Process for reducing the permeability of polyalkene layers to hydrocarbons

The invention relates to a process for reducing the permeability of a polyalkene layer to liquid and gaseous hydrocarbons by adhering to the polyalkene layer a two-component polyurethane lacquer composed of a polyester polyol and a prepolymer of an aromatic diisocyanate.

Such a process is known from DE-A-3009412. In that process the surface to be provided with the lacquer is roughened in advance mechanically or chemically in order to promote the adhesion.

It has been found that, for instance, petrol tanks produced from polyethylene and provided with a coat of lacquer in the manner described in DE-A-3009412 do at first indeed show a reduced permeability to hydrocarbons, but after a number of temperature fluctuations between +100 °C and -40 °C on petrol-saturated test specimens an increased permeability is shown again. This phenomenon might be attributed to the polyurethane resin or to the adhesion between the coat of lacquer and the substrate or to a combination of the two.

The object of the invention is to provide a process for reducing the permeability of a polyalkene layer to liquid and gaseous hydrocarbons by coating the polyalkene layer with a polyurethane lacquer which, compared with the process known in the art, produces an improved result in terms of permeability even after a great many temperature fluctuations.

This has been achieved by choosing as the polyester polyol component of the lacquer the reaction product of an esterification of a dicarboxylic acid and an excess of a trivalent alcohol, followed by a ring-opening polymerization of a lactone, and by subjecting the surface of the polyalkene layer to be provided with the lacquer to an oxidative treatment.

The oxidative treatment applied preferably consists in flaming. The fact is that with this kind of lacquer this method has been found to produce very favourable results. The application of an intermediate layer (primer) improving the adhesion may result in an improvement of the whole system.

In all likelihood the lactone plays a decisive part in the chosen polyester polyol, which may be due to the high functionality (number of OH groups available) in the polyester polyol, so that a high degree of crosslinking can be achieved.

It is assumed that both the special polyester polyol component of the lacquer and the exceptionally good adhesion of the lacquer to the oxidized substrate occuring in combination with it produce the results shown in the example below.

Exampel I

One-litre bottles weighing 50 grammes produced by blow moulding from high-density polyethylene (STAMYLAN 6761 of DSM) are subjected externally by flaming to an oxidative treatment and are successively provided externally by spraying with a coat of lacquer about 50 µm thick. The lacquer to be be sprayed on consists of a mixture of a prepolymer of a polyol and an aromatic diisocyanate (URADUR P 49 of DSM Resins) and the reaction product of an esterification of isophthalic acid and trimethylol propane, [CH₃ CH₂ C (CH₂0H)_3], and the subsequent ring-opening polymerization of monomeric caprolactone (URALAC 331 of DSM Resins). The bottles are subsequently dried at a temperature of 100 °C for 30 minutes.

After this treatment the bottles are filled to 50 % of their capacity with a test petrol CEC RF-01 according to DIN 70030 Teil 1 and sealed airtight. Non-treated bottles, too, are filled with this test petrol to 50 % of their capacity. The treated and non-treated bottles are stored for 120 days at 20 °C. After that the average percentage of the loss in weight of the petrol of the treated and non-treated bottles is determined. It is found that the bottles treated according to the invention show a loss in weight of about 5 X, whereas the non-treated bottles show a loss in weight of about 32 X.

The improvement of the treated bottles, more than sixfold according to the above test, constitutes a result that cannot be reached with the process known so far. Probably the special polyester polyol component of the lacquer plays a double role in the realization of this result, in which not only the poor permeability of the applied lacquer as such, but also the adhesion to the flamed polyethylene surface may be important. The fact is that tests in which bottles treated according to the invention are subjected to a great many cold-heat fluctuations referred to above have been found not to show any appreciable change in the adhesion between lacquer and substrate, which is unlike the processes known in the art in which the coat of lacquer is found to become detached. Neither is any appreciable reduction of the adhesion observed if the bottles treated according to the invention are subjected to the hot water test for three months. This test, which is known to be a test method having a very rigorously unfavourable effect on the adhesion, comprises the immersion of a test specimen in water of 80 °C for a certain length of time.

The process can yet be improved by adding a pigment such as aluminium or carbon black as shown by examples II and III below.

Example II

The process of example I is repeated. The difference is that aluminium powder is added to the polyester polyol component in an amount of 14.2 grammes to 100 grammes dry binding agent.

After 120 days the loss in the weight of the test petrol is found to be less than 4 7.

Example III

The process of example I is repeated. The difference is that carbon black is added to the polyester polyol component in an amount of 9.9 grammes to 100 grammes dry binding agent. After 120 days it is found that the loss in the weight of the test petrol is less than about 1 % (wt).