A star may remain in a state of hydrostatic equilibrium for most of its lifetime, but different physical processes act over different characteristic timescales. Once equilibrium is disturbed—by exhaustion of fuel, loss of pressure support, or radiative imbalance—the rate of change is governed by these intrinsic timescales.

Three important timescales determine the pace of stellar evolution:

• Thermal (Kelvin–Helmholtz) timescale — governs how long a star can radiate away its thermal energy if nuclear fusion ceases.
• Dynamical timescale — measures how fast a star would collapse under gravity if pressure vanished.
• Diffusion timescale — quantifies how long it takes photons to random-walk from the stellar core to the surface.

The thermal timescale \( \tau_{\mathrm{th}} \) (also called the *Kelvin–Helmholtz timescale*) is the time required for a star to radiate away its total internal (thermal) energy \(E_k\) at its luminosity \(L\):

$$ \tau_{\mathrm{th}} \approx \frac{\sum E_k}{L}. $$

From the virial theorem, the total kinetic energy is half the magnitude of the gravitational potential energy:

$$ \sum E_p = -2 \sum E_k = -\frac{G M^2}{R}, $$

where \(M\) is the stellar mass, \(R\) its radius, and \(G\) the gravitational constant. Substituting gives

$$ \tau_{\mathrm{th}} \approx \frac{G M^2}{R L}. $$

If nuclear reactions in the Sun stopped today, it would radiate away its stored thermal energy in roughly 30 million years (30 Myr). However, we know the Sun’s age exceeds 4.5 billion years (4.5 Gyr)—older than Earth itself—implying that nuclear fusion provides a continuous source of internal energy far exceeding the Kelvin–Helmholtz limit.

—

The dynamical timescale \( \tau_{\mathrm{dyn}} \) is the time a star would take to collapse under its own gravity if pressure were suddenly removed.

Consider a small test mass \(m\) on the stellar surface. Its inward acceleration due to gravity is

$$ a = \frac{F}{m} = \frac{G M}{R^2}. $$

If we (unrealistically) assume \(a\) remains constant during the fall, then the free-fall time for a distance \(R\) follows from \(s = a t^2 / 2\):

$$ \tau_{\mathrm{dyn}} = \sqrt{\frac{R}{a}} = \sqrt{\frac{R^3}{G M}}. $$

Expressing the mean density as \( \rho = M / R^3 \), we obtain the elegant form

$$ \tau_{\mathrm{dyn}} = (G \rho)^{-1/2}. $$

For the Sun (\(\rho_\odot \approx 1.4\times10^3\ \mathrm{kg\,m^{-3}}\)), \(\tau_{\mathrm{dyn}} \approx 50\ \text{minutes}\), reducing to about 20 minutes when accounting for internal density variations.

Exercise:
A white dwarf has \(M \approx M_\odot\) but \(R \approx 0.01 R_\odot\).
Estimate its density and corresponding \(\tau_{\mathrm{dyn}}\).
What about a neutron star with nuclear density \( \rho \sim 10^9\ \mathrm{kg\,m^{-3}} \)?
How short can \(\tau_{\mathrm{dyn}}\) become?

—

The diffusion timescale \( \tau_{\mathrm{diff}} \) measures how long a photon takes to travel from the stellar core to the surface. Because photons are constantly scattered or absorbed and re-emitted (e.g. by Thomson scattering), their path is a random walk rather than a straight line.

Let the step size be the mean free path \(l\). In a one-dimensional random walk of \(N\) steps, the root-mean-square (rms) displacement is

$$ x_{\mathrm{rms}} = \sqrt{N}\, l. $$

To traverse a distance \(x\), the number of steps required is

$$ N \approx \left(\frac{x}{l}\right)^2. $$

In three dimensions, replacing \(x\) by the stellar radius \(R\):

$$ N \approx \left(\frac{R}{l}\right)^2. $$

Since each step takes a time \(l/c\) (where \(c\) is the speed of light), the total diffusion time is

$$ \tau_{\mathrm{diff}} = \frac{N l}{c} = \frac{R^2}{c l}. $$

To compute \(l\), we use the Thomson cross-section for electron scattering:

$$ \sigma_T = \frac{8\pi}{3} r_e^2, $$

where the classical electron radius

$$ r_e = \frac{1}{4\pi \epsilon_0} \frac{e^2}{m_e c^2}. $$

For a fully ionized hydrogen plasma, the electron number density equals the proton number density \(n_e = n_p = \frac{M / R^3}{m_p}\), giving

$$ l = \frac{1}{\sigma_T n_e} = \frac{m_p R^3}{\sigma_T M}. $$

Substituting this into the earlier expression yields

$$ \tau_{\mathrm{diff}} = \frac{\sigma_T}{c m_p} \frac{M}{R}. $$

For solar parameters, this gives approximately 10,000 years, and accounting for absorption–reemission processes, the actual photon escape time is closer to 20,000 years.

—

The Sun’s luminosity can be roughly estimated from its internal energy and photon diffusion time:

$$ L_\odot \approx \frac{a T^4 (4\pi R_\odot^3 / 3)}{\tau_{\mathrm{diff}}} \approx 10^{27}\ \mathrm{W}, $$

or about 1 ronnawatt (RW)—an octillion watts.

Interestingly, photons have about a thousand times less energy than typical particles in the solar plasma. Thus, if the Sun were composed entirely of photons, it would take roughly a thousand times longer to lose all its energy—comparable to the thermal timescale.

—

• The thermal timescale (\(\tau_{\mathrm{th}}\)) determines how fast a star cools without fusion.
• The dynamical timescale (\(\tau_{\mathrm{dyn}}\)) sets the limit for structural collapse or oscillations.
• The diffusion timescale (\(\tau_{\mathrm{diff}}\)) explains the long delay between core energy generation and surface radiation.
• For the Sun, these timescales are hierarchically ordered:

$$ \tau_{\mathrm{dyn}} \ll \tau_{\mathrm{diff}} \ll \tau_{\mathrm{th}} \ll \tau_{\mathrm{nuc}} $$ showing why stars appear so stable over billions of years.

1. Derive the ratio of \(\tau_{\mathrm{dyn}}\) to \(\tau_{\mathrm{th}}\) for the Sun and interpret its meaning.
2. What physical processes dominate in objects where \(\tau_{\mathrm{dyn}} \approx \tau_{\mathrm{th}}\)?
3. How would the diffusion timescale change for a red giant compared to a main-sequence star?
4. Why is the Kelvin–Helmholtz timescale much shorter than the nuclear timescale?
5. Explain how the virial theorem connects gravitational and thermal energies in estimating \(\tau_{\mathrm{th}}\).