Where the Classical Account Fails
Four confident predictions. Four disagreements with observation.
Explore atomic stability, discrete spectra, heat capacities and blackbody radiation—and the scales that make quantum theory necessary.
Classical atom · unstable
New rules for a deeper reality.
Four predictions. Four failures.
Classical models make testable claims. The evidence tells us where those models stop working.
1Atomic stability
A classical orbiting charge radiates, loses energy and spirals inward. Starting near 53 pm gives a formal lifetime of about 16 ps.
2Atomic spectra
A shrinking Coulomb orbit sweeps continuously through frequencies. Hydrogen instead has reproducible, discrete lines.
3Heat capacities
Equipartition predicts 3R for a harmonic solid and 7R/2 for a fully active diatomic molecule. Vibrational modes can contribute less at low temperature.
hν ≳ kBT
Level spacing matters. “Low temperature” is relative to the mode.
Compare the predictions →4Blackbody radiation
Rayleigh–Jeans theory assigns kBT to every cavity mode. Its total energy diverges; observed thermal spectra peak and fall.
uν ∝ ν²
The failure occurs at high frequency, not at the long-wave limit.
Explore the spectrum →The shared scale—not a universal switch
Planck’s constant connects these failures, but each approximation has its own conditions.
Energy spacing
Thermal excitation of a mode depends on its level spacing relative to thermal energy. If hν ≪ kBT, many levels are accessible.
hν / kBT
S / h
The classical limit
High temperature relative to a mode’s spacing and large characteristic action often support classical approximations. They do not make classical physics exactly true for every observable.
Why size alone is not enough
A macroscopic solid can have quantum heat capacity. Conversely, highly excited atomic states can exhibit semiclassical behaviour for selected observables. Coherence, quantum statistics, tunnelling and the preparation of the state matter too. Ask which model, which observable and which dimensionless ratio is being tested.
Classical hydrogen collapse
An accelerating charge radiates. The planetary model therefore cannot explain stable ground-state hydrogen.
Fixed proton, slowly shrinking circular orbit, nonrelativistic radiation. Real ground-state hydrogen does not follow this path.
Classical prediction
P = e²a²/(6πε₀c³)
r³ = r₀³ − 4rₑ²ct
τ = r₀³/(4rₑ²c)
The formal collapse time scales with the cube of the initial radius.
The model fails near its endpoint; a zero radius is an extrapolation, not a physical event.
As r decreases, f ∝ r⁻³ᐟ² increases and λ decreases: a continuous chirp, not discrete emission lines.
Derivation & limits of the model
For a circular Coulomb orbit, a = e²/(4πε₀mₑr²) and E = −e²/(8πε₀r). Substitute the acceleration into Larmor’s formula, then use dE/dt = −P to get dr/dt = −4rₑ²c/(3r²). Integration gives r³ = r₀³ − 4rₑ²ct.
At r₀ = 52.9 pm, the initial orbital frequency is about 6.58 × 10¹⁵ Hz and the radiation wavelength is about 45.6 nm—not the 121.6 nm quantum Lyman-α transition. As r shrinks, frequency rises. The slow, nonrelativistic approximation eventually breaks down; the formal collapse time is not a reliable model of the final impact.
Quantum stationary states are not miniature radiating planetary orbits. Real excited states can emit photons during transitions, while the ground state has no lower atomic state to decay into.
Hydrogen line spectrum
Sharp, repeatable wavelengths require more than a continuously shrinking orbit.
Rydberg’s empirical pattern
1/λ = RH(1/n₁² − 1/n₂²)
n₂ > n₁ ≥ 1. The visible Balmer lines shown have n₁ = 2 and n₂ = 3, 4, 5, 6. Wavelengths are in vacuum; air wavelengths differ slightly.
The formula organises the observations. A quantum energy-level model later explains the differences: hν = Eupper − Elower.
What the line spacing tells us
The Balmer frequencies are not equally spaced successive harmonics. As n₂ increases, the lines crowd towards the series limit at about 364.7 nm. A classical periodic orbit has a fundamental frequency and harmonics; its continuously changing radius does not select these fixed term differences.
