Exploiting the interactions between the ruthenium Hoveyda–Grubbs catalyst and Al-modified mesoporous silica: the case of SBA15 vs. KCC-1

2nd generation Hoveyda–Grubbs catalyst immobilized onto well-ordered 2D hexagonal (SBA15) and 3D fibrous (KCC-1) mesostructured silica displaying tetra-coordinated Al–H via Surface Organometallic Chemistry (SOMC).


General Procedure
All experiments were carried out under controlled atmosphere. Treatments of the surface species were carried out using high vacuum lines (10 -5 mbar) and glove box techniques. SBA15 was prepared according to literature. 1 Diisobutylaluminum hydride (DIBAL), diethyldiallyl-malonate (DEDAM), 2 nd generation Hoveyda-Grubbs catalyst was purchased from Sigma Aldrich. Dichloromethane (DCM), DEDAM were dried over activated molecular sieve and degassed through freeze pump thaw cycles.
TEKPol 2 was dried under high vacuum (10 -5 mbar) and the solvents were stirred over calcium hydride and then distilled in vacuum.

Support Preparation
The reaction of mesoporous silica (dehydroxylated at 700 ºC at 10 -5 mbar for 30 h) with DIBAL was carried out in a double Schlenk tube. 0.5 g of SBA-15 700 or KCC-1 700 , was reacted with 1 equivalent of DIBAL (1M in hexane) followed by addition of 3 ml of dry n-pentane. The evolved gases were collected in a 10 L flask, and the white powder was washed with dry pentane (*2) to eliminate unreacted DIBAL. The solid was dried under vacuum (10 -5 mbar) for 12 h. The resulting solid was introduced in a glass reactor (275 mL) and heated up to 400°C (8°C/h) for 1h under dynamic vacuum (10 -5 mbar). The final amount of the aluminum grafted was 4 wt. % with an amount of Carbon of 3.51 wt. % (C/Al of 2).

Catalyst Preparation
In a double Schlenk, 250 mg of A2, B2 or C2 (1 eq., n [Al] = 0.37 mmol) was reacted at room temperature with 40 mg of complex HG-II 3 (M w = 626. 62 g.mol -1 0.2 eq., n [HG-II] = 0.065 mmol) in dry CH 2 Cl 2 (8 mL). The reaction mixture was stirred for 3 hours. After contacting the supports A2, B2 and C2 with dissolved [Ru]complex HG-II, the green color of the solution immediately vanished and the mesoporous silica turned brown. In comparison, previous attempts to graft HG-II on SBA15 700 using DCM revealed that the main part of HG-II remained dissolved in DCM (green solution). Only a small part of the catalyst stacked on the mesoporous silica support, which remained light green after washing. Using non-polar solvents (as toluene) 4 lowers the affinity of the catalysts for the solvent and improves its physisorption to silica oxide support.
After 3 hours the supernatant became transparent indicating the end of the reaction. After 5 h, the mixture was filtrated and the powder was washed (three times). Then, the brown powder was dried overnight and given to further investigations.

Gas Phase Analysis
After completion of the gas phase was investigated by gas chromatography on an Agilent 6850 gas chromatograph with split/split-less injector and FID. 10 μL were injected by the hot needle technique (thermospray) at an injector temperature of 250˚C using the split mode (split ration 50:1, 100 mL/min split flow). A HP-Al/KCl 50 m x 0.32 mm; 8.00 μm capillary column coated with a stationary phase divinylbenzene/ethylene glycol dimethacrylate was used with nitrogen as carrier gas at 18.3 Psi pressure.
Each analysis was carried out with the same conditions: a flow rate of 2 mL/min, an isotherm at 80 ˚C for 12 min, then at 170˚C for 3 min. The detector sets with a data rate of 5 Hz and a minimum peak width of 0.04 min.

DFT calculations
To understand the structure of the catalyst after grafting HG-II on the support 1, DFT calculations were performed with the generalized gradient approximation (GGA) functional with the Gaussian09 software using the BP86 level of theory of Becke and Perdew. [5][6][7] The electronic configuration of the molecular systems is described with the standard split-valence basis set with a polarization function of Ahlrichs and co-workers for H, C, N and O (SVP keyword in Gaussian). 8 For Ru, the small-core, quasi-relativistic Stuttgart/Dresden effective core potential, with an associated contracted valence basis set (standard SDD keyword in Gaussian09) was used. 9-10 The reported energies were obtained via single-point calculations on the BP86 optimized geometries using the M06 functional and triple-ζ basis set for main-group atoms (TZVP keyword in Gaussian09). 11-12 The influence of the solvent (DCM) was included in these single-point energy calculations by using the polarization continuum solvation model PCM. 12

