**2.1.2 Extrinsic semiconductors**

The *ad hoc* introduction (doping) of impurities into the semiconductor lattice allow to modulate its electrical properties: these doped semiconductors are generally called "extrinsic semiconductors". Extrinsic semiconductors with a larger electron concentration than hole concentration are known as "*n*-type" semiconductors, whereas those with a larger hole concentration than electron concentration are known as "*p*-type" semiconductors. The energy band diagrams for *n*-type and *p*-type semiconductors are given in Fig. 1. In the figure, EA and ED represent respectively the energy level of the acceptor and the donor specie. Assuming a complete ionization of the donor atom at a given temperature, we see that the free electrons coming from the donor atoms fill the state close to the bottom of the CB. Thus, it is more difficult for the electrons in the VB to cross the band-gap only by thermal agitation. Therefore, the number of holes in the VB decreases. This means that, if we

It was 1782 when Alessandro Volta used for the first time the term "semiconducting". One century later, in 1883, Michael Faraday documented the first observation of a semiconductor effect. Faraday noted that the electrical resistance of silver sulphide decreased with increasing temperature, showing a behaviour opposite to that of metals (Łukasiak &

Nowadays, everywhere we look, we can certainly see articles that contain semiconductors. The modern microelectronics applications are based on semiconductor technologies. Moreover, building materials, health care products, materials for special applications and much more, often employ semiconductors due to their chemical and physical properties. It is sufficient to write "semiconductor applications" into any internet search engine to get

By definition, a semiconductor is characterized by the absence of a continuum between the states (as for metals) but shows a band structure. The filled levels, called "valence band" (VB), are an energetically closely spaced array of orbitals composed by the valence electrons of the material. A similar, higher energetic, spaced array is formed by the unoccupied orbitals and it is called "conduction band" (CB). The gap existing between the top edge of VB (Ec) and the lower edge of CB (Ec) is called "band-gap" (Chattopadhyay & Rakshit, 2010).

Pristine semiconductors are generally called "intrinsic semiconductors". In an intrinsic semiconductor (IS) the Fermi-level (EF), which can be the highest occupied energy level at T = 0 K, lies near the middle of the band-gap and the corresponding Fermi-Dirac distribution *f(E)* is plotted in Fig. 1. At T = 0 K the probability of an electron to occupy a state in the CB is zero and the VB is totally full (see the thick black line). At room temperature (blue sigmoid curve) some electrons can jump into the CB and fill the states closed to the bottom of the CB. In this case the tail of *f(E)* is extended into the CB, thus, there is a probability to have electrons there. In this condition an equal number of holes exists close to the top of VB. EF is now the energy level at which the probability of occupancy is half and it lies near the middle

The *ad hoc* introduction (doping) of impurities into the semiconductor lattice allow to modulate its electrical properties: these doped semiconductors are generally called "extrinsic semiconductors". Extrinsic semiconductors with a larger electron concentration than hole concentration are known as "*n*-type" semiconductors, whereas those with a larger hole concentration than electron concentration are known as "*p*-type" semiconductors. The energy band diagrams for *n*-type and *p*-type semiconductors are given in Fig. 1. In the figure, EA and ED represent respectively the energy level of the acceptor and the donor specie. Assuming a complete ionization of the donor atom at a given temperature, we see that the free electrons coming from the donor atoms fill the state close to the bottom of the CB. Thus, it is more difficult for the electrons in the VB to cross the band-gap only by thermal agitation. Therefore, the number of holes in the VB decreases. This means that, if we

**2. Semiconductors** 

Jakubowski, 2010).

many tens of millions of responses.

**2.1.1 Intrinsic semiconductors** 

**2.1.2 Extrinsic semiconductors** 

of the band-gap.

**2.1 Electrical behavior of semiconductors** 

consider EF as the energy for which the occupancy probability is 0.5, EF must move closer to the CB for a *n*-type semiconductor (Fig. 1 scheme on the right). Similarly, EF must move closer to the VB for a *p*-type semiconductor (Fig. 1 scheme on the left). In the extreme cases, in which the doping level is very high, the Fermi levels move into the CB for *n*-type semiconductor and into the VB for *p*-type semiconductor.

