Enhanced Solar Photothermal Catalysis over Solution Plasma Activated TiO2

Abstract Colored wide‐bandgap semiconductor oxides with abundant mid‐gap states have long been regarded as promising visible light responsive photocatalysts. However, their catalytic activities are hampered by charge recombination at deep level defects, which constitutes the critical challenge to practical applications of these oxide photocatalysts. To address the challenge, a strategy is proposed here that includes creating shallow‐level defects above the deep‐level defects and thermal activating the migration of trapped electrons out of the deep‐level defects via these shallow defects. A simple and scalable solution plasma processing (SPP) technique is developed to process the presynthesized yellow TiO2 with numerous oxygen vacancies (Ov), which incorporates hydrogen dopants into the TiO2 lattice and creates shallow‐level defects above deep level of Ov, meanwhile retaining the original visible absorption of the colored TiO2. At elevated temperature, the SPP‐treated TiO2 exhibits a 300 times higher conversion rate for CO2 reduction under solar light irradiation and a 7.5 times higher removal rate of acetaldehyde under UV light irradiation, suggesting the effectiveness of the proposed strategy to enhance the photoactivity of colored wide‐bandgap oxides for energy and environmental applications.

1      Table S1. Comparison of photocatalytic activity of CO 2 reduction over TiO 2 -based  UV-vis diffuse reflectance (DR) spectra were collected on a PerkinElmer UV Win Lab spectrophotometer, and BaSO 4 was used as a reference. The crystal structure and phase identification were characterized by X-ray diffraction (XRD, Rigaku, D/max-2500 X-ray diffractometer) using Cu Kα radiation (λ = 1.5406Å). The PL and Raman spectra were The procedure for measurement of ERDT according to the method developed by Ohtani et al [1][2][3] was briefly described as following: A stainless-steel sample holder was filled with

Photocatalytic and photothermalcatalytic reduction of CO 2
Catalytic reduction of CO 2 with H 2 O was conducted in a stainless autoclave reactor (100 mL) with a quartz window on the top. Solar light and visible light were provided by a Hayashi LA 410 Xenon lamp (150 W). The solar light intensity was 100 mW cm -2 . Visible light was acquired with a 420 nm long-pass filter with light intensity of 100 mW cm -2 . In the photocatalytic test, 20 mg of catalyst was ultrasonically dispersed in 1 mL of deionized water and placed in the reactor. Then, the autoclave was sealed, and the internal air was degassed quickly and completely using high-purity CO 2 (99.999%) for 20 mins at room temperature and atmosphere pressure. In the photothermalcatalytic test, the reactor was heated to 120 o C, while other conditions kept the same as photocatalytic test. The gaseous mixture was qualitatively and quantitatively analyzed using a Shimadzu 2014C GC instrument. The isotopic experiment was carried out to check the product of CO 2 reduction using 13 CO 2 (Aldrich Comp. 99%) as reaction gas. The experiment was performed on gas 6 chromatography-mass spectrometry (Agilent 6890 GC/59973MSD (EI), HP-MOLESIEVE equipped with Micropacked column).

Catalytic degradation of gaseous acetaldehyde
The In the photothermalcatalytic test, the reactor was heated to 343 K, while other conditions were kept the same as that in photocatalytic test.

The equation for AQE calculation
Single electron energy = ℎ /λ N e × number of produced molecule= xn×NA A: lighting area h=6.626×10 -34 J·s c=3×10 8 m/s λ: Incident wavelength (365 nm, 400 nm, 450 nm and 500 nm) x: the number of electron transfer n: amount of substance

Calculation method
All the DFT calculations were performed by Vienna Ab initio Simulation Package (VASP) [4] with the projector augmented wave (PAW) method [5] and the generalized gradient approximation (GGA) of Perdew-Burke-Ernzerhof (PBE) [6] functional. The energy cutoff for the plane wave basis set was set as 450 eV. The force on each atom less than 0.03 eV/ Å was adopted as convergence criterion during the geometry optimization. The self-consistent calculations apply a convergence energy threshold of 10 -4 eV. The supercell was constructed by a three-layer 3×1 TiO 2 (101) slab with 108 atoms and a 15 Å vacuum, in order to avoid the interaction between periodic structures. The 2×2×1 Monkhorst and Pack [7] k-point sampling was used.
The free energies of the CO 2 reduction steps (CRR) were calculated by the equation: [8] , where ΔE DFT is the DFT electronic energy difference of each step, ΔE ZPE and ΔS are the correction of zero-point energy and the variation of entropy, respectively, which are obtained by vibration analysis, T is the temperature (T = 300 K).
According to previous study, [8,9] corrections of -0.51, -0.08 and +0.13 eV have been used to eliminate PBE error for gas-phase CO, H 2 and CO 2 , respectively.The climbing-image nudged elastic band (CI-NEB) method implemented in VASP was used to search the saddle points and minimum energy pathways.        On the basis of finding the most stable O vacancy configuration and referring to previous work [10] , we carried out the calculation of bulk hydrogen doping, and the optimized structure is shown in Figure S13.           For photocatalysis and photothermal catalysis, TiO 2 (AB) with increasing time of SPP treatment shows a volcano removal of acetaldehyde and evolution of CO 2 . As shown in Figure S31, T-2 h shows the best photoactivity, while the photoactivity of T-4 h decreases.   The photocatalytic activity of untreated and treated semiconductor oxides was assessed by monitoring the degradation of acetaldehyde gas under UV light irradiation, and the course of degradation of acetaldehyde and the evolution of CO 2 were displayed in Figure S39. The activity follows the order to that of photocatalysis: treated semiconductor oxides > untreated semiconductor oxides. The activity of photothermalcatalysis show the same trend with that of photocatalysis. Therefore, SPP is also generally applicable to other semiconductor oxides. Ag /TiO2 500 W Xe lamp (λ＞420 nm) CO2 saturated H2O CH3OH [33] Additionally, the products of C1 are dependent on the experimental conditions. In general, the reported researches on CO 2 reduction with H 2 O can be classified into two types (summarized in