechin tannins are more reactive in this respect. The high polymer product of condensation with oxazolidine might be held between the collagen fibres by physical interaction. This possibility is limited by the use of hide powder, reducing the difficulty of diffusion. If there is blocking of the polar basic groups of collagen, hydrophobic interactions might be formed, which would be resistant to the hydrogen bond breaking agents, see below. b. Deaminated collagen How the tannin is fixed on the collagen can be further examined by washing experiments conducted on deaminated collagen as the substrate for tanning. From the literature, about 85% of the amino groups and some of the guanidyl groups are removed in deaminated hide powder²⁶. Many studies have shown that removing the amino groups from hide powder does not affect the shrinkage temperature of vegetable tanned leather: the combination of tannins is reduced by about 25%, but aldehyde tanning is completely prevented. From the washing experiments of deaminated hide powder, most of the combined tannins can be removed before or after oxazolidine reaction, which implies that the amino groups of collagen play a key role in the combination tanning process. Without the presence of side chain amino groups, the tannin combination with oxazolidine is significantly reduced, see Table 4. This is explained by the loss of the binding sites for crosslinking the tannin molecules to the protein. After removing the amino groups of collagen, only 6-10% of tannins are combined through oxazolidine crosslinking; this means that this amount of tannin in the combination tannage is fixed via oxazolidine at other reactive groups, such as the guanidyl groups. In Table 5, the highest shrinkage onset temperatures are achieved using gallocatechin structured polyphenols tannins: green tea, pecan, myrica and mimosa. There is some loss of hydrothermal stability after washing with the hydrogen bond breaker but high stability remains; this indicates the importance of covalent bonding in the mechanism. Without the presence of the amino groups, to create the covalent links between the polyphenol and the collagen, high shrinkage temperature cannot be achieved; this is shown by the loss of hydrothermal stability when combination tanned deaminated collagen is washed with hydrogen bond breakers, also shown in Table 6. In combination tanned leather, the following crosslinking patterns coexist; combined, all these reactions create a network around the collagen. But which is the determining mode for producing the high hydrothermal stability of combination tanned leather? * Type I: collagen + tannin + collagen Solo vegetable tanning has a minimal stabilising effect, achieved through multiple hydrogen bond and hydrophobic bond crosslinking, perhaps with some covalent fixation by quinoid reaction. * Type II: collagen + oxazolidine + collagen Solo aldehydic tanning confers only moderate hydrothermal stability, whether or not it is theoretically capable of crosslinking. * Type III: collagen + oxazolidine + tannin The two component reaction of the tannins would imply merely an additive effect of aldehydic and vegetable tanning. * Type IV: collagen + oxazolidine + tannin + oxazolidine + tannin + oxazolidine + collagen This reaction involves covalent binding of oxazolidine to collagen, including a covalent polymerising reaction with polyphenol, thereby covalently binding a tannin matrix to collagen in a concerted interaction. The DSC studies clearly show that the washing process has no influence on the natural, intact collagen, because the Ts are relatively unaffected. A similar result was obtained with oxazolidine tanned collagen; the covalent crosslinking function is unchanged by washing with 50% acetone. From the results presented above, some general observations can be made, which lead to a view of the mechanism of the combination tanning reaction. For vegetable tanned collagen, washing with 50% acetone strips out most of the tannin, reducing the Ts almost to that of natural collagen. However, for the vegetable-oxazolidine tanned collagen, although washing reduces the fixed tannin to 50-60%, the Ts is not much affected. This means the covalently fixed portion of tannin (types I, III and IV) contribute to the tanned collagen's high hydrothermal stability; this is about 30% of the total amount of tannin used in the vegetable tannage. Deamination reduces the fixation of tannin on collagen to 60%. While there is not a big reduction of Ts in vegetable tanned collagen, it reduces the Ts of combination tanned collagen to the same value as for solo vegetable tanned collagen. The loss of amino groups does not change the hydrothermal stability of the vegetable tannage substantially. It does affect the outcome of the crosslinking reaction with oxazolidine. Washing tanned deaminated collagen with acetone strips out the tannin completely and reduces the hydrothermal stability to that of the original collagen. Although a high molecular polymer condensate may be formed by polyphenol and oxazolidine within the collagen structure (type IV), this is not necessarily the key point for high hydrothermal stability. For the smaller molecular weight tanning materials, such as phloroglucinol and green tea