Water molecule switching heterogeneous proton-coupled electron transfer pathway

Figuring out the specific pathway of semiconductor-mediated proton-coupled electron transfer (PCET) driven by light is essential to solar energy conversion systems. In this work, we reveal that the amount of adsorbed water molecules determines the photo-induced PCET pathway on the TiO2 surface through systematic kinetic solvent isotope effect (KSIE) experiments. At low water content (<1.7 wt%), the photo-induced single-proton/single-electron transfer on TiO2 nanoparticles follows a stepwise PT/ET pathway with the formation of high-energy H+/D+–O 
<svg xmlns="http://www.w3.org/2000/svg" version="1.0" width="13.200000pt" height="16.000000pt" viewBox="0 0 13.200000 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(0.017500,-0.017500)" fill="currentColor" stroke="none"><path d="M0 440 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z M0 280 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z"/></g></svg>
 C or H+/D+–O–C intermediates, resulting in an inverse KSIE (H/D) ∼0.5 with tBu3ArO· and KSIE (H/D) ∼1 with TEMPO in methanol-d0/d4 systems. However, at high water content (>2 wt%), the PCET reaction follows a concerted pathway with a lower energy barrier, leading to normal KSIEs (H/D) ≥ 2 with both reagents. In situ ATR-FTIR observation and DFT calculations suggest that water molecules' existence significantly lowers the proton/electron transfer energy barrier, which coincides with our experimental observations.


Materials
Commercial titanium dioxide nanoparticles (TiO 2 , 30 nm, anatase), deuterium oxide (D 2 O, 99 atom % D), and methanol-d 4 (CD 3 OD, 99.8% D) were purchased from Sigma-Aldrich Company Ltd. Methanol (CH 3 OH) was purchased from Sinopharm Chemical Reagent Co. Ltd. The toluene-d 8 used in 1 H NMR experiments was from Innochem. 2,4,6-tri-tert-butylphenol (97%) was purchased from Acros. The corresponding phenoxyl radical t Bu 3 ArO· was prepared according to the reference. [1] All reagents used in the synthesis were analytical grade without further purification. Deionized water, with a resistivity of 18 MΩcm, was used throughout the experiments.

Preparation of TiO 2 nanoparticles with different water content
In a typical preparation procedure, untreated TiO 2 nanoparticles were calcined in a tube furnace at 773K for 30 min. After the temperature was cooled to room temperature, dried TiO 2 samples were transferred to a humidity chamber with different air humidity degrees (0%~100%) and kept for 30 minutes.

KSIE(H/D) experiments
In a typical procedure, 20 mg anatase was dispersed in 8 mL methanol-d 0 . The suspension was purged by Argon for 10 min and transferred to the Ar glovebox. After that, 0.5 mL t Bu 3 ArO· in acetonitrile of 2.5 mmol/L was added to it. Finally, the resulting suspension was sealed into a unique tube for an online ESR experiment in 90s under UV irradiation (365 nm, 68 mW/cm 2 ). Deuterated methanol-d 4 was used to repeat the experiment above under otherwise identical conditions. The settings for the ESR spectrometer were as follows: center field, 3400 G; sweep width, 400/800 G; microwave frequency, 9.52 GHz; field modulation frequency,100 kHz.
The electron concentration of trapped electrons on reduced TiO 2 samples was measured by Fe (III)-1, 10phenanthroline titration spectrometric method. 1, 10-phenanthroline spectrometric measurement is a simple and widely used method for the measurement of Fe (II) ions. Here, we use the Fe (III) solution to titrate the trapped electrons on TiO 2 samples that quantitatively lead to the produce of Fe (II) ion, then used 1, 10phenanthroline to measure the concentration of Fe (II) ions. Thus, we can quantitatively obtain the concentration of trapped electrons. The concentration of Fe(NO 3 ) 3 solutions employed in this measurement is 10 -3 M. 0.2% 1, 10-phenanthroline water solution and pH = 4.6 HAc-NaAc buffer solution were previously prepared for use. The pH = 4.6 HAc-NaAc buffer solution was prepared by dissolving 135 g sodium acetate and 120 mL acetic acid into 500 mL water solution. Before the titration, the Fe(NO 3 ) 3 solution is purged by Argon for 30 min to remove oxygen. The titration was conducted in the Argon glove box.
In a typical procedure, 2.5 mL reduced TiO 2 solution was taken in the Argon glove box and mixed with 2.5 mL Fe(NO 3 ) 3 solution. 5 min later, the resulting mixed solution was taken out and then centrifuged to conduct the spectrometric measurement. The spectrometric measurement was conducted in air atmosphere. 1.5 mL centrifuged supernatant taken from the Argon glove box was mixed with 1.5 mL pH = 4.6 HAc-NaAc buffer solution, and then added with 1 mL 0.2% 1, 10-phenanthroline water solution to obtain a red solution. Waiting 5 minutes to obtain a stable state, the resulting solution was transferred in a quartz cuvette and measured on a Hitachi U3900 spectroscopy. The spectrum at 0 minutes was used as the background. Obtained absorbance value was compared with the standard fitting line to obtain a certain Fe (II) concentration.

