In this paper we discuss the motion of the vapor formed during the evaporation of a solid by a continuous radiation spectrum. The vapor is assumed to be heated by this radiation to a temperature T much higher than the phase-transition temperature T
v and much higher than the temperature T
i at which significant ionization of the vapor begins. in the case, T
v and T
i can be neglected (as can the heat of evaporation Q
v and the energy Q
i expended on ionization). As a result of this motion, the vapor has a density ? much lower than the density
ρ 0 of the solid. It can therefore be assumed that the heating wave moves through an absolutely cold and infinitely dense gas. At the same time, the vapor temperature is assumed low enough that reradiation can be neglected. The radiation-absorption coefficient η for the ionized vapor can be described by a power-law dependence on T and ? for certain ranges of T, ε, and the photon energy ε. In this case, the motion of the gas is a self-similar problem. The spectrum and angular distribution of the incident radiation [φ (ε, θ)] and the η and ε dependences can be arbitrary. A system of ordinary differential equations is found and solved. Intense radiation incident on a solid surface will evaporate the solid. If the absorption coefficient η of the vapor and the flux density q of the radiation are high enough, the escaping vapor will be heated to a high temperature in a relatively short time. This temperature will not only be much higher than the evaporation temperature T
v, but it will also be higher than the “ionization temperature” T
i. If the internal energy per unit mass of the vapor is much higher than the heat of evaporation Q
v and the energy Q
i expended on ionization, and if the vapor density ? is much lower than the initial density
ρ 0 as a result of its escape, then the problem of the motion and heating of the vapor can be simplified through the assumptions.
(0.1) $$T_v = T_i = Q_v = Q_i = 0,\rho _0 = \infty (v_0 = 1/\rho _0 = 0)$$ (here and below, v is the specific volume). We can therefore assume that the heating wave moves through an infinitely dense and absolutely cold gas. In the region of multiple and complete ionization, the ionized-vapor absorption coefficient η, associated with free-free electron transitions in the field of ions, and bound-free transitions from the higherlying states of atoms and ions, has an approximately power-law dependence on T and ? [1], or on p and ? (p is the pressure):
Here k and K are numerical coefficients which depend on the substance and on the ranges of T, ?, and ε in which (0.2) is used. For a completely ionized gas, we have α=3/2, β=1,
a=?5/2, b=?3/2, and
\( - \bar 1/2\) when ε?T; or α=3/2, β=1,
a=3/2, and b=?1/2 when when ε ? T. We assume that (0.2) holds for any T, for approximation (0.1). We assume the ratio of specific heats γ to be constant for a certain temperature range in the range of multiple and complete ionization. With these simplifying approximations, the problem of the planar, transient flow of a gas heated by a beam of monochromatic radiation is a self-similar problem. It has been studied in [2,3]. It is shown below that the analogous problem of the motion of a gas heated by a nonequilibrium continuous radiation spectrum is also self-similar. For a partially ionized gas, approximation (0.2) is usualy satisfied only for the long-wavelength part of the incident spectrum. For the short-wavelength part of the spectrum (that is, for photons whose energy is close to or greater than the ionization potential characteristics of the ions for the given temperature range, and which are capable of direct photoionization of these ions from the ground or first excited status), the absorption coefficient is usually much smaller (by several orders of magnitude). This “hard” radiation penetrates a short distance into the solid, causing intense heating of a thin surface layer of small mass. An afterionization wave propagates through the substance, moving under the influence of the radiation flux in the hard part of the spectrum; if the temperature of the surface layer is close to the source temperature T
e, and reradiation becomes important, there will also be a thermal wave [1]. Since the energy expended in heating is large in these waves, their propagation velocity is small (in comparison with that of the wave of evaporation, initial ionization, and heating of the plasma by the long-wavelength part of the spectrum), even if the hard and soft parts of the incidence spectrum have comparable energies (E
h and E
