August Kekulé

How to use this course

Every organic chemist draws a molecule as letters joined by lines, with each carbon carrying four bonds and benzene as a hexagon. That habit comes largely from August Kekulé. Between 1857 and 1865 he argued that carbon is tetravalent, that carbon atoms bond to one another in chains, and that benzene is a closed ring of six carbons. Together these ideas turned organic chemistry from a catalogue of formulas into a science of structure that could predict how many isomers a compound should have.

Each technical lesson opens with a paragraph for the general reader, then gives the key ideas in boxes, the reasoning behind them, and exercises with folded solutions. The lessons follow the history: the type theory Kekulé inherited (Lesson 04), his structural theory of 1857–58 (Lessons 05–06), the benzene theory of 1865 and its testing (Lessons 07–08), and the twentieth-century explanation of why benzene behaves as it does (Lesson 09).

Timeline

Darmstadt and Giessen, 1829–1852

Friedrich August Kekulé was born in Darmstadt on 7 September 1829. His family were upper-middle-class civil servants of the Grand Duchy of Hesse. The acute accent on the final e had been added by his father during the French occupation of Hesse in the Napoleonic period. August, who never used his first given name, was a gifted schoolboy at the Grand Ducal Gymnasium, good at languages and at drawing, and he planned to become an architect.

In the autumn of 1847 he entered the University of Giessen to study architecture. Giessen was then the most famous chemistry school in the world. Justus von Liebig had built there a teaching laboratory where students learned analysis by doing it, and he lectured with great force. Kekulé heard him in his first semester and changed subjects.

Chemists of his generation knew the composition of compounds, but few asked how their atoms might be arranged. Kekulé came to the subject with a draughtsman’s habit of thinking in drawings and spatial arrangements. Biographers have often connected this training with his later structural theory, though the connection is a matter of interpretation.

The doctorate

After four years at Giessen he completed a short compulsory military service and, on Liebig’s advice, went to Paris in 1851 to widen his chemical education. He returned to take his doctorate at Giessen in the summer of 1852 with a thesis on amyloxide sulfuric acid and some of its salts, a compound of amyl alcohol and sulfuric acid. The work was competent analytical chemistry in the Giessen style and gave no hint of the theorist to come.

Paris, Reichenau, London, 1851–1855

The year in Paris in 1851–52 shaped Kekulé’s chemistry more than the doctorate did. He attended the lectures of Jean-Baptiste Dumas and came to know Charles Gerhardt and Adolphe Wurtz. Gerhardt was then completing his theory of types, which classified organic compounds by analogy with a few simple inorganic molecules (Lesson 04). Gerhardt was a difficult man and an outsider in French chemistry, but Kekulé found his ideas congenial, and they became the starting point for his own.

After the doctorate, Kekulé spent about a year as private assistant to Adolph von Planta, a wealthy Swiss chemist who worked in a laboratory at the castle of Reichenau, near Chur in the canton of Graubünden. The work was routine analysis. He then moved to London, where from 1853 to 1855 he worked as assistant to John Stenhouse at the medical school of St Bartholomew’s Hospital.

Williamson and London

The decisive contact in London was Alexander Williamson of University College. Williamson had shown in 1850 that ethers are formed by replacing both hydrogens of water with organic groups, and he had made water the model for a whole family of compounds. Kekulé extended the idea in 1854 by preparing thioacetic acid, replacing oxygen by sulfur with phosphorus pentasulfide, and arguing that hydrogen sulfide could serve as a type just as water did. In this paper he began to speak of atoms of different “basicity”, that is, of different combining capacity.

Kekulé later told how, riding on the top of a London omnibus one summer evening, he fell into a reverie in which atoms danced before his eyes, the larger ones taking the smaller ones in their grip and dragging chains of atoms behind them. Historians date the episode, if it happened, to the late summer of 1855. He said that he spent part of that night sketching what he had seen, and that this was the origin of the structure theory. The story was told 35 years later (Lesson 11), and it should be read as a recollection, not a record.

