Reversible alkene binding and allylic C–H activation with an aluminium(i) complex

The monomeric molecular aluminium(i) complex 1 [{(ArNCMe)2CH}Al] (Ar = 2,6-di-iso-propylphenyl) reacts with a series of terminal and strained alkenes including ethylene, propylene, allylbenzene and norbornene to form alkene bound products.


General Experimental Section
All manipulations were carried out using standard Schlenk-line and glovebox techniques under an inert atmosphere of argon or dinitrogen. A MBraun Labmaster glovebox was employed, operating at < 0.1 ppm O2 and < 0.1 ppm H2O.
Solvents were dried over activated alumina from an SPS (solvent purification system) based upon the Grubbs design and degassed before use. Glassware was dried for 12 h at 120 °C prior to use. Benzene-d6 was stored over 3Å molecular sieves and distilled prior to use. NMR-scale reactions were conducted in J. Young's tap tubes and prepared in a glovebox.
All heating mentioned was done using silicone oil baths. 1 H (tetramethylsilane; 0 ppm) and 13 C (tetramethylsilane; 0 ppm) spectra were obtained on BRUKER 400 MHz or 500 MHz machines unless otherwise stated; all peak intensities are derived from internal standard peaks with values quoted in ppm. Data was processed using the MestReNova or Topsin software. C IV refers to quaternary carbons.
CHN analysis were not possible on the metallocyclopropane complexes 2b-g due to their potential to liberate the alkene under thermal conditions. Due to this thermal instability multinuclear NMR data were collected at either 273 or 298 K.

Allylic sp 3 C-H Activation
Complex 1 (15 mg, 0.03 mmol) was dissolved in benzene-d6 (0.6 mL) and the solution transferred into a Young's tap NMR tube. The sample was then degassed (freeze/pump/thaw) three times, and the ethylene was added to the tube at approximately 1 bar pressure. The formation of compound 2c was confirmed by 1 H NMR spectroscopy. The sample was then heated, in the presence of an excess of propylene, at 353 K overnight. The solvent was removed and X-ray quality crystals were isolated from concentrated n-hexane at -35 °C.
C-H Activation of Hex-1-ene: Complex 1 (5 mg, 0.01 mmol) and hex-1ene (14 µL, 0.1 mmol) were dissolved in benzene-d6 (0.6 mL) and the solution transferred into a Young's tap NMR tube. The formation of compound cis/trans-2d was confirmed by 1 H NMR spectroscopy. The sample was then heated, in the presence of an excess of the alkene, at 353 K overnight. Two new species were observed to form in a 1:0.6 ratio by 1 H NMR spectroscopy, however the diagnostic peaks of these products overlapped, preventing identification of the major isomer.

Variable Temperature NMR of metallocyclopropanes
Complex 1 (15 mg, 0.3 mmol) and an excess of the alkene (0.1 mmol) were dissolved in benzene-d6 (0.6 mL) and the solution transferred into a Young's tap NMR tube. The formation of the previously characterized (vide supra) metallocyclopropane was confirmed by 1 H NMR spectroscopy. The excess alkene was removed in vacuo after which time the sample was dissolved in toluene-d8 (0.6 mL) and left to equilibrate at 298 K for 24 hours. Variable temperature 1 H NMR analysis was performed on the sample over the temperature range 343-373 K. At temperatures lower than 353 K it was found the be necessary to hold the sample at temperature for a prolonged period to allow equilibrium to be reached (e.g. for a sample of compound 2a, this was found to take 15 minutes at 343 K).

Van't Hoff Analysis
The Van't Hoff equation was used in order to determine ΔH and ΔS from the slope and the intercept of the plot of ln(Keq) versus 1/T. ΔG was calculated according to Gibb's Free Energy equation.

S12
The variable temperature 1 H NMR spectrum of compound 2b showed the formation of 1 and free ethylene. However, a significant amount of degradation or further reaction also occurs (note: Compound 2b further reacts to form 3), therefore it was not possible to perform van't Hoff analysis on the sample. Figure S5: Variable temperature 1 H NMR spectra of 2b (343-373 K) in toluene-d8.

