A theory of burning that lasted a century,
broken by a man with a very good balance.

The path

1. Act 1

   What burning is

   Everything that burns was thought to lose something. It gains weight instead.

   1667 – 1785
   1. A theory that workedPhlogiston explained burning, rusting, breathing and smelting with one idea. It is worth understanding why it was believed before enjoying why it fell.Becher and Stahl on calcination, 1667 – 1703; Jean Rey weighs a calx, 1630
   2. Lavoisier weighs everythingSeal the vessel, weigh it, burn the metal inside, weigh it again. Nothing changes — until you open it.Lavoisier's sealed retort, Paris, 1774
   3. One fifth of the airThe gas that does the burning, isolated by two men who did not believe what they had made.Priestley's burning lens on mercuric oxide, 1774; Lavoisier's twelve-day mercury run, 1777
   4. Water is not an elementTwo gases burn together and make it; hot iron takes it apart again. One of the four classical elements does not survive the decade.Cavendish 1784; Lavoisier & Meusnier decompose water over hot iron, 1785
2. Act 2

   Fixed ratios

   Compounds combine in small whole-number proportions. What could make that true?

   1794 – 1811
   1. Always the same recipeCopper carbonate from a mine and copper carbonate from a bench have identical composition. Eight years of argument establish it.Proust against Berthollet, Madrid, 1794 – 1806
   2. The ratio of the ratiosTwo oxides of carbon, and the second holds exactly twice the oxygen of the first. Not approximately twice.Dalton's carbon oxides and marsh gas, 1803 – 1808
   3. Dalton draws the conclusionWhole-number ratios are what you get if matter comes in indivisible pieces that combine one-to-one, one-to-two, and no other way.A New System of Chemical Philosophy, 1808
   4. The volumes are simple tooTwo volumes of hydrogen and one of oxygen give two volumes of steam. Dalton refused to believe it, and he was the one who was wrong.Gay-Lussac's combining volumes, 1808
   5. Avogadro's repair, ignored for fifty yearsEqual volumes hold equal numbers — and hydrogen gas is a pair of atoms, not one. Both halves were needed, and neither was accepted.Avogadro's reading of Gay-Lussac, 1811
3. Act 3

   Counting the invisible

   How do you weigh a thing you cannot see, and how do you know it is there at all?

   1833 – 1909
   1. Electricity comes in portionsThe same quantity of charge always deposits the same quantity of matter. A unit of charge is hiding in the result, fifty years before anyone finds it.Faraday's laws of electrolysis, Royal Institution, 1833 – 1834
   2. The congress that fixed the numbersBy 1860 chemists could not agree on the formula for water. A pamphlet handed out at the door settled the atomic weights in an afternoon.Cannizzaro's vapour densities, Karlsruhe, 1860
   3. Proof that atoms are realPollen grains jitter under a microscope. Counting how they settle gives the number of molecules in a gram, and three unrelated methods agree.Perrin measures Avogadro’s number, Paris, 1908 – 1909
   4. From grams to countsWhat the mole actually buys you, and why chemistry is the science of doing arithmetic on things you can never see one of.The 2019 redefinition: Avogadro’s number is now exact by fiat
4. Act 4

   The table

   Sort the elements sensibly and gaps appear. Are the gaps real?

   1817 – 1914
   1. Pattern-hunting that half workedFifty years of nearly-right arrangements, including one its author was laughed at for.Döbereiner's triads 1817; Newlands' octaves 1865
   2. Mendeleev leaves holes and fills themHe predicted the density, melting point and oxide formula of an element nobody had seen. Then two of them were found.Predicted 1869; gallium found 1875, germanium 1886
   3. An entire column nobody had noticedNitrogen from air weighed one part in two hundred more than nitrogen from chemicals. That discrepancy was argon.Rayleigh & Ramsay chase a density anomaly, 1892 – 1898
   4. The table is ordered by charge, not weightX-ray frequencies give every element an integer. The integers fix the two places where the table had been wrong.Moseley's X-ray spectra, Manchester and Oxford, 1913 – 1914
5. Act 5

   Why the table has that shape

   Periodicity is a fact about measurement. What is it a fact about the atom?