Equipartition versus thermal excitation
Classical energy sharing misses the effect of discrete vibrational energy levels.
Einstein solid · θ = 1320 K
Ideal diatomic gas · vibration
Count the classical terms
One quadratic term contributes ½kBT. Three harmonic directions in a solid contain six terms, giving Cv = 3R. An intact diatomic molecule with translation, rotation and vibration fully active gives Cv = 7R/2.
Freeze-out is gradual
θ = hν/kB; x = θ/T
Cv,osc / kB = x² exp(x)/(exp(x) − 1)²
For x ≫ 1, excitation is suppressed. At x = 1, the heat capacity is about 92% of the classical limit, while the thermal energy is only about 58% of kBT.
What the simplified curves leave out
The solid curve is an Einstein model with one chosen characteristic temperature, 1320 K. It illustrates suppression and the high-temperature 3R limit, but is not labelled as measured diamond data. The low-temperature behaviour of a three-dimensional insulating crystal is better captured by acoustic modes in the Debye model.
The gas curves include a 5R/2 translation/rotation baseline and one vibrational mode. They omit rotational freeze-out, interactions and dissociation. They must not be extrapolated to describe a real gas down to absolute zero or arbitrarily high temperature.
The ultraviolet catastrophe
The Rayleigh–Jeans law is the long-wavelength approximation to Planck’s law.
Wavelength-space peak: λmax = 2897.77 μm K / T = 0.500 μm. A frequency-space peak is a different maximum; do not convert one by simply taking c/λmax.
Frequency-space energy density
uν,RJ = 8πν²kBT/c³
uν,P = (8πhν³/c³)/(exp(hν/kBT) − 1)
Both are energy per volume per frequency: J m⁻³ Hz⁻¹. Rayleigh–Jeans diverges when integrated over all frequencies.
Keep the spectral variable straight
The graph uses wavelength-space radiance Bλ, not uν. Bλ,P = (2hc²/λ⁵)/(exp(hc/λkBT) − 1); Bλ,RJ = 2ckBT/λ⁴. Changing the spectral variable requires a Jacobian.
Per-frequency and per-wavelength graphs have different shapes and different peak positions.
Interactive classical-to-quantum boundary lab
Change the scales. Compare two explicit thermal predictions, then inspect the separate action ratio.
S is a separate illustrative input, not calculated from the radiation.
Thermal mode test
x = hν/(kBT) = 1.6
Planck / Rayleigh–Jeans energy per mode
40.5%Oscillator heat capacity / classical value
81.1%Energy excludes zero-point energy. Heat capacity is its temperature derivative, so these two percentages differ.
Action-scale check
S/h = 15.1
A large characteristic action can support a semiclassical approximation. It is not a universal certificate that all quantum effects vanish.
Coherence, tunnelling, degeneracy, the observable and the state also matter. The lab’s two ratios test different assumptions.
Catch the common trap
“Quantum” does not simply mean “small”.
Macroscopic does not guarantee classical
A diamond on a laboratory bench has many atoms, yet its high-frequency lattice vibrations can contribute much less than the classical heat-capacity prediction at room temperature.
Classical is a controlled approximation
Highly excited atomic states can have transitions approaching classical orbital frequencies. That correspondence does not restore a classical account of the stable ground state.
Practice & worked examples
Three quantitative checks, from line patterns to the limits of a classical atom.
Easy · Hydrogen spectral lines
The Balmer lines have vacuum wavelengths 656.5, 486.3, 434.2 and 410.3 nm. Are these successive harmonics of one orbital frequency?
View solution →
- Use ν = c/λ: approximately 4.567, 6.165, 6.905 and 7.307 × 10¹⁴ Hz.
- The successive gaps are 1.598, 0.740 and 0.402 × 10¹⁴ Hz. They are not equally spaced, unlike successive harmonics. This does not test every conceivable selection of harmonics; it rejects the simple successive-harmonic explanation.