DRIFTS Measurement
In-situ diffuse reflection infrared Fourier transform spectroscopy (DRIFTS) experiments were implemented on a Thermo Scientific Nicolet iS10 spectrometer equipped with a mercury cadmium telluride (MCT) detector.
In a typical procedure, the catalysts were housed in Harrick Praying Mantis high-temperature reaction tank with a ZnSe window. The reaction chamber was heated to 773K, and the stabilized system was used as the background spectrum. A series module was used to observe the changes in the cooling process on the TiO 2 surface.

TGA experiments
Thermogravimetric analysis (TGA) was performed on Per-Kin-Elmer Pyris 1 and TGA 7 thermogravimetric analyzers in N 2 atmospheres. Different TiO 2 solid samples were heated from 25 °C to 700 °C at a 20 °C/min rate.

NMR 1 H Measurement
After reducing t Bu 3 ArO·/TEMPO with TiO 2 methanol suspension under UV irradiation, the solvent was removed by purging with Ar. Then the residue was transferred to the Ar glovebox and dissolved in toluene-d 8 for 1 H-NMR measurement.

ATR-FTIR experiments
ATR-FTIR measurement was employed to detect the surface species and their changes with the in-situ reaction process on TiO 2 films. The ATR-FTIR spectroscopy experimental setup was similar to that described in the references. [2] The instrumental setup consisted of a Harrick Horizon multiple internal reflection accessories coupled to a 1 mL flow-through cell containing an AMTIR(As/Se/Ge) crystal on the bottom plate and a quartz window on the top plate. Eleven infrared bounces were allowed using a 45° internal reflection element (50×10×2 mm 3 ). The FT-IR measurements were performed on a Nicolet iS10 FTIR with an MCT detector. In a typical photocatalysis procedure, a layer of methanol-d 0 /d 4 was dripped onto the surface of the AMTIR crystal that was coated with a TiO 2 (Wetted by water or deuterium water vapor before calcination) film. Argon degassed the apparatus for 30 min, and the crystal was then scanned to obtain the background spectrum. Time-resolved in-situ FTIR data was then collected, turning up the 365 nm LED lamp.

Suspension degree measurement
The titanium dioxide of 20 mg is evenly scattered in methanol, and the ultrasonic is 10min evenly. The obtained solution is immediately placed in the rack. At this time, the state is set to the background, using a fiber optic photometer to measure its luminous flux, the obtained spectrum is the increase in luminous flux caused by in-situ deposition within a same measurable time.

Computational methods
The photocatalytic properties of TiO 2 were investigated by the Vienna Ab-initio Simulation Package (VASP) with the revised Perdew-Burke-Ernzerh functional of (RPBE) of the generalized gradient approximation (GGA). The Van Der Waals correction using DFT-D2 method of Grimme was considered in the calculations.
The Hubbard U term (DFT+U) was added on O 2p orbitals at the value of 6.3 eV and Ti 3d orbitals at the value of 4.2 eV. The interaction between ionic core and valence electrons was obtained from PAW pseudopotential. TiO 2 surface was described by its typical (101) Figure S1. 1 H-NMR spectra of single-proton/single-electron transfer products (a) t Bu 3 ArOH and (b) TEMPOH (After the reaction, the TiO 2 was filtered out, and the solvent was removed under anaerobic condition, then the residue was dissolved in toluene-d 8 ).             Figure S14. In-situ DRIFT spectra to observe the change of surface water adsorption on TiO 2 nanoparticles during (a) heating process from 298 K to 773 K and (b) colling process from 773 K to 298 K. The equilibrium state at 773 K was used as a blank background baseline. Figure S15. Thermogravimetric analysis (TGA) spectra to determine the water content in pristine TiO 2 and the 773 K calcined counterpart. Figure S16. Thermogravimetric analysis (TGA) spectra to determine the water content in TiO 2 after being treated in humid conditions (0%-100%) for 30 min. Figure S17. Control experiments of Fe(III)-1,10-Phenanthroline titration measurements. The UV/Vis absorbance spectra collected during titration of the trapped electron samples with different light-irradiation time were carried out on (a) dry TiO 2 (water content~1.3%) -CH 3 OH; (c) dry TiO 2 (water content~1.3%) -CD 3 OD; (b) wet TiO 2 (water content~3.8%) -CH 3 OH; (d) wet TiO 2 (water content~3.8%) -CD 3 OD, and the KIE value obtained by electron number fitting were (e) dry TiO 2 , (f) water treatment TiO 2 , background electrons before 0 min were deducted.   Scheme S1. Schematic diagram of CPET and PT/ET reaction pathways of the single-proton/single-electron transfer on TiO 2 with TEMPO as the acceptor.