The Stellar Opacity The mean absorption coefficient, κ, is not a constant; it is dependent on frequency, and is therefore frequently written as κ ν. Inside a star, several different sources of opacity are important, and each has its own frequency dependence. To compute the appropriate average opacity, κ recall that Fick s law states that the flux at any point is related to the energy density U by F ν = D U = 1 3 vl n = 1 3 and that, when integrated over all energies, c κ ν ρ U F = 4acT 3 3 κρ dt dr (3.1.4) But how is κ related to κ ν? To answer this question, we start by explicitly writing in the frequency dependence of U, i.e., U ν = 4π c B ν(t ) = 4π hν 3 { } 1 c c e hν/kt 1 and compute the monochromatic flux distribution F ν = D U ν = c { } 1 4π 3ρ κ ν c B ν(t ) = 4π 3ρ 1 db ν κ ν dt dt dr The total flux, integrated over all frequencies is then F = 4π 3ρ dt dr 1 db ν κ ν dt Comparing (6.1.1) and (3.1.4), we see that dν (6.1.1) 1 κ = π act 3 1 db ν dν (6.1.) κ ν dt
or, since db ν dt dν = d dt B ν dν = db dt = ac π T 3 1 κ = 1 db ν κ ν dt dν (6.1.3) db ν dt dν The average opacity that you should use is therefore that which is harmonically weighted over the temperature derivative of the Planck curve. This is called the Rosseland mean opacity. Note that the function weights the high frequencies more than the low frequencies. To see this, we can take the temperature derivative of the Planck function db ν dt = h ν 4 c kt = e hν/kt { e hν/kt 1 } ( k 3 T ) ( x 4 e x c h {e x 1} ) (6.1.4) where x = hν/kt, and find its peak by taking the derivative. Then ( k 3 T ) { } x 4 e x + 4x 3 e x c h (e x 1) x4 e x (e x 1) 3 = x 3 e x (e x 1) 3 (xe x x + 4e x 4 xe x ) = x 3 e x (e x 1) 3 (4e x 4 x xe x ) = Mathematically, this works out to x 3.83. In other words, the high energy photons count much more than low energy photons.
Although in real models, the Rosseland mean opacity is computed numerically, it is instructive to evaluate the form of κ when κ ν ν n. From the definition of the Rosseland mean 1 κ h c kt h c kt ν n+4 ν 4 e hν/kt ( e hν/kt 1 ) dν e hν/kt ( e hν/kt 1 ) dν which, in terms of x is or 1 κ ( ) n+4 kt x n+4 e x ( ) kt h (e x 1) dx h ( ) 4 kt x 4 e x ( ) kt h (e x 1) dx h 1 κ ( kt h ) n x n+4 x 4 e x (e x 1) dx e x (e x 1) dx T n (6.1.5) Thus, if the opacity law is κ ν n, the temperature dependence of the Rosseland mean opacity will be κ T n. This follows since the mean opacity must drop as the number of high energy photons increases.
