Electron Photoemission into Solutions

Events of electron photoemission from a metal into an aqueous solution had first been documented in 1966 by Geoffrey C. Barker and Arthur W. Gardner on the basis of indirect experimental evidence. The formation of solvated electrons in nonaque-ous solutions (e.g., following the dissolution of metallic sodium in liquid ammonia) had long been known, but it was only in the beginning of the 1950s that their existence in aqueous solutions was first thought possible. It is probably for this reason that even nowadays in aqueous solutions we more often find the term solvated than hydrated electrons. 

When photons are absorbed in a metal, the ensemble of electrons are excited and some of the electrons are promoted to higher energy levels. The excited state is preserved in the metal for only a short time, and the system returns rapidly to its original state. When the photon energy hv is higher than the metal s electron work function, in the solution at a given potential, individual excited electrons 

The further fate of the solvated electrons depends on solution composition. When the solution contains no substances with which the solvated electrons could react quickly, they diffuse back and are recaptured by the electrode, since the electrochemical potenhal of electrons in the metal is markedly lower than that of solvated electrons in the solution. A steady state is attained after about 1 ns) at this time the rate of oxidahon has become equal to the rate of emission, and the original, transient photoemission current (the electric current in the galvaihc cell in which the illuminated electrode is the cathode) has fallen to zero. Also, in the case when solvated electrons react in the solution yielding oxidizable species (e.g., Zn + Zn ), 

Steady photoemission currents can be realized when acceptors (scavengers) for the solvated electrons are present in the solution. A good scavenger should be nonelectroactive at the potenhal of interest, should react quickly with solvated electrons, and the reaction products should be either nonelectroactive or reducible. A reachon with acceptors implies that the current of reoxidation of the solvated electrons becomes lower, and thus a steady photoemission current appears. The acceptors most often used are nitrous oxide, N2O, and hydroxonium ions, HjO. In the former case, OH radical is produced in the scavenging process, which undergoes further reduction on the electrode, thus doubling the photocurrent  

Water molecules do not react with the solvated electrons in the time scale of charge recapture, and hence do not fnnction as acceptors. 

---

If the electrode process is not complicated by adsorption or other phenomena, depending directly on the nature of the electrode, then the rate of the charge transfer stage proper should be independent of the electrode material. Indeed, this type of independence has been earlier ascertained experimentally for the electroreduction of anions and the electron photoemission into solution, fo such processes should belong also the cathodic generation of solvated electrons by the primary mechanism. 

PHOTOEXCITATION OF METALS (ELECTRON PHOTOEMISSION INTO SOLUTIONS)... 

Electron work functions of metals in solution can be determined by measurements of the current of electron photoemission into the solution. In an electrochemical system involving a given electrode, the photoemission current ( depends not only on the light s frequency v (or quantum energy hv) but also on the potential E. According to the quantum-mechanical theory of photoemission, this dependence is given by... 

The basic law of electron photoemission in solntions which links the photoemission current with the light s frequency and with electrode potential is described by Eq. (9.6) (the law of five halves). This eqnation mnst be defined somewhat more closely. As in the case of electrochemical reactions (see Section 14.2), not the fnll electrode potential E as shown in Eq. (9.6) is affecting the metal s electron work function in the solution bnt only a part E - / ) of this potential which is associated with the potential difference between the electrode and a point in the solntion jnst outside the electrode. Hence the basic law of electron photoemission into solntions should more correctly be written as... 

Semiconductor electrodes exhibit electron photoemission into the solution, like metal electrodes, but in addition they exhibit further photoelectrochemical effects due to excitation of the electrode under illumination. The first observations in this area were made toward the middle of the twentieth century. At the end of the 1940s,... 

Fig. 30. Diagram of transitions related to photoemission into solution 1—photoexcitation of an electron and photoemission, 2—thermalization of the photoemitted electron in the solution, 3—solvation of the thermalized electron, and 4—trapping of the solvated electron by acceptor A in the solution. de)oc is the lower edge of the band of delocalized states in the solution, solv is the energy level of the solvated electron, and EA is the acceptor energy level.

Gurevich has developed a theory for the case of electronic photoemission into concentrated electrolyte solutions. The approach is analogous... 

Electron photoemission from an electrode into an electrolyte solution, which yields solvated (hydrated) electrons, can be regarded as a particular case of reactions with photoexcitation of the electrode (Section 29.2). 

The theoretical developments in the above areas were influenced, to a considerable extent, by concepts borrowed from semiconductor physics and the physics of surfaces. Other fields of photoelectrochemistry of semiconductors were affected to a greater degree by progress achieved in the study of metal electrodes. Here we mean photoemission of electrons from semiconductors into solutions and electroreflection at a semiconductor-electrolyte interface. 

All the above-considered photoelectrochemical phenomena are based on the transition of light-excited electrons into a localized state in the solution, namely at the energy levels associated with individual ions or molecules. However, the phototransition is also possible when the electrons pass into a qualitatively different delocalized state in the solution it is this type of phototransition that represents photoemission (Barker et al, 1966). The emitted delocalized electron in the solution is then thermalized and localized to form a solvated (hydrated in aqueous solution) electron. The energy level, which corresponds to the solvated electron, lies below the bottom of the band of permitted delocalized states in the solution. Finally, the electron may pass from the solvated state to an even lower local energy level associated with an electron acceptor in the solution (see Fig. 30). 

An important advantage of the photoemission method is the fact that the electrons emitted into the solution have a known or the studied energy distribution with a strictly definite boundary energy Em.lx188 ... 

