Differentiating surface titanium chemical states of anatase TiO2 functionalized with various groups† †Electronic supplementary information (ESI) available. See DOI: 10.1039/c7sc04828a

The local electronic effects on surface Ti, caused by adsorbates on TiO2 facets, are probed experimentally (using probe-assisted NMR spectroscopy) and theoretically (using DFT).

XPS measurement. XPS measurements were recorded on a Thermo Scientific K-Alfa XPS instrument equipped with micro-focused monochromated Al X-ray source. The source was operated at 12 keV and a 400 micron spot size was used. The analyzer operated at the analyzer energy (CAE) of 200 eV for survey scans and 50 eV for detailed scans. Charge neutralization was applied using a combined low energy/ ion flood source. The data acquisition and analysis were conducted with CasaXPS (Casa software Ltd.). The peak position was referenced to C1s peak of the carbon tape at 285.00 eV.
TMP-adsorbed sample preparation. About 150 mg of TiO 2 sample was placed in a homemade glass tube and activated at 150 o C for 2 h under vacuum (10 -1 Pa). After cooling down to room temperature, 300 μmol/catalyst g (calculated by the pressure and volume of isolated system) of TMP was then introduced. Wait ~10 min for the reach of equilibrium between TMP and catalyst surface. Extra TMP molecules were removed by vacuum system. These steps were repeated three times to ensure the fully adsorption of TMP on catalyst surface. The sample tube was then flame sealed for storage and transferred to Bruker 4 mm ZrO 2 rotor with a Kel-F endcap in a glove box under nitrogen atmosphere before NMR measurement. See our previous report for more details. [3.4] NMR measurement. The solid state magic angle spinning (MAS) NMR experiments were carried out using a Bruker Avance III 400WB spectrometer at room temperature. Magic angle spinning speed of 12 kHz and high power decoupling (HPDEC) sequence were adopted here.
Considering the long relaxation time of 31 P nuclei in NMR experiment, 30° pulse with the width of 1.20 μs, 15 s delay time was used. The radiofrequency for decoupling was 59 kHz.
The spectral width was 400 ppm, from 200 to −200 ppm. The number of scanning was 800.
The 31 P chemical shifts were reported relative to 85% aqueous solution of H 3 PO 4 , with NH 4 H 2 PO 4 as a secondary standard (0.81 ppm). See our previous report for more details. [3.4] NMR spectrum deconvolution. All raw TMP NMR spectra were deconvoluted using the software 'peakfit v4.12'. We employed "gauss area" and ensured all results with R 2 value > 0.98. Notice that the raw spectra data of samples with the same treatment/modification show a standard deviation of ±1 ppm in chemical shift position. For example, during the spectral deconvolution of 2HF and 6HF samples, we have fixed two positions of Ti 5c (101) and Ti 5c (001) at -31 ppm and -22.5 ppm within ±1 ppm uncertainty. Also, for S-Na-PD, S-Na- DFT calculations. Projector-augmented waves (PAW) [5,6] generalized gradient approximation (GGA) [7] was employed in DFT calculations. In the plane wave calculations, cutoff energy of 500 eV was applied and automatically set by the total energy convergence calculation for anatase TiO 2 (001) and TiO 2 (101) slab system. The primitive unit cell of TiO 2 was constructed to consist of tetragonal anatase TiO 2 structure containing eight O atoms with four Ti atoms; the system was then allowed to reach its lowest energy configuration by a relaxation procedure. The k-point grid determined by the Monkhorst-Pack method was 7 × 7 × 3 for bulk calculations in this study. The calculated lattice parameters of TiO 2 were 3.776 × 3.776 × 9.486 Å, which was in good agreement with the experimental value (3.785 × 3.785 × 9.514 Å). [8] For the modeling of both (001) and (101) surfaces, we adopted a slab containing twelve layers of Ti-O units. The surface was constructed as a slab within the three dimensional periodic boundary conditions. This model was separated from their images in the direction perpendicular to the surface by a 14 Å vacuum space. The bottom three layers were kept fixed to the bulk coordinates; full atomic relaxations were allowed for the top nine layers.
During calculations, a 3 × 3 × 1 k-Point mesh was used in the 4 × 4 super cell. A suitable dimension of supercell (11.328 × 11.328 × 26.255 Å 3 ) was found to perform the adsorption of trimethylphosphine (TMP) on TiO 2 (001). Supercell with dimension 10.885 × 11.328 × 23.353 Å 3 was used for TiO 2 (101). The atoms in the cell were allowed to relax until the forces on unconstrained atoms were less than 0.01 eV/Å. The adsorption energy, E ad , is defined as the sum of interactions between the capping molecule and slab system, and it is given as E ad = E total -E TiO2(001) -E TMP , where E total , E TiO2(001) and E TMP are the energy of total system, TiO 2 (001) slab and TMP molecule, respectively. The negative sign of E ads corresponds to the energy gain of the system due to molecular adsorption. The calculation of TMP-TiO 2 (101) was carried out similarly. To calculate the effects of various adsorbates to surface Ti chemical states among facets, we placed F-, OH-, SO 4 molecules on TiO 2 (001) and TiO 2 (101) and then calculated the corresponding E ad of the TMP on the given surfaces. For the F case, two H (for charge balance) and two F atoms were introduced on both the TiO 2 (001) and TiO 2 (101) slabs.
1~3 F atoms were also placed on TiO 2 (001) slab to evaluate the coverage effect on the E ad of TMP. For the OH case, two OH groups were generated by the surface hydrolysis of Ti-O-Ti bonds on both slabs. For the SO 4 case, two H atoms (for charge balance) and a SO 4 molecule binds to two Ti 5C sites via its two oxygen atoms were introduced on both the slabs. To simulate solid-state NMR environment, the plane wave DFT code, VASP, was also adopted to calculate adsorption energies for TiO 2 systems using the linear response method. [9,10] The five

