= Solution
The <specific intensity> of a <blackbody> is direction-independent over its outward hemisphere. Its outward flux per unit frequency is therefore
$$
F_\nu=\int_{\rm outward} B_\nu\cos\theta\,d\Omega
=2\pi B_\nu\int_0^{\pi/2}\cos\theta\sin\theta\,d\theta=\pi B_\nu.
$$
Multiplying by the stellar surface area gives the <spectral luminosity>
$$
\boxed{L_\nu=4\pi^2R_s^2B_\nu.}
$$
Using the <Planck law>, the <hydrogen-ionizing photon production rate> follows by dividing each spectral contribution by its <photon> energy and integrating above the threshold:
$$
Q_H=\int_{I_H/h}^\infty\frac{L_\nu}{h\nu}\,d\nu
=\frac{8\pi^2R_s^2}{c^2}\int_{I_H/h}^\infty\frac{\nu^2}{e^{h\nu/kT_s}-1}\,d\nu.
$$
With $x=h\nu/(kT_s)$ and $y=I_H/(kT_s)$, this becomes
$$
Q_H=\frac{8\pi^2I_H^3}{h^3c^2}\frac{R_s^2}{y^3}
\int_y^\infty\frac{x^2}{e^x-1}\,dx.
$$
For the initial hydrogen-only <photon> balance, assume an <ionization-bounded nebula> at steady <photoionization equilibrium>, negligible dust absorption and ionizing-photon leakage, negligible <collisional ionization>, and local reabsorption of <photons> from direct ground-state recombinations. A ground-state recombination then causes another <photoionization> and produces no net removal of a proton. Net removal is counted by <Case B recombination>, whose coefficient is $\beta=\sum_{n\geq2}\alpha_n$. Integrating the local balance gives
$$
\boxed{\int N_eN_+\beta\,dV=Q_H
=\frac{8\pi^2I_H^3}{h^3c^2}\frac{R_s^2}{y^3}
\int_y^\infty\frac{x^2}{e^x-1}\,dx.}
$$
The stellar <temperature> $T_s$ sets the <photon> supply; the gas <temperature> sets the recombination coefficients. They are not generally equal. For uniform <hydrogen> and constant gas <temperature>, the <Strömgren radius> is
$$
R_H=\left(\frac{3Q_H}{4\pi\beta_HN_eN_H}\right)^{1/3},
$$
with $N_e=N_H$ in fully ionized pure <hydrogen>.
For the hydrogen-helium mixture, define the <blackbody ionizing photon function>
$$
F(y)=\int_y^\infty\frac{x^2}{e^x-1}\,dx,\qquad
Q(E_0)=\frac{8\pi^2R_s^2(kT_s)^3}{h^3c^2}F(E_0/kT_s).
$$
The three relevant thresholds are approximately $13.6$, $24.6$ and $54.4\,\mathrm{eV}$, for $\mathrm H^0$, $\mathrm{He}^0$ and $\mathrm{He}^+$ respectively. Higher stellar <temperature> increases the relative supply of the harder <photons>. These cumulative <photon> rates cannot be assigned independently to the three species: a <photon> above a <helium> threshold can instead ionize <hydrogen>, and <helium> recombination radiation can also ionize <hydrogen>. The absorption cross-sections and the diffuse radiation determine the effective allocation.
A useful sharp-front model of <nested hydrogen-helium ionization zones> has an inner $\mathrm{He}^{++}$ sphere of radius $R_2$, a $\mathrm{He}^+$ shell ending at $R_1$, and an outer hydrogen-ionized region ending at $R_H$, with $R_2\leq R_1\leq R_H$. Let $n_H,n_{He}$ be uniform nuclear number densities. The <electron number densities> in these three regions are $n_H+2n_{He}$, $n_H+n_{He}$ and $n_H$. For constant gas <temperature>, integrating the recombination rates gives
$$
\begin{aligned}
Q_H^{\rm eff}&=\frac{4\pi}{3}\beta_H n_H\{n_HR_H^3+n_{He}(R_1^3+R_2^3)\},\\
Q_{HeI}^{\rm eff}&=\frac{4\pi}{3}\beta_{HeI}n_{He}(n_H+n_{He})(R_1^3-R_2^3),\\
Q_{HeII}^{\rm eff}&=\frac{4\pi}{3}\beta_{HeII}n_{He}(n_H+2n_{He})R_2^3.
