Lili Zhao

The Catalytic Role of N-Heterocyclic Carbene in a Metal-Free Conversion of Carbon Dioxide into Methanol: A Computational Mechanism Study

A density functional theory study at the M05-2X(IEFPCM, THF)/6-311+G**//M05-2X/6-31G* level has been conducted to gain insight into the catalytic mechanism of the first metal-free N-heterocyclic carbene (NHC)-catalyzed conversion of carbon dioxide into methanol. Among the various examined reaction pathways, we found that the most favorable leads to the experimentally detected intermediates, including formoxysilane (FOS), bis(silyl)acetal (BSA), silylmethoxide (SMO), and disiloxane (DSO). However, our study also revealed that formaldehyde (CH(2)O), generated from the dissociation of BSA into DSO and CH(2)O via a mechanism somewhat similar to the Brook rearrangement, should be an inevitable intermediate, although it was not reported by the experimentalists. When NHC catalyzes the reactions of CO(2)/FOS/CH(2)O with silane, there are two activation modes. It was found that NHC prefers to activate Si-H bonds of silane and push electron density to the H atoms of the Si-H bonds in favor of transferring a hydridic atom of silane to the electrophilic C center of CO(2)/FOS/CH(2)O. This holds true in particular for the NHC-catalyzed reactions of silane with FOS/CH(2)O to produce BSA/SMO. The preferred activation mode can operate by first passing an energetically unfavorable NHC-silane local minimum via pi-pi interactions or by directly crossing a transition state involving three components simultaneously. The activation mode involving initial coordination of NHC with the electrophilic C atom of CO(2)/FOS/CH(2)O is less favorable or inoperable. The predicted catalytic mechanism provides a successful interpretation of the experimental observation that phenylsilane is more efficient than diphenylsilane in performing the conversion.

Transfer Hydrogenation of Imines with Ammonia–Borane: A Concerted Double-Hydrogen-Transfer Reaction†

Ammonia–borane (H3N-BH3, AB) is considered a feasible material for chemical hydrogen storage owing to its ideally very high storage capacity (19.6 weight% H) and thus has attracted much attention. Dehydrogenations of AB were accomplished either thermally or by transition metal catalysis. Considering AB as a significantly polarized molecule, we reasoned that it could be dehydrogenated by direct reaction with a similarly polarized unsaturated compound by the rarely explored reaction mode of double H transfer (Scheme 1).

Computationally Designed Metal-Free Hydrogen Activation Site: Reaching the Reactivity of Metal−Ligand Bifunctional Hydrogenation Catalysts

In this study, a strategy to design a metal-free hydrogen activation site has been proposed. On the basis of our so-called sp(3) carbon bridged FLPs (Frustrated Lewis Pairs), we first hypothesized that a more reactive activation site should arrange the nitrogen lone pair and the boron vacant orbital to lie in the same plane face-to-face, because such orbital orientations can simultaneously enhance the interaction between the nitrogen lone pair and the H(2) sigma* antibonding orbital and the interaction between the boron vacant orbital and the H(2) sigma bonding electrons. To verify that such an active site is achievable, we then computationally designed molecules and studied their reactions with hydrogen. The energetic results show the designed molecules are indeed more reactive than the sp(3) carbon bridged FLPs. Some of the hydrogen activations reach kinetics and thermodynamics comparable with those of the hydrogen activations mediated by the well-known metal-ligand bifunctional hydrogenation catalysts. The designed molecules could be the targets for experimental synthesis. The pattern of the proposed active site can be based to design similar molecules for metal-free hydrogenations.

