Al(ORF)3 (RF = C(CF3)3) activated silica: a well-defined weakly coordinating surface anion

Weakly Coordinating Anions (WCAs) containing electron deficient delocalized anionic fragments that are reasonably inert allow for the isolation of strong electrophiles. Perfluorinated borates, perfluorinated aluminum alkoxides, and halogenated carborane anions are a few families of WCAs that are commonly used in synthesis. Application of similar design strategies to oxide surfaces is challenging. This paper describes the reaction of Al(ORF)3*PhF (RF = C(CF3)3) with silica partially dehydroxylated at 700 °C (SiO2-700) to form the bridging silanol 
<svg xmlns="http://www.w3.org/2000/svg" version="1.0" width="23.636364pt" height="16.000000pt" viewBox="0 0 23.636364 16.000000" preserveAspectRatio="xMidYMid meet"><metadata>
Created by potrace 1.16, written by Peter Selinger 2001-2019
</metadata><g transform="translate(1.000000,15.000000) scale(0.015909,-0.015909)" fill="currentColor" stroke="none"><path d="M80 600 l0 -40 600 0 600 0 0 40 0 40 -600 0 -600 0 0 -40z M80 440 l0 -40 600 0 600 0 0 40 0 40 -600 0 -600 0 0 -40z M80 280 l0 -40 600 0 600 0 0 40 0 40 -600 0 -600 0 0 -40z"/></g></svg>
 Si–OH⋯Al(ORF)3 (1). DFT calculations using small clusters to model 1 show that the gas phase acidity (GPA) of the bridging silanol is 43.2 kcal mol−1 lower than the GPA of H2SO4, but higher than the strongest carborane acids, suggesting that deprotonated 1 would be a WCA. Reactions of 1 with NOct3 show that 1 forms weaker ion-pairs than classical WCAs, but stronger ion-pairs than carborane or borate anions. Though 1 forms stronger ion-pairs than these state-of-the-art WCAs, 1 reacts with alkylsilanes to form silylium type surface species. To the best of our knowledge, this is the first example of a silylium supported on derivatized silica.

Solid-state NMR experiments at 9.4 T at Iowa State University were performed on a Bruker Avance III HD spectrometer equipped with wide-bore magnet. Experiments were performed at an MAS frequency ( rot ) of 25 kHz using a 2.5 mm triple-resonance probe. 1D 1 H NMR spectra were acquired using the DEPTH pulse sequence 4 comprising of a 90° excitation pulse and followed by two successive 180° pulses for background suppression at 100 kHz radiofrequency (RF) field. The 1 H{ 27 Al} RESPDOR 5-6 experiment was performed with the dipolar recoupling 4 2 1 sequence 7 on the 1 H channel applied with a radiofrequency (RF) field of twice the MAS frequency (2 ×  rot ). The saturation pulse was applied on the 27 Al channel at 80 kHz RF field with a duration of 60 s (1.5 ×  rot ,  rot = 1/ rot ). The experiment was performed in an interleaved manner where a control dataset is obtained without the pulse on the 27 Al channel for every recoupling duration. Numerical simulations of 1 H-27 Al RESPDOR were performed with SIMPSON v4.2.1 [8][9][10] . The 1 H{ 27 Al} RESPDOR curve shown in the main text compares S/S 0 with numerical simulations performed with a saturation factor (f) =0.55 and different 1 H-27 Al dipolar coupling constants/internuclear distances. The curve corresponding to a 1 H-27 Al distance of 2.46 Å shows the best agreement with experiment, consistent with the DFT calculated structure of 1. Numerical simulations were performed in SIMPSON with the start operator set to I 1x and the detect operator set to I 1p . Powder averaging was performed using the 'rep320' crystallite orientation file comprising of 320 (, ) pairs. 16  angles were used. An ideal 1 H 180° pulse was used, whereas the 27 Al saturation pulse used an 80 kHz RF field and a duration of 60 s (1.5 ×  rot ) to mimic experimental conditions. The 27 Al C Q and were set to 15.2 MHz and 0.0, respectively. The relative orientations (Euler angles) of the 27 Al quadrupole and CSA tensors and the 1 H-27 Al dipole vector were set according to the DFT optimized structure of 1.
The proton detected 27 Al→ 1 H D-RINEPT experiment [11][12] was performed with a 0.1 s recycle delay, 4096 scans, 100 kHz indirect spectral width and 92 t 1 increments. The STATES-TPPI procedure was used to achieve sign discrimination and obtain absorptive peaks in the indirect dimension. Rotor synchronized dipolar recoupling was applied on the 1 H channel with RF 4 2 1 set to 2 ×  rot . 4 s central transition (CT) selective 90° pulses were applied on the 27 Al channel. RAPT pulses 13 were applied on the 27 Al channel prior to the D-RINEPT transfer step using 38 s frequency switched WURST (wideband, uniform rate, smooth truncation) pulses separated by a 2 s delays at 31 kHz RF field.
All solid-state NMR were processed using Topspin v3. 6.1. 27 Al analytical simulations were performed using ssNake v1.1. 14 Synthesis and characterization of 1 -3 Synthesis of 1: SiO 2-700 (2 g, 0.52 mmol OH) and PhF-Al[OC(CF 3 ) 3 ) 3 ] (480 mg, 0.58 mmol) were transferred to one arm of a double-Schlenk flask inside an argon-filled glovebox. Perfluorohexane (ca. 10 ml) was transferred under vacuum to the flask at 77 K. The mixture was warmed to room temperature and gently stirred for two hours. The clear solution was filtered to the other side of the double Schlenk. The remaining solid was washed by condensing solvent from the other arm of the double Schlenk at 77 K, warming to room temperature, stirring for 2 minutes, and filtering the solvent back to the other side of the flask. This was repeated two times. The solid was dried under diffusion pump vacuum for 1 hour. The white material was stored in a glovebox freezer at -20 °C.     The 2D 27 Al→ 1 H D-RINEPT spectrum shows that the acidic proton at 5.0 ppm correlates with a broad 27 Al NMR signal at 50 ppm (C Q = 15.7 MHz), which is assigned to 1. The observed C Q of this site is consistent with the 14.1 T measurements shown in Table S1. The INEPT spectrum also shows an intense correlation between a 1 H NMR signal at 3.0 ppm and a sharper 27 Al NMR signal at 73 ppm (C Q = 10.0 MHz). This signal is assigned to a higher symmetry 27 Al species that forms during the course of sample rotation, most likely because of partial hydrolysis of 1 in the imperfectly sealed 2.5 mm rotors. Consistent with this interpretation, the 1 H DEPTH spectrum of 1 obtained immediately at the start of MAS experiments ("fresh") and after 19 hours of continuous MAS ("19 hours later") shows a clear increase in total 1 H integrated signal intensity, suggesting ingress of water into the rotor ( Figure S6C). The 27 Al RAPT spin echo spectrum of the "fresh" sample was obtained immediately after starting MAS, however, acquisition of the spectrum required ca. 3 hours, during which partial hydrolysis likely occurred ( Figure S6B). A second 27 Al spin echo spectrum was then obtained ("3 hours later") and the intensity of the broad 27 Al signal was observed to decrease slightly, while the narrower 27 Al signal increased slightly. All of these observations are again consistent with partial hydrolysis of 1 in the rotor.

