PROCESS FOR PREPARING POLYISOCYANATES CONTAINING URETHANE GROUPS

Abstract

The present invention relates to a process for preparing a polyisocyanate composition containing urethane groups, and to the polyisocyanate compositions containing urethane groups that are obtained or obtainable by this process.

Claims

1. A process for preparing polyisocyanates containing urethane groups by reacting an excess of diisocyanate with a composition comprising at least one polyhydroxy compound, wherein the reaction is operated continuously in a reaction system which comprises at least two residence apparatuses.

2. The process according to claim 1, wherein the diisocyanate is 2,4-toluene diisocyanate or a mixture of 2,4 toluene diisocyanate with up to 35 wt % of 2,6-toluene diisocyanate, based on the total weight of the mixture.

3. The process according to claim 1, wherein the polyhydroxy compound is selected from the group consisting of di- to tetrahydric alcohols having a molecular weight of 62 to 146 g/mol and polyether polyols prepared from them by addition reaction of ethylene oxide and/or propylene oxide and having a molecular weight, calculable from hydroxy group content and hydroxyl functionality, of 106 to 600 g/mol.

4. The process according to claim 1, wherein the polyhydroxy compound is trimethylolpropane and/or diethylene glycol.

5. The process according to claim 1, wherein the composition consists of trimethylolpropane and/or diethylene glycol.

6. The process according to claim 1, wherein the reaction system comprises a stirred tank cascade having 2 to 10 stirred tanks.

7. The process according to claim 1, wherein the reaction system comprises at least one stirred tank and a tubular reactor connected downstream to the at least one stirred tank.

8. The process according to claim 1, wherein the reaction system is charged with diisocyanate and heated at 50 to 120° C. before the continuous reaction started by introducing continuous feed streams of the diisocyanate and the composition comprising at least one polyhydroxy compound into the reaction system.

9. The process according to claim 1, wherein during the continuous operation, the diisocyanate, before being added to the reaction system, has a temperature ≤40° C.

10. The process according to claim 1, wherein the composition comprising at least one polyhydroxy compound, before being added to the reaction system, has a temperature ≤65° C.

11. The process according to claim 1, wherein the diisocyanate and the composition comprising at least one polyhydroxy compound are passed into the first residence apparatus in an NCO:OH equivalent ratio of between 2.5:1 and 20:1.

12. The process according to claim 1, wherein a first substream of the composition comprising at least one polyhydroxy compound is metered into the first residence apparatus and a further substream of the composition is metered into at least one further residence apparatus.

13. The process according to claim 1, wherein the reaction system consists of n residence apparatuses and the composition comprising at least one polyhydroxy compound is metered into the first n−1 residence apparatuses.

14. A polyisocyanate containing urethane groups obtained by a process according to claim 1.

Description

EXAMPLES

Example 1 (Batch, Comparative Example)

[0035] A jacketed 15 L stirring vessel was charged with 10 kg of toluene diisocyanate (mixture of 80% of the 2,4-isomer and 20% of the 2,6-isomer) and this initial charge was heated to 85° C. Then, with stirring and over the course of 2.5 h, 1.67 kg of a polyol mixture consisting of trimethylolpropane and diethylene glycol in a molar ratio of 3:2 (trimethylolpropane:diethylene glycol) was metered in and the reaction mixture was stirred for a further 30 minutes. Then unreacted toluene diisocyanate was distilled off under vacuum conditions in a thin-film evaporator. The bottom product obtained was a colourless resin (resin yield 64%), which was thereafter diluted with ethyl acetate to a solids content of 75%. This gave a product having a viscosity of 1440 mPas, a residual monomer content of 0.31 wt % and an NCO content of 13.2%.

Example 2 (Stirred Tank Cascade, Inventive)

[0036] The reaction system used is a cascade of 4 stirred tanks each with a capacity of 400 L. Before starting the continuous reaction, all 4 stirred tanks were charged with toluene diisocyanate (a mixture of 80% of the 2,4-isomer and 20% of the 2,6-isomer) and these initial charges were heated to 72° C. Then, for preparing the polyisocyanate, continuous streams of toluene diisocyanate and polyol mixture of the same composition as in Example 1 were dosed in a mass ratio of 10:1 into the first reactor of the cascade while the temperature inside the reactor was maintained at 72° C. via the jacket. The temperature of the diisocyanate stream was 28° C., the polyol stream was at 60° C. The overall feeding rate was 1.3 m.sup.3/h, resulting in an average residence time in the cascade of about 1.2 h. At these conditions a conversion of >99.9% was achieved, based on the OH groups of the polyol mixture. Following discharge from the last stirred tank, unreacted toluene diisocyanate was distilled off under vacuum conditions in a thin-film evaporator. The bottom product obtained was a colourless resin (resin yield 41%), which was thereafter diluted with ethyl acetate to a solids content of 75%. The product had a viscosity of 1465 mPas, a residual monomer content of 0.29 wt % and an NCO content of 13.3%.