polyphenol, high hydrothermal stability can still be achieved in the combination tannage. Simple phenols such as phloroglucinol have little tanning ability and the stability of collagen molecules⁵ may even be lowered because phenols can break hydrogen bonds. Green tea tannin, which is a mixture of flavanol monomers in which epigallocatechin gallate is prevalent²⁷, shows tanning ability that results in Ts 70°C. After washing with 50% acetone, nearly all of the phloroglucinol or tea polyphenol is removed from the collagen and Ts reverts to the original value. There is a change after cross-linking with oxazolidine, because both reagents exhibit high Ts, which is not changed significantly by the washing process. If the 'polymerisation' hypothesis is right, the condensation products of small molecular weight phenols and oxazolidine could be regarded as large molecular weight tannins and thus should behave as vegetable tannins, exhibiting loss of hydrothermal stability. The contrary observation demonstrates that high hydrothermal stability depends on the crosslinking reaction of oxazolidine between collagen and tannin, crosslink type IV. Among the tannins tested, the gallocatechin tannins pecan and myrica show the strongest synergistic effect in combination tannage. This can be explained by previous work4, in which it was shown that the flavanol unit structure is the key point for the tannin-oxazolidine cross-linking reaction. The hydrothermal stability obtained from these two tannins is higher than from mimosa or quebracho. Therefore, the cross-linking happening between polyphenol and oxazolidine, type III, does benefit the overall hydrothermal stability but is not the essential factor. Hence, it is apparent that the combination reaction for high hydrothermal stability relies on the creation of a crosslinked polyphenol polymer matrix, which interacts in a concerted manner with the collagen triple helices. However, it is an essential feature of the reaction that the concerted reaction should include covalent bonding between the matrix and the collagen. Hydrothermal isometric tension studies The relationship between cross-linking models and hydrothermal stability can be demonstrated quantitatively and qualitatively by HIT^8-13. Note, the HIT extrapolated onset temperatures in Figure 3 are close to the values obtained by DSC. The slope of the HIT curve at increasing tension during the heat shrinking transition represents the crosslink density or rigidity of collagen fibres and the decreasing tension after the transition represents the stability of these cross-links. High shrinkage temperature does not necessarily mean high crosslink density in collagen material. Ts are more controlled by the stability of crosslinks rather than their numbers. From Figure 3, the tension generated by combination tanned collagen is quite different to the original collagen or vegetable tanned collagen. When the tannins are mainly bound to collagen by hydrogen bonds or other weak bonds, tension reduces after the maximum point as the temperature increases. However, in the case of oxazolidine tanned collagen or vegetable-oxazolidine combination tanned collagen, tension continues to increase. This indicates the crosslinking modes in these two tanned materials is similar but different to vegetable tanned collagen. The covalent bonds in the combination tanned leathers are, therefore, the main reason for high hydrothermal stability in those leathers. From Table 7, the slope of contraction process in tea tanned leather is much lower than for vegetable tanned leather, which indicates the stability of the collagen fibres is lower. This further confirms that high crosslinking density does not necessarily lead to high hydrothermal stability; instead, effective crosslinking should be from the appropriate molecular size and appropriately reactive sites of the tanning agent. This means we can use low molecular weight fractions of condensed tannins with oxazolidine to produce higher stability leather, the defects from traditional vegetable tannins could be avoided. Conclusion The mechanism of high hydrothermal stability combination tanning depends on the following factors: * Preferably starting with gallocatechin type of plant polyphenol * Crosslinking of polyphenol by oxazolidine in situ in collagen * Concerted interaction between the polymeric matrix and the collagen structure * Covalent linking of the matrix to the collagen The mechanism does not assume the necessity for using conventional plant tannins. Indeed, these studies have shown the efficacy of the low molecular weight fraction of tannins, known as the non-tans. Using these monomeric or dimeric flavonoid species has two advantages for the leather industry: * problems of penetration of the primary component of the tannage are eliminated * the components of the tannage may be used as precise reagents, not used in excess, which is a typical feature of vegetable tanning. The mechanism for this organic tannage also reflects the requirements for high hydrothermal stability mineral tanning: it has been shown that enhancing the interaction between the inorganic matrix and the collagen increases the shrinkage temperature²⁸. In this way, the tanning reactions are precisely the same.