Old quantum theory

• 1. Introduction

  ›

  Old quantum theory stands at one of the strangest thresholds in the history of science: a body of knowledge that worked, often brilliantly, yet was never actually correct. From 1900 to 1925, physicists assembled a patchwork of rules, corrections, and inspired guesses that let them calculate the behavior of atoms with surprising precision. The rules did not form a coherent theory. They were, in the blunt assessment of those who came after, a set of heuristic corrections grafted onto classical mechanics.

  What launched this era was a single act of intellectual desperation. In 1900, Max Planck was trying to explain how a perfectly absorbing object, a so-called black body, emits and absorbs light. His solution introduced something he called the quantum of action. It fit the data. It made no physical sense he could explain. And it set off twenty-five years of improvisation that would eventually, under the hands of Werner Heisenberg and Erwin Schrödinger, become modern quantum mechanics.

  The questions this story raises are not small ones. How do you build a productive scientific framework on rules you know are incomplete? Why did some physicists succeed spectacularly while others hit walls? And what does it mean when a theory that cannot explain half of what it touches still manages to reveal the shape of the periodic table?

• 2. Planck's Spark

  ›

  Max Planck's 1900 paper on the emission and absorption of light introduced what he called the quantum of action, a minimum packet of energy that nature seemed to trade in. The idea was radical enough that it sat quietly for several years before Albert Einstein picked it up in 1907.

  Einstein applied quantum principles not to light but to the motion of atoms in solids, publishing a model that explained why specific heats of materials behave so strangely at low temperatures. His paper on this came to the attention of Walther Nernst, and it brought Einstein fully into the circle of physicists wrestling with these problems. Peter Debye then extended the approach in 1912 with a more detailed treatment of quantized oscillators at different frequencies, giving the field its first quantitatively reliable model of solid specific heats.

  The practical stakes were real. James Clerk Maxwell had noticed in the 19th century that classical mechanics predicted a constant specific heat for solids regardless of temperature. Experiments showed that specific heats dropped sharply at low temperatures and reached zero at absolute zero, which is the third law of thermodynamics. Classical physics had no answer for this. Einstein's 1906 proposal that atomic motion is quantized resolved the contradiction directly, making it the first application of quantum ideas to mechanical systems.

  The harmonic oscillator, the simplest system in the old quantum theory, illustrated the mechanism clearly. At high temperatures, the average energy of an oscillator matches the classical prediction: kT, where k is the Boltzmann constant. At low temperatures, when kT falls below the energy of a single quantum, the oscillator cannot absorb even one quantum and stays in its ground state, storing almost nothing. The specific heat collapses. This behavior, inexplicable in classical terms, followed naturally from Planck's quantization rule.

• 3. Bohr, Sommerfeld, and the Atom

  ›

  In 1910, Arthur Erich Haas published a paper developing J. J. Thomson's atomic model that included quantization of electronic orbitals, anticipating Niels Bohr's far more celebrated model by three years. John William Nicholson went further, creating an atomic model that quantized angular momentum. Bohr cited him in his own 1913 paper.

  Bohr's 1913 model of the hydrogen atom explained the line spectrum by restricting electrons to specific orbits and applying what he called the correspondence principle: quantum results, in the limit of large quantum numbers, must reproduce classical predictions. The model was a success, but it was limited to circular orbits and to hydrogen.

  Arnold Sommerfeld extended Bohr's framework in the years that followed, generalizing the quantum rule to any integrable system and allowing electron orbits to be ellipses rather than circles. Sommerfeld's crucial contribution was quantizing the z-component of angular momentum, a phenomenon called space quantization, or in German, Richtungsquantelung. The result was the Bohr-Sommerfeld model, which introduced quantum degeneracy and came considerably closer to the modern quantum mechanical picture than Bohr's original version had.

  The model had a philosophical puzzle built into it. The z-axis against which angular momentum was measured was chosen arbitrarily by the physicist. How could nature care about a direction that a human had simply picked? The paradox of space quantization was real and unresolved within the old theory. Modern quantum mechanics dissolved it by showing that states of definite angular momentum in one orientation are quantum superpositions of states in other orientations, so no axis is privileged.

  Sommerfeld's enhancements ultimately provided the foundation for two of the old quantum theory's greatest achievements: Edmund Stoner's determination of the modern form of the periodic table, and the Pauli exclusion principle.

• 4. A Quarter-Century of Mixed Results

  ›

  Throughout the 1910s and into the 1920s, physicists attacked problem after problem with old quantum tools, and the outcomes were uneven. Molecular rotation and vibration spectra were successfully understood. Planck introduced the zero-point energy. Sommerfeld semi-classically quantized the relativistic hydrogen atom.

  Hendrik Kramers explained the Stark effect, the splitting of spectral lines in an electric field. Satyendra Nath Bose and Einstein together developed Bose-Einstein statistics, the rules governing particles now called bosons. Einstein also refined the quantization condition itself in 1917, in a paper that would later prove foundational to a method called the Einstein-Brillouin-Keller approach.

  But the theory's failures were just as instructive. It could not calculate the intensities of spectral lines. It failed on the anomalous Zeeman effect, where electron spin could not be ignored. Most strikingly, it could not handle chaotic systems, those in which trajectories are neither closed nor periodic. A two-electron atom, which is classically chaotic in the same way as the gravitational three-body problem, was beyond its reach.