Pyridine adsorption
Pyridine Transmission IR spectra were obtained by use of a Nicolet FT-IR 6700 spectrometer equipped with an DTGS -KBr detector at a 4 cm -1 resolution with the help of specially designed Pyrex made IR cell capable for heating and evacuation; which is similar to IR cell mentioned by E. Parry. 13 32 scans were collected to obtain each spectrum. The spectra were obtained in frequency range 400-4000 cm -1 . For FT-IR measurement, the sample in the form of a thin self-supporting wafer with a thickness less than 1 mm was mounted on pyrex cell holder. This support sample can slide back and front into the cell at proper position in front of CaF 2 windows to record the spectra. Thin wafers of materials, A2, B2 and C2 supports (10 mg) were prepared by using a Carver hydraulic pellet, making unit located within glovebox under argon atmosphere with moisture level below 0.1 ppm. For wafer formation, a pressure of approximately 12,000 psi was applied. Initially, an FT-IR spectrum of clean dehydrated support was measured before pyridine adsorption treatment at room temperature for each sample.
Initially, Pyridine AR (≥99.5) (Sigma Aldrich) were purified, dried and degased using freeze-thaw pump degassing process. Then pyridine was passed over support wafer under static vacuum and allowed to adsorb. The physisorbed pyridine was removed by heating support wafer at 110°C about 1 hour and then spectra were recorded on FT-IR instrument to check pyridine adsorption. Later, spectra for given wafer recorded after heating wafer under vacuum (10 -6 mbar) bar for 1h at 200, 300 and 400 °C. The peaks areas at 1455 and 1550 cm -1 were used after linearization and deconvolutions. Immobilization of organometallic complex HG-II Figure S3. FT-IR spectra of material (A2), after grafting of HG-II (A3) and the subtraction (A3) -(A2) (red) in the 3900-1400 cm −1 region at 25 °C.

Elemental analyses
Elemental analyses (Ru, Cl, N, C) were performed at Mikroanalytisches Labor Pascher (Germany) and Varian 720-ES ICP-Optical Emission Spectrometer (Ru) after the samples preparation by microwave digestions on and on Milestone ETHOS 1. Analysis of C and N were done on Flash 2000 Elemental Analyzer from Thermo Scientific.  A 2 ms contact time was used for cross polarization experiments. The MAS frequency was 10 kHz. All 13 C NMR spectra were referenced to adamantane with the higher frequency peak set to 38.48 ppm with respect to TMS (0 ppm).

Nitrogen adsorption/desorption measurements
Typical isotherms of all materials are shown in Figure S6-8. It is a classical type IV isotherm, which bears a characteristic H1 hysteresis loop. A type IV isotherm indicates the monolayer adsorption followed by multilayer formation and capillary condensation. 16 Figure S7. Nitrogen adsorption/desorption isotherms at 77 K of A0 (blue), A2 (red) and A3 (green).

Transmission-electron microscopy (TEM).
The primary particle size and morphology of A3 and B3 were examined by transmission electron microscopy (TEM). This task was accomplished by using a double aberration corrected (double Cscorrected) microscope of model Titan 80-300 ThemisZ from Thermo-Fisher Scientific (Waltham, MA).
Samples were loaded onto holey-carbon coated copper grids for TEM analysis. The samples were placed onto grids in a solvent-free but ambient environment. Conventional bright-field TEM (BF-TEM) imagingtechnique was performed by operating the scope at the accelerating voltage of 300 kV. Furthermore, the microscope was set to a range of magnifications during the image acquisition process. No nanoparticles of Ru were obtained, proving that Ru was not decomposed during the grafting. It is pertinent to note that the X-ray energy dispersive (EDS) analysis of samples was also done for the determining the overall Structures of A3 and B3 were found to be well preserved after their loading with HG-II as presented in the main paper ( Figure 6). For A3, both channels and walls of the silica particles could be identified easily. The presence of Ru in the channels was proved by the Ru-maps and Al-maps. The measurements on the white dot matched with channel width and hence it corroborates our argument that the Ru was present inside the channels. For B3, the Ru maps and Al-maps proved the presence of Ru throughout the external surface.
EDS spectra for A3 and B3 can be found in Figure S10 and S11. Both spectra confirm the presence of Ru, Cl, Al, Si, O, C (for edges see Table S3), which are the main components of the catalyst A3 and B3.  The low ruthenium (Ru) signal on the elemental map indicates that only a small fraction of Ru is presented in both samples, A3 and B3. By element quantification, we found a Ru/Al ration of 0.05 (Table S4 and Table S5).

Catalytic Tests Propene Metathesis
Catalytic propylene metathesis was performed on a micropilot (PID Eng&Tech) equipped with a stainless steel reactor at atmospheric pressure. The catalyst powder was placed in the reactor supported by preliminary dried quartz glass wool. Conversions were calculated based on carbon mass balance.

RCM Metathesis
The activity of A3 and B3 were examined in ring closing metathesis of the benchmark substrate diethyldially malonate (DEDAM). The reaction was performed under inert Argon atmosphere in a Schlenk.