Reverse alchemy: replacing precious platinum with ignoble iron

Homogeneous catalysis, in which the catalyst is mixed directly in with the reaction components, sees widespread use in industrial settings. The catalysts themselves are often complex organometallic compounds that contain a precious metal atom/ion—platinum, rhodium, palladium, rhenium—at their molecular center. 

From an engineering standpoint, a reactor for a homogeneously catalyzed reaction can often be described as a catalyst recovery system first, reactor second. The high cost of these precious metals means that recovery and reuse of the catalyst is essential to making the reactions economic. 

A report published in last week's edition of Science discusses the work of a team of chemists who are looking at ways of obviating the need for the precious metals, replacing them with their more ordinary relatives. The paper focuses on chemistry that is important to the silicone industry.

The authors point out that in 2007, the worldwide silicone industry used 5.6 metric tons (12,345 lbs) of platinum in its various reactions, and a lot of it was lost in the product stream and could not be recovered. As of publication time, platinum is trading at $1,623 per troy ounce—that means the silicone industry, assuming their usage levels are the same as 2007, will spend just shy of $300 million on platinum alone this year. The authors point out that platinum is becoming rarer and rarer and finds itself in other high-tech uses such as fuel cells, so its price will continue to rise and see wild fluctuations.

The focus of this work is hydrosilylation reactions, which link a tertiary silane (something of the form R₁R₂R₃-Si-H, where each R represents some organic functional group) to an existing carbon-carbon double bond. Normally, these reactions involve a platinum catalyst, but the researchers managed to substitute an iron-based one. They found that "in many examples, the new base-metal (iron) catalysts offer advantages in both activity and selectivity over currently employed precious compounds." Basically, their new catalyst worked as well or better than what we're currently using in each of the reactions they looked at.

As a base test case, the team studied the formation of 3-octyl-1,1,1,3,5,5,5-heptamethyltrisiloxane (for those who despised organic chemistry like I did, this is basically a silicone compound terminated by seven methyl groups hanging off the end of an octane chain). It's a compound commercially used in agricultural and cosmetic products. The reagents used to synthesize this compound were (Me₃SiO)₂MeSiH, termed MD'M, and 1-octene. In the presence of the iron catalyst and fairly mild conditions (barely above room temperature), 2000 ppm of the catalyst would yield over 98 percent conversion to the desired product.

From a chemistry standpoint this is great news—near complete conversion is considered a positive, and there was no evidence of any other side reaction products. That means no additional purification before adding the results to commercial products.

The researchers also compared the results to those obtained using the more traditional Karstedt's catalyst (PDF). That requires a reactor temperature of 72 ^oC and a 30 ppm precious metal catalyst. It only went to around 80 percent conversion and, even worse—from a chemical engineering standpoint—it produced a sizable quantity of byproducts from alkene hydrogenation, isomerization, and dehydrogenation. (The authors note, however, that researchers have discovered ways to suppress these side reactions using a different type of precious metal catalyst.)

The authors looked at a second class of reactions: the hydrosilylation (same idea as before) of hydrido- and vinyl-functionalized silicon fluids. These reactions result in a cross-linked silicon polymer (think something akin to vulcanized rubber) that finds uses in the coatings industry. Here, the reaction product is highly viscous and has a morphology that makes recovering much, if any, of the catalyst a major technical challenge.

Here again, the researchers found that the addition of only 500ppm of one of the iron catalysts in a silicone fluid solution resulted in a nicely crosslinked polymer within two hours. As an added bonus, this process does not require the use of toluene, which is needed in the existing reaction to to get the fluid and catalyst to play nicely with one another. Analysis of the final product showed that it was structurally identical to those prepared commercially using platinum catalysts.

Based on the Science article, it seems like this class of bis(imino)pyridine iron dinitrogen complexes provides a promising route for industrially relevant reactions. The simple fact that it replaces a high cost material with a readily available and cheap metal is a boost right away. We still need to know how readily (affordably) the iron catalysts can be made and how easily they can be separated back out from the product stream. Even if it is cheaper then platinum, no one wants to see their catalyst only get used once and then be lost forever.

Science, 2012. DOI: 10.1126/science.1214451  (About DOIs).

Listing image by Illustration by Paul J. Chirik, et al.