ELECTRON SCATTERING When free electrons have thermal motions that are much less than their rest mass energy, i.e., m e c kt, or T 5.9 1 9 K, the cross-section for scattering of photons off electrons is frequency independent. The opacity per unit mass of material in this case is κ e = n eσ e ρ where σ e is the Thomson cross-section of the electron σ e = 8π ( ) e 3 m e c =.665 1 4 cm and n e is the density of electrons n e = ρn A i ( xi A i ) f i Z i (5.1.4) If the gas is fully ionized, i.e., if the ionization fractions f i 1 (as it will be in stellar interiors), and if we approximate the atomic weight (A i ) of a metal as twice its atomic number (Z i ), then n e = ρn A i Z i x i A i ρn A { X 1 + Y + Z where X, Y, and Z are the mass fractions of hydrogen, helium, and metals, respectively. Since X + Y + Z = 1, n e = ρn A { 1 + 1 X } (6.1.6) and κ e = N A σ e (1 + X)/ =. (1 + X) cm g 1 (6.1.7) Note again that as long as the ionization is complete (and matter is not degenerate), the result is independent of temperature and density. }
FREE-FREE ABSORPTION The energy emitted per unit volume by a single electron scattered by ions of charge Z (as a function of velocity, v) is de dωdv dt = 16πe6 3 3c 3 m ev n en I Z g ff (v, ω) where e is the charge of the electron, and g ff is the gaunt factor. To get the emissivity of thermal bremsstrahlung as a function of frequency, we integrate this over a Maxwellian distribution, noting that to get a photon with frequency ν, the incident electron must have m e v / hν. Thus, v min = (hν/m e ) 1/ and ϵ ff (ν) = 4πj ff = v min de ( dωdv dt 4π me ) 3/ v e m ev /kt dv πkt = 5 πe 6 3m e c 3 ( ) 1/ π g ff n e n I Z e hν/kt 3m e kt Next, to go from emission to absorption, we can use the fact that in thermodynamic equilibrium, Kirchoff s law must hold, hence the source function is S ν = j ν /a ν = B ν (T ), where a ν, is the absorption coefficient per electron (with units of inverse distance). Thus a ff (ν) = ρκ ν = 4e6 3m e hc ( ) 1 π gff n e n I Z ν 3{ 1 e hν/kt } 3m e kt (6.1.8) or, upon substituting ρ = n e µ e /N A and ρ = n I µ I /N A
κ ν = 4e6 3m e hc ( ) 1 π NA Z g ff ρ T 1/ ν 3{ 1 e hν/kt } 3m e k µ e µ I (6.1.9) The first term in the equation gives the total absorption coefficient; the exponential term at the end represents a negative contribution from stimulated emission. If we combine the constants, letting C = ( ) 1/ 4e6 π 3m e hc N A = 1.34 1 56 (cgs units) 3m e k then κ ν = C ( ) Z g ff ρ T 1 ν 3 { 1 e hν/kt } (6.1.1) µ e µ I To calculate the Rosseland mean free-free opacity, the inverse of this expression, weighted by the derivative of the Planck Function, must be integrated over all frequencies. This is tedious, but note that if we neglect the effects of stimulated emission, the frequency dependence of the opacity is ν 3, which, through (6.1.5), implies a temperature dependence of T 3. This, plus the T 1/ dependence which is already in the equation, gives a total temperature dependence of T 7/. Numerically, κ ff 1 3 Z µ e µ I ρ T 7/ cm g 1 (6.1.11) Opacities of the form κ ρ T 7/ are called Kramers opacities.