- Use the first line with n₂ = 3 and n₁ = 2: R_H c = (36/5)ν₃ ≈ 3.288 × 10¹⁵ Hz.
- For n₂ = 6, ν₆ = R_H c(1/4 − 1/36) ≈ 7.307 × 10¹⁴ Hz, giving 410.3 nm. The series converges to about 364.7 nm, not to indefinitely increasing harmonic frequency.
A difference of atomic energy terms explains the pattern; successive orbital harmonics do not.
Medium · Diatomic vibration freezes out
At 298 K, compare ideal diatomic oscillators with vibrational wavenumbers 2359 cm⁻¹ and 560 cm⁻¹. Assume translation and rotation are active.
View solution →
- Classical equipartition gives 3 translational + 2 rotational + 2 vibrational quadratic terms: Cv = 7R/2 = 29.10 J mol⁻¹ K⁻¹.
- Convert cm⁻¹ to m⁻¹. With hc/kB = 0.0143878 m K, the vibrational temperatures are θv ≈ 3394 K and 806 K.
- The vibrational contribution is Cv,vib/R = x² exp(x)/(exp(x) − 1)², x = θv/T. This is a heat-capacity ratio, not the thermal-energy ratio.
- At 298 K the ratios are approximately 0.0015 and 0.56. Adding the 5R/2 translation/rotation baseline gives Cv ≈ 20.8 and 25.5 J mol⁻¹ K⁻¹. These are harmonic-model estimates, not tabulated measurements.
- For T ≫ θv the vibration contributes R and Cv approaches 7R/2, provided the molecule remains intact and the ideal-gas assumptions still apply.
Higher-frequency vibrations are harder to thermally excite; mass or object size alone is not the criterion.
Hard · Can a larger classical atom survive?
Using τ = r₀³/(4rₑ²c), find the starting radius giving a formal classical lifetime of 1.0 s. Compare its binding energy with ground-state hydrogen.
View solution →
- r₀ = (4rₑ²cτ)⅓ = 2.12 × 10⁻⁷ m, or about 0.21 μm. This is roughly 4000 times the Bohr radius.
- The initial orbital frequency is f = (c/2π)√(rₑ/r₀³) ≈ 2.59 × 10¹⁰ Hz.
- The binding energy is |E| = e²/(8πε₀r₀) ≈ 5.44 × 10⁻²² J = 0.0034 eV. That is far below the 13.6 eV ground-state binding of hydrogen.
- At 300 K, kBT ≈ 0.0259 eV. This comparison indicates susceptibility to thermal disruption; it is not, by itself, a calculation of an ionisation rate.
- Increasing the radius delays the formal collapse but loses the observed binding energy. The nonrelativistic circular-orbit extrapolation still fails near collapse, and it does not describe stationary quantum states.
A larger orbit does not rescue a classical account of stable, tightly bound ground-state hydrogen.
Quick check
Five questions · progress is for this visit only
Summary & next steps
Keep the evidence, the model and the approximation separate.
- The radiating planetary atom predicts instability and a continuous chirp, not stable atoms with discrete lines.
- Equipartition and Rayleigh–Jeans fail when relevant quantum energy spacings cannot be treated as negligible.
- Planck’s constant introduces an energy and action scale. Useful classical limits are conditional, not an exact universal boundary.
- Check units, the spectral variable, the model’s assumptions and its limiting cases.
Sources, modelling choices & further reading
All curves and moving diagrams on this page are calculations or schematics. They are not presented as newly measured experimental data. SI constants use NIST/CODATA values; the Rydberg calculation uses the hydrogen reduced-mass constant and neglects fine structure.
- NIST · Fundamental physical constants
- NIST · Hydrogen line measurements (visible values in air)
- University of Texas · Planck radiation and the ultraviolet catastrophe
- University of Texas · Specific heats of solids
- University of Texas · Larmor radiation
Reviewed for this lesson, 28 September 2026. This is introductory university support, not certification of alignment to every school syllabus. Prerequisites: circular motion, energy conservation, oscillations, basic thermal physics and exponentials.