ABSTRACT
Abstract Heterogenization of homogeneous metal complex catalysts represents one way to improve the total turnover number for expensive or toxic catalysts. Two case studies in catalyst immobilization are presented here. Immobilization of Pd(II) SCS and PCP pincer complexes for use in Heck coupling reactions does not lead to stable, recyclable catalysts, as all catalysis is shown to be associated with leached palladium species. In contrast, when immobilizing Co(II) salen complexes for kinetic resolutions of epoxides, immobilization can lead to enhanced catalytic properties, including improved reaction rates while still obtaining excellent enantioselectivity and catalyst recyclability. Introduction Supported metal complexes (1) have been studied for many years due to their potential for combining the best attributes of both homogeneous and heterogeneous catalysis – high reaction rates and selectivities coupled with easy catalyst recovery. Unfortunately, in many cases, immobilized metal complex catalysts display the disadvantages of each class of catalysts, poor recyclability due to catalyst leaching, low reaction rates due to diffusional limitations, and poor selectivities due to the presence of multiple types of active sites. Indeed, although hundreds of different metal complexes have been immobilized on virtually every type of known catalyst support, supported metal complex catalysts still are relatively poorly understood compared to the more typical homogeneous (e.g. soluble metal complexes) and heterogeneous (e.g. supported metals) catalysts that dominate commercial processes. To this end, we have undertaken a detailed, long-term investigation of two families of supported metal complex catalysts, supported Pd pincer complexes for use in C-C couplings such as Heck and Suzuki reactions and supported Co salen complexes for epoxide ring-opening reactions. These two catalyst systems represent interesting targets for detailed study. Pd(II) pincer palladacycles have been proposed in the literature to be well-defined, stable Pd(II) catalysts that are active in Heck or other coupling reactions (2-7), potentially via a controversial Pd(II)-Pd(IV) catalytic cycle (8). Here we summarize our studies of supported Pd(II) PCP and SCS pincer
complexes on both insoluble (porous silica) and soluble supports (polymers). Co(III) salen complexes represent powerful enantioselective catalysts for epoxide ringopening reactions (9). In this case, the design of the proper support is of paramount importance, as the transition state of the reaction involves two Co salen centers, and hence the supported system must be able to accommodate such a transition state. Here we explore our use of different soluble polymeric supports of differing backbone and side-chain structure and evaluate the role of the support on the catalytic properties in the hydrolytic kinetic resolution of rac-epichlorohydrin. Results and Discussion A variety of Pd(II) palladacycle complexes have been reported over the past decade for applications in Heck, Suzuki and other coupling reactions (10). These precatalysts appeared quite stable under reaction conditions and little evidence was observed for the formation of Pd(0), the usual form of active palladium in these coupling reactions. For this reason, a new catalytic Heck cycle was hypothesized to account for the catalytic activity observed when using these precatalysts a Pd(II)- Pd(IV) cycle, rather than the usual Pd(0)-Pd(II) cycle. Over the last 5-7 years, it has been systematically shown that bidentate palladacycles based on SC, NC and PC ligands (Figure 1) decompose to liberate soluble, ligand-free Pd(0) that is the active catalyst (11-17). However, up until 2004, tridentate palladacycle such as Pd(II) pincer complexes of NCN, SCS, or PCP ligation (Figure 1) had still been thought to potentially be stable complexes that catalyze the Heck coupling by a Pd(II)-Pd(IV) cycle (18-19). Here we share our recent results that conclusively show that SCS (2021) and PCP (22) Pd(II) pincer complexes decompose to liberate soluble catalytic Pd(0) species, and that solid-supported Pd(II) pincers simply represent reusable soluble precatalyst sources, rather than the previously hypothesized stable recyclable catalysts (20-23).
