Thesis (M.S.)--Florida State University, 2005. / Advisor: Dr. Marie E. Krafft, Florida State University, College of Arts and Sciences, Dept. of Chemistry and Biochemistry. Title and description from dissertation home page (viewed Sept. 19, 2005). Document formatted into pages; contains 147 pages. Includes bibliographical references.

Ruthenium carbene complexes, with the general structure, [LL’Ru=CHR], are commonly known as Grubbs type catalysts, named after the discoverer of these metathesis catalysts. The discovery was quite revolutionary, since the catalysts proved to be easy to handle, tolerant towards various functional groups and more stable with regard to air and water than previous transition metal catalysts. Another important advantage was that all types of olefin metathesis reactions could be initiated without the help of co-catalysts or promoters. Today Grubbs type catalysts find wide application in especially organic and synthetic chemistry. A well-known example is the SHOP-process which produces long chain -olefins, while other important applications include the synthesis of macro-cyclic and cyclic olefins. The current study involved experimental and theoretical work to investigate various aspects comprising synthetic procedures, reactivity, kinetics, geometry and electronic properties of the complexes. Results are discussed briefly in the following paragraphs. The first aim of the project was to synthesise a Grubbs type catalyst. Initial efforts were focused on the preparation of a first generation catalyst through various methods. This included modifying the reported method for the synthesis of [(PPh3)2Cl2Ru=CH-CH=CMe2] to yield [(PPh2Cy)2Cl2Ru=CHCH= CMe2] instead; a phosphine exchange reaction with the complex [(PPh3)2Cl2Ru=CH-CH=CMe2] and free phosphine PPh2Cy; and utilising the analogue arsine ligand, AsPh3, to synthesise [(AsPh3)2Cl2Ru=CHCH=CMe2]; but unfortunately no success was achieved. However, it was possible to synthesise a novel second generation Grubbs type catalyst, [(IMesH2)(PPh2Cy)Cl2Ru=CHPh], through the phosphine exchange reaction of [(IMesH2)(NC5H5)2Cl2Ru=CHPh] and PPh2Cy. The new complex was tested in kinetic reaction studies and phosphine exchange reactions. Results showed that [(IMesH2)(PPh2Cy)Cl2Ru=CHPh] was catalytically active for the ring closing metathesis of commercial diethyl diallylmalonate. The reaction was first order with regard to the olefin, contrary to the second order kinetic results reported for similar reactions catalysed by first generation Grubbs catalysts. The phosphine exchange reactions were very successful and a rate constant could be determined. The rate constant was independent of the free phosphine concentration and activation parameters had relatively large, positive values; results indicative of a dissociative mechanism. These findings are in correlation with literature reports. A kinetic investigation was done on the catalyst-olefin coordination involving the functionalized olefins vinyl acetate, allyl acetate and allyl cyanide; and the first generation Grubbs catalyst, [(PCy3)2Cl2Ru=CHPh]. A two-step rate law, similar to an interchange mechanism, was determined. Phobcat, [(PhobCy)2Cl2Ru=CHPh], is modified first generation Grubbs type catalyst with rigid bicyclic phosphine rings which was recently developed by the Sasol Homogeneous Metathesis Group. In the current study Phobcat was compared to Grubbs1-PCy3 in the cross metathesis reaction of 1-octene. Results showed that Phobcat was up to 60% more active and had a 5 hour longer lifetime than Grubbs 1-PCy3. Theoretical studies were done on the three functionalized olefins of the earlier experimental study to gain fundamental understanding of steric and electronic influences on these catalyst-olefin systems. Without exception, coordination via the heteroatom of the olefin was significantly more favourable than coordination via the double bond functionality. This result indicates that metathesis of these olefins is highly unlikely, since the stable heteroatom coordination will suppress the parallel Ru=C/C=C interaction which is compulsory for the metathesis reaction. Orbital studies highlighted the difference between coordination of acetate and cyanide, but no trend of an electronic nature could be recognised. / Prof. A. Roodt

