Carbonylation Process of MDI: Aniline Oxidative Carbonylation Process

In the mid-1980s, Japan's Asahi Kasei Corporation improved the non-phosgene method by using aniline as a raw material and a palladium-iodide compound catalytic system. Currently, this method has become an important direction in the development of non-phosgene processes.

(1) Carbonylation

Aniline undergoes catalytic oxidative carbonylation with carbon monoxide, ethanol, and oxygen to produce ethyl phenyl carbamate (EPC).

There are few reports on the oxidative carbonylation catalytic system. Asahi Kasei Corporation developed a heterogeneous catalytic system with palladium metal (the main catalyst) and iodide compounds (co-catalysts). This catalytic system enables the catalytic oxidative carbonylation of aniline to produce a high yield of phenyl carbamates with high selectivity, without generating a large number of difficult-to-separate by-products.

In the presence of metal or iodides (such as Nal), the carbonylation reaction occurs at 150-180℃ and 5-8 MPa, with a residence time of 2 hours, achieving an EPC yield of over 95% and selectivity of over 97%. The iodide compounds can be easily recovered by countercurrent extraction with water.

In the traditional method of carbonylation using nitrobenzene reduction, palladium chloride (main catalyst) and Lewis acids (co-catalysts) are used. The Lewis acids must contain chloride ions and a metal (such as FeOC1, FeCl3, or CuCl2) to facilitate the reduction-oxidation reaction. When reacting with PdCl2 and FeCl3, palladium and iron exist both in the solid phase and in the liquid phase. The dissolved palladium cations are reduced to palladium metal, which is then oxidized back to palladium ions by FeCl3, while FeCl3 is reduced to FeCl2, with some forming metallic iron. The reaction mixture is dark black, containing not only metals and metal compounds but also very difficult-to-separate by-products, making the separation of EPC and the recovery of catalysts from the reaction mixture very challenging.

(2) Condensation

EPC undergoes catalytic condensation with formaldehyde aqueous solution to produce methylene diphenyl diaminocarbamate (MDU). The reaction process is divided into two stages: condensation and rearrangement. The first step involves interfacial reactions between two dispersed liquid phases: the organic phase (containing EPC) and the aqueous phase (containing formaldehyde and sulfuric acid), resulting in intermediate products with methylene-amino bonds, such as N-benzyl compounds. Although MDU is produced in large quantities, the reaction is incomplete, retaining intermediate products. This issue is independent of the catalyst type used, meaning intermediates inevitably form and can negatively impact the subsequent thermal decomposition step.

To eliminate these compounds, an intramolecular rearrangement method has been proposed, but it requires very strong acids, such as concentrated sulfuric acid or tricyanomethanesulfonic acid, which are difficult to separate. Asahi Kasei Corporation uses an intermolecular transfer reaction between the intermediate and EPC in the second step, easily converting it to MDU using liquid carboxylic acids, preferably with a pKa value of no less than 4.

The first stage of the condensation reaction uses 40%-60% sulfuric acid as a catalyst at 60-90℃ and atmospheric pressure. Over 40% of EPC is converted into MDU (65%-75%), intermediates (20%-30%), and trisubstituted phenyl carbamate (3%-5%). The reaction effluent easily separates into two phases: an organic phase containing the product and an aqueous phase containing the catalyst.

In the second stage of the condensation, the intermolecular transfer reaction between the intermediate and EPC occurs in the presence of carboxylic acid at 60-90℃. After 30 minutes of reaction, the intermediates are almost completely converted to MDU, resulting in a selectivity of over 95% for MDU. Carboxylic acid can be easily recovered by distillation.

(3) Decomposition

MDU undergoes thermal decomposition to obtain MDI and ethanol, with ethanol recycled for carbonylation. The decomposition reaction proceeds in stages: releasing half of the ethanol from MDU to form mono-carbamate mono-isocyanate (MMI), followed by releasing the remaining alcohol from MMI to convert it to MDI. It is best to use a solid catalyst that does not dissolve in the reaction mixture.

Thermal decomposition is carried out in a solvent at 230-280℃ and 1-3 MPa. After 20 minutes of reaction, the condensation product (mainly containing MDU) converts into 93%-95% MDI, 2%-3% tri-phenyl isocyanate (MTI), and 3%-4% carbodiimide (HN=C=NH) compounds. Ethanol is almost entirely recovered and reused in the carbonylation reaction.

The characteristics of this thermal decomposition step are as follows: high yield of pure MDI, no side reactions, few by-products, fast and complete reaction, and no residual carbamates.

Using aniline as a raw material, the carbonylation process is an oxidative carbonylation reaction rather than a reductive carbonylation reaction (the latter using nitrobenzene as raw material), resulting in a fast reaction rate. In terms of proportions, the relative molecular mass of aniline is only 1/1.32 that of nitrobenzene, making its usage economical.

The aniline method also has the following features: each reaction step has high yield and selectivity; the concentration of diisocyanate exceeds 93%, ensuring high selectivity for MDI; and the separation and recovery of catalysts are easy.

The MDI product obtained by Asahi Kasei Corporation has an NCO mass fraction of 32%-33%, a viscosity of 8-11 mPa·s at 50℃, and a light yellow appearance. Notably, the product contains 93%-95% MDT, with very little polymerized MDI. The MDI contains 92%-97% 4,4'-MDI and 3%-8% 2,4'-MDI. The product has a small amount of carbodiimide compounds (3%-4%) formed during decomposition due to decarboxylation. Unlike the phosgene method, another feature of this method is that the product does not contain hydrolyzable chlorides.