Copper modified-TiO₂ catalysts for hydrogen generation through photoreforming of organics. A short review

Highlights

• First literature survey on Cu–TiO₂ catalysts for H₂ generation via photoreforming.

• Comparison of preparation methods, organics and pH used, and efficiencies recorded.

• Estimation of different light absorption responses depending on the catalysts used.

• Classification of Cu active species identified and the reaction mechanisms reported.

Abstract

An intense scientific activity was recorded during the last several years in the field of preparation, characterization and use of copper-based TiO₂ photocatalysts for hydrogen generation through photocatalytic reforming of organics. Different copper species were used dissolved in aqueous solution or incorporated on the TiO₂ surface as single co-catalyst or in the presence of a second catalyst (e.g., graphene, carbon fibers) to (1) effectively separate the electron–hole pairs, thus reducing the occurrence of the recombination reaction, and (2) extend the light absorption to the visible range of the solar spectrum. Many organic species (e.g., methanol, glycerol, formic acid) were proposed as sacrificial agents for hydrogen generation, although the prevailing idea is that of using organic compounds currently found in industrial wastewaters. The pH value was recognized as a fundamental variable in photocatalytic H₂ generation via copper modified-TiO₂ catalysts. A positive effect to promote hydrogen generation was associated to an increase in pH until moderate alkaline values. On the other hand, a release in the solution of cupric ions and a consequent decrease in photocatalytic activity were observed when decreasing pH. A relevant lack of information was recorded about the efficiencies of hydrogen generation which were reported only in few papers. Therefore, this critical literature review has been performed with the aim of providing a complete background to select the most efficient approaches and eventually promote new competitive systems for hydrogen generation via photoreforming for industrial applications.

Introduction

Due to its high energy content and the absence of toxic or greenhouse-responsible emissions during its combustion, hydrogen is considered as an important energy carrier for the future [1]. Unfortunately, since it is present on earth in combination with other elements, its availability for industrial applications can only be the result of production processes starting from substances which contain it. As illustrated in Fig. 1, the latter substances may be represented, today, mainly by fossil fuels (coal, oil, natural gas) and biomasses [2] and, in the future, by organic substances eventually contained in civil or industrial wastewaters, or by the water itself [3]. Although hydrogen generation from fossil fuels is already a mature technology, the use of these raw materials is unappealing for the future due to their unrenewable nature and the emission of greenhouse gases (carbon dioxide) to ensure the energy required for hydrogen production processes. The last problem may also limit the interest in biomass, unless the energy required for its pyrolysis process could be captured from the sun [4], [5].

The production of hydrogen from organic substances contained in wastewaters or from water could be achieved, at least in principle, by means of photocatalytic processes capable of exploiting the solar radiation arriving daily on earth. The photocatalytic hydrogen generation can be obtained substantially following two different approaches: 1) photocatalytic water splitting [6], [7], [3] and 2) photocatalytic reforming of organics [8], [9]. The first method relies on the capability of water to be reduced and oxidized by reacting with photogenerated electrons and positive holes (generated by illuminating the semiconductor) in presence of selected co-catalysts respectively [6], [7], [3].

The role of co-catalysts is that of favoring the oxidation/reduction steps mainly by acting as reaction sites for oxidants and reductants, promoting the charge separation and the transport of charge carriers. In this way, they contribute significantly to enhance the overall reaction (water decomposition), which gives rise to the formation of hydrogen and oxygen gases (Fig. 2).

The second approach is based on the ability of some organic species, named sacrificial agents, to donate electrons to the positive holes of the illuminated photocatalyst and to be oxidized generating protons ions, while the latter are reduced by photogenerated electrons forming hydrogen in presence of proper co-catalysts [8], [9] (Fig. 3).

Both processes can be carried out through different experimental setups. The first one is based on a system in which the photocatalyst is merely suspended in the solution, the second method makes use of a photoelectrochemical cell [10], with the catalyst immobilized on a photoanode. Referring to the latter method under illumination, water (in photosplitting) or the organic compound (in photoreforming) are oxidized at the photoanode, whereas protons or water are reduced at a cathode in a second compartment with the two electrodes being electrically connected (Fig. 4).

However, with photoreforming the use of sacrificial organics is required and this represents an additional cost in the process. The possibility of using wastewater streams containing selected organics can be considered, which yields to a combined process of wastewater treatment with simultaneous hydrogen generation [11], [12]. It is interesting to observe that a combination of the two approaches (photosplitting and photoreforming) is made possible by means of some inorganic electron donors (such as I⁻ or Fe²⁺), whose oxidized forms may be reduced on a second photocatalyst, allowing the simultaneous oxidation of water and oxygen formation [13]. For example, in the case of iodide, two photocatalysts may be properly chosen and suspended in the same solution: (i) a first one allowing the reduction of protons (or water) along with I⁻ species oxidation to iodate, and (ii) a second one on which iodate may be reduced again to iodide while water is oxidized to O₂. With this approach, the overall reaction is still that of water photosplitting without any sacrificial organic consumption [14].

