In the leather industry the vast majority of finishing systems are water-borne and are most often based upon either aqueous polyurethanes (polyurethane dispersions) or aqueous acrylics (acrylic latexes) as the principal polymeric binders. Often the two types are used in combination.
For many high performance applications, such as upholstery leather finishing, crosslinking is essential and the most commonly encountered crosslinkers are medium oligomeric molecular weight, water-dispersible polyisocyanates. Some are supplied in solvent at 50-90% solids; others are solvent-free, viscous liquids.
Water-dispersible carbodiimides, also of a molecular weight that justifies calling them oligomers, are of increasing interest as crosslinkers for leather finishes.
Water-dispersible isocyanates became available just as automotive upholstery finishing began to change from being dominated by isocyanate-cured solvent-borne materials (vinyls and vinyl-urethanes) in colour and top coats towards all-aqueous finishes. Water-based finishes had to be crosslinked to achieve the required performance targets, just as was the case for their solvent-based predecessors, but the isocyanates employed to crosslink solvent-borne vinyls and vinyl-urethanes were not readily dispersible in water.
In any case, isocyanates were so well-known for reacting readily with water that many in the field thought them to be inherently incompatible with aqueous finishing. Some twenty years after an enterprising chemist made the revolutionary decision to make and evaluate water-dispersible polyisocyanate crosslinkers to cure his newly-developed all-aqueous finishing formulae, such water-dispersible isocyanates have become truly ubiquitous in a whole host of industries.
At about the same time as leather finishers were experimenting with water-dispersible polyisocyanates to crosslink aqueous leather finishes, water-dispersible polycarbodiimides also appeared. Carbodiimides are already much less reactive with water than isocyanates, though not truly inert. Thus, making them water-dispersible was a sound concept.
Whereas the water-dispersible isocyanates gained instant acceptance in the aqueous finishing of leather, where isocyanates had already been used successfully to crosslink solvent vinyls and solvent-borne vinyl-urethanes, water-dispersible polycarbodiimides gained no such easy acceptance. First they were promoted to quite different markets, principally as alternatives to aziridine crosslinkers which they could not equal in performance and, second, when tried against water-dispersible isocyanates for leather finishing, they brought no new benefits nor could they be used interchangeably with isocyanates to achieve equivalent performance.
The requirements, and with that the sophistication and performance of leather finishes, have steadily advanced over the years and environmental consciousness has intensified. The recent availability of water-dispersible, fully aqueous, ie zero VOC, polycarbodiimides, has brought about renewed interest in polycarbodiimide crosslinkers for finishing leather. This article will try to make some comparisons between the two as crosslinking chemistries in the context of aqueous leather finishing.
What are polymers; what is crosslinking?
Polymers are large long-chain molecules assembled from small sub-units (monomers). Acrylic latexes and polyurethane aqueous dispersions are the main water-borne polymers used in leather finishing today. Desirable properties originate from the entanglement of long polymer chains, from the ability of long chain polymers to form a three-dimensional network, or from both entanglement and network formation occurring at the same time.
A crosslinker can react with polymers to make short, medium, and even long chains longer, thereby increasing chain entanglement. Adjacent chains may tie together to generate a three-dimensional network and even without the direct reaction of a crosslinker with a polymer, the crosslinker can condense with itself to form a polymer network interwoven with the other polymer chains. That is a special case of entanglement and results in the formation of interpenetrating polymer networks (IPNs).
Crosslinks can involve covalent bonding, ionic or hydrogen bonding, interactions, associations into domains akin to crystallites, or other physical impediments to chain mobility of which an IPN is a case in point. Both chain entanglements and network formation (including IPNs) create harder, stronger, tougher coating films.
Crosslinking, network formation particularly, has the potential to reduce elongation as it simultaneously increases tensile strength and polymer hardness. For this reason, over-crosslinking of leather finishes should be avoided, especially if hard-won victories in the wet-end to achieve leather softness and flexibility are to be preserved throughout finishing operations. This is a distinguishing characteristic of leather finishing, and the management of crosslinking a coating for such flexible, extensible substrates is different from what might be appropriate for coatings on rigid substrates, where generally higher crosslink densities are employed than would be acceptable for a leather finish.
Isocyanate and carbodiimide crosslinkers operate by different chemical mechanisms
The chemistry of polycarbodiimide crosslinking mainly involves the reaction of carboxylic acid (-COOH) groups which are present in nearly all acrylic latexes and polyurethane dispersions used for leather finishing. Carboxylic acid groups react with carbodiimide functionality (-N=C=N-).