Heidelberg, 1856–1858

Returning to Germany, Kekulé habilitated at Heidelberg in 1856 and became a Privatdozent, an unsalaried lecturer paid by student fees. Heidelberg’s chemistry was dominated by Robert Bunsen, a great experimentalist with little interest in theory. Kekulé fitted out a small private laboratory of his own and taught organic chemistry to a handful of students. Among those who worked with him was the young Adolf von Baeyer, with whom he argued long and hard about chemical theory.

In these two years he wrote his most important theoretical papers. A paper of 1857 in Liebig’s Annalen, on so-called conjugated compounds and the theory of polyatomic radicals, proposed that carbon is tetravalent and added a fourth type, marsh gas \(\mathrm{CH_4}\), to Gerhardt’s three. A second paper on the constitution of mercury fulminate showed the new ideas at work. In May 1858 the Annalen published his paper on the constitution and metamorphoses of chemical compounds and the chemical nature of carbon. It added the second essential step: carbon atoms bond to one another, forming chains (Lesson 05).

A month later Archibald Scott Couper, a Scot working in Wurtz’s laboratory in Paris, published the same idea independently in the Comptes rendus and the Annales de chimie et de physique, with diagrams in which lines joined atoms. Couper had asked Wurtz to present his paper to the Academy of Sciences, and Wurtz delayed. Kekulé’s paper appeared first, and the priority went to him. That autumn, on 8 October 1858, Kekulé was appointed professor of chemistry at the University of Ghent in Belgium, at 29.

Types, radicals and formulas

For the general reader. In the 1850s chemists could measure what a compound was made of, but they disagreed even on how to write its formula. One school grouped organic compounds into families, each modelled on a simple molecule such as water or ammonia. Kekulé learned this “theory of types” in Paris and London. His structural theory grew directly out of it.

Two systems of weights

Many chemists of the 1840s and 1850s used equivalent weights with \(\mathrm H=1\), \(\mathrm C=6\), \(\mathrm O=8\). In that system water was HO and acetic acid \(\mathrm{C_4H_4O_4}\). Gerhardt and Auguste Laurent argued for \(\mathrm C=12\), \(\mathrm O=16\), which gave water as \(\mathrm{H_2O}\) and made the formulas of organic compounds consistent with their vapour densities. Kekulé adopted the Gerhardt values, but chemists did not settle the question until after the Karlsruhe Congress of 1860 (Lesson 06).

Gerhardt’s types

Williamson’s ether synthesis of 1850 was the key evidence for the water type. Potassium ethoxide and ethyl iodide give diethyl ether, and potassium ethoxide with methyl iodide gives the mixed ether ethyl methyl ether. A single oxygen holds two groups, so the oxygen atom must be “dibasic”, as Williamson and Kekulé put it.

From types to valence

Read the types as a list of atomic capacities: hydrogen and chlorine combine with one partner, oxygen with two, nitrogen with three. Kekulé’s thioacetic acid of 1854 showed sulfur behaving like oxygen. The marsh-gas type assigned carbon four partners. Once each atom has a fixed combining capacity, the types become unnecessary, because any molecule can be built by satisfying the capacities of its atoms. This is the step Kekulé took in 1858.

Tetravalence and the carbon chain

For the general reader. Carbon is special because its atoms can join one another to form long chains and rings, which is why there are millions of organic compounds. Kekulé’s 1858 paper made two claims: each carbon atom has four bonds to give, and carbon atoms can use some of those bonds on each other. From these two rules alone he could explain the formulas of whole families of compounds.

Hence the saturated hydrocarbons have the formula \(\mathrm{C}_n\mathrm{H}_{2n+2}\), as observed for methane, ethane, propane and the paraffins. Kekulé used the same count to explain series in which successive members differ by \(\mathrm{CH_2}\), a regularity Gerhardt had called homology. Each added carbon brings four affinities and uses two to join the chain, leaving two for hydrogen.

Degree of unsaturation

Any ring or multiple bond lowers the hydrogen count by two compared with the open saturated chain. Allowing for nitrogen (valence 3) and halogens (valence 1), the number of rings plus π bonds in a formula \(\mathrm{C}_c\mathrm{H}_h\mathrm{N}_n\mathrm{O}_o\mathrm{X}_x\) is

Oxygen and other divalent atoms do not change \(U\), because a divalent atom inserted into a chain uses both of its bonds. Kekulé did not write this formula, but it is a direct consequence of his counting, and he used the equivalent reasoning for ethylene, \(\mathrm{C_2H_4}\), and for benzene.