S9
ASO + n-octyl 3 N (2): ASO (200 mg, 0.05 mmol ≡Si-OH---Al(OR F ) 3 ) was loaded into a teflon -valved flask. Pentane (2 ml) was vacuum transferred to the solid at -196 °C using a high vacuum line. In a N 2 filled glovebox, trioctylamine (12 µL, 0.03 mmol) was added to the slurry. The reaction was gently stirred for 30 minutes then the solution was removed by cannula under argon flow. The solid was washed 2 X more by vacuum transferring in more pentane (2 mL     , S12, and S17; and b.) taken from the simulation of the MAS spectrum in Figure S4.
Stability of 1 in common solvents: 1 (50 mg) was loaded into a teflon -valved NMR tube then solvent (0.5 mL) was vacuum transferred over the solid.  17 The geometries of all structures were optimized in Gaussian 09 with the B3LYP functional and 6-31G(d,p) basis set. 18 Frequency calculations at this level of theory produced no imaginary frequencies, indicating a ground-state energy minima equilibrium structure. The NMR parameters of 1 were calculated in using the GIAO method at the M06L/Al

Gas-Phase Acidity:
Gas-Phase Acidity (GPA) calculations were performed with the calculated geometries in Gaussian 09 at the BP86/def2-TZVP level of theory. The geometries of HBr and H 2 SO 4 were optimized at the B3LYP / 6-31G** level of theory. Bromine was described with the SDD basis set. The gas-phase acidity (GPA) is defined as:

ΔG (H + + A -) -ΔG (HA) = GPA in kJ/mol
The GPA produced at the B3LYP / 6-31G** level of theory did not reproduce experimental trends. With these geometries, other levels of theory were explored, particularly the BP86 functional, which has been known to produce accurate GPA values. 19 The experimental value as well as the theory that best reproduces the trend is used in the text and bolded in Table S2.

NMR Calculations:
The NMR parameters of 1-DFT and 3-DFT were calculated using either B3LYP or M06-L functionals. For optimizations at M06-L / 6-31G**, the scaling factor of 0.952 for M06-L / 6-31+G** from Truhlar and coworkers is used. 20 Geometry optimizations of 1-DFT and 3-DFT using the 6-31G** basis set showed that both functionals give optimized structures with very similar ν OH and Al-OH distances, indicating that these geometries of these species are similar at these levels of theory. However, the 27 Al C Q was underestimated using this basis set. We screened basis sets and functional combinations and found that M06L with a 6-311G** basis set on Al, 6-31G** on other atoms, gave results that most closely agreed with experiment. These data are summarized in Table S3. All combinations of basis set and functional give 1 H NMR chemical shifts for the bridging silanol close to experiment. Table S3. DFT Calculated Parameters for 1-DFT.
Similarly, the δ( 29 Si) was not well reproduced with B3LYP. Combinations of basis sets and functionals to calculate the 29 Si NMR chemical shift of 3-DFT are given in Table S4. Similar to the data shown in Table 3, we found that M06L with a 6-311G** basis set on Al, 6-31G** on other atoms, gave results that most closely agreed with experiment.