Fig. 1. Energy diagram and Fermi distribution function *f(E)* for semiconductors

#### **2.2 Electrical properties of illuminated** *n***-type semiconductors**

The application of several *n*-type semiconductors on photocatalysis has experienced an exponential growth in the last 20 years (Fox & Dulay, 1993; Hashimoto et al., 2005; Hoffmann et al., 1995; Maldotti et al., 2002a). Let we consider now the influence of photoirradiation onto *n*-type semiconductor properties (Fox, 2001).

When a semiconductor surface is brought into contact with an electrolyte containing a redox couple, interfacial electron transfer can transpire, to equilibrate the solution phase potential and the energy levels of the bulk semiconductor. As electron exchange takes place across the interface, equilibration with the solution occurs and the bulk EF moves to the solution-phase equilibrium potential, whereas the band-edge positions at the surface remain fixed at their original values (Fig. 2). Thus, in moving inward from the semiconductor-electrolyte

Fig. 2. Band bending into *n*-type and *p*-type semiconductors. The vertical lines represent semiconductor / solution interface.

Semiconductors in Organic Photosynthesis 83

For practical applications, because the simultaneous presence of cathodic and anodic site onto the same particle, the semiconductor photocatalyst can be considered as a "short circuit" electrochemical cell, when prepared by deposition island of an inert metal onto its

These metalized particles bear regions for both photoactivated oxidation of organic substrates adsorbed at the semiconductor-electrolyte interface and collection of the photogenerated electrons on the metal island. Thus, as in a electrochemical cell, these metalized powders include sites for photoinduced oxidation and reduction, analogous to the working anode and cathode. On platinized TiO2 powders, however, no external current

Anyway, even pristine wide band-gap semiconductors often show photoactivity if the two redox sites can be separated; metalized semiconductors are essential to promote some redox

In the last decades photocatalytic processes induced by semiconductors have attracted the great interest, due to their low environmental impact. Nowadays, many *n*-type semiconductors are studied and applied in several application fields, such as energy production, smart-materials technology, environment depollution, chemical synthesis,

processes, such as selective hydrogenations of alkenes and alkynes (Fox, 2001).

Fig. 4. Interfacial electron transfer on semiconductor surface

Fig. 5. Irradiation of metalized semiconductor particle

flow accompanies these transformations.

**3. Applications of** *n***-type semiconductors** 

surface (Fig. 5) (Bard, 1979).

interface, the bands bend. Charge injection in this region of bent bands forces electrons to move in the opposite direction from holes: in a *n*-type semiconductor, electrons move away from the surface and toward the bulk of the semiconductor while the holes move toward the surface. Thus, band bending assists in improving the efficiency of photoinduced charge separation. As a result of band bending, the surface of an irradiated *n*-type semiconductor becomes electron-deficient and acts as a photoanode toward an oxidizable adsorbate.

#### **2.3 Redox reactions onto irradiated** *n***-type semiconductors**

When a semiconductor surface is irradiated by light of an energy higher than the band-gap, a band to band transition takes place. An electron (e-) moves from the filled VB to the empty CB, leaving an electron hole (h+) in the VB (Fig. 3).

Fig. 3. Photo-irradiation of a semiconductor particle

Depending on their own band-gap, for each semiconductor exists a minimum light frequency to promote the electron jump. Thus, the absorption of a photon induces a charge separation, generating a highly energetic electron and creating a lower-energy vacancy in the valence band. This intermediate species is called "electron-hole pair". In this excited state, the recombination of the electron-hole pair, due to back electron transfer, is of course strongly thermodynamically favored. But, because of the absence of states in the semiconductor band-gap, quantum restrictions slow recombination by back electron transfer and the lifetime of the photogenerated electron-hole pair is sufficiently long to allow charge trapping. Thus, the life time of VB hole, in principle, can be enough long to act as an anode, whereas, similarly, the CB can act as cathode. As a consequence, the excited semiconductor is able to promote redox reactions on substrates that may be present on its surface or in solution. In *n*-type semiconductors, the surface becomes electron-deficient and acts as an oxidative site toward oxidizable organic substrates. If an adsorbed substrate has an appropriate redox potential, it can trap one of the charge carriers faster than the electronhole recombination and photoinduced chemistry will be observed. As represented in Fig. 4, the e- promoted in the CB is transferred to the adsorbed acceptor (A) and the VB hole is filled by an e- coming from the adsorbed donor species (D). The rate of the electron-transfer processes compete with spontaneous back electron transfer of the electron-hole pair.