Kinetic Analysis
Complex 1 (5 mg, 0.01 mmol) was dissolved in toluene-d8 (0.6 mL) and five equivalents of allylbenzene (7.5 µL, 0.05 mmol) were added along with a small amount of ferrocene (2 mg), to be used as an internal standard. The solution transferred into a Young's tap NMR tube and the formation of compound 2f was confirmed by 1 H NMR spectroscopy.
Kinetic analysis was conducted using compound 1 at 18.7 mM concentration in toluene-d8 over a range of temperatures (343 -363 K). 1 H NMR spectra were recorded at regular intervals throughout the reaction, and the data was processed using Topspin and MNova software. The ferrocene was used as an internal standard to ensure that the overall concentration of substrates remained consistent throughout the reaction. The data was modelled using Copasi software consider two pre-equilibria scenarios: model A 2f = 1 + alkene -> 4f and model B 1 + alkene = 2f -> 4f. While these models can not be distinguished experimentally, model A is unambiguously favoured by the DFT calculations (vide infra). Time (s) S14    The C18-based 2,6-diisopropylphenyl group in the structure of 2d was found to be disordered. Two orientations were identified of ca. 80 and 20% occupancy, their geometries were optimised, the thermal parameters of adjacent atoms were restrained to be similar, and only the non-hydrogen atoms of the major occupancy orientation were refined anisotropically (those of the minor occupancy orientation were refined isotropically). The C31-based hexyl group was found to be disordered, and three orientations were identified of ca. 43, 34 and 23% occupancy. The geometries of the three orientations were optimised, the thermal parameters of adjacent atoms were restrained to be similar, and all of the non-hydrogen atoms were refined isotropically. The C40-based included toluene solvent molecule was found to be disordered across a centre of symmetry, and two unique orientations were identified of ca. 33 and 17% occupancy (with two further orientations of the same occupancies being generated by operation of the inversion centre). The geometries of both unique orientations were optimised, the thermal parameters of adjacent atoms were restrained to be similar, and all of the atoms of both orientations were refined isotropically.

S19
The X-ray crystal structure of 3 The structure of 3 was found to sit across a centre of symmetry at the middle of the C4Al2 ring. The Al-H hydrogen atom in the structure of 4c was located from a ΔF map and refined freely. The C31-based allyl group was found to be disordered, and three orientations were identified of ca. 42, 39 and 19% occupancy. The geometries of the three orientations were optimised, the thermal parameters of adjacent atoms were restrained to be similar, and all of the non-hydrogen atoms were refined isotropically. The allylbenzene group was found to be disordered with major and minor occupancies ca. 66% and 34%. Non-hydrogen atoms of the major occupancy orientation were refined anisotropically (those of the minor occupancy orientation were refined isotropically).

Computational Details
DFT calculations were run using Gaussian 09 (Revision D.01) 4 using the M06L Minnesota density functional. Al centers were described with Stuttgart SDDAll RECPs and associated basis sets and the 6-31G** basis sets were used for all other atoms. 5,6,7 Alternate functions were also investigated: ωB97X, M062X, B3PW91.
Geometry optimisation calculations were performed without symmetry constraints. Frequency analyses for all stationary points were performed using the enhanced criteria to confirm the nature of the structures as either minima (no imaginary frequency) or transition states (only one imaginary frequency).
Intrinsic reaction coordinate (IRC) calculations were used to connect transition states and minima located on the potential energy surface allowing a full energy profile (calculated at 298.15 K, 1 atm) of the reaction to be constructed. 8,9 Free energies reported within the main text are corrected for the effects of benzene solvent (ε=2.2706) using the using the polarizable continuum model (PCM). 10 In addition, single point dispersion corrections were applied to the ωB97X optimised geometries (dispersion corrected ωB97X-D functional) and to the B3PW91 functional (dispersion corrected GD3BJ functional). 11 QTAIM calculations were run with the AIMAll package.