An O-acyl urea is created which rearranges to form an N-acyl urea. Since the polycarbodiimide contains several -N=C=N- groups, one polycarbodiimide molecule can react, tie together or link carboxyl groups on different polymer chains forming either a longer chain, a crosslink and/or a network.
Reaction can be quite fast under ambient or mild thermal cure conditions and although this might be thought to be ideal it is not yet entirely so in practice. In most leather finishes, there is no significant amount of hydroxyl or other reactive functionality capable of reacting with isocyanate.
So the reaction when isocyanate crosslinkers are employed is probably one of simple moisture-cure, believed to occur after the isocyanate oligomer crosslinker has penetrated into the latex, polyurethane dispersion particle or polymer.
This results in the formation of an interpenetrated polymer network (IPN) in which the isocyanate polymerises in and around and in the presence of an acrylic or polyurethane but does not actually react with the polymer to which this extreme form of entanglement is associated.
Hotpotting and potlife
In tannery use, both types of crosslinker can be hotpotted, ie mixed into the aqueous finish in advance of being applied. The alternative to hotpotting is in-line mixing immediately ahead of the application device, usually a spray gun, thus obviating any need for concern about potlife.
The reaction of carbodiimides with water is quite slow. That of isocyanate with water is not so slow. In practice polycarbodiimide crosslinkers can exhibit quite good potlife in aqueous systems, unlike polyisocyanates.
Figure 1 illustrates typical stability curves for selected polycarbodiimides and polyisocyanates. Plotted is the percentage of isocyanate or carbodiimide functionality remaining as a function of time compared with that initially present in the fully formulated (hotpotted) top coat.
Data for only the first 24 hours is shown and clearly after the 4 to 6 hours of useful life of a hotpotted isocyanate crosslinker system, much isocyanate is still available. The carbodiimides are more stable and the hotpotted water-borne carbodiimide has a useful potlife of at least days.
Determining the correct amount of crosslinker: dosage ladders
Experience teaches that for soft coating systems such as those in use for leather finishing, especially where application is over an extensible material which in the case of leather is also fibrous, dosage ladders are often the best experimental technique for arriving at optimum crosslinker levels.
In the case of carbodiimide crosslinking, the use of stoichiometry, that is calculating the amount of carbodiimide required based on polymeric acid available for reaction, leads to the use of levels that are often too high and not economical. Stoichiometry is not a factor at all for isocyanates since, as mentioned earlier, there is little or no co-reactive functionality usually present for typical polymers used to finish leather.
Setting up a dosage ladder is simple. Start with a low level of crosslinker and work up, tracking performance along the way. For carbodiimides especially, more is not always best and usually not even as good as using a smaller amount. Optimum dosage ranges tend to be broad.
When comparing a high-solids carbodiimide crosslinker with a low-solids one, correct for the solids if the difference is large. A small difference in crosslinker solids, say 40% compared with 50%, may not make any adjustment worthwhile. Typical carbodiimide use levels to start are 2 to 4% of a 40 to 50% solid content of polycarbodiimide crosslinker on the finish formula. Rarely is over 5% needed. 8 to 10% is a good maximum not to exceed.
Isocyanate crosslinkers generally have higher minimum thresholds, say 6 to 8% of a 100% solids polyisocyanate on finish formula. This higher level is understandable if the isocyanate is not just linking chains but where it must be provided in sufficient quantity to wrap around existing polymer chains where it will eventually moisture-cure to form its interpenetrating network.
Fate of excess crosslinker
What happens at high crosslinker dosages? In the case of isocyanate crosslinkers, whatever the amount used, all of it gets consumed through moisture-cure. Thus residual isocyanate is left unreacted in the coating. This can be detected and quantified using surface infrared spectral techniques. This residual isocyanate always disappears in time because of reaction with water in the finish formula, in the drying tunnel, or with atmospheric moisture afterwards.
An anomaly
What happens as time goes by after leather is dried and allowed to stand in the case where excess polycarbodiimide is used? As already mentioned, polycarbodiimides are much less reactive with water than polyisocyanates, something which should be obvious from the ability to design and deliver a special class of water-borne polycarbodiimide crosslinker.
The same surface infrared spectral technique used to detect residual isocyanate is equally useful to track unreacted carbodiimide functionality.
