History of Atomic Theory

The history of atomic theory is commonly taught as an example of the self-correcting nature of science. I wanted to extend the standard list of models to get the full picture of how we came to understand matter at the most fundamental levels.

Early Ideas: Antiquity and Middle Ages

Fundamental Elements (Empedocles, ~450 BC). Ancient Greeks believed all matter was composed of four fundamental elements (earth, fire, air, water) as continuous substances, as well as a fifth element permeating the space outside of Earth (‘quintessence’, the luminiferous aether). There were also similar ideas in ancient China and India regarding fundamental elements.

Atomism (Democritus, Aristotle and Epicurus, ~400 BC). The concept that every substance is made of indivisible parts (‘atomos’). The concept of the aether was retained, passed on by Aristotle and developed further in ~300 BC by Epicurus. In ~50 BC (Ancient Rome), Lucretius brought back atomism. After the fall of the Roman Empire, Western thought on the nature of matter regressed and did not develop in any significant way until the late Renaissance.

Alchemy (Islamic scholars, 1100s). During the Islamic Golden Age, alchemy (a blend of mysticism and proto-chemistry) was developed, with the fundamental ‘principles’ being sulfur, mercury and salt. Alchemy was considered more an art than a science, and was widely practised throughout Asia and spread to Europe in the late Middle Ages under scholasticism. The majority of thought in these times concerned metaphysics (e.g. souls, existence) and not matter. Many literary writers (Dante, Chaucer) considered alchemy to be fraudulent and/or nonsensical.

Corpuscular theory (Newton, Gassendi, Boyle, Descartes and Lemery, 1600s). During the Scientific Revolution, Western natural philosophers tried to rationalise properties of matter. They considered atoms to have different shapes and hook together, which modified Democritus’ atomism and considered light to be made of particles (corpuscles).

Phlogiston theory (Becher, 1600s and Stahl, 1723). A type of matter invoked to explain the mass loss observed in flammability, combustion and corrosion. This was disproven in the 1770s with the discovery of oxygen by Lavoisier.

Development of Atomic Theory

Dalton’s Model (Dalton, 1803). He conceptualised atoms as hard spheres in different forms (elements), that atoms can join together to form compounds, and that chemical reactions involve rearranging the atoms. This was capable of explaining proportionality and stoichiometry in compounds. Dalton and his contemporaries did not suggest that atoms physically took this form, instead using it as a visual aid.

Kinetic Theory of Gases (Avogadro, 1811). Avogadro considered gases as lots of molecules made of small numbers of atoms, building on Bernoulli’s hydrodynamic theory and reconciling it with Gay-Lussac’s law.

Periodic Table (Mendeleev, 1871). By the 1830s, the molecular formulas of many simple molecules had been correctly identified, and in 1871, Mendeleev’s periodic table of elements was published, recognising periodicity and predicting properties of undiscovered elements.

Disproof of the Aether (Michelson and Morley, 1887). The Michelson-Morley experiment disproved the existence of the luminiferous aether, requiring that light can travel in a vacuum. Although Maxwell’s theory of electromagnetism assumed the aether’s presence, it was not required for the theory to be self-consistent.

Plum Pudding Model (Thompson, 1904). Proposed after his discovery of the electron with cathode ray tube experiments in 1897, with atoms as a large positively charged body with small negative electrons distributed inside. The model was unable to predict most physical properties (e.g. mass, volume, charge, magnetic moment), and so some physicists still doubted that atoms existed (one of the last being Mach, 1910), though chemists had come to accept their existence in some form since the mid 1800s.

Saturnian Model (Nagaoka, 1904). Nagaoka opposed the plum pudding model on the grounds that opposite charges in direct contact would be too unstable. He instead proposed that a large number of electrons orbited a central positive charge in a single near-continuous rotating ring, like the rings of the planet Saturn.

Nuclear Model (Rutherford, 1911). Proposed after his discovery of the proton with his alpha particle scattering experiment, with a small positive nucleus and electrons around.

Planetary Model (Bohr, 1913). Modified the nuclear model to add that the electrons orbited the nucleus in discrete shells. The Bohr model was able to approximately predict the spectral lines of hydrogen, but lost accuracy for other elements.

Octet Rule (Lewis, 1916). Provided a rationale for the Lewis structure of covalent molecules, in which every atom tended to have a total of 8 valence electrons.

Bohr-Sommerfeld Model (Sommerfield, 1916). Modified Bohr’s model to use elliptical electron orbits instead of circles, with one focus at the nucleus. This more accurately matched the fine structure spectral lines of the hydrogen atom and made use of ‘old quantum theory’, when the need for quantisation had become clear but no coherent quantum theory had yet been developed.

