In previous studies1-6, a collagen condensate tanned with formaldehyde shows a consistent tanning action. The Ts of the tanned samples was 94-96°C, which is a good value for a non-chrome tanning salt. The values can be further improved by pretanning with condensed tannins or a resourcin pretreatment5.

With these conditions, Ts>100°C are reached and the leathers show a good reactivity towards retanning, dyeing and fatliquoring agents. Such a behaviour is the consequence of the condensate’s amphoteric characteristics.

At acid pH, the positive charge centres link the anionic products successively, allowing the short radii forces, such as van der Waals or dipolar forces, to develop their action. The mechanism is similar to that of chrome leathers.

The fullness of the leathers that result is particularly appreciable. It has been shown that to reach the same fullness, the percentage of retanning agent can be reduced by 60% compared with that normally used for chromed leathers.

The dyeing uniformity, softness and the resistance to washing are good.

The condensate used is not a single molecular weight compound. Indeed, ultrafiltration and steric exclusion chromatography investigations6 suggest five fractions, having molecular weights corresponding to 262,000-270,000, 78,250, 26,870, 3,900 and 1,000 Daltons (Da) respectively.

These fractions were singularly utilised for the study of their tanning capability with reference to a control constituted by a mixture of the various fractions in the same ratio as the original condensate6. Surprisingly, it was shown that all fractions achieved the same Ts.

But the values were lower than that registered for the use of the mixture or for the full condensate. This suggests a synergistic effect. The results suggested selective reactivity of the different fractions with respect to the various structual units of the hide: fibres, fibrils and protofibrils.

In fact, it is impossible for the aggregates with the biggest dimensions (~270,000, 79,000 or 27,000 Da) to penetrate into the inter-protofibrillar spaces in order to crosslink. Therefore, it was considered that the three fractions with high molecular weights developed their tanning action by crosslinking at distances of several hundred or thousand Å.

Such a theory finds some support in the observation about the different actions, developed by the single fractions with respect to the unrefined condensate on the increase of Ts. In essence, when tanning with a single fraction, this can form crosslinks with only one structural unit. In the other cases, co-operative crosslinks occur, producing the synergistic action.

Nevertheless, in the previous study6, the highest weight fraction (I fraction), disclosed by steric exclusion chromatography, showed a long tail, suggesting some quantities of the II and III fractions. This study intends to demonstrate that structural units crosslinking at hundreds or thousands of Å are involved in increasing the hydrothermal stability.

The I fraction was submitted to several ultrafiltration cycles in order to isolate a fraction which steric exclusion chromatography suggested possessed a symmetical molecular weight distribution, ie there was no tail. It was employed in tanning sheepskins with a reference being the unpurified I fraction.


The tanning was carried out, as in the previous studies, utilising New Zealand depickled and repickled sheepskins.

The collagen condensate was prepared following methodology described elsewhere1. The tanning solution was prepared by dissolving 300g of condensate under agitation in water. The solution was neutralised to pH7.5 using HCl, diluted to 1000ml and filtered in order to remove undissolved matter.

The ultrafiltration apparatus used has already been described3. The pressure during the developing of the process was kept at 3Bar and the flux at 10-20ml/h. A membrane of YM100 type was used for the separation of the low molecular weight fractions.

The ultrafiltration concentrate was collected, diluted to 1000ml and submitted to a second ultrafiltration cycle by utilising the same membrane. The operation was repeated three times, and the product was concentrated under vacuum, dried and employed in the tanning experiments. In order to characterise the fraction, steric exclusion chromatography was utilised. The analytical conditions are described elsewhere3.

The tanning experiments were carried out, utilising the different products (unrefined condensate (C0) and I fraction (C1) and various purified I fractions (C2-C4)), using the process in Table 1. C0 and C1 were the controls.


The shrinkage temperatures and the fixed quantities of the product were determined on the leathers. The fixed amounts were calculated by the decrease of nitrogen in the final liquor.

The graphs C0, C1, C2, C3 and C4 (Figure 1) report the profiles of the steric exclusion chromatograms of the unrefined condensate and the concentrates coming from different ultrafiltration passages. The increase in the number of ultrafiltration cycles makes the profile more symmetric. The fourth concentrate chromatogram expresses an excellent symmetry, which confirms the purification.

Figure 2 expresses the average molecular weight of the various concentrates and of the unrefined condensate as calculated by chromatographic analysis. The average of the concentrates constituents increases with increasing ultrafiltration cycles from 200,000 Da (unrefined condensate) to an average of 400,000 Da after the fouth filtration. The tanning behaviour of the C4 concentrate at the various pH values has been compared with the C1 concentrate and C0, the unrefined condensate.

Figure 3 gives the Ts of leathers registered after 18h of processing. The results are no different from those of the pickled samples when the tanning is carried out at pH2.5. As already described1,2, this confirms the absence of crosslinks in such conditions, the crosslinks not being formed until the tanning liquor pH is 4.5.

The leathers obtained using fraction C4 are rather horny but as said before, the molecular weights average of the fraction is 400,000 Da. Under these conditions, penetration into the substructure of the skin becomes impossible. Pure steric hindrance considerations are, therefore, suggested as the reason for the small quantities of C4 fixed to the collagen molecule. It is believed that the crosslinks form between the fibres in this case.

One theory is that interunit crosslinks occur, localised at distances corresponding to hundreds or thousands of Å. This could explain the increase in the shinkage temperature and the poor fixation.

The consistent difference, between the Ts of the leathers treated by C4 and the one not purified, does not mean that the crosslinked fibres are less stable with respect to the other case but that the differences, very probably, are due to the fact that only a small part of the purified fraction has the capability of penetrating the skin to form crosslinks.

In the tanning process, the formation of the crosslinks is not always within the substructure of the skin. The tanning process, in some cases, can be produced by crosslinks from all structural units, which then act cooperatively.

Thus the use of particular tanning agents, which have the capability of maximising the use of all the structural units, could lead to a hydrothermal stability much higher than the ones achieved so far.

Apart from the discussed collagen condensates, many other synthetic organic compounds have the latent capability to crosslink with collagen units. It would be sufficient only to adequately modulate the ratio between the various molecular weight of the components in order to reach the conditions of the best penetration and reactivity towards the biopolymer units.


The study of the tanning activity of the C4 purified fraction of the collagen condensate suggests the formation of crosslinking interunits being very much bigger than the protofibrils. The molecular weight distribution of C4 suggests that for steric reasons, the penetration of this particular tanning agent below the protofibril level is impossible.

In essence, the increase of the shrinkage temperature occurring by using the C4 fraction can be only a consequence of the crosslinks formed between the fibres.

The tanning capability is less significant than that of the unpurified fraction. The differences are due, in part, to the absence of the lower molecular weight components, which are able to interact in the free spaces of the fibrils.

These conclusions suggest that the tanning capability of the original condensate could be notably improved by adequately controlling the conditions of its production. That could be verified, for example, by eliminating the formation of very high molecular weight components which do not penetrate the hide structure.

The transformation of these constituents into suitable compounds to pass through the inter-protofibrils and/or interfibrils and/or inter-fibre spaces and to react, would generate a higher number of crosslinks and, therefore, could lead to a higher increase in Ts.

Developing these results could be achieved by applying the ideas to other products of organic syntheses, such as polymers or resins of urea-formaldehyde, dicyandiamide-formaldehyde, etc.