Flow Chemistry with Supported Ionic Liquid Phase Grubbs-type Catalysts

Scholz, J,; Loekman, S.; Szesni, N.; Hieringer, W.; Görling, A.; Haumann, M.; Wasserscheid, P.  “Ethene-Induced Temporary Inhibition of Grubbs Metathesis Catalysts.”  Adv. Synth. Catal.  2011, 353, 2701-2707.

An interesting article emerged late last year in which the authors wanted to obtain some kinetic data employing immobilized Grubbs catalysts in continuous flow gas-phase reactions of unfunctionalized olefins. They immobilized a variety of Grubbs catalysts onto calcined silica using the SILP (supported ionic liquid phase) technique. This technique involves supporting the catalyst in a thin film of ionic liquid on a porous support. These types of materials had previously been reported and were shown to be effective catalysts for RCM in batch reactions with an ability to be recovered and recycled a few times.1

When the SILP materials were employed under continuous flow conditions using fixed-bed reactors to study the self-metathesis of propene (products = ethylene and 2-butene with an equilibrium conversion of 34%) a very rapid decrease in propene conversion was observed, indicative of catalyst decomposition or deactivation. In an attempt to determine the effect that the ionic liquid had on this phenomena, a series of different SILP’s were prepared with different ionic liquid anions; however, no direct correlations between the anion’s coordination strength or the thermal/hydrolytic stability were observed. Upon further cogitation, an inverse correlation was observed between the stability of the SILP catalyst and the solubility of ethylene within the specific ionic liquid (i.e. the least stable SILP catalyst is composed of the ionic liquid that has the highest ethylene solubility).

In an attempt to validate the hypothesis that ethylene was responsible for the dramatic decrease in catalytic activity, a series of experiments were performed. First off, in a set of independent experiments, the catalyst bed was first subjected to a flow of ethylene gas for varying amounts of time before propene was introduced to the system. This set of experiments revealed that indeed, the catalyst beds that were subjected to larger amounts of ethylene had diminished initial activity upon the switch to propene. However, in a somewhat intriguing observation, it was seen that after the switch to propene, the catalyst actually regained some activity, reaching a maximum conversion shortly after the propene switch, before observing the same decrease in propene conversion described previously. This was suggestive that some sort of catalyst deactivation and subsequent re-activation process was in operation.

The authors performed some DFT calculations on possible reaction intermediates, and observed that the lowest energy species were ethylene-based metallacycles that would not be involved in the productive metathesis catalytic cycle (e.g., ruthenacycle F in the proposed catalytic cycle shown below). This effectively serves to remove “active” catalyst from the productive cycle, thus, diminishing the apparent overall catalytic activity.  In this case, ethylene is responsible for shifting the metallacycle equilibrium to an inactive or dormant state without irreversible catalyst deactivation (i.e., catalyst decomposition).

To test this second hypothesis, an experiment was devised where a metathesis reaction having a negligible contribution from ethylene was monitored using this fixed-catalyst-bed protocol. Flowing a C4 gas mixture of 1-butene (6%) and 2-butenes (94%) diluted in an inert butanes atmosphere through the fixed-bed with a short residence time of 8 s resulted in the productive metathesis reaction generating products 1-propene and 2-pentene with an equilibrium conversion of 96%. In impressive nature, over the course of 20 h, there was no apparent catalyst deactivation for this process with the conversion remaining constant at 96%. At this stage, a 50:50 mixture of ethylene and 2-butene was then passed through the catalyst bed. As expected, the conversion of 2-butene decreased dramatically over time due to the catalyst deactivation (dormant state) in the presence of ethylene. When the substrate stream was switched back to the initial diluted 1-butene/2-butenes mixture, the equilibrium ratio of 96% conversion was quickly achieved, indicating catalyst re-activation. To push the limits, this feedstock switch was repeated two additional times, with equilibrium being reached quickly for each switch back to the diluted C4 stream. Very impressively, the total time on-stream for this catalyst bed was greater than 500 h without any observed decrease in the catalyst efficiency for the conversion of the C4 stream. However, the authors point out that an irreversible effect is observed, having a negative impact on the conversion of 2-butene, each time the stream is cycled back to the ethylene/2-butene mixture.

This is a very interesting phenomenon that was observed, and it will be crucial to determine the exact nature of this ethylene induced “deactivation” as ethylene plays a major role in a number of olefin metathesis processes as either a reactant or reaction by-product. Do similar processes occur in homogeneous catalysis? Is this isolated to heterogeneous SILP-based systems? What about other types of heterogeneous catalyst systems? These are important questions that will need to be addressed.

1 (a) Hagiwara, H.; Okunaka, N.; Hoshi, T.; Suzuki, T.  “Immobilization of Grubbs Catalyst as Supported Ionic Liquid Catalyst (Ru-SILC).”  Synlett.  2008, 1813-1816.  (b) Hagiwara, H.; Nakamura, T.; Okunaka, N.; Hoshi, T.; Suzuki, T.  “Catalytic Performance of Ruthenium-Supported Ionic-Liquid Catalysts in Sustainable Synthesis of Macrocyclic Lactones.”  Helv. Chim. Acta  2010, 93, 175-182.

Be Sociable, Share!

Leave a Reply

Your email address will not be published. Required fields are marked *