Photocatalysis at the Semiconductor Surface Principles

A necessary condition for the absorption of the photons by the semiconductor particle is that the energy of the photon should exceed the energy of the bandgap of the concerned semiconductor [38]. This threshold wavelength for the absorption, Xbg, can be expressed according to the equation

where Xbg is the threshold wavelength of a photon and £bg is the bandgap energy. Table 1 lists the A.bg and Ehg values for some important semiconductors as determined by the flat band condition. These values for the bandgap energies may be altered when the semiconductor surface is in contact with an electrolyte solution [18,38]. The band edge positions and the bandgap energies of a few semiconductors of importance in the photocatalytic degradation studies are shown in Figure 3. The data refer to the conditions where the semiconductors are in contact

Table 1 Threshold Wavelengths and Bandgap Energies for Some Semiconductors

Bandgap

Threshold

Semiconductor

Energy (eV)

wavelength (nm)

Sn02

3.5

354

ZnO

3.2

388

SrTiOj

3.2

388

Ti02

3.0

413

CdS

2.4

517

Fe203

2.1

590

Si

1.1

1128

Source: Finklea [38].

Source: Finklea [38].

with an aqueous solution of pH 1. These values are derived from the flat band potential measurements [18].

The knowledge of the band edge position is particularly useful in the discussion of photo-catalysis [18,21,39]. In Figure 3, the standard potentials for several redox systems are also listed. The relative positions of the standard potentials and the band edge positions are indicative of the thermodynamic limits for the photochemical reactions at the surface of the illuminated semiconductor particles. For example, if an oxidation of the species in the electrolyte is to be performed, the valence band edge position of the semiconductor must be positioned below the relevant redox level. Thus it can be seen that colloidal TiOz will be a strong oxidizing system.

The free energy of the charge carriers generated by photoexcitation of semiconductors is directly related to the chemical potential. In the dark, under thermal equilibrium the chemical potential of the electron is equal to that of the hole and corresponds to the Fermi level of the solid. But under illumination, the system departs from the equilibrium, and the chemical potentials of electrons and holes are no longer equal as they are under equilibrium, nonirradiated conditions. As a result, the Fermi level splits into two quasi-Fermi levels, one for the electron and one for the hole. The chemical potentials become functions of the nonequilibrium concentrations of electrons and holes. These concentrations are dependent on the absorbed light intensities. These differences are useful in redox reactions involving the electroactive species in the medium.

The transfer of mobile charge carriers between the semiconductor and the electrolyte is an important step in the photocatalytic degradation of toxic species. When an electroactive species is present in the electrolyte solution, the charge transfer can take place directly across the semiconductor-solution interface. This will create a space charge layer at the interface, and the valence and conducting bands will be bent. This will affect the effectiveness of the redox processes possible in the presence of the illuminated semiconductor particles. If the majority carriers are depleted from a colloidal semiconductor in solution and the particles are too small to develop a space charge layer, the electric potential difference resulting from the transfer of a charge from the semiconductor to the solution must drop in the Helmholtz layer (neglecting diffusion layer contributions). As a consequence, the positions of the band edges of the semiconductor particle will shift. For example, if after photoexcitation of a colloidal n-type particle, holes are transferred rapidly to an acceptor in solution while electrons remain in the particle, a negative shift in the conduction band edge at the surface will take place.

In the case of colloidal semiconductors, the band bending is small, and charge separation occurs via diffusion. The absorption of light leads to the generation of electron-hole pairs in

Figure 3 Bandgap energies and band edge positions for some semiconductors in contact with an electrolyte solution at pH 1. (From Réf. 18.)

the particle that are oriented in a spatially random fashion along the optical path. These charge carriers subsequently recombine or diffuse to the surface, where they undergo chemical reactions with suitable solutes or catalysts deposited onto the surface of the particle.

The depletion layer of the semiconductor-solution interface, sometimes referred to as the Schottky barrier, plays an important role in light-induced charge separation. The local electrostatic field present in the space charge layer serves to separate the electron-hole pairs generated by illumination of the semiconductor. For «-type materials, the direction of the field is such that holes migrate to the surface, where they undergo a chemical reaction, while the electrons drift through the bulk to the back contact of the semiconductor and subsequently through the external circuit to the counter electrode. Charge carriers that are photogenerated in the field-free space of the semiconductor can also contribute to the photocurrent. In solids with low defect concentrations, the lifetime of the electron-hole pairs is long enough to allow for some of the minority carriers to diffuse to the depletion layer before they undergo recombination.

Since in colloidal semiconductors the diffusion of charge carriers from the interior of the particle surface can occur more rapidly than the recombination, it is feasible to obtain quantum yields for photoredox processes approaching unity. Whether such high efficiencies can really be achieved depends upon the rapid removal of at least one type of charge carrier, i.e., either electrons or holes, upon the arrival at the interface.

The theroretical basis of semiconductor photochemistry is well established [12,18]. For reactions occurring at the semiconductor-liquid interface, either the semiconductor itself or any of the adsorbed species may act as the light harvester. If the semiconductor is the absorbing unit, a light photon will promote an electron from the valence band to the conduction band,

combination

ph cu

ph cu

Migration ^ >0

Diffusion

Valence Band

Electrolyte

Figure 4 A schematic presentation of photocatalyzed oxidation at an illuminated, /¡-type semiconductor. (From Ref. 18.)

thereby generating a hole in the valence band. After these charges migrate to the surface of the particles, the reaction of the photogenerated electron with a reducible absorbed species and/or the reaction of the hole with an oxidizable species can take place. Figure 4 is a simple representation of the photocatalyzed reaction at the surface of an «-type semiconductor particle [18].

These principles have found extensive application in the field of light-induced generation of fuels and that of organic transformations, but only recently in the field of environmental chemistry and pollution control. In fact, with the exception of a few reports concerning some contaminants, photoeffects at semiconductor interfaces have been largely ignored by environmental chemists [7],

0 0

Post a comment