Safe reactors with no gas

Hydrogenation reactions are extremely important in the pharmaceutical and fine chemical industries - adding hydrogen across a double bond can create chiral molecules.

Hydrogenations are, however, difficult reactions, because hydrogen is barely soluble in the solvents used in the processes. One way of improving the solubility is to replace the organic solvent with supercritical carbon dioxide. Hydrogen dissolves completely in supercritical carbon dioxide, but there's a problem - pressures of around 100bar are required, and handling hydrogen at this pressure is extremely dangerous.

Now, a research group at the University of Nottingham team, led by Martyn Poliakoff, has devised a new type of reactor that renders high-pressure gas handling unnecessary. And the London-based HEL Group has been granted an exclusive world-wide license to market the technology the University has developed.

Rather than using compressors, the new equipment generates the high-pressure gas at the inlet of the reactor itself, using a simple liquid, formic acid, as a starting material.

There's no need to store, meter or control any gases, so the equipment is simpler than a conventional supercritical CO2 rig, and the possibility of gas leaks is also reduced.

The system works by passing formic acid over a catalyst containing 5% platinum, which causes it to decompose into carbon dioxide and hydrogen.

Another liquid, ethyl formate, is also passed over the catalyst, decomposing to form carbon dioxide and ethane, which have similar properties, both becoming supercritical under the same conditions. Altering the flow-rate of the formic acid and formate allows the researchers to change the proportion of hydrogen in the gas stream to that required for the hydrogenation reaction.

'Thus, the creation of a supercritical mixture of H2/CO2 is reduced to the pumping of two common liquids into small pieces of heated tubing,' explained Jasbir Singh of HEL Group.

Supercritical reactors already have an important environmental advantage over conventional solvent-based reactions - to recover the product, operators merely have to reduce the pressure. This changes the supercritical CO2 back into a normal gas, and it evaporates leaving the product behind. By contrast, conventional reactions require many process steps to separate the product from the solvent - and then the solvent must be treated or disposed of itself.

The Nottingham/HEL reactor is a lab-scale unit, but the process has been trialled at a larger scale. Specialities producer Thomas Swan has built a pilot plant using the process to hydrogenate isophorone to trimethyl cyclohexanone (TMCH), which is used to make styrene.

This is a tricky reaction under normal conditions, because it tends to generate a large number of by-products which reduce the useful yield of the process and push the costs up.

The Thomas Swan plant handles 100kg/hr of material, 400 times larger than the laboratory model. However, it replicates the laboratory results exactly - it produces TMCH alone, with no side-products, and exceeded all the quality parameters for the product, such as colour and acidity, without the need for any downstream processing.

The system shows great promise for further supercritical processes, says Singh, because it can be adapted for other reactions. If supercritical CO2 alone is required, the formic acid feed can be omitted. This creates the potential to replace a large number of solvent-based reactions with 'green' solventless alternatives, without the risk of handling high-pressure gases, he says.