s). Also, the intense reradiation by the thermal wave in the hard part of the spectrum increases its propagation velocity. Finally, the energy in the short-wavelength part of the spectrum may in general be small because of self-adsorption in the source itself (for example, adsorption of the short-wavelength radiation in the cold working gas ahed of a shock wave front in an explosive source [4]). Accordingly, the heating waves for the various parts of the source spectrum may propagate differently Since the mass of the surface layer heated by the short-wave-length part of the spectrum is small, the pressure produced as a result of of the disintegration of the surface layer is small when E
h is of the order order of E
s or, especially, when E
h?E
s; that is, the hydrodynamic effects of the heating and surface-layer disintegration on the motion and and heating of the deep layers heated by the “basic” part of the spectrum can also be neglecred. The high temperature and low density of this layer only facilitate the penetration of the long-wavelength part of the spectrum into the deeper layers; however, because of the small mass of this layer, even this phenomenon has little effect on the hydrodynamic processes in the deeper layers. Accordingly, Eq. (0.2) can frequently be assumed valid for the basic part of the spectrum in the case of a partially ionized gas, also; the rest of the spectrum may simply be neglected. These restrictions on the applicability of the self-similar problem are generally removed in the case of a completely ionized gas. A state close to that of complete ionization arises when two ionization potentials typical of a given temperature range are greatly different (this occurs, for example, in the case of the alkaline metals, and also when one atomic shell has been essentially ionized, while another has not yet started to be ionized; e. g., the L- and K-shells or the M- and L-shells). We consider here the case in which the heating is caused by nonequilibrium radiation, that is, radiation such that the intrinisic radiation of the vapor may be neglected. This is a valid assumption when the vapor temperature is considerably below the source temperature T
e, or, more accurately, when the following condition holds (for a Planckian source spectrum):
(0.3) $$W\sigma T_e^4 \chi \left( {\frac{{\varepsilon _1 }}{{T_e }},\frac{{\varepsilon _2 }}{{T_e }}} \right) \gg \sigma \Upsilon ^4 \chi \left( {\frac{{\varepsilon _1 }}{T},\frac{{\varepsilon _2 }}{T}} \right)$$ Here W is the source-radiation dilution coefficient due to geometric factors, σ is the Stefan-Boltzmann constant, ?
1 and ?
2 are the boundaries of the “basic part” of the spectrum, and χ is the fraction of the spectral energy of a Planckian source with a temperature T
e or T for photous with energies ?
1≤?≤?
2. We note that the boundaries ?
1 and ?
2 for the source and vapor-radiation spectra are sometimes slightly differnt, but condition (0.3) can be easily modified for this situation or for a non-Planckian source spectrum. For our problem, the radiation intensity J=J (m, t, ε, θ) is a function of four variables: the time t, the Lagrangian mass coordinate m, the photon energy ε, and the angle θ between the direction of motion and the beam direction. The intensity J
o=J (o, t, ε, θ) of the radiation incident on the boundary m=0 is assumed to be a given function. In the self-similar problem, J can be represented as
(0.4) $$J = t^\lambda J(mt^{ - n} ,\varepsilon ,\theta )$$ . This can be done (when conditions (0.1)–(0.3) are satisfied) when J
o can be represented by
(0.5) $$J_0 = t^\lambda \psi (\varepsilon ,\theta )(\varepsilon _1 \leqslant \varepsilon \leqslant \varepsilon _2 ,\theta _1 \leqslant \theta \leqslant \theta _2 )$$ If the source spectrum is Planckian, condition (0.5) requires that T
e=const. In this case, the power-law time dependence of the intensity J
o may reflect, for example, motion of the radiation source toward the irradiated surface; in this case, however, the limiting angle θ
2 of the incident radiation also changes (usually, θ
1=0). As before, the problem is self-similar if these angles θ
2(t) are always small; that is, if the radiation is almost completely unidirectional. The arbitrary nature of the function ψ(ε, ч), which shows the spectrum and angular distribution of thesource radiation, and the arbitrary nature of the function φ (ε), which shows the dependence of the absorption coefficient on the photon energy, permit us to analyze the effects of these functions on the heating and motion of the substance for the case of the self-similar solution.
相似文献