Intermolecular Forces (Debye, 1912, Keesom, 1921, London, 1930). Described electrostatic forces between ions, permanent dipoles and induced dipoles. London used perturbation theory on the quantum model to derive the London dispersion force between nonpolar atoms.

The Modern Theory of Atoms and Matter

Quantum Mechanical Model (de Broglie, Heisenberg, Schrödinger and Born, 1926). The development of quantum mechanics gave rise to the quantum model of the atom, with electrons as probability density clouds occupying orbitals around the nucleus with discrete energy levels.

Relativistic Quantum Model (Dirac, 1928). Formulated the first quantum theory that incorporated Einstein’s special relativity, describing electrons as excitations of a spinor field (fundamental particles with intrinsic spin) and photons as excitations of an electromagnetic field. This also predicted antimatter, with the positron being discovered in a cloud chamber by Anderson in 1932.

Band Theory (Bloch, 1928). Developed the band theory of network solids by applying quantum mechanics to periodic lattices, laying the ground for solid state and semiconductor physics.

Crystal Field Theory (CFT) (Bethe and van Vleck, 1929). Developed to describe ionic bonding in metal-inorganic complexes using the concepts of atomic orbital theory.

Hybridisation Theory (Pauling, 1931). Described covalent bonding in terms of atomic orbitals.

Molecular Orbital (MO) Theory (Hund, Mulliken and Hückel, 1932). Described covalent bonds as in-phase and antiphase overlaps between atomic or hybridised orbitals. Mulliken and Hund invented LCAO MO theory while Hückel applied MO theory to conjugated $π$ systems.

Neutrons and Isotopes (Chadwick, 1932). Neutrons were discovered in the nucleus by observing radiation produced when beryllium was bombarded with alpha particles.

Resonance Bonding (Pauling, 1933). Used the idea of quantum superposition to describe resonance in molecules. Soviet scientists rejected this theory until the late 1950s due to their Marxist-Leninist philosophy, seeing resonance and quantum phenomena as anti-materialist and anti-deterministic, criticising it as metaphysical.

Valence Shell Electron Pair Repulsion (VSEPR) Theory (Sidgwick and Powell, 1940, and Gillespie and Nyholm, 1957). Proposed that molecular geometries can be predicted by considering valence electron domains, incorporating hybridisation theory, formalised in 1957.

Quantum Electrodynamics (QED) (Feynman, Schwinger and Tomonaga, 1940s). Building on the work of Dirac, the first quantum field theory was developed to be compatible with special relativity, capable of describing all electromagnetic phenomena from first principles. QED is fundamentally required to explain the Van der Waals’ force between neutral molecules, as well as subtleties such as the Lamb shift of the hydrogen atom.

Ligand Field Theory (LFT) (Griffith and Orgel, 1957). Combined MO theory and CFT, accurately describing the partial covalent bonding in organometallic compounds. LFT made extensive use of the modern group-theoretic approaches to molecular symmetry and spectroscopy.

Molecular Dynamics (MD) (Fermi, Pasta, Ulam, Tsingou…, 1950s-60s). Computational methods applied to statistical mechanics allowed for classical simulations of ensembles of molecules. Early experiments led to fortuitous discoveries in applied mathematics (soliton theory). Used to study the mechanics of macromolecules and solid-state materials.

Density Functional Theory (DFT) (Kohn, Hohenberg and Sham, 1965). A computational method to solve for the numerical energy levels of molecular orbitals in molecules and solid state matter based on quantum mechanics. It is also used to compute the band structure of semiconductors.

Quarks and Quantum Chromodynamics (QCD) (Gell-Mann, 1964 and Fritzsch, 1972). Deep inelastic scattering experiments led to the proposal that protons and neutrons are composed of quarks. The quantum field theory of quarks and gluons was formed in 1972, which could model nucleons and radioactivity from first principles with the strong and weak nuclear forces. In 1974, Wilson invented ‘lattice QCD’, a computationally feasible method of numerically solving QCD, and this was used to calculate the nuclear magnetic moment of the proton (causes the hyperfine structure of the hydrogen atom) with extreme precision on a CERN supercomputer in 1982.

Scanning Tunneling Microscopy (STM) (Binnig and Rohrer, 1980s): Allowed direct observations of individuals atoms and molecules for the first time, using scanning tunneling microscopy (STM) in 1981 and then atomic force microscopy (AFM) in 1985 to probe their electrostatic surfaces.