If the trapping of both VB hole and CB e are faster than the spontaneous back electron transfer, singly oxidized and reduced intermediates are formed on the photocatalyst surface. Because these adsorbed species are free to move along the surface, novel chemistry is likely to ensue. This is a unique characteristic of photochemical catalysts, that are able to promote oxidations and reductions on the same surface.

interface, the bands bend. Charge injection in this region of bent bands forces electrons to move in the opposite direction from holes: in a *n*-type semiconductor, electrons move away from the surface and toward the bulk of the semiconductor while the holes move toward the surface. Thus, band bending assists in improving the efficiency of photoinduced charge separation. As a result of band bending, the surface of an irradiated *n*-type semiconductor

When a semiconductor surface is irradiated by light of an energy higher than the band-gap, a band to band transition takes place. An electron (e-) moves from the filled VB to the empty

Depending on their own band-gap, for each semiconductor exists a minimum light frequency to promote the electron jump. Thus, the absorption of a photon induces a charge separation, generating a highly energetic electron and creating a lower-energy vacancy in the valence band. This intermediate species is called "electron-hole pair". In this excited state, the recombination of the electron-hole pair, due to back electron transfer, is of course strongly thermodynamically favored. But, because of the absence of states in the semiconductor band-gap, quantum restrictions slow recombination by back electron transfer and the lifetime of the photogenerated electron-hole pair is sufficiently long to allow charge trapping. Thus, the life time of VB hole, in principle, can be enough long to act as an anode, whereas, similarly, the CB can act as cathode. As a consequence, the excited semiconductor is able to promote redox reactions on substrates that may be present on its surface or in solution. In *n*-type semiconductors, the surface becomes electron-deficient and acts as an oxidative site toward oxidizable organic substrates. If an adsorbed substrate has an appropriate redox potential, it can trap one of the charge carriers faster than the electronhole recombination and photoinduced chemistry will be observed. As represented in Fig. 4,

 promoted in the CB is transferred to the adsorbed acceptor (A) and the VB hole is filled by an e- coming from the adsorbed donor species (D). The rate of the electron-transfer

transfer, singly oxidized and reduced intermediates are formed on the photocatalyst surface. Because these adsorbed species are free to move along the surface, novel chemistry is likely to ensue. This is a unique characteristic of photochemical catalysts, that are able to promote

are faster than the spontaneous back electron

processes compete with spontaneous back electron transfer of the electron-hole pair.

becomes electron-deficient and acts as a photoanode toward an oxidizable adsorbate.

**2.3 Redox reactions onto irradiated** *n***-type semiconductors** 

CB, leaving an electron hole (h+) in the VB (Fig. 3).

Fig. 3. Photo-irradiation of a semiconductor particle

If the trapping of both VB hole and CB e-

oxidations and reductions on the same surface.

the e-

Fig. 4. Interfacial electron transfer on semiconductor surface

For practical applications, because the simultaneous presence of cathodic and anodic site onto the same particle, the semiconductor photocatalyst can be considered as a "short circuit" electrochemical cell, when prepared by deposition island of an inert metal onto its surface (Fig. 5) (Bard, 1979).

Fig. 5. Irradiation of metalized semiconductor particle

These metalized particles bear regions for both photoactivated oxidation of organic substrates adsorbed at the semiconductor-electrolyte interface and collection of the photogenerated electrons on the metal island. Thus, as in a electrochemical cell, these metalized powders include sites for photoinduced oxidation and reduction, analogous to the working anode and cathode. On platinized TiO2 powders, however, no external current flow accompanies these transformations.

Anyway, even pristine wide band-gap semiconductors often show photoactivity if the two redox sites can be separated; metalized semiconductors are essential to promote some redox processes, such as selective hydrogenations of alkenes and alkynes (Fox, 2001).