Solvated electrons were considered to be exotic over several decades and that is why comparatively little time was devoted to their study. However, the situation drastically changed early in the 60 s when the crucial role of solvated electrons in radiation chemistry was reliably ascertained. This triggered numerous theoretical and experimental studies into the structure and properties of solvated electrons and their solutions, the pathways of formation of solvated electrons, the kinetics and the mechanism of reactions involving their participation. The findings of these studies have been summarized in several monographs and in the proceedings of international conferences dedicated to the physico-chemistry of solvated electrons (the last, 6th Weyl Symposium was held in 1983) therefore, these will not be spjecific-ally considered in this overview. We shall juxtapose these with electrochemical data, of course. The same relates also to electron photoemission from metal into solution, to which monographs are devoted. 

A theory of the photoelectrochemical kinetics of the hydrogen evolution reaction at the metal-solution interface was developed by Bockris, Khan, and Uosaki. This theory takes into account photoemission into the solution, the reflective properties of the electrode material, absorption of light by the electron, and the excitation probability from photon-electron interactions. 

A quantum yield of about 0.07 electrons emitted into the solution per absorbed photon was found. Interestingly, this value exceeds, by several orders of magnitude, the yields encountered in photoemission experiments with compact semiconductor electrodes [53]. This result indicates that the particle size may be important and indeed it was found that the e absorption occurs only with small particles (nm range) and the absorption coefficient increased with decreasing par-... 

There are two important differences between photoemission into vacuum and into condensed media. Firstly, the work function in the latter case usually (but not necessarily) is lowered because of an interaction of the photoelectron with the solvent. Further, chemical reactions are possible with deliberately introduced scavengers or trace impurities present in an electrolyte. Secondly, the presence of an electrical double layer introduces a potential drop across the Helmholtz plane and diffuse layer. In some earlier theories, the electrical double layer in the presence of solute molecules was thought to screen effectively the positive charge remaining after the electron is emitted (unlike the case of emission into vacuum), thereby reducing image effects and changing the theoretical form of the photocurrent rate expressions. 

The solution of Eq. (27) for (m) is known as the Jost solution, and this can be substituted into Eq. (24) to obtain an expression for For the case of photoemission into vacuum, it is found that, indeed, the relation = const, holds, as originally postulated phenomenologically by Fowler. The magnitude of reflects the presence of image forces and is characteristic of the external media (if other than vacuum). Physically, this result has been derived without any assumptions concerning the electron dispersion law for metals. ... 

At any particular potential, the thermodynamic work needed to transfer an electron from a metal surface into solution is determined only by the magnitude of the potential, regardless of the nature of the metal (the actual photoemissive work function may be slightly larger, depending on the internal momentum distribution, but in what follows we shall equate the thermodynamic and photoemissive work functions). This independence of the threshold potential has been confirmed for liquid (e.g.. Ref. 36), and solid metals (e.g.. Ref. 40). At the electrified interface, it is well known that the electrochemical potential of species /, / , is given by... 

There are two broad aspects of studies of photoemission from electrodes into condensed media—the description of the physical mechanisms per se and the investigation of the subsequent interactions of the photoemitted electron in the solution phase. In this chapter, the theory presented has dealt almost exclusively with the first of these topics, i.e., the various attempts at an exact, quantitative description of the microscopic interactions occurring in photoemission. The two concluding sections below contain brief summaries of the current status of each of these two fields of research. 

The photocurrent 1 due to photoemission of electrons from simple metals into an electrolyte solution in the near-the-threshold frequency range is given by the relation (Gurevich et al., 1967)... 

Thus hole or electron transfer can follow a number of pathways across the semiconductor/electrolyte interface. First, one can have direct oxidative or reductive charge transfer to solution species resulting in desired product formation. Second, one can have direct charge transfer resulting in surface modification, such as oxide film growth on GaP or CdS in aqueous PECs. Finally, one can have photoemission of electrons or holes directly into the electrolyte. All of these processes provide some information about the electronic structure of the interface. 

Table 1 summarizes the basic relationships that link energy characteristics of excess electrons with the values measured by the aforementioned methods (see also Fig. 1). In the equations given therein, i.e. in Eqs. (5) and (6) w , w , and w denote respectively metal-to-vacuum, metal-to-solution, and solution-to-vacuum photoemission work functions AT is the Volta potential difference for a metal-solution system Eg is the equilibrium potential of the electrode in solvated electron solution and il(RE) is the Fermi level of the reference electrode. Equation (6) is approximate (see above) because the solvated electron entropy has not been taken into consideration. The main error in equating the heat of electron solvation and the activation energy of the thermoemission current for the solvated electron solution is caused by the variation in the solution s surface potential with temperature apparently, here specific adsorption of solvated electrons (or of an alkali metal) on the solution/vapour interface makes major contribution to the surface potential . This error can be probably neglected if measurements are taken in very dilute solutions (<10 mol/1, see ) of the alkali metal. This follows from the dependence measured in between thermoemission current and the concentration of sodium in hexamethylphosphotriamide. 

Now we return to the dependence of the current of photoemission, from sodium solutions in hexamethylphosphotriamide into vapour phase, on quantum energy. According to Ref. the second peak in the photocurrent vs. energy curve has its origin in some complex containing Na. In Ref. it has been proposed that it is the associate (of type Na", for instance) that acts as emitter. (In no case, significant concentration of some other complexes containing Na and solvated electrons... 

---