Noble Metal Modified TiO2 Photocatalysts for Solar Hydrogen Production: Roles of Metal Co-catalysts and Reaction Media

Abstract

M/TiO2 (M = Pd, Pt or Au) photocatalysts demonstrate excellent activity for hydrogen (H2) production in alcohol-water mixtures under UV excitation. Their activity depends on two main factors: i) the metal co-catalyst (M) and ii) the reaction medium (alcohol sacrificial reagent). This thesis systematically evaluates the effect of metal co-catalyst type, co-catalyst metal loading, co-catalyst particle size and dispersion and also the effect of the sacrificial agent (alcohol type and concentration) on the activity of M/TiO2 photocatalysts for H2 production in alcohol-water mixtures under UV irradiation. A series of M/TiO2 photocatalysts with metal loading 0-4 wt.% were prepared by the deposition-precipitation with urea method, followed by calcination at 350 oC and/or H2 reduction at 500 oC using the commercial semiconductor photocatalyst, Degussa P25 TiO2 (85 wt.% anatase + 15 wt.% rutile) as the support. TEM analyses showed that the Pd and Pt co-catalysts were highly dispersed over the TiO2 support evidencing a strong metal-support interaction (MSI), with an average particle size of ~2 and 1.5 nm, respectively. For Au/TiO2, the MSI was weaker and the average metal nanoparticle size was larger (~5 nm) and the nanoparticles more spherical in shape. XPS, EXAFS and UVVis analyses confirmed the presence of predominantly zero valent metal nanoparticles on the surface of the photocatalysts. Photoluminescence data established that the metal co-catalysts effectively suppressed electron-hole pair (e--h+) recombination in TiO2 following UV photoexcitation. The noble metal co-catalysts also function as active centres for H2 evolution during alcohol photoreforming. In order to understand the effect of metal co-catalysts on hydrogen production, a series of metals were deposited on P25 TiO2. The amount of Pd, Pt and Au was varied, analyzed and tested in ethanol-water mixtures (80:20 volume ratio) under UV-fluxes comparable to that of sunlight (365 nm, 5-6 mW cm-2). The selection of P25 TiO2 as the support allowed meaningful comparison with the work of other groups. It was observed that optimal H2 production rates were achieved at metal loadings of 0.25-1 wt.% for Pd/TiO2 (40-45 mmol g-1 h-1); 0.5 wt.% for Pt/TiO2 catalysts (43 mmol g-1 h-1); and 1-2 wt.% for Au/TiO2 catalysts (32 mmol g-1 h-1). Higher metal loadings did not enhance the H2 rate confirming the existence of an optimal loading. At all metal loadings, the H2 production rates (in mmol g-1 h-1) for the M/TiO2 (M = Pd, Pt, Au) photocatalysts under UV excitation decreased in the order Pd/TiO2 > Pt/TiO2 > Au/TiO2 >> TiO2. Intrinsic properties of each metal such as dispersion, work function and the d-band centre position explains this trend. Considering the role of metal particle size, hydrogen production rates were found to be one order of magnitude higher per Au particle when compared to Pt and Pd. The activity of the 0.5 wt.% Pd/TiO2, 0.5 wt.% Pt/TiO2 and 1 wt.% Au/TiO2 photocatalysts were subsequently evaluated over a wide range of ethanol concentration from 0 to 100 vol.%. An ethanol to water volume ratio of 90:10 afforded the highest H2 evolution rates in the case of Pd and Pt photocatalysts (51 and 47 mmol g-1 h-1, respectively), whilst the activity of 1 wt.% Au/TiO2 was highest at an 80:20 volume ratio. Reaction products of ethanol photorefroming were analyzed for the M/TiO2 photocatalysts at a fixed metal loading of 1.5 × 10-4 molM (0.5 wt.% Pd/TiO2, 1 wt.% Pt/TiO2 and 1 wt.% Au/TiO2 photocatalysts) in a 10:90 volume ratio of ethanol:water in order to gain more information about the reaction mechanism. The evolved gases followed the trend H2 >> O2 > CO2 > hydrocarbons. The gas phase H2:CO2 ratio was 15-23 for all photocatalysts confirming that the evolved H2 was produced by water splitting and ethanol photo-oxidation. The effect of the reaction medium on the performance of M/TiO2 (M = Pd, Pt, Au) photocatalysts for H2 production was also explored. The photocatalytic activities of 0.5 and 1 wt.% Pd/TiO2, 1 wt.% Pt/TiO2 and 1 wt.% Au/TiO2 were evaluated in a wide range of alcoholwater mixtures (fixed alcohol concentration of 10 vol.%) under UV irradiation (365 nm, 5 mW cm-2). H2 production rates in the alcohol-water mixtures were dependent on (i) the metal cocatalyst; and (ii) the alcohol type. The co-catalyst activity followed the order Pd > Pt  Au when normalized by catalyst mass (in mmol g-1 h-1). The highest H2 production rates were achieved for the 1 wt.% Pd/TiO2 photocatalyst in glycerol-water mixtures (48 mmol g-1 h-1) and ethylene glycol-water mixtures (45 mmol g-1 h-1). Regarding the effect of the reaction medium, H2 production rates decreased in the order glycerol > ethylene glycol > 1,2-propanediol > methanol > ethanol > 2-propanol > tert-butanol >> water. For each M/TiO2 photocatalyst, correlations were established between the rate of H2 production and specific alcohol properties, especially alcohol polarity and the exponential of the energy change for the alcohol oxidation reaction, exp( ( 2) ) o ox o  EVB TiO E . The latter was rationalised in terms of simple electron transfer reactions between a donor (alcohol) and an acceptor (valence band holes of TiO2). The photoreaction products in the gas phase from methanol, ethylene glycol and glycerol photoreforming over M/TiO2 (M = Pd, Pt, Au) photocatalysts. The gas phase H2:CO2 ratio ranged from 6 to 10 depending on the alcohol used. Relative H2 and CO2 evolution rates indicated that hydrogen is produced from both water splitting and the alcohol. Finally, the effect of alcohol concentration on the rate of H2 production over 0.5 wt.% Pd/TiO2 was investigated in detail. Results revealed that low alcohol concentrations (e.g. ~5 vol.%) were optimal for H2 production rates in ethylene glycol-water and glycerol-water mixtures, whilst a concentration of around 40 vol.% was optimal for methanol-water mixtures. Results of this thesis provide valuable new insights into the factors influencing H2 yields via alcohol photoreforming over M/TiO2 (M = Pt, Pt, Au) photocatalysts under UV excitation.