BOUND-FREE ABSORPTION For a hydrogenic atom, the cross section for absorption from state (n, l) to the continuum is σ bf (Z, n, l) = 9 π 7 m e e 1 Z 4 3 g bf 3 c h 6 n 5 (πν) 3 = 64π4 m e e 1 3 Z 4 3 c h 6 n 5 g bf ν 3 (6.1.1) where the energy of the incident photon must be great enough to ionize the atom, i.e., hν > χ n = π m e e 4 Z h n From this, it is clear that the opacity law for a single species of atom will have a number of absorption edges below which the absorption cross section will drop quickly. The total bound-free opacity, of course, will be the sum over all elements, Z, all ionization states, i, and and all excitation states, n. In the hydrogenic ion approximation, the opacity is κ ν = Z σ bf (Z, n, l) n ( Ni,n N i ) Z ( ) Ni Ni Z ρn A A Z (6.1.13) This summation, along with the process of taking the Rosseland mean, will act to smear out many of the edges. The bound-free opacity must be computed numerically, but we can get a rough idea of how bound-free opacity behaves by making the assumption that in a star s interior, all ions will be very highly ionized and thus will have at most one electron. The Saha equation states that the number of atoms in ionization state i is related to the number in ionization state i + 1 by N i = N e N i+1 u i (T ) u i+1 (T ) ( h πm e kt ) 3/ e χ i/kt (5..5)
where u(t ) are the partition functions and χ i is the ionization energy. If the gas is highly ionized, N i+i 1, u i+1 = 1, and u i =, and ( ) h 3/ N i N e e χ i/kt πm e kt We can then estimate the number of these atoms in excitation state n from the Boltzmann equation. For hydrogenic ions, the statistical weight of a level is n, so if we let χ i,n = χ i + χ n, then the total number of absorbing ions is N i,n N e n ( h πm e kt ) 3/ e χ i,n/kt The total absorption coefficient is therefore a ν σ bf N i,n ρx Z N A A Z 64π4 m e e 1 3 Z 4 3 c h 6 n 5 16πe6 Z g bf 3m e c h nν 3 g bf ν 3 ( π m e e 4 16πe 6 Z g bf 3m e c h n 6πm e k h kt ( ρ NA X ) ( Z h n µ e A Z πm e kt Z ) ρ NA X Z n µ e A Z ( ) 3 1 6πm e kt e χ i,n/kt ) 1 e χ i,n /kt ( ) ( χi,n kt e χ i,n/kt ρ NA X ) Z T 1 ν 3 µ e A Z and the opacity is κ bf = a ν /ρ z,n C ( ) ( χi,n kt e χ i,n/kt Z ) X Z µ e A Z ρ T 1 ν 3 (6.1.14)
where we have dumped the many constants into the variable C (and have ignored the stimulated emission term). Now examine the factor χ i,n /kt. If for a particular n and Z, χ i,n kt, then that term will contribute little to the opacity, since the species will have too few bound electrons. Similarly, if χ i,n kt, there will not be many photons of the requisite energy to be absorbed, and the contribution to the opacity will again be small. Thus, only those terms with χ i,n kt will contribute. This leaves κ ρ T 1/ ν 3, which is the same as free-free opacity. Bound-free absorption will therefore also follow the Kramers opacity law. A rough numerical estimate for the Rosseland mean for bound-free opacity is κ bf 1 5 Z(1 + X) ρ T 7/ cm g 1 (6.1.15) Note that this is a factor of 1 greater than the Rosseland mean for free-free opacity.
To first order, the cross-section for ionization of most species goes as ν 3 for frequencies above the ionization energy.
BOUND-BOUND ABSORPTION At extremely high temperatures, T > 1 6 K, the radiation field will consist mostly of ionizing photons, hence bound-bound absorption will be minimal (< 1%). However, at cooler temperatures the onset of bound-bound transitions in the UV and far UV can double the stellar opacity. Obviously, the computation of bound-bound opacity must be done numerically, since it involves the summation of millions of individual absorption lines. It should be noted, however, that in the interior of a star, the pressure (and thermal) broadening terms will be substantial, so the net result will be continuum absorption, rather than a series of absorption lines. Also, the expression for the absorption coefficient of hydrogenic bound-free transitions (6.1.1) is the same as that for bound-bound transitions. Thus, we can expect the same ν 3 dependence and therefore a Kramerlike opacity law. The strength of the total absorption, however, will be critically dependent on temperature.