Catalytic metal-free ketone hydrogenation: a computational experiment

A computational study has been carried out to examine if the metal-free catalyst (1) designed for imine hydrogenation is able to hydrogenate ketones, using the cyclohexanone (3) and its derivatives (4-6) as ketone models. The catalytic cycle includes two major steps: hydrogen activation and hydrogen transfer. The concerted pathway in the hydrogen transfer step is preferred over the stepwise pathway. The two separated steps for hydrogen activation and hydrogen transfer can benefit the hydrogen addition to the substrates (e.g., ketones) which do not have strong Lewis base centres, because the substrates need not to be involved in the hydrogen activation. In general, the larger the steric effect of the substrate is, the less severe the side reactions become, and the more difficultly the desired reaction occurs. The energetic results show that the hydrogenations of 3-5 are kinetically and thermodynamically feasible under ambient conditions, but the hydrogenation of 6 is less energetically favourable. Therefore, it is important to establish a proper balance between promoting the desired reaction and meanwhile avoiding the undesired reactions. The issue of the resting state, caused by forming stable alkoxide complexes like in the ketone hydrogenation catalyzed by the metal-ligand bifunctional catalysts, is also discussed.

Insight into the relative reactivity of “Frustrated Lewis pairs” and stable carbenes in activating H2 and CH4: A comparative computational study

Computational study has been conducted to gain insight into the relative reactivity of stable carbenes (1 and 2) and typical frustrated Lewis pairs (FLPs, 3-6) in activating H(2) and CH(4). For the FLP H(2) activations, despite the quite different basicities of the Lewis base components, they have comparable reactivities. The unexpected relative reactivity can be attributed to the following two factors: (i) the vacant carbene C: p(π) orbital, which is important when carbene works alone but does not participate in the FLP activation; and (ii) the electrostatic interaction between the Lewis base center and the approaching H atom which plays an important role and can either favor or disfavor a reaction. These explanations are also applicable to methane activations. The study brings two messages to the experimentalists for constructing FLPs: (i) it is recommended to use P- and N-centered Lewis bases to construct FLPs for H(2) activation because using more reactive components does not benefit the activation; and (ii) the FLPs are less reactive in activating CH(4) than H(2). In addition, using more reactive carbenes as Lewis bases in FLPs does not necessarily benefit the methane activation.

Designing Metal-Free Catalysts by Mimicking Transition-Metal Pincer Templates

Whereas nature often promotes reactions by utilizing cheap and abundant light transition metals (TMs; e.g., Fe, Mn, Ni, Cu, and Zn) in enzyme active sites, the majority of man-made catalysts are based on precious heavy TMs (e.g., Ru, Rh, Ir, Pd, and Pt), despite high costs, limited availability, and contamination problems. Development of cheap, green, and effective catalysts (without precious TMs or even TMs at all) is at the forefront of chemical research. The recent discovery of metal-free reversible hydrogen activation by Stephan and co-workers facilitates direct catalytic hydrogenation. On the basis of the frustrated Lewis pair (FLP) principle, 3] we designed metal-free hydrogenation catalysts computationally. The metal-free catalyst 1 c] and the well-known metal–ligand bifunctional hydrogenation catalyst 2 have related electronic and geometric structural features (Scheme 1). Indeed, increasing experimental evidence demonstrates the resemblance between the reactivity of TM-free compounds and TM complexes. Herein, we computationally explore the design of metal-free catalysts by directly mimicking TM catalysts. The M05-2X DFT functional was specifically developed to target nonbonding interactions. We previously calibrated the good performance of the functional in describing weak bonding interactions and applied it to understand the catalytic role of N-heterocyclic carbenes in the metal-free transformation of carbon dioxide into methanol. In our design of FLP-based hydrogen-activation molecules and hydrogenation catalysts, we found that the energetic results given by M05-2X were in good agreement with those provided by CCSD(T) or MP2 calculations. M05-2X (as implemented in the Gaussian 03 program) was thus selected for use in all of the DFT calculations. All of the geometries were optimized and characterized as minima or transition states (TSs) at the M05-2X/6-31GACHTUNGTRENNUNG(d,p) level. The energies were then refined by using M05-2X/6-311++GACHTUNGTRENNUNG(2d,p)// M05-2X/6-31GACHTUNGTRENNUNG(d,p) single-point computations. The M05-2X/ 6-31G ACHTUNGTRENNUNG(d,p) wave functions of the TSs were confirmed to be stable, which indicates the reliability of our single-referencebased calculations. The harmonic frequencies at the same level were employed for zero-point energy corrections and thermal and entropy corrections at 298.15 K and 1 atm. Bulky solvation effects were simulated by using the IEFPCM model with benzene as a representative solvent. The corresponding free energies are discussed below, unless otherwise specified. For clarity, we use Lewis structure drawings to depict the electronic structures of molecules in the main text. The optimized geometries and their Cartesian coordinates are provided in the Supporting Information. We selected a new type of pincer catalyst (3 in Scheme 2) recently developed by Milstein and co-workers as a template to derive TM-free analogues. Such TM pincer catalysts have intriguing electronic structures (see below) and exhibit high s-bond activation reactivity (e.g., involving H H, b] C H, N H, and O H 14b] bonds). They promote the direct synthesis of amides and imines from alco[a] G. Lu, H. Li, L. Zhao, F. Huang, Prof. Dr. Z.-X. Wang College of Chemistry and Chemical Engineering Graduate University of Chinese Academy of Sciences Beijing,100049 (P.R. China) Fax: (+86) 10-88256674 E-mail : zxwang@gucas.ac.cn [b] Prof. P. v. R. Schleyer Center for Computational Chemistry, University of Georgia Athens, Georgia 30602 (USA) [] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201002631. Scheme 1. Comparison of the metal-free (1) and TM (2) hydrogenation catalysts.

