Calcium Hydride Catalysts for Olefin Hydrofunctionalization: Ring‐Size Effect of Macrocyclic Ligands on Activity

Abstract The fifteen‐membered NNNNN macrocycle Me5PACP (Me5PACP=1,4,7,10,13‐pentamethyl‐1,4,7,10,13‐pentaazacyclopentadecane) stabilized the [CaH]+ fragment as a dimer with a distorted pentagonal bipyramidal coordination geometry at calcium. The hydride complex was prepared by protonolysis of calcium dibenzyl with the conjugate acid of Me5PACP followed by hydrogenolysis or treating with nOctSiH3 of the intermediate calcium benzyl cation. The calcium hydride catalyzed the hydrogenation and hydrosilylation of unactivated olefins faster than the analogous calcium complex stabilized by the twelve‐membered NNNN macrocycle Me4TACD (Me4TACD=1,4,7,10‐tetramethyl‐1,4,7,10‐tetraazacyclododecane). Kinetic investigations indicate that higher catalytic efficiency for the Me5PACP stabilized calcium hydride is due to easier dissociation of the dimer in solution when compared to the Me4TACD analogue.

The mixture was degassed applying three "freeze-pump-thaw" cycles and charged with ethylene (1 bar) at 0 °C. The in-situ generated ethyl complex was analyzed by 1 H NMR spectroscopy. For the 13 C NMR measurement, the sample was put into a precooled (0 °C) NMR spectrometer and measured at −20 °C.   1-octene (2 mg, 18 µmol) and 1,4-(SiMe 3 ) 2 C 6 H 4 (3 mg, 13 µmol) in 0.1 mL of THF-d 8 were added and the mixture analyzed by NMR spectroscopy. Figure S45: In-situ 1 H NMR of the reaction of 3b with 1-octene in THF-d8 (*) at 25 °C. # denotes 1,4-(SiMe3)2C6H4. Figure S46: Cutout of the in-situ 1 H NMR spectra of the reaction of 3b-d2 (left) and 3b (right) with 1-octene in THF-d8 at 25 °C. For hydrogenation experiments, 0.1 mL of the catalyst stock solution and the olefin stock solution were transferred into a J. Young-style NMR tube and 0.4 mL pure THF-d 8 was added.
The reaction mixture was degassed applying three "freeze-pump-thaw" cycles and charged with H 2 (1 bar). Progress of the reaction was monitored by 1 H NMR spectroscopy.
For hydrosilylation experiments, 0.1 mL of the catalyst stock solution, the olefin stock solution and the silane stock solution were transferred into a J. Young-style NMR tube and 0.3 mL pure THF-d 8 was added. The reaction progression was monitored by 1 H NMR spectroscopy.

Kinetic Investigations
Stock solutions for hydrosilylation experiments with catalyst 3b were prepared as described in For each catalytic experiment, the corresponding amount of the three stock solutions was added to a J.Young-style NMR tube. For experiments with a ten-fold excess of olefin or silane, these substrates were weighed out and added separately. In all experiments, THF-d 8 was added until the total reaction volume reached 0.6 mL and progress of the reaction was monitored by 1 H NMR spectroscopy.
[3b] (mol  Figure S50: Effect of catalyst concentration on hydrosilylation catalysis mediated by 3b. a) Reaction profile for different catalyst concentrations. b) to e) VTNA plots for different partial reaction orders of 3b. c) shows that a 0.5 th partial reaction order for 3b gives a fitting plot.

S45
3.3.4.2 Partial reaction order of the alkene Table S2: Variation of 1-octene concentration for determination of the partial reaction order of the alkene.
[3b] (mol   Figure S52: Effect of silane concentration on hydrosilylation catalysis mediated by 3b. a) Reaction profile for different silane concentrations. b) to e) VTNA plots for different partial reaction orders of n-octylmethylsilane. c) shows that a 0 th partial reaction order for the silane gives a fitting plot.   Figure S53: Effect of catalyst concentration on hydrosilylation catalysis mediated by 6. a) Reaction profile for different catalyst concentrations. b) to e) VTNA plots for different partial reaction orders of 6. c) shows that a 0.5 th partial reaction order for 6 gives a fitting plot. S48 3.3.5.2 Partial reaction order of the alkene Table S5: Variation of 1-octene concentration for determination of the partial reaction order of the alkene.
[6] (mol   Figure S55: Effect of silane concentration on hydrosilylation catalysis mediated by 6. a) Reaction profile for different silane concentrations. b) to e) VTNA plots for different partial reaction orders of n-octylmethylsilane. c) shows that a 0 th partial reaction order for the silane gives a fitting plot.

S50
3.3.6 Determination of k obs . Figure S56: Determination of kobs for 3b (top) and 6 (bottom). The slope of the linear regression line corresponds to kobs of the reaction. The kinetic data is taken from the reactions carried out with 5 mol% of catalyst and 0.1 M of substrate in 0.6 mL of THF-d8.

Crystal Structure Determinations
X-ray diffraction data were collected at 100 K on an Eulerian 4-circle diffractometer STOE STADIVARI in ω-scan mode with Mo-Kα (1b) or with Cu-Kα (3b, 6). The crystal structures were solved by direct methods using SHELXT [S7] and all refinements were carried out against F 2 with SHELXL [S8] as implemented in the program system Olex2. [S9] The molecular cation in 3b shows crystallographic 1 symmetry around Wyckoff position 1g. The crystal packing of 3b contains crystallographically independent, non-coordinated molecules of THF. The lattice of 6 also contains non coordinated THF molecules of THF. Because these could not be reliably refined, they were not included in the structure factor calculation with their atom positions and the SQUEEZE routine was applied using the program "solvent mask" as implemented in Olex2.
In 3b, disorder was found for the atoms C5 and C6 of the ligand Me 5 PACP, for a methyl group of the borate (C35), and for the atoms C56 as well as C61, C62, C63 within non-coordinated THF. This disorder could be resolved well with split positions. Due to disorder in a THF group, one distance restraint (concerning the atoms C61B and C62B) was applied using the command DFIX as implied in SHELXL. In 6, distance restraints were included in the refinement of the disordered carbon atoms of a Me 4 TACD ligand (C1-C8) as well as for the bond C37 between C38 (the latter being refined with split positions).
Non-hydrogen atoms were refined with anisotropic displacement parameters. Restraints were employed using the command RIGU as implied in the program SHELXL for the refinement of the anisotropic displacement parameters of atoms C28 and C29 in 3b. Hydrogen atoms were included in idealized positions. Only the hydrogen atom of the NH unit in 1b (H1), the hydride atom in 3b (H1), as well the hydride atoms in 6 (H1 and H2) were located in a Fourier difference map and were refined with isotropic displacement parameters. Refinement results are given in Table S7. Graphical representations were performed with the program DIAMOND. [S10] Crystals of 4 were obtained at -40 °C from THF/n-pentane under an atmosphere of ethylene within eight weeks. X-ray diffraction data were collected with a crystal of dimensions 0.

Computational Details
Calculations were carried out at the DFT level using the hybrid functional B3PW91 [S11] with the Gaussian 09 [S12] suite of programs. Polarized all-electron triple-ζ 6-311G(d,p) [S13] basis sets were used for Ca, C, H, O and N. Geometry optimization was carried out without any symmetry restriction. The nature of the extrema (minimum or transition state) was verified with analytical frequency calculations. The NBO analysis [S14] was finally carried out on the optimized geometry.