Role of Atomicity and Interface on InOx-TiO2 Composites: Thermo-Photo Valorization of CO2

The synthesis, physicochemical, and functional properties of composite solids resulting from the surface spread of oxidized indium species onto nanoplatelets of anatase were investigated. Both the size and the interaction between the indium- and titanium-containing components control the functional properties. In the reduction of CO2 to CO, the best samples have an indium content between ca. 2 and 5 mol % and showed an excess rate over the photo and thermo-alone processes above 33% and an energy efficiency of 1.3%. Subnanometric (monomeric and dimeric) indium species present relatively weak thermal catalytic response but strong thermo-photo promotion of the activity. A gradual change in functional properties was observed with the growth of the indium content of the solids, leading to a progressive increase of thermal activity but lower thermo-photo promotion. The study provides a well-defined structure–activity relationship rationalizing the dual thermo-photo properties of the catalysts and establishes a guide for the development of highly active and stable composite solids for the elimination and valorization of CO2.


Catalytic set-up and details
. Upper, Left and Center: Photocatalytic annular reactor scheme; side and front views: (1) Gas inlet, (2) gas outlet, (3) UV lamp, (4) catalyst (brown) sample, (5) cartridge heater.  Radiation flow on the surface of the sample (red),  n radiation flow from the lamps (blue).Upper, Right: Center of coordinates located at the sample (defined by coordinates x s , y s , z s ).Down, Coordinate system to define the integration limits of the radiation Model.(Left) φ  and φ  .
The thermo-photo activity of the samples for methanol reforming was tested using a gas-phase continuous flow annular thermo-photo-reactor (pyrex) schematically depicted in Figure S1.The catalyst (ca.0.2 mg cm −2 ) was deposited onto the inner tube (ca.15 cm; 0.8 cm diameter) as a thin layer from a suspension in ethanol.As previously detailed, a continuous rotation of the inner In this work we measure catalytic output with the help of three observables, the reaction rate, the quantum efficiency and the global energy balance of the reaction.The reaction rate (r) measures the number of hydrogen production molecules per surface area and time unit, but to analyze the thermo-photo production of hydrogen we define an "excess rate" (r e ) measured through Equation S1.
Such "excess" rate measures the potential synergy occurring between both energy sources in the thermo-photo catalytic process.Synergy is thus measured as the excess (i.e.positive value) over the additive effect of light and heat in the reaction rate.
The second is the Quantum Efficiency (QE) parameter for hydrogen production.QE is defined by Equation S2 [2].
In this equation, r is the reaction rate and  , the average local superficial rate of photon absorption.Here, for the calculation of the quantum efficiency, we will use two different reaction rates, the normal one and the excess one.The use of the latter would allow to measure an "excess" quantum efficiency.
The rate of hydrogen production is measured using mass spectrometry and gas chromatography as previously outlined and normalized using the BET surface area of the sample.The local superficial rate of photon absorption ( , ) is defined by Equation S3.It follows from the equation corresponding to a pure photo-catalytic process but with an additional term that accounts for the losses coming from charge emission with temperature [3].In this equation,   is the fraction of light absorbed by the sample,   the radiation flux at each position (x ≡   ,   ,   ) of the catalytic film, and T e is the thermal emission loss terminus.
To obtain the radiation flow on the surface of the samples, we calculate first the impinging radiation flux from the lamps ( n ).Considering the coordinate system presented in Figure S1 and the geometry of the reactor (annular multi-lamp), the   can be determined by Equation S4 [1].Where , ,  are the coordinates of the points located on the surface of the catalytic film, and   ,   ,   which are the coordinates of the points located on the surface of the lamp.R type variables correspond to the radius of the cylinder supporting the sample (R) or of the lamp (R L ), see Figure S1.Angular variables (Ѳ,φ) are defined as described in Figure S1.Integration limits of Equation S4 are summarized in Equations S5-S12 and can be graphically visualized in Figure S1. Where: Finally, the   x/y components (see Figure S1; Equation S13) can be determined using   and a radiation balance, which considers the main optical (Transmittance, F i , and Reflectance, R i ) events occurring in all components of the reactor placed between the emission source and catalyst, i.e. glass and reaction media, as well on the catalytic film.
,  =  ( q n , F i ,R i );i = catalyst, glass, reaction media (S13) A detailed description of the mathematical formulation to provide   as a function of   (Equation S13) and the transmittance/reflectance optical measurements for each component of our reactor system can be found elsewhere [1,4].As examples, the F i parameters are displayed in Figure S2 for selected samples as a function of the illumination wavelength.The T e is a loss term and can be calculated using Equation S14.This Equation considers the emission of a body in a medium can be calculated using Plank´s law [3].The radiation intensity per surface area unit is [3,5]: Where γ is the photon frequency, h is the Plank´s constant, c is the speed of light, n is the refraction index of the solid, k is the Boltzmann´s constant, T is the temperature of the sample and  is the absorption efficiency that acts as an emissivity type factor as discussed in refs.4,5.
This T e term is negligible at the temperatures of this work as it only makes a maximum correction 4 parts per million to the local superficial rate of photon absorption values.This is at least 3-4 orders of magnitude below the standard error of the  , coefficient.Such a result is somehow expected as emission losses in titania-based (the dominant component) materials are known to occur at higher temperatures than here used [6].Finally, an energy balance of the thermo-photo process is carried out to compare with the simple sum of the thermal and photo processes.Taking the equilibrium nature of the reaction [7], the denominator of equation 2 of the main text is obtained as: The Δ f H • for CO, H2O, H2 and CO2 are -110.6,-242, 0 and -393.8 kJ mol − 1 , respectively.In this work, CO is obtained with nearly 100 % selectivity.Thus, the speed of formation of all molecules involved in the reaction is considered identical.

Catalytic and Characterization Results
Table S1  The statistical significance of the fitting outcome using a standard F-test of the variance is given by: F = ((WRSS2-WRSS1)/(P2-P1)) / ((WRSS2/(n-P2)) (S16) where WRSSi is the weighted sum of squared residuals of model i, n are the number of data points, and pi are the number of parameters corresponding to the three and four shells fittings; with p2 > p1.The F statistic will have an F distribution, with ((p2−p1)/2, (n−p2)/2) degrees of freedom.Under the null hypothesis, model number 2 (the one with 4 shells) does not provide a significantly better fit than model 1 (the one with 3 shells).
For our 10 InTi sample (data presented in Figure S6), the F-test takes the value of 4.057.This vale indicates that the 4-shell is statistically significant at a probability of 97.3 %.See main text for further details.
photo-catalytic tests, the film was heated using a cartridge heater.The temperature of the layer

Figure S2 .
Figure S2.Fraction of light absorbed, transmitted or reflected by selected samples.

Figure S5 .Figure
Figure S5.Primary particle size distributions for the In component for selected InTi samples.Pre-and post-reaction samples are presented.
. Comparison of activity parameters.The table includes literature reports considering the thermo-photo reverse water gas shift (RWGS) reaction.Results for samples with an asterisk consider activity parameters for the formation of CO but other products are detected.Isotopic switching experiment using 12 C/ 13 C marked CO 2 .Thermo-photo conditions at 250 ℃ using the 10InTi sample.An arbitrary initial point for the time axis is presented.