Metal-free catalysts for hydrogenation of both small and large imines: a computational experiment.

This study extends our previous work of using π-FLP strategy to develop metal-free hydrogenation catalysts. Using small MeN=CMe(2) imine (im1) as a model, we previously designed cat1 and cat2 catalysts. But it is unclear whether they are capable of catalyzing the hydrogenations of bulky imines. Using tBuN=C(H)Ph (im2) as a representative of large imines, we assessed the energetics of the cat1- and cat2-catalyzed im2 hydrogenations. The predicted energetics indicates that they can still catalyze large imine hydrogenations with experimentally accessible kinetic barriers, although the energetics becomes less favorable. To improve the catalysis, we proposed new catalysts (cat3 and cat4) by tailoring cat1 and cat2. The study indicates that cat3 and cat4 could have better performance for the hydrogenation of the bulky im2 than cat1 and cat2. Remarkably, cat3 and cat4 are also found suitable for small imine (im1) hydrogenation. Examining the hydrogen transfer substeps in the eight hydrogenations involved in this study, we observed that the mechanism for the hydrogen transfer step in the catalytic cycles depends on the steric effect between catalyst and substrate. The mechanism can be switched from stepwise one in the case of large steric effect (e.g.im2/cat2) to the concerted one in the case of small steric effect (e.g.im1/cat3). The new catalysts could be better targets for experimental realization because of their simpler constructions.

Computational Design of Metal-Free Molecules for Activation of Small Molecules, Hydrogenation, and Hydroamination

Hydrogen activation is a key step in hydrogenation reactions which are widely used in both laboratory synthesis and the chemical industry. Traditionally, it was often considered that only transition metal complexes/systems are able to activate hydrogen and to catalyze hydrogenations. This view has been changed recently; more and more metal-free molecules/systems have been found capable of activating hydrogen. Among these developments, the frustrated Lewis pairs (FLPs) are of particular significance, not only because they exhibit high reactivity toward hydrogen as well as other small molecules, but also because some of them can perform direct catalytic hydrogenations, which pave the way to the development of cheaper and greener hydrogenation catalysts. Inspired by the FLP principle, we used quantum mechanics computations to design molecules for H2, CH4, and NH3 activation and catalysts for hydrogenation of imines, ketones, and alkenes. While our designed molecules are awaiting experimental preparation, the active sites in our designed molecules anticipated the features appeared in the compounds synthesized later by experimentalists. This chapter reviews our computational explorations to enrich FLP chemistry.