H OPACITY A free electron passing by a neutral hydrogen atom can produce a dipole moment in the atom. At that time, an extra electron can attach itself to the system; the result is an H ion. This ion is fragile it has an ionization energy of only.754 ev, and a large cross section for absorption, either through bound-free absorption or free-free absorption H + hν H + e (v) H + e (v) + hν H + e (v) In the envelopes of cool stars, H can be the dominant source of opacity. The contribution of H to the star s opacity will depend on the number of H ions, which can be calculated from the Saha equation. n H = n ( ) e u H h 3/ e χ/kt (5..5) n H u H πm e kt where u H = 1, u H =, and χ =.754 ev. If x is the fraction of ionized hydrogen (x for cool stars), and a ν is the frequency dependency of H absorption, then the opacity due to H is κ H = a ( ) ( ) νn H nh aν = n H (1 x) ρ ρ n H = a ν XN A (1 x) n e 4 ( h πm e kt ) 3 e χ/kt = a ν XN A (1 x) 4 P e µ e kt ( h πm e kt ) 3 e χ/kt (6.1.16)
Note that the total opacity is directly proportional to the electron density. In extremely metal poor stars, the free electrons must come from ionized hydrogen, hence the opacity will rise with temperature, until the number of neutrals starts to decline. However, for normal metal rich stars, common, low ionization metals (such as Ca, Na, K, and Al) contribute to the supply of free electrons, thereby lessening this dependence. Consequently, H opacity is much stronger in metal rich stars than in metal poor stars, and is stronger in dwarfs than in giants (due to the pressure term). The determination of the frequency dependent cross-section of the H ion involves the quantum mechanical computation of overlapping wave functions. It is not a power law: it increases linearly at short wavelengths until 85 Å, then declines until the ionization threshold at 16, 5 Å. After that, that opacity increases again due to free-free opacity. As a result, the Rosseland mean of H opacity is not a Kramers law opacity. Very roughly κ H.5 1 31 (Z/.) ρ 1/ T 9 cm g 1 (6.1.17)
The combined bound-free and free-free opacity from the H ion at T = 63 K. The y-axis plots the cross-section ( 1 6 per neutral H atom per unit electron pressure, P e = n e kt ). From Mihalas, Stellar Atmospheres (nd edition), pg. 13.
This figure illustrates the importance of the various opacities, as a function of temperature and density, for a normal (Population I) mix of abundances. The lines are labeled in units of electron scattering opacity, κ = κ e =.(1 + X). From Hayashi, Höshi, & Sugimoto 196, Suppl. Prog. Theo. Phys. Japan, Vol..
The Los Alamos National Labs radiative opacities for X =.7, Y =.8, and Z =.. The dashed line shows the half-ionization curve for pure hydrogen.
The Rosseland mean opacity (in cm g 1 ) as a function of density (in g cm 3 ) and temperature for X =.739, Y =.4, and Z =.4. The range of values plotted is applicable for the outer regions of stars; the dots represent values in the solar model.
Tabulated Opacities When computing real stellar models, simple power-law representations of the Rosseland mean opacity are inadequate. Instead, one uses tables created either by The Opacity Project or OPAL http://cdsweb.u-strasbg.fr/topbase/theop.html http://opalopacity.llnl.gov One chooses the table appropriate for the metallicity being considered, and then interpolates in density and temperature (or related variables). Note that most of these tables assume a fixed metal distribution (i.e., the amount of given element, such as oxygen, scales with that of iron). However, a few do allow for independent changes in CNO, or α-process elements.
The Eddington Luminosity If energy transport at the surface of a star is through radiation, then equation (3.1.4) holds, and the star s luminosity is related to its temperature gradient by F = L T 4πR = 4ac 3κρ T 3 dt dr (6..1) or L T = 16πac 3κρ R T 3 dt dr (6..) Now assume that the dominate source of pressure is from radiation. (At the surface of a bright star, this will certainly be true.) For radiation pressure, P rad = 1 3 at 4 (6..3) which implies and dp dt = 4 3 at 3 (6..4) L T = ( 16πacR T 3 3κρ ) dt dr ( 3 ) dp 4aT 3 dt = 4πc κρ R dp dr (6..5) Now if the star is in hydrostatic equilibrium, dp dr = GM T R ρ (..) which implies L T = 4πc GM T κ (6..6)
This is the Eddington luminosity. If the luminosity of the star is greater than this, the surface cannot be stable, and will be ejected from the star via radiation pressure. Note that equation (6..6), coupled with the mass-luminosity relation on the main sequence, implies that there is an upper limit to the mass of main sequence stars. If, for main sequence stars, L T M 3.5 and electron scattering is the dominant source of opacity at the surface, then ( M M ).5 < 4πGc M (6..7).(1 + X) L This equation implies that the maximum mass of a main-sequence star is M max 6M. Stars more massive than this must have strong stellar winds and will lose mass rapidly.