\end{aligned}
$$
Here the effective budgets mean net species-specific ionizations after eliminating local ground-state recycling. Their calculation must include <photon> competition and any interspecies diffuse contribution; the <helium> coefficients label the recombining <ion>'s emitted spectrum. Once these budgets are specified, solve the third equation for $R_2^3$, the second for $R_1^3-R_2^3$, and the first for $R_H^3$.
For an isolated <helium> front with approximately common <electron number density>, the useful scaling is
$$
\left(\frac{R_{He}}{R_H}\right)^3
\simeq\frac{Q_{He}^{\rm eff}}{Q_H^{\rm eff}}
\frac{n_H\beta_H}{n_{He}\beta_{He}}.
$$
Thus the small <helium> abundance partly compensates for the smaller hard-photon supply. A cooler central star produces little $\mathrm{He}^{++}$ and can have a substantially smaller helium-ionized zone. A sufficiently hot star can ionize <helium> throughout most of the <hydrogen> region. If independent radius estimates place a <helium> front beyond the <hydrogen> front, the independent <photon> allocations are inconsistent: the fronts and transfer must be treated together. \b[The radii depend on <photon> hardness, abundance, <electron number density> and recombination rates, not on thresholds alone.] The sharp-front equations are an approximation; they do not assert three exact independently conserved <photon> budgets for a real mixed nebula.
For a nonuniform nebula, an observed <recombination line> with <effective recombination coefficient> $\alpha_\ell^{\rm eff}$ has <luminosity>
$$
L_\ell=h\nu_\ell\int N_eN_{\rm parent}\alpha_\ell^{\rm eff}\,dV.
$$
If gas <temperature> and density permit a known approximately constant ratio $\beta/\alpha_\ell^{\rm eff}$, the integrated line gives
$$
Q^{\rm eff}=\frac{\beta}{\alpha_\ell^{\rm eff}}\frac{L_\ell}{h\nu_\ell}.
$$
The unknown density distribution cancels because the same recombination integral appears in both quantities. For a stellar continuum measurement at a nonionizing frequency $\nu_o$,
$$
F_{\nu_o,*}=\pi B_{\nu_o}(T_s)\frac{R_s^2}{d^2}.
$$
When the inferred recombination budget traces the full appropriate ionizing-photon supply, define $\Phi_{E_0}(T)=\int_{E_0/h}^\infty B_\nu(T)/(h\nu)\,d\nu$ and obtain the <Zanstra method> equation
$$
\boxed{\frac{F_\ell}{h\nu_\ell F_{\nu_o,*}}\frac{\beta}{\alpha_\ell^{\rm eff}}
=\frac{\Phi_{E_0}(T_s)}{B_{\nu_o}(T_s)}.}
$$
Both $R_s$ and the distance $d$ cancel. <Hydrogen> and <helium> recombination lines probe different hardness thresholds, so their ratios and the stellar continuum constrain $T_s$ and test the assumed spectrum and trapping. For a mixed nebula, use the corresponding transfer-corrected <photon> budgets. <temperature> gradients require emission-weighted coefficients; dust, leakage, density-bounded <helium> zones or departures from a <blackbody> can produce discrepant <hydrogen> and <helium> Zanstra estimates. Line fluxes alone without an appropriate budget or continuum normalization need not uniquely determine $T_s$.
Finally, locate the derived $T_s$ and <luminosity> on the <Hertzsprung-Russell diagram>. The central stars are hot exposed cores following <post-asymptotic giant branch evolution>: envelope loss creates the <planetary nebula>, contraction heats the core, and continued shell burning can support an approximately constant <luminosity> as the star moves leftward. Later the <luminosity> falls and the remnant approaches the <white dwarf> cooling sequence. Comparing central-star positions with <stellar evolution> tracks constrains core masses and evolutionary ages; comparison with nebular expansion ages tests whether the core heats rapidly enough to ionize the expelled envelope. \b[The central-star locus traces the transition from an envelope-losing giant to a compact stellar remnant.]
Back to article page