NLSPLINE=.TRUE.
All simulation graphics in this work were generated using GaussView version 3.0.

Degree of surface F/SO 4 modification
The calculation of F-modified TiO 2 (6HF) case are elaborated as an example below: According to Scheme 1b, the concentration of F attached Ti 5C can be obtained 1 : 1 from the protonation of TMP by neighboring BrÖnsted acid proton. Therefore, the total concentration of F-Ti 5C on (001) and (101) facets of 6HF sample is 89.4 μmol/g (Table S3).
The concentration of rest surface Ti 5C atoms (without F attachment) can be obtained from the NMR signal in Lewis acid range (the formation of TMP-Ti 5C ,  Table S4. Noted that S-Na-0HF wasn't consider here due to the generation of Ti 5C (101)-OH during the extensive hydrolysis. Figure S1. TEM (first column) and HRTEM (second column) images of as-prepared (a) 0HF, (b) 2HF, and (c) 6HF and their corresponding AB (face length) and AE (thickness) value (see Figure S3 for the surface area calculation of (001) and (101) facets) (50 particles are used in each histogram). [4] Figure S2. XRD spectra of as-prepared anatase 0HF, 2HF and 6HF. Figure S3. Simulated shape of the TiO 2 anatase single crystal and the equation for the surface area calculation of (001) and (101) facets (AB and CD are considered of the same value as face length herein; AE is equal to the thickness, θ of 68.3° is the angle between (001) and (101)). [2]       (c) on bridging hydroxyl proton (Brønsted acid, BA) site, the formation of TMPH + complex).
The δ 31 P of adsorbed TMP spans over a wide range (-20~-58 ppm) when interacting with various metal cations on different solid acids ( Figure S9a), whereas a TMPH + ionic complex formed when a TMP molecule adsorbs onto a bridging hydroxyl proton tends to give rise to a 31 P resonance in a much narrower range of -2 to -5 ppm ( Figure S9c). [11] Therefore, Brønsted (proton donor) and Lewis acid (electron acceptor) sites presented in a solid acid catalyst can be readily distinguished using 31 P NMR of adsorbed TMP. On the other hand, TMP on an isolated hydroxyl proton surface usually gives a signal at higher field (~-61 ppm, Figure   S9b). [11] Figure S10. Comparison of TMP-adsorbed 0HF, 2HF and 6HF samples with various surface coverage of (a) F and (b) SO 4 . Their coverages are summarized inTable S4.    Table S1. Samples preparation conditions and the calculated percentage of exposed (101) and (001) facet. Table S2. Particle size (calculated from the full width at half-maximum of the (101) peak in Figure S5 using Scherrer equation) and BET surface area data of 0HF, 2HF, 6HF and their corresponding calcination(Cal-)/NaOH wash(Na-) treatments. Table S3. Summary of the qualitative (chemical shift) and quantitative (peak area) of each deconvoluted peak in the region of Brønsted acid site (−2 to −5 ppm) and Lewis acid site (−20 to −58 ppm). [4] The concentration of adsorbed TMP on each site was calculated according to corresponding peak area (*Adsorbed TMP molecules in μmol/g) Table S4. Summary of the total concentration of surface Ti 5C and the surface F/SO 4 coverage of 2HF and 6HF samples. 1 Concentration of F/SO 4 obtained by NMR BA signal (μmol/g). 2 Total concentration of surface Ti 5C obtained by NMR BA and LA signals (μmol/g). 3 Percentage of F/SO 4 coverage on surface Ti 5C . Table S5. Experimental δ 31 P and calculated adsorption energy of TMP (both PAW-GGA and linear reponse) on Ti 5C (001) promoted with -F, -SO 4 , -O-and -OH groups. Table S6. Experimental δ 31 P and calculated adsorption energy of TMP (both PAW-GGA and linear reponse) on Ti 5C (101) promoted with -F, -SO 4 , -O-and -OH groups.