  The discovery of the electron's spin added its own confusion: half-integer quantum numbers appeared where only integers had been expected, and the old rules had no clean place for them. The framework was straining against its own foundations.

  In 1924, Niels Bohr, Kramers, and John C. Slater proposed what became known as the BKS theory, treating atomic systems as quantum mechanical while keeping the electromagnetic field classical. It was a creative attempt to patch the divide. The Bothe-Geiger coincidence experiment rejected it.

• 5. De Broglie and the Wave Interpretation

  ›

  In 1905, Einstein had noticed something peculiar about the entropy of quantized electromagnetic field oscillators in a box: at short wavelengths, it matched the entropy of a gas of point particles. He concluded that light had attributes of both waves and particles, that an electromagnetic standing wave of frequency v could be thought of as consisting of n photons each carrying energy hv. The name "photon" itself came later, coined by Gilbert N. Lewis in a letter to Nature.

  Einstein could not explain how photons related to the underlying wave. That connection waited until 1924, when Louis de Broglie, then a PhD candidate, proposed that all matter, electrons as well as photons, is described by waves. His key insight was a reinterpretation of the old quantum condition: the requirement that a quantity be an integer multiple of Planck's constant was really a requirement for constructive interference. Matter waves make standing waves only at discrete frequencies, at discrete energies. Quantized orbits were not an arbitrary rule. They were the natural consequence of wave behavior.

  Einstein extended this formulation mathematically, noting that the phase function for de Broglie's waves should be identified with the solution to the Hamilton-Jacobi equation. William Rowan Hamilton himself, working in the 19th century, had believed this equation was a short-wavelength limit of some deeper wave mechanics. Schrödinger found the wave equation that matched the Hamilton-Jacobi equation for the phase, producing what is now called the Schrödinger equation in 1926.

  Heisenberg, working from Kramers' analysis of transition probabilities expressed as Fourier components of orbital motion, had reformulated all of quantum theory in his 1925 Umdeutung paper. He and Max Born and Pascual Jordan built matrix mechanics from those transition rules. The two approaches, wave mechanics and matrix mechanics, appeared to be entirely different until Schrödinger and others proved they predicted the same experimental results. Paul Dirac proved in 1926 that both could be derived from a more general method he called transformation theory.

• 6. The Quantization Condition

  ›

  The technical core of the old quantum theory was a single rule, now called the Wilson-Sommerfeld quantization condition, proposed independently by William Wilson and Arnold Sommerfeld. It stated that for each coordinate of a classical system, the integral of the momentum over one full period of the corresponding motion must equal an integer multiple of Planck's constant h.

  This integral is a quantity called action, measured in units of h. The physical meaning of Planck's constant as a quantum of action was central to why the old quantum theory's practitioners called it what they did.

  For the rule to work, the classical motion had to be separable: there had to exist a set of coordinates in which the motion decomposed into independent periodic components. The different periods did not need to match or be commensurate with each other, but separability was required. This restriction was the source of the theory's failure on chaotic systems, where no such separation exists.

  The conceptual justification for the rule rested on two pillars. One was the correspondence principle: quantum results must approach classical predictions at large quantum numbers. The other was the principle of adiabatic invariance, developed by Hendrik Lorentz and Einstein, which held that the quantities being quantized must be invariants under slow changes in the system's parameters. Given Planck's quantization rule for the harmonic oscillator, either principle was sufficient to identify the correct classical quantity to quantize in a general system, up to an additive constant.

  The additive constant was not a minor detail. The old quantum theory consistently missed the ground-state energy, the zero-point energy that modern quantum mechanics requires. The difference between old and new was exactly half a quantum, a small gap that pointed toward the deeper framework waiting to replace it.

• 7. After 1925

  ›

  The old quantum theory did not simply stop. In the 1950s, Joseph Keller updated the Bohr-Sommerfeld quantization method using Einstein's 1917 interpretation, producing what is now called the Einstein-Brillouin-Keller method. The approach retained practical value even after full quantum mechanics was established.

  In 1971, Martin Gutzwiller addressed the old theory's most persistent failure directly. He recognized that the Einstein-Brillouin-Keller method only works for integrable systems, those whose motion is orderly and separable. For chaotic systems, he derived a semiclassical way to quantize using path integrals, extending the reach of semiclassical methods into territory the old quantum theory could never enter.

  The broader verdict on the old quantum theory is that it is now understood as a semiclassical approximation to modern quantum mechanics. Its limitations are still under investigation. It can describe atoms with more than one electron, including helium, and it handles the ordinary Zeeman effect, though not the anomalous version involving spin. The tools it built, the quantization condition, the correspondence principle, Kramers' transition matrix, all contributed directly to the machinery of the theory that replaced it.

  Kramers' prescription for calculating transition probabilities, extended in collaboration with Heisenberg, was the direct ancestor of matrix mechanics. De Broglie's wave reinterpretation of the quantization condition set Schrödinger on the path to his wave equation. The old quantum theory's most important legacy may be that it failed productively, revealing through each limitation exactly what a complete theory would need to provide.