The Ultraviolet Catastrophe

The ultraviolet catastrophe is the high-frequency divergence produced when classical electromagnetic mode counting is combined with classical equipartition. It is one of the cleanest ways to see why blackbody radiation forced a change in microscopic state counting.

The phrase labels the divergence retrospectively: the completed Rayleigh–Jeans coefficient dates to 1905, after Planck’s formula. Historically, blackbody radiation developed through Kirchhoff’s universality, Wien’s scaling laws, measurements over broader frequency ranges, Rayleigh and Jeans’s classical limit, and Planck’s radiation law. The “catastrophe” is the sharp mathematical diagnosis of the classical high-frequency failure.

Classical Mode Density

Consider electromagnetic radiation in a large cavity. Classical Maxwell theory allows standing-wave modes. Counting the allowed wave vectors and including two transverse polarizations gives the number of modes per unit volume between frequency $\nu$ and $\nu+d\nu$:

$$g(\nu)\,d\nu = \frac{8\pi\nu^2}{c^3}\,d\nu.$$

For example, periodic boundary conditions give one allowed wave vector per volume $(2\pi)^3/V$ in wave-vector space. A spherical shell has volume $4\pi k^2dk$; multiplying by two polarizations gives $g(k)dk=k^2dk/\pi^2$ per physical volume. Substitution of $k=2\pi\nu/c$ gives the formula above. Surface corrections become negligible for a large cavity.

The factor $\nu^2$ is the important part. In three dimensions, the number of modes grows like the surface area of a sphere in wave-vector space. Higher frequencies therefore have more available modes.

This mode count is not the part that fails. The same density appears in the successful Planck law. The failure comes from how classical statistical mechanics assigns energy to those modes.

Equipartition Intuition

Each classical electromagnetic normal mode behaves like a harmonic oscillator. For a classical harmonic oscillator at temperature $T$, equipartition assigns an average energy

$$\bar E_{\rm cl}=k_B T.$$

Multiplying the mode density by this average energy gives the Rayleigh–Jeans spectral energy density:

$$u_\nu^{\rm RJ}(\nu,T)\,d\nu = \frac{8\pi\nu^2}{c^3}k_B T\,d\nu.$$

At low frequency this is right. The problem is that classical equipartition assigns the same average energy to every mode, no matter how high the frequency.

High-Frequency Divergence

The total radiation energy density predicted by Rayleigh–Jeans is obtained by integrating over all frequencies:

$$U_{\rm RJ}(T) = \int_0^\infty u_\nu^{\rm RJ}(\nu,T)\,d\nu.$$

With a finite high-frequency cutoff $\nu_{\max}$, the integral is

$$U_{\rm RJ}(T;\nu_{\max}) = \frac{8\pi k_B T}{c^3} \int_0^{\nu_{\max}}\nu^2\,d\nu = \frac{8\pi k_B T}{3c^3}\nu_{\max}^3.$$

As $\nu_{\max}\to\infty$, this diverges:

$$U_{\rm RJ}(T;\nu_{\max})\to\infty.$$

That is the ultraviolet catastrophe. The word “ultraviolet” refers to the high-frequency, short-wavelength side of the spectrum. The word “catastrophe” refers to the fact that the theory predicts infinite energy density in a finite-temperature cavity.

The curves are calculated from the stated formulas. In the lower panel, $x=h\nu/(k_BT)$: Rayleigh–Jeans ($x^2$) lies above Planck ($x^3/(e^x-1)$) and quickly leaves the vertical plotting range. Wien ($x^3e^{-x}$) approaches Planck from below.

Why Experiment Disagrees

Real blackbody spectra do not grow without bound. They rise at low frequency, reach a temperature-dependent peak, and then fall rapidly at high frequency. The total thermal energy density is finite and scales as $T^4$, as the integral in Planck’s radiation law shows.

The disagreement is structural:

Feature	Rayleigh–Jeans prediction	Observed spectrum
Low frequency	$u_\nu\propto\nu^2T$	Correct limiting behavior
Peak	No finite peak	Has a finite peak
High frequency	Grows without bound	Falls rapidly
Total energy	Infinite	Finite

This is why the blackbody problem was not a small correction to a nearly right curve. The low-frequency success and high-frequency failure had to be reconciled in one formula.