In general, an efficient photocatalyst needs have the following characteristics [15]:

• to be capable to absorb in the UV–Vis region of the solar spectrum and to actually use this energy to generate electron–hole pairs. In this region, about 50% of solar energy is concentrated and, in particular, about 45% is in the visible range [15]. The availability of photocatalysts capable of absorbing in the visible range of the electromagnetic spectrum may, in principle, guarantee the capture and storage in the photoproducts of a significant part of the energy emitted by the sun, provided that the energy absorbed is actually used to generate charge transporters (photogenerated electrons and holes);

• to be capable to immediately separate these pairs transferring electrons and holes at the liquid-semiconductor junction, where they participate to half-reactions; it is well known that photogenerated electrons and positive holes can recombine:

  $${\text{e}}^{-}+{\text{h}}^{+}\to \text{heat}$$

• thus reducing the efficiency of the process of interest, including hydrogen generation. The incorporation in the photocatalyst of species capable of promoting this separation is necessary to achieve a significant efficiency. It has been reported that for any photocatalyst, to be considered commercially viable, it has to display an efficiency of overall energy capture of about 15% in the visible region of the electromagnetic spectrum [16];

• to be characterized by an electronic structure which makes the half-reactions of interest thermodynamically feasible. As indicated in Fig. 2, Fig. 3, the charge transporters formed upon the absorption of the radiation are allocated on electronic bands which are characterized by different potentials. In Fig. 5, the importance of the position of energy bands in the semiconductor photocatalyst (e.g., TiO₂) is highlighted. As a matter of fact, the capability of photogenerated electrons to reduce protons (or water) is strictly related to the position of the conduction band (CB) potential in the photocatalyst hosting them, which must be lower than that of the H⁺/H₂ couple, as shown in Fig. 5. At the same time, the potential of the valence band (VB), in which the holes are present, has to be higher than that of the H₂O/O₂ couple for water photosplitting, or suitable for the organic species (e.g., methanol and formaldehyde) oxidation in photoreforming (Fig. 5);

• to be characterized by surface active sites that make possible the occurrence of these reactions.

Metal oxides of transition elements whose cations show a d⁰ or d¹⁰ configuration (Ti⁴⁺, Zr⁴⁺, Nb⁵⁺, Ta⁵⁺, W⁶⁺, Ce⁴⁺, Ga³⁺, In³⁺, Ge⁴⁺, Sn⁴⁺, Sb⁵⁺) have been so far successfully adopted for water splitting tests [3].

Among these species, TiO₂, in different crystalline forms, pure or properly modified, is one of the most investigated photocatalysts due to its great availability and low cost [15]. However, the properties of TiO₂, both in anatase or rutile forms, for water photosplitting or photoreforming, are not particularly inspiring since this solid is characterized by an electronic band structure which allows to absorb radiation only in UV range and with a significant occurrence of the recombination reaction, which greatly contributes to lower the efficiency of the process. Many of these aspects are treated in details in several excellent reviews which have been published so far mainly for water photosplitting [17], [18], [13], [3].

In the preparation of TiO₂-based catalysts, for both methods of hydrogen generation, the general approach followed by researchers is the use of some co-catalysts, mainly represented by noble metals, combined with the solid semiconductor [19], [8]. These species, once deposited on the semiconductor surface, are believed to act as electron traps which significantly reduce the parasitic recombination reaction between photogenerated electrons and holes, responsible for the low efficiencies recorded on pure TiO₂ (Fig. 6), and to kinetically favor the reduction of water (in the case of splitting) or of proton ions (photoreforming).

Another noticeable strategy to prevent the e⁻-h⁺ recombination reaction is that of coupling photocatalyst particles with photosensitizers [20]. As shown in Fig. 7, in these systems photogenerated electrons from the excited state of the sensitizers (S) can be quickly transferred to the CB of TiO₂, while the positive holes remain in the sensitizer, leading to an effective charge separation [20].

Moreover, a growing interest is also nowadays recorded among researchers to extend the TiO₂ absorption capability to capture the more abundant fraction of the solar radiation arriving on earth represented by that contained in the visible light range [3]. To this purpose, several approaches have so far been adopted to prepare some photocatalysts for hydrogen generation (e.g., for water photosplitting). Among these methods the use of metal or non-metal doping to obtain narrower band gaps, or that of dye sensitizers, have often been reported [21], [22], [23], [24], [25].

In drawing up this review, a literature survey disclosed the presence of more than 45 papers in which hydrogen generation was attempted through photocatalytic reforming of organic species, by means of TiO₂-based catalysts modified with copper or used in the presence of copper salts (Fig. 8). It is surely surprising to discover that 93% of these papers appeared in the literature only during the period 2008–2013 as an indication – probably – of a growing interest in the use of TiO_2, to which a more cheap element (copper) was added as co-catalyst than the more expensive Pt, Pd or rare elements.

Therefore, in the present work a critical analysis of these papers has been carried out by comparing mainly the different ways of preparation, the mechanisms proposed, the efficiencies recorded, the extension of TiO₂ light absorption capabilities, the sacrificial organics adopted and the pH of the test solutions, with the aim of collecting relevant references and stimulating the development of new ideas for the solution of problems which still limit the full exploitation on industrial scale of photocatalytic reforming for hydrogen generation.