   1897 – 1930
   1. Almost all of it is emptyAlpha particles fired at gold foil mostly sail through, and one in eight thousand comes straight back.Geiger & Marsden, Manchester, 1909 – 1911
   2. You can see the shells in the dataStrip electrons off an atom one at a time and the energy needed jumps, hugely, at exactly the places the table says it should.Successive ionisation energies; photoelectron spectroscopy, 1907 onwards
   3. What an orbital is, and is notNot a track. A standing wave with a shape, and the shapes are where the blocks of the table come from.Solutions of the hydrogen atom, 1926; imaged directly 2013
   4. Size, grip and reactivityAtomic radius, ionisation energy and electronegativity all trend the same way, and all for the same reason.Measured radii from X-ray crystallography; Pauling’s 1932 electronegativity scale
6. Act 6

   What a bond is

   Atoms stick together. Nothing in classical physics says they should.

   1916 – 1939
   1. The shared pairWhy methane is CH₄ and never CH₅ — a rule that worked for forty years before anyone could say why it was true.Valence saturation; Lewis’s 1916 paper on the cubical atom
   2. There is no molecule of saltX-rays show sodium chloride is a lattice with no pairs in it. Two kinds of bond, told apart by measurement rather than assertion.Bragg’s structure of NaCl, 1913; Born–Haber cycles, 1919
   3. The first bond ever calculatedTwo hydrogen atoms, one equation, and a binding energy that came out right. Chemistry becomes a branch of physics, in principle.Heitler & London compute H₂, Zurich, 1927
   4. Why molecules have shapesWater is bent at 104.5 degrees and methane is a tetrahedron. Both angles were measured before either was explained.Bond angles from infrared and microwave spectroscopy; dipole moments
   5. The ring that is not alternatingAll six carbon–carbon bonds are the same length, and the molecule is 150 kJ/mol more stable than it has any right to be.Hydrogenation enthalpies, 1930s; Lonsdale’s X-ray structure of hexamethylbenzene, 1929
7. Act 7

   Why reactions go

   Some reactions run downhill, some need pushing, and some get cold as they happen.

   1840 – 1935
   1. Adding up the heatThe heat released by a reaction does not depend on the route taken, which lets you measure changes you cannot perform.Hess’s law of constant heat summation, 1840
   2. The reaction that gets coldDissolve ammonium nitrate and the beaker chills your hand — and it happens anyway. Heat cannot be what decides.Endothermic spontaneity; Gibbs’ free energy, 1876
   3. Reactions that stop halfwayForward and back at the same rate, and a quotient that always lands on the same number however you get there.Guldberg & Waage’s mass action, 1864; Le Chatelier, 1884
   4. Bread from airEverything in Act 7 pushed against everything else, and two men found the one corner of the space where it works. It now feeds about half the world.Haber’s bench synthesis 1909; Bosch scales it, 1913
   5. The rate law is not the equationHow fast a reaction goes does not follow its balanced equation — and that mismatch is how mechanisms are discovered.Hughes & Ingold separate SN1 from SN2 by kinetics, 1933 – 1935
   6. Lowering the hillA catalyst changes the road, not the destination. What that means, and why enzymes are so much better at it than platinum.Arrhenius, 1889; Sabatier’s hydrogenations, 1897; Michaelis & Menten, 1913
8. Act 8

   Seeing molecules

   For a century structure was inferred. Then it became something you could look at.

   1912 – now
   1. Diffraction gives you coordinatesSpots on a plate, an equation with one unknown, and the first structure of anything.Laue 1912; W. L. Bragg solves NaCl and diamond, 1913 – 1914
   2. Molecules have fingerprintsEthanol gives three peaks in the ratio 3:2:1, and you can read the structure straight off the chart.Arnold, Dharmatti & Packard resolve the ethanol spectrum, Stanford, 1951
   3. Structures nobody could have guessedPenicillin, vitamin B₁₂ and insulin, solved from diffraction spots by hand and by the earliest computers.Hodgkin: penicillin 1945, B₁₂ 1956, insulin 1969
   4. A photograph of a moleculeIn 2009 an atomic force microscope resolved the individual bonds in a single pentacene molecule. Dalton’s indivisible pieces, in a picture.Gross et al., IBM Zurich, 2009
   5. What chemistry cannot yet doPrediction from first principles, catalysis by design, and the folding problem that turned out to be tractable after all.CASP14 and AlphaFold2, 2020