Why Quantum Suppression Solves It

Planck’s law replaces the classical average mode energy $k_B T$ with

$$\bar E_{\rm q}(\nu,T) = \frac{h\nu}{\exp(h\nu/k_B T)-1}.$$

At low frequency, $h\nu\ll k_B T$, this becomes approximately $k_B T$, so Rayleigh–Jeans is recovered. At high frequency, $h\nu\gg k_B T$, it behaves like

$$\bar E_{\rm q}(\nu,T) \approx h\nu \exp\!\left(-\frac{h\nu}{k_B T}\right).$$

Thus the high-frequency spectrum behaves like a power of $\nu$ multiplied by an exponential suppression:

$$u_\nu(\nu,T) \sim \nu^3 \exp\!\left(-\frac{h\nu}{k_B T}\right) \qquad (\nu\to\infty).$$

The exponential wins over the polynomial growth in the mode density, so the total energy remains finite.

The physical interpretation is simple but profound: high-frequency modes are not easily thermally excited because their energy spacing is large compared with $k_B T$.

Historical Caution

The ultraviolet catastrophe is a useful derivation for students, but it can make the history sound too linear. The phrase became a compact retrospective label for the classical failure. Around 1900, the problem was embedded in a broader web of empirical spectra, thermodynamic arguments, and competing formulas.

Use the term to name the divergence. Do not use it to imply that the entire quantum revolution followed from one dramatic plot of a wrong curve. Blackbody radiation was one route into quantization, alongside the photoelectric effect, atomic spectra, heat capacities, matter waves, and spin evidence.

Common Mistakes

• Blaming Maxwell’s equations alone; the divergence comes from Maxwell mode density plus classical equipartition.
• Forgetting that Rayleigh–Jeans is correct at low frequency.
• Treating the ultraviolet catastrophe as the same thing as all of blackbody radiation history.
• Assuming a finite experimental cutoff would solve the theoretical problem; the theory itself has no natural high-frequency cutoff.
• Saying Planck simply “cut off” the modes; Planck changed the thermal occupation of high-frequency modes.
• Thinking the exponential suppression is arbitrary rather than tied to the energy scale $h\nu$.

Exercises

1. If a cutoff model agrees with the measured energy density at temperature $T_0$, can the same fixed cutoff reproduce the $T^4$ law at other temperatures?

Solution

Start from

$$u_\nu^{\rm RJ}(\nu,T) = \frac{8\pi\nu^2}{c^3}k_B T.$$

Integrating up to $\nu_{\max}$ gives

$$U_{\rm RJ}(T;\nu_{\max}) = \frac{8\pi k_B T}{c^3} \int_0^{\nu_{\max}}\nu^2\,d\nu = \frac{8\pi k_B T}{3c^3}\nu_{\max}^3.$$

For a fixed cutoff this scales as $T$, so doubling temperature only doubles the energy. The measured blackbody energy grows by $2^4=16$. A fixed cutoff cannot repair the temperature dependence; choosing a cutoff proportional to $T$ would add a new unexplained assumption.

2. Why does Planck’s high-frequency form remain integrable?

Solution

At high frequency, Planck’s spectrum behaves like

$$u_\nu(\nu,T) \sim \nu^3 e^{-h\nu/k_B T}.$$

An exponential decay dominates any finite power of $\nu$, so the integral over high frequencies converges. This is the mathematical reason quantum suppression removes the ultraviolet divergence.

3. Explain why the ultraviolet catastrophe is not simply “Maxwell was wrong.”

Solution

The electromagnetic mode density from Maxwell theory is still used in Planck’s law. The failure comes from combining that mode density with classical equipartition, which assigns average energy $k_B T$ to every mode. Quantum theory changes the thermal occupation of high-frequency modes while preserving the classical mode density in the appropriate cavity setting.

References

• D. J. Griffiths, Introduction to Electrodynamics, 4th ed., Cambridge University Press, 2017.
• J. H. Jeans, “On the Partition of Energy between Matter and Æther,” Philosophical Magazine 10, 91–98 (1905), DOI: 10.1080/14786440509463348.
• T. S. Kuhn, Black-Body Theory and the Quantum Discontinuity, 1894-1912, University of Chicago Press, 1987 (with a new Afterword).
• Lord Rayleigh, “Remarks upon the Law of Complete Radiation,” Philosophical Magazine 49, 539-540 (1900).
• R. K. Pathria and P. D. Beale, Statistical Mechanics, 3rd ed., Academic Press, 2011.
• M. Planck, “Ueber das Gesetz der Energieverteilung im Normalspectrum,” Annalen der Physik 309, 553-563 (1901), DOI: 10.1002/andp.19013090310.