Exchange of one PCy3 unit of the classical Grubbs catalyst 1 by N-heterocyclic carbene (NHC) ligands leads to "second-generation" metathesis catalysts of superior reactivity and increased stability. Several complexes of this type have been prepared and fully characterized, six of them by X-ray crystallography. These include the unique chelate complexes 13 and 14 in which the NHC- and the Ru-CR entities are tethered to form a metallacycle. A particularly favorable design feature is that the reactivity of such catalysts can be easily adjusted by changing the electronic and steric properties of the NHC ligands. The catalytic activity also strongly depends on the solvent used; NMR investigations provide a tentative explanation of this effect. Applications of the "second-generation" catalysts to ring closing alkene metathesis and intramolecular enyne cycloisomerization reactions provide insights into their catalytic performance. From these comparative studies it is deduced that no single catalyst is optimal for different types of applications. The search for the most reactive catalyst for a specific transformation is facilitated by IR thermography allowing a rapid and semi-quantitative ranking among a given set of catalysts.

Instrumentation and Spectra Formats. NMR: Spectra were recorded on a Bruker AC 200, AMX 300, DPX 300, AMX 400 or DMX 600 spectrometer in CDCl 3 unless stated otherwise. Chemical shifts (8) are given in ppm relative to TMS, coupling constants (J) in Hz. The multiplicity in the 1 3 C NMR spectra refers to the geminal protons (DEPT). IR: Nicolet FT-7199, wavenumbers in cm-1. MS: Finnigan MAT 8200 (70 eV); HR-MS: Finnigan MAT SSQ 7000 (70 eV). Elemental analyses: Dornis & Kolbe, MUlheim.

Whereas recent synthetic studies concerning Rh-catalyzed olefin hydrogenation based on BINOL-derived monodentate phosphites have resulted in an efficient and economically attractive preparative method, very little is known concerning the source of the unexpectedly high levels of enantioselectivity (ee often 90-99%). The present mechanistic study, which includes the NMR characterization of the precatalysts, kinetic measurements with focus on nonlinear effects, and DFT calculations, constitutes a first step in understanding this hydrogenation system. The two most important features which have emerged from these efforts are the following: (1) two monodentate P-ligands are attached to rhodium, and (2) the lock-and-key mechanism holds, in which the thermodynamics of Rh/olefin complexation with formation of the major and minor diastereomeric intermediates dictates the stereochemical outcome. The major diastereomer leads to the favored enantiomeric product, which is opposite to the state of affairs in classical Rh-catalyzed olefin hydrogenation based on chiral chelating diphosphines (anti lock-and-key mechanism as proposed by Halpern).

An optimized and large scale adaptable synthesis of the ruthenium phenylindenylidene complex 3 is described which employs commercially available diphenyl propargyl alcohol 5 as a stable and convenient carbene source. Previous ambiguities as to the actual structure of the complex have been ruled out by a full analysis of its NMR spectra. A series of applications to ring closing metathesis (RCM) reactions shows that complex 3 is as good as or even superior to the classical Grubbs carbene 1 in terms of yield, reaction rate, and tolerance towards polar functional groups. Complex 3 turns out to be the catalyst of choice for the synthesis of the enantiopure core segment 77 of the marine alkaloid nakadomarin A 60 comprising the ADE rings of this target. Together with a series of other examples, this particular application illustrates that catalyst 3 is particularly well suited for the cyclization of medium-sized rings by RCM. Other key steps en route to nakadomarin A are a highly selective intramolecular Michael addition setting the quaternary center at the juncture of the A and D rings and a Takai-Nozaki olefination of aldehyde 73 with CH2I2, Ti(OiPr)4 and activated zinc dust.