57th Annual Report on Research 2012 Under Sponsorship of the ACS Petroleum Research Fund

Fossil fuels account for roughly 80% of mankind's current energy portfolio with roughly one third coming from petroleum based sources. Petroleum resources are especially important to the transportation sector and are driving annual atmospheric emissions of 2.5 Gt of carbon in the form of CO₂. Given the evidence linking anthropomorphic carbon emissions to climate change, the long term use of petroleum resources will be reliant on the development of systems capable of sequestering and recycling CO₂. The reduction of CO₂ to CO would allow for the mitigation of atmospheric CO₂ levels while producing a versatile energy rich commodity chemical. This reaction is a proton-coupled multiple electron transfer process (2e^– + 2H⁺). Since CO is an important chemical for the production of synthetic petroleum and hydrocarbons using Fischer-Tropsch methods our work seeks to uncover the fundamental science needed to drive the conversion of CO₂ to liquid fuels.

For the reaction CO₂ + 2e^– + 2H⁺ → CO + H₂O the standard potential is E⁰ = –0.53 V at pH = 7.0, however, most previously studied CO₂ reduction catalysts display poor selectivities, limited stabilities and require large overpotentials to effect CO₂ activation at an appreciable rate. These shortcomings can be overcome by judicious ligand and catalyst design. For example, the instability of some of the most successful CO₂ reduction catalysts is due to the high lability of the ligand scaffolds used to support low-valent catalytic centers. The sluggish kinetics observed for such platforms can be attributed to the fact that these systems do not minimize the large nuclear reorganization energy associated with binding of CO₂ to reduced metal centers. In confronting these issues, we have developed a new kinetically robust platform for CO₂ conversion and are designing platforms with hydrogen-bonding functionalities to stabilize M–CO₂^– adducts and promote the controlled delivery of protons to drive C–O bond cleavage to efficiently generate CO and H₂O. In pursuing these research opportunities, we have identified the following objectives.

  The development of transition metal carbene complexes for CO₂ activation has been a major area of focus in our lab.[1] In carrying out this work, we have developed complexes of Ni and Pd supported by the Pyridyl-DiCarbene (PDC^R) ligand scaffolds depicted in Figure 1A. The two NHC moieties on the ligand backbone support the electron-rich Ni and Pd metal centers needed for catalysis and are suitable for activating CO₂. Moreover, the modular synthesis of these compounds allows for the steric and electronic properties of these architectures to be tuned with ease and fidelity. Crystal structures for a homologous set of these palladium CNC-pincer complexes are shown in Figure 1B. Inspection of these ORTEP diagrams clearly illustrates how the size and accessibility of the molecular cleft is attenuated by variation of the pincer R-group. Attenuation of the steric bulk about the pincer periphery significantly impacts the kinetics and efficiacy and catalysis at the metal center.

  We have successfully shown that the palladium systems of Figure 1A are competent electrocatalysts for reduction of CO₂. Our work in this area was recently recognized by Oak Ridge Associated Universities and highlighted by Nature Magazine.[2] Shown in Figure 2 are the cyclic voltammetry traces recorded for [Pd(PDC^Me)MeCN](PF₆)₂ in MeCN. The CV trace recorded under N₂ (Figure 2, blue) reveals the redox activity of the Pd-pincer architecture with reduction waves centered at Eº ~ –1.0 V and –1.3 V (vs. Ag/AgCl), which correspond to Pd^II/I and Pd^I/0 couples, respectively. There is also a small peak at Eº ~ –1.7 V, which corresponds to reduction of an electrogenerated dimer. Notably, the size of this final reduction wave is small compared to that observed for related palladium complexes supported by phosphine ligands.[3] Suppression of [Pd^I(PDC^Me)]⁺ dimerization by our inert pincer ligands is an important development, since bimetallic palladium species are not competent catalysts for CO₂ reduction. Accordingly, our systems offer improvement over previously studied complexes that display low TON for CO₂ conversion due to this dimerization process.

  Minimization of Pd^I—Pd^I formation is reflected by the efficacy with which [Pd(PDC^Me)MeCN](PF₆)₂ activates CO₂. CVs recorded for this complex under 1 atm of carbon dioxide result in a more pronounced Pd^I/0 wave at –1.3 V, consistent with reduction of CO₂ by the electrogenerated Pd⁰ (Figure 2, red). Repetition of this experiment in the presence of a weak acid results in further enhancement of this redox wave. This observation is consistent with the need for proton transfer to complete the PCET conversion of CO₂ to CO (Eq. 1) and is evidenced by the broad electrocatalytic wave observed between –1.35 and –1.9 V (Figure 2, dashed red). 

  The pincer constructs of Figure 1 proficiently catalyze the electrochemical conversion of CO₂ to CO. Bulk electrolysis of solutions of these complexes in acidic MeCN under an atmosphere of CO₂ has allowed for the products of the electrocatalytic reduction to be ascertained by gas chromatography. Our Pd(PDC) architectures display exquisite selectivity for CO production over other reduced carbon species including CH₃OH, HCO₂H and H₂CO. Moreover, we have shown that H₂ is not formed as an unwanted side product via direct reduction of protons by the Pd catalyst. Accordingly, our novel Pd(PDC) construct displays far higher stability, selectivity and efficiency compared to homologous Pd-phosphine complexes.

References: