Secondary Electron Emission

the emission of electrons by the surface of a solid object when it is bombarded by electrons. It was discovered in 1902 by the German physicists L. Austin and H. Starke. The electrons bombarding the object are called primary, and the emitted electrons are called secondary. Some of the primary electrons are reflected without energy loss (elastically reflected primaries), and the remaining electrons are reflected with a loss of energy; the energy is mainly consumed in exciting the electrons of the solid object to higher energy levels. If the electrons’ energy and momentum prove to be sufficiently great to surmount the potential barrier on the surface of the object, they leave the surface (“true” secondaries). All three groups of electrons are present in the recorded electron flow (see Figure 1).

Figure 1. Distribution of secondaries according to energies: (I) elastically reflected primaries, (II) inelastically reflected primaries, (III) “true” secondaries. Ƹp is the primary electron energy.

In thin films, secondary electron emission is observed not only from surfaces being bombarded (reflection emission; see Figure 2, a) but also from the opposite surface (leakage emission; see Figure 2, b).

Figure 2. Secondary electron emission: (a) reflection, (b) leakage

Secondary electron emission is quantitatively characterized by the coefficient of secondary electron emission, σ = i_y/i_p where i_s is the current formed by secondaries and i_p is the primary-electron current; the coefficients of elastic electron reflection, r =i_r/i_p, and inelastic electron reflection, η = _iη/i_p; and the coefficient of emission of true secondaries, δ = i_δ/i_p(i_r, i_η,) and iδ are flows that correspond to elastically reflected, inelastically reflected, and true secondaries, respectively, where i_s, = i_r + i_n + i_δ).

The coefficients σ, r, η, and δ depend on both the primary-electron energy Ep and their angles of incidence and on the chemical composition, method of production, and state of the irradiated sample’s surface. In metals, where the conduction-electron density is high, the secondaries produced have a low probability of escaping. In dielectrics, where the concentration of conduction electrons is low, the probability of the escape of secondaries is greater. In addition, the probability of electron escape depends on the height of the potential barrier on the surface. As a result, in a number of nonmetallic substances (oxides of alkaline-earth metals; alkali-halide compounds), or > 1 (see Figure 3). For specially prepared effective emitters (intermetallic compounds of the antimony-alkali-metal type, the specially activated alloys CuAlMg, AgAlMg, AgAlMgZi, and others), cr »1. In metals and intrinsic semiconductors, the value of σ is comparatively small (see Figure 4). In carbon (carbon black) and oxides of the transition metals, σ < 1; they can be used as antiemission coatings.

Figure 3. Dependence of the coefficient of secondary electron emission ղ on the energy ɛ_p of the primaries

With an increase of the energy E_p of primaries, σ at first increases (see Figures 3 and 4); this takes place until the excitation of the electrons of the solid occurs near the surface at a distance less than their path length. Upon further increase of E_p, the total number of excited electrons continues to increase, but most of them originate at a greater depth, and the number of electrons that escape decreases. The increase of σ with the angle of incidence of the primaries is explained similarly.

Figure 4. Dependence of the coefficients ղ and n. on the energy £_p of the primaries for certain metals

Single crystals are anisotropic with respect to electron motions. During electron motion along channels formed by closely packed chains of atoms, the probability of electron scattering and atom ionization increases (channeling). Electron diffraction in the crystal lattice is also observed. As a result of this, the dependence of σ, η, and r on the angle of incidence of the primaries, as well as the curves σ(E_p), r(E_p), and η(E_p) for single crystals, have a complex shape, with several maxima and minima (see Figure 5). The coefficients of σ, η, r, and δ usually given for polycrystals are values averaged over various directions.

Figure 5. Dependence of σ,η, and r on the angle of incidence Φ of primaries for single silicon crystals: £_P = 1,000 eV; the dotted line is the dependence σ (Φ) for a silicon film.

Secondary electron emission occurs within a time interval of less than 10^-12 sec—that is, it is a practically inertialess process.

The study and utilization of secondary electron emission in strong electrostatic fields and ultrahigh-frequency electrical fields has taken on its own significance. A strong electrical field (10⁵ 10⁶ volts per cm) in a dielectric leads to an increase of cr to 50-100 (secondary electron emission intensified by a field). In addition, in this case cr depends essentially on the porosity of the dielectric layer, since the presence of pores increases the effective surface of the emitter, and the field promotes the “extraction” of slow secondaries which, colliding with the walls of the pores, can cause secondary electron emission with cr > 1 and the emergence of electron avalanches. The development of avalanches under certain conditions leads to self-sustaining cold emission, which continues for many hours after the cessation of electron bombardment.

Secondary electron emission is used in many electrical vacuum devices for amplifying electron currents (photoelectric multipliers, image amplifiers, and so on) and for recording information in the form of potential relief on the surface of the dielectric (electron-beam instruments). In some devices, secondary electron emission is a “harmful” effect (the dynatron effect in electron tubes and the appearance of electric charge on the surface of glass and dielectrics in electrical vacuum devices).

In a high-frequency electric field, E = E₀cos⍵t, avalanchelike electron multiplication (secondary-electron resonance) is observed on the electrode surfaces because of secondary electron emission. This phenomenon was discovered by H. E. Farnsworth in 1934. For resonance to occur, it is necessary that the time between two successive collisions of electrons with the electrode surfaces be equal to an odd number of half-periods of the high-frequency field E (synchronism condition). Here the electrons in the field can take on an energy for which a > 1. Electron multiplication occurs on the surfaces of two electrodes between which a high-frequency electrical field is applied, or on a single sur-face placed in crossed electrical and magnetic fields (see Figure 6,b). The rapid increase in electron concentration is limited by the growth of the space charge, which violates the synchronism condition. Secondary electron resonance plays a significant role in the emergence of a dense volumetric cathode charge in magnetrons and amplitrons, as well as in the operation of dynamic photoelectric multipliers. On the other hand, this phenomenon can be the cause of unstable operation of these devices and can limit their power output.

Figure 6. Electron multiplication: (a) in a high-frequency electrical field and (b) in crossed electrical (E) and magnetic (H) fields. The field H is perpendicular to the plane of the diagram; the arrows indicate electron trajectories.