Dye Molecules Adsorbed on the Electrode and in Solution

As demonstrated in the previous section, photocurrents were observed upon excitation of dyes in a system where the dye was dissolved in the electrolyte. Since an excited molecule has a limited lifetime, the question arises regarding whether 

Since in most photoelectrochemical experiments, the O2 concentration was relatively high, the triplet lifetime was certainly never greater than 10 s. The diffusion length L of an excited molecule can be calculated by using Eq. (2.26) as given by 

It is also an interesting question whether the electron transfer between an excited dye molecule and a semiconductor electrode occurs via the singlet or the triplet of the molecule. This depends on thermodynamic and kinetic factors, as follows  

Since the optical energy is smaller for a triplet (Sq Tj) compared with an excited singlet (Sg - S ), the reduction potential of a triplet molecule is less negative and the oxidation potential is less positive than those of an excited singlet molecule. An electron transfer can only occur if the corresponding energy states still overlap with the conduction or valence band of the semiconductor. 

The intersystem crossing rate (rate constant k- ) has to be compared with that of electron transfer between the excited molecule in its Sj state and the semiconductor. 

As demonstrated in the previous section, photocurrents were observed upon excitation of dyes in a system where the dye was dissolved in the electrolyte. Since an excited molecule has a limited lifetime, the question arises regarding whether only adsorbed molecules or also those in the solution are involved in the charge transfer process. This depends on the nature of the excited state, i.e. whether the excited molecule remains in its singlet state or is converted into the triplet state before the electron transfer occurs. As already mentioned in Section 10.1., the singlet lifetime is mostly in the order of ltr to 10 s, whereas the triplet lifetime ranges from about 10 s up to some 10 s. 

This conclusion has been confirmed by several experimental results. First of all, the excitation spectra of the photocurrent are frequently shifted compared with the absorption spectra measured with dye solutions, as shown for oxazine adsorbed on an n-type Sn 2 electrode ( g = 2.22 eV) in Fig. 10.9 [22]. This red shift corresponds to an energy difference of about 0.1 eV, which indicates a strong interaction between Sn 2 and oxazine (compare also with Section 10.2.4). The photoelectrochemical experiments were performed with very dilute solutions and the anodic photocurrent was found to saturate at a dye concentration of about 2 x 10 M [21], Probably, just one complete monolayer was formed at this concentration. 

In other cases, there is experimental evidence in the excitation spectra for aggregate formation such as dimers (double peak) which are not visible in the absorption spectra of the solutions with low dye concentrations. This has been confirmed by absorption measurements of adsorbed dye layers (crystal violet on ZnO electrodes [23]. Even polymers can be formed by adsorbed dye molecules, as found, for example, with pseu-doisocyanine on ZnO [24]. One example is given in Fig. 10.10. This is a well-known phenomenon for cyanine dyes [25] where polymer bands have been found at high dye concentrations. 

In most sensitization experiments it has been observed that the sensitization current decreased with long exposure to light. This result indicates that the reaction product, i.e. the oxidized or reduced dye, remained adsorbed at the electrode surface and was only slowly exchanged for other dye molecules from the solution. This result is also of interest from another point of view. At an early stage of investigation in this field, the 

---