Kekulé and Couper

Couper’s paper of June 1858 reached the same two postulates and went further in notation. He wrote formulas with dotted or straight lines between atoms, the ancestor of the modern structural formula. He also allowed carbon a combining power of two as well as four. Kekulé gave no diagrams in 1858, and his first graphical formulas, in the Lehrbuch of 1859, used elongated “sausage” shapes for atoms. Couper published nothing further after a breakdown; his work was rescued from obscurity in the early twentieth century by Kekulé’s former student and biographer Richard Anschütz. Aleksandr Butlerov introduced the phrase “chemical structure” in 1861 and argued, more firmly than Kekulé, that a compound has one structure that determines its properties.

Structural formulas, isomers, atomic weights

For the general reader. A structural theory is only as good as the formulas it starts from, and in 1860 chemists still disagreed about the relative weights of atoms. Kekulé helped call the first international chemistry conference at Karlsruhe to settle the matter. There a paper by Stanislao Cannizzaro showed how to fix atomic weights reliably. With correct weights, structural formulas could predict which isomers exist.

The Karlsruhe Congress

Kekulé, Wurtz and Karl Weltzien sent out invitations, and about 140 chemists met in Karlsruhe on 3–5 September 1860. Kekulé gave an opening address. The debates on the meaning of atom, molecule and equivalent ended without a vote, but on the last day copies of Cannizzaro’s 1858 Sunto were handed out. Lothar Meyer later wrote that reading it cleared his doubts. Within a decade most chemists used \(\mathrm C=12\), \(\mathrm O=16\).

Structural isomers

With tetravalent carbon and correct weights, chemists could enumerate the possible structures for a formula. The theory predicted, for example, that there should be two propyl alcohols, one with OH on an end carbon and one on the middle carbon. Hermann Kolbe predicted the secondary alcohols on a similar basis, and Charles Friedel made isopropyl alcohol in 1862. Each success strengthened the theory. What the theory did not yet address was the arrangement of atoms in space: two compounds with the same connections but mirror-image arrangements remained unexplained until van ’t Hoff and Joseph Le Bel in 1874.

The benzene ring, 1865

For the general reader. A large family of compounds, from coal-tar dyes to oil of bitter almonds, contain at least six carbon atoms and behave differently from ordinary chains. They were called aromatic. In January 1865 Kekulé proposed that they all share a core of six carbon atoms closed into a ring, benzene, with alternating single and double bonds. The idea organized the chemistry of the dye industry for the next century.

The puzzle

Benzene, \(\mathrm{C_6H_6}\), had been isolated by Michael Faraday in 1825. Its formula gives \(U=4\), so it should be highly unsaturated. Yet benzene resists addition reactions and undergoes substitution, keeping its six carbons intact through reaction after reaction. Every aromatic compound had at least six carbons. Its substitution products were remarkably few: there was only one bromobenzene, one phenol, one nitrobenzene.

Count the affinities: each carbon has 4, uses 1 for H, and 3 within the ring (one single and one double link, shared with its two neighbours). The six carbons thus use \(6\cdot3/2=9\) bonds in the ring: 6 sigma links and 3 double-bond second links, consistent with \(U=4\) (one ring plus three double bonds).

Kekulé first published the idea in French in the Bulletin de la Société Chimique de Paris in January 1865, and at length in German in the Annalen in 1866. His early figures drew the ring with his sausage-shaped atoms; the regular hexagon soon became the standard symbol. Johann Josef Loschmidt, in his privately printed Chemische Studien of 1861, had drawn benzene as a circle, and Couper had proposed other cyclic structures for aromatic compounds in 1858. Neither developed the idea into a theory of isomers.

Predictions

Equivalence of the six positions implies one monosubstituted derivative \(\mathrm{C_6H_5X}\), and three disubstituted derivatives \(\mathrm{C_6H_4X_2}\), with substituents at positions 1,2 (ortho), 1,3 (meta) and 1,4 (para). These counts matched what chemists found, and Kekulé made them the test of his theory.

