July 25, 2011
A little hydrogen boosts selectivity to propylene oxide. Propylene oxide (PO) process technology has undergone dramatic changes over the past several years. Beginning in 1974, the PO process that was generally considered to be the low-cost method was the so-called propylene oxide–styrene monomer (POSM) process. In it, 2.3 tons of styrene is produced for each ton of PO. Therefore, POSM technology users are in the styrene business whether they want to be or not.
This type of process is called a “2-for-1” process—nice in concept, but in practice it only works if the markets for both products are growing at about the same rate. In POSM, styrene was often in oversupply, resulting in very low prices and causing a loss in overall profitability for POSM producers.
Over the past several years, two new processes for making PO with no coproduct have been commercialized. One is Sumitomo’s cumene-based process, and the other is based on H2O2. The “holy grail” of the PO business, however, is a process for direct oxidation of propylene by oxygen to PO. This route has eluded commercialization because of low yields.
M. Haruta and colleagues found that the direct oxidation of propylene with oxygen can be carried out with very high selectivity if a catalytic amount of hydrogen is used. To demonstrate their method, the inventors first passed propylene, oxygen, and argon (1:1:17 vol/vol/vol) over a potassium-modified gold-containing titanium silicalite catalyst (Au/TS-1-K1) with a gaseous hourly space velocity of 4000 h–1 at 200 °C. After 2 h, propylene conversion was only 0.33%, and selectivity to PO was 11.8%. CO was the main product (56.5% selectivity).
The experiment was repeated with 1 vol% hydrogen added to the gas mixture. Propylene conversion and PO selectivity increased dramatically to 2.3% and 70%, respectively, and CO selectivity decreased to 24%. This process has promise, but propylene conversion must be increased to make it commercially viable. (Tokyo Metropolitan University and National Institute of Advanced Industrial Science and Technology [Tokyo]; US Patent 7,973,184, July 5, 2011; Jeffrey S. Plotkin)