MA Emilio Schmidt

The isothermal self-gravitating cylinder

Sought is the stationary density profile \(\rho\) of an infinitely long, isothermal, self-gravitating gas cylinder. To do this, the hydrostatic Euler equation has to be solved, which is given by:

\begin{align} \boldsymbol{\nabla} P = \boldsymbol{f}_\text{ext} \end{align}

Here, \(P\) is the pressure field and \(\boldsymbol{f}_\text{ext}\) is the force density acting on the gas. In the considered case, this is the gas's own gravity. Since the gravitational field is a conservative force field, it can be expressed by a potential \(\phi_\text{ext}\):

\begin{align} \boldsymbol{f}_\text{ext} = - \rho \boldsymbol{\nabla} \phi_\text{ext} \end{align}

Inserting yields:

\begin{align} \boldsymbol{\nabla} P = - \rho \boldsymbol{\nabla} \phi_\text{ext} \end{align}

The pressure field of the gas under consideration can be described by the equation of state of an isothermal gas, which is given by:

\begin{align} P = \frac{R T}{M} \rho \end{align}

Here, \(R\) denotes the universal gas constant, \(T\) the spatially homogeneous temperature distribution, and \(M\) the molar mass of the gas. Inserting into the hydrostatic Euler equation yields:

\begin{align} \phantom{\Rightarrow}\;& \frac{R T}{M} \boldsymbol{\nabla} \rho = - \rho \boldsymbol{\nabla} \phi_\text{ext} \\[6pt] \Leftrightarrow\;& \frac{1}{\rho} \boldsymbol{\nabla} \rho = - \frac{M}{R T} \boldsymbol{\nabla} \phi_\text{ext} \\[6pt] \Leftrightarrow\;& \boldsymbol{\nabla} \ln(\rho) = - \frac{M}{R T} \boldsymbol{\nabla} \phi_\text{ext} \\[6pt] \Rightarrow\;& \Delta \ln(\rho) = - \frac{M}{R T} \Delta \phi_\text{ext} \end{align}

It proves useful to use the Poisson equation of Newtonian gravity, which is given by:

\begin{align} \Delta \phi_\text{ext} = 4 \pi G \rho \end{align}

Inserting yields:

\begin{align} \phantom{\Rightarrow}\;& \Delta \ln(\rho) = - \frac{4 \pi G M}{R T} \rho \, , \quad \text{def.:} \, a := \frac{4 \pi G M}{R T} \\[6pt] \Leftrightarrow\;& \Delta \ln(\rho) = - a \rho \end{align}

For an infinitely long, straight, self-gravitating gas cylinder without an externally imposed azimuthal or axial structure, there is symmetry with respect to rotation about the cylinder axis and translation along the axis. The underlying equations share this symmetry, which is why corresponding derivatives disappear in the considered equation:

\begin{align} \phantom{\Rightarrow}\;& \frac{1}{r} \partial_r \big(r \partial_r \ln(\rho)\big) = - a \rho \\[6pt] \Leftrightarrow\;& \partial_r \bigg(r \frac{\partial_r \rho}{\rho}\bigg) = - a r \rho \\[6pt] \Leftrightarrow\;& \frac{\partial_r \rho}{\rho} - r \bigg(\frac{\partial_r \rho}{\rho}\bigg)^2 + \frac{\partial_r^2 \rho}{\rho} = - a r \rho \\[6pt] \Leftrightarrow\;& \rho \partial_r \rho - r (\partial_r \rho)^2 + r \rho \partial_r^2 \rho = - a r \rho^3 \end{align}

To be continued ...