Example 3 (Tubular Reactor, Comparative Example)

[0037] A heated reaction tube (internal diameter 80.8 mm, length 200 m) was used for the reaction of toluene diisocyanate (mixture of 80% of the 2,4-isomer and 20% of the 2,6-isomer) and a polyol mixture of the same composition as in Examples 1 and 2, at 72° C. Different mass ratios were tested, with the overall metering rate being selected in each case such that the residence time of the reaction mixture in the reaction tube corresponded to the residence time from Example 2. Complete conversion of the OH groups was ensured in this way. Independently of the mass ratio, which was varied between 6:1 and 10:1 (toluene diisocyanate:polyol mixture), however, it was not possible to obtain a colourless resin as a bottom product following removal of the unreacted excess of toluene diisocyanate. In every case, a visible, yellowish discoloration occurred. Furthermore, after just 2 days, solid deposits were found in the entry region of the reaction tube.

Example 4 (Stirred Tank Cascade with Polyol Split, Inventive)

[0038] The reaction was performed in analogy to Example 2, with the following modifications being made: the mass ratio between toluene diisocyanate and polyol was lowered to 8.9:1 and the polyol mixture was metered half into the first and half into the second stirred tank in the cascade. The crude product was worked up as in Example 2. The bottom product obtained was a colourless resin (resin yield 46%), which was thereafter diluted with ethyl acetate to a solids content of 75%. The product had a viscosity of 1450 mPas, a residual monomer content of 0.30 wt % and an NCO content of 13.2%.

Discussion of the Examples

[0039] While the batch reaction (Example 1) does lead to a high resin yield, deviations in quality occur naturally in the course of production of a plurality of successive batches. For example, increased tendency to crystallization was observed, implying reduced shelf life for certain batches, presumably attributable to the non-steady-state regime. As the reaction progresses, there is a change in the composition in the reactor, and the associated increase in viscosity impairs the removal of heat. Furthermore, by comparison with a continuously operated process, a batch process necessitates larger apparatus and a greater holdup, since the space-time yield is reduced for example as a result of set-up times. For industrial production in particular, batch processes quickly reach their limits, since the surface-to-volume ratio is unfavourable in larger apparatus. Ultimately, the metering rate would have to be reduced in order to control the reaction temperature, thus leading to a further deterioration in the space-time yield. Also this batch reaction process is not as energy efficient as the inventive continuous process. The reason being that all diisocyanate has to be heated to reaction temperature and then the reaction requires high amounts of cooling energy to maintain the temperature despite the exothermic reaction. In contrast, in the inventive processes, only a small portion of the diisocyanate has to be heated in order to initiate the reaction. During continuous operation, the heat of reaction is absorbed by the reactants which are fed at a lower temperature.

[0040] A continuous reaction in a tubular reactor is likewise unrewarding, as shown by Example 3. All in all, while the tubular reactor does have a large surface area for the removal of heat, a large part of the heat of reaction is liberated right at the start of the reaction. The deposits observed in the entry region show that within this region there were undesirably high reaction temperatures and hence unwanted secondary reactions going as far as to build up the polymer. In theory, a very high NCO:OH ratio should overcome this issue but at the cost of a low resin yield and consequently high amounts of diisocyanate that have to be removed during distillation.

[0041] A single continuously operated stirred tank was not used to carry out the reaction, since in a set-up of that kind, naturally, it is not possible to achieve complete conversion. In order to come as close as possible to this full conversion, the reactor selected would have to be very large, leading in turn to difficulties with mixing and with heat removal. Furthermore, reaction in a single continuously operated stirred tank is known to result in a broad residence time distribution, thus promoting unwanted secondary reactions.

[0042] Reaction in a stirred tank cascade, in contrast, is characterized by effective heat removal and constant reaction conditions. Furthermore, this reaction regime enables the metered introduction of precooled starting materials, meaning that the cooling performance of the reactor itself is no longer a limiting factor. The resin yield in such a system is indeed somewhat lower, but this is outweighed by the advantages. The reduced yield can also be counteracted by metering the polyol component not all into the first reactor, but instead distributing it over a number of reactors, as shown by Example 4.