Figure 2 shows Fourier Transfer Infrared (FTIR) scans of freshly top coated leather after one pass through a typical drying tunnel, followed by two weeks standing at ambient temperatures. Only the slightest trace of carbodiimide remains after two weeks, whereas some three quarters might be expected to be left over based on the large excess used.
Usually less than stoichiometric carbodiimide is used than is required to react with all of the carboxylic acid present but in this case, since a large excess was used, clearly the excess has, after two weeks, fully reacted.
This same behaviour that excess polycarbodiimide is eventually consumed in cases where excess polycarbodiimide crosslinker was used was observed with several other coating formulations and also with some unformulated acrylic latex polymers and polyurethane dispersions, so the loss of excess carbodiimide is not due to non-polymeric ingredients in the formulated finish.
The time required for all of the excess carbodiimide to disappear varies from situation to situation. In one case the result was spectacular, as shown in Figure 3. Indeed this particular polymer lost excess carbodiimide functionality faster than any other polymer tested and produced a significant improvement in tensile properties at the same time as the excess continued to react off.
The specific details of transformations that are taking place remain a subject of continuing research and there is a possibility that in some instances rapid loss of carbodiimide functionality in excess may be a self-condensation similar to what happens in the case of polyisocyanates.
Useful potlife versus survival of crosslinker functionality
A top coat formula was made up with excess carbodiimide (carbodiimide to carboxyl stoichiometry >>1). FTIR techniques were used to gauge the loss of carbodiimide over time. The formulation, allowed to age at ambient temperature, was periodically sampled for analysis and for application as top coat over base coated automotive upholstery leather.
The resulting finished leather was then aged at room temperature, and tested at intervals, initially daily, to determine how many days of room temperature curing were needed to pass a simple wet abrasion test consisting of 500 Veslic abrasion cycles (wet pad/dry leather) at 10% extension of the leather and 1kg load.
Ageing at room temperature for finished leather to achieve this result gets longer the older the age of the hotpotted formula containing carbodiimide crosslinker.
The results are shown in Figure 4. They clearly demonstrate a useful potlife, in this case just short of two weeks, potlife being judged to be expired after the leather required more than two days of room temperature ageing after the top coat was applied to pass the wet abrasion test.
This experiment clearly shows that just tracking stability of the chemical functionality in hotpotted formulated finishes, the analytical technique here based on FTIR spectroscopy, does not tell the whole story. One must also track performance. This was not a normal mix, formulated as it was with high -N=C=N-/COOH stoichiometry.
For a normal mix under normal situations, exceeding a few days as a useful potlife would not be advisable. Under some situations reinocculation with additional carbodiimide can be done. However, such reinoculation with polyisocyanate crosslinkers is not possible.
Higher performance multifunctional carbodiimides
High performance, multifunctional water-dispersible polycarbodiimides in solvent have become available and they find most current use in non-leather finishing markets. Some success has been achieved in the leather industry as a result of the higher degree of crosslinking which is what differentiates this product class from traditional or new water-borne polycarbodiimides. Whereas polyisocyanates are generally considered to outperform polycarbodiimides, the higher performance polycarbodiimides narrow that gap.
The future for polycarbodiimides
Polycarbodiimides are admittedly less aggressive crosslinkers than polyisocyanates but they have practical applications in leather finishing. In base coats, at a low level of 2 to 4%, they can be used to advantage.
Improved stacking as a result of less surface tack out of the dryer is one important benefit often cited. An additional benefit is that it lets the finisher build properties upward from the base coat rather than relying exclusively on heavy crosslinking of the top coat to achieve the desired wear properties.
It should be recognised that at a level of just 2 to 4%, in base coats only, or in base coats and as an adjunct in top coats to assist polyisocyanates, carbodiimides offer an economical way to achieve performance improvement.
One important advantage of aqueous polycarbodiimides is that, while they can be used in-line as is near to necessity with polyisocyanates, unlike polyisocyanates the long potlife recommends them for hotpotting.
For the future, new high performance multifunctional polycarbodiimides suitable for leather finishing are expected to elevate carbodiimide crosslinking to greater prominence.
Polycarbodiimide and polyisocyanate crosslinkers can be used together.
What are the advantages of doing so? There are thought to be two complementary reactions taking place when the two are used together. The high reactivity of the polycarbodiimide with carboxylic acid promotes an early cure, enhancing early development of the properties of the finish while the slower moisture-cure of the polyisocyanate creates a tighter, IPN-type of crosslink.