This study has focused on the design and preparation of bi– and tridentate N–Heterocyclic Carbene (NHC) ligands in order to investigate the effect of a multidentate approach to the formation, stability and catalytic activity of coordination complexes. Chapters 1 – 3 provide background information of relevant catalysis, carbene and coordination chemistry, followed by previous work performed within our research group. In Chapter 4 attention is given to the synthetic aspects of the research conducted, comprising two distinct approaches to the preparation of unsymmetrical saturated and unsaturated NHCs. Firstly, an investigation of the saturated NHC ligands yielded three novel, unsymmetrical pro–ligands, viz., two halopropyl imidazolinium salts and a bidentate hydroxypropyl imidazolinium salt. Secondly, eight imidazolium salts have been generated, including a hydroxypropyl analogue and novel decyl and tridentate malonyl derivatives. These compounds were prepared using microwave–assisted methodology for the alkylation of N– mesitylimidazole – an approach that drastically reduced reaction times (from 8 hours – 7 days to ca. 0.5 – 2 hours) and facilitated isolation of the imidazolium salts. Many of the compounds prepared in this study are novel and were fully characterized using HRMS and 1– and 2–D NMR analysis. Coordination studies using a selection of the prepared pro–ligands afforded an alkoxy–NHC silver derivative and four novel Ru–complexes, viz., Grubbs II–type Ru–complexes containing:– chloropropyl imidazolinylidene; propenyl imidazolylidene; and bidentate alkoxypropyl imidazolylidene ligands. Furthermore, a well–defined benzyl mesitylimidazolylidene Ru–complex has been isolated, which exhibited good stability in air. DFT–level geometry–optimization studies, using the Accelrys DMol3 package have given valuable insights into the likely geometries of the prepared and putative catalysts.

Olefin metathesis is a valuable synthetic tool, widely used in several fields of science. Due to the importance of this transformation several contributions have been made in this field in order to understand mechanistic aspects, reactivity and applicability of this process. In this topic, ruthenium indenylidene complexes have shown great activity and stability in metathesis, making them very valuable pre-catalysts. However, several aspects of these pre-catalysts have not been evaluated yet. For example, even though reports of active second generation ruthenium indenylidene complexes bearing bulky N-heterocyclic carbenes are present in the literature, no studies have been done to understand how steric hindrance affects the process. For these reasons, [RuCl₂(IPr*)(PPh₃)(3-phenylindenylidene)] (IPr*-PPh₃) and [RuCl₂(IPr*)(Py)(3-phenylindenylidene)] (IPr*-Py), bearing the very bulky ligand, IPr* have been synthesised and compared with [RuCl₂(IPr)(PPh₃)(3-phenylindenylidene)] (IPr-PPh₃) and the new [RuCl₂(IPr)(Py)(3-phenylindenylidene)] (IPr-Py). Another important aspect, presented in this thesis, is the investigation of the stability of indenylidene pre-catalysts in alcohol solvents. Surprisingly, several different decomposition processes occur depending on the starting complex and the alcohol used. Mechanistic investigation into this decomposition, allowed us to develop a better understanding of this process, and to predict the decomposition product based on the environment. In particular, this study revealed that [RuCl(η⁵-3-phenylindenyl)(PPh₃)₂] (Eta-5) is accessed from [RuCl₂(3-phenylindenylidene)(PPh₃)₂] (M₁₀) via a novel indenylidene to η⁵-indenyl rearrangement. This formal decomposition product has been found to be active in at least 20 different catalytic transformations, rendering it a versatile catalytic tool.

Ruthenium-based homogenous catalysis is a broad and extremely useful branch of transition metal catalysis. Surely, the most famous example is olefin metathesis, for which Yves Chauvin, Robert Grubbs and Richard Schrock were awarded the 2005 Chemistry Nobel Prize. Although some of the most well-known catalysts are widely used and considered benchmark catalysts, the research around this topic has not stopped. The modification of known systems to achieve better performance and better understanding of the catalytic mechanism is very important and an example of such modification is reported in this thesis. The newly synthesised catalysts were compared to the parent commercially available catalyst showing better reactivity. Ruthenium catalysis, though, is not limited to olefin metathesis and C-H activation, for example, it has become a useful approach to the functionalisation of organic molecules. In this field, the deuteration of C-H bonds is an interesting transformation, which has many applications. The synthesis of new hydridosilylruthenium complexes and their application in the deuteration of a variety of substrates is reported in this manuscript. The unprecedented synthesis of tetradeuterated Ketoprofene is also reported. Recently, ruthenium-based catalysts have found application in the dehydrogenation of suitable compounds, such as formic acid, ammonia-borane and other hydrogen-rich substances. The driving force behind these discoveries is the use of H2 as an energy vector in place of fossil fuels. A hydrido-ruthenium catalyst was shown to catalyse the decomposition of formic acid in CO2 and H2 and to catalyse the reduction of olefinic substrates. The released CO2 from the reaction did not interfere with the fuel cell due to its inertness. This property makes its employment as C1 source very challenging, although its use would also be extremely attractive because of the abundance of this gas. In these regards, both frustrated Lewis pairs (FLPs) and gold catalysts have shown interesting reactivity in the activation of CO2. A new FLP and a silica supported gold catalyst were synthesised to test them in CO2 activation and the results are reported in this manuscript.