Counting isomers and the oscillation hypothesis

For the general reader. Kekulé’s ring could be tested by counting. If it was right, chemists should find exactly three forms of every disubstituted benzene. Albert Ladenburg, who had worked in his Ghent laboratory, pointed out a flaw: a ring with fixed alternating bonds should give two different ortho compounds, depending on whether the substituents straddle a single or a double bond. Only one was ever found. In 1872 Kekulé answered that the bonds do not stay fixed.

Counting with symmetry

If all six positions are equivalent, the symmetry group of the regular hexagon, the dihedral group \(D_6\) of order 12, acts on substitution patterns. Two patterns describe the same compound if a symmetry carries one to the other.

Colour each of the six positions H or X. A symmetry with cycles of lengths \(\ell_1,\dots,\ell_k\) on the vertices fixes \(2^k\) colourings. For \(D_6\): the identity has 6 cycles; rotations by \(\pm60^\circ\) have 1, by \(\pm120^\circ\) have 2, by \(180^\circ\) has 3; three reflections through opposite vertices have 4 cycles each, and three through opposite edge midpoints have 3. So

The 13 patterns split by number of X as \(1,1,3,3,3,1,1\), matching one parent, one mono-, three di-, three tri-, three tetra-, one penta- and one hexa-substituted benzene.

Körner’s absolute method (1874)

Knowing that three disubstituted isomers exist does not say which is which. Wilhelm Körner, who had worked with Kekulé in Ghent, gave a method: introduce a third substituent and count products. From 1,2-dibromobenzene one gets two tribromo (or bromo-nitro) derivatives, from 1,3 three, from 1,4 only one.

Ladenburg’s objection and the oscillation hypothesis

In 1869 Ladenburg observed that in a ring with fixed alternating bonds, 1,2-disubstitution across a double bond differs from 1,2-disubstitution across a single bond (sometimes labelled 1,2 and 1,6). Only one ortho isomer of each kind was known. Ladenburg himself proposed a triangular prism structure with six equivalent carbons and only single bonds. James Dewar, who had also worked in Kekulé’s Ghent laboratory, had described several other \(\mathrm{C_6H_6}\) structures in 1867, including the bicyclic one now called Dewar benzene.

The hypothesis appeared in a paper on condensation products of aldehyde in the Annalen of 1872. It is not the modern theory, but it captured the essential point: no single valence-bond drawing describes benzene, and the real molecule has six equivalent bonds.

Benzene after Kekulé

For the general reader. Kekulé’s two alternating structures were a brilliant guess, but chemistry needed physics to explain them. In the twentieth century, X-rays showed benzene to be a flat regular hexagon, and quantum mechanics showed that its six bonding electrons are shared around the whole ring. That sharing lowers the energy and explains why benzene resists the addition reactions typical of double bonds.

Evidence

In 1929 Kathleen Lonsdale’s X-ray study of hexamethylbenzene showed the ring to be a planar regular hexagon. All six C–C bonds in benzene have length about 139–140 pm, between a typical single bond (154 pm) and double bond (134 pm). Thermochemistry gives the energetic signature. Hydrogenating cyclohexene releases about 120 kJ/mol; three isolated double bonds would release about 360 kJ/mol; benzene releases only about 208 kJ/mol. The difference, about 150 kJ/mol, is the aromatic stabilization.

Resonance

In 1933 Linus Pauling and George Wheland described benzene as a quantum-mechanical superposition dominated by the two Kekulé structures. The molecule does not oscillate between them, as Kekulé had suggested; its single ground state is a combination of both, lower in energy than either.

Hückel theory

Derivation. The Hückel matrix is \(H=\alpha I+\beta A\), where \(A\) is the adjacency matrix of the \(N\)-cycle. \(A\) is circulant, so its eigenvectors are \(c_j=e^{2\pi ijk/N}\) with eigenvalues \(e^{2\pi ik/N}+e^{-2\pi ik/N}=2\cos(2\pi k/N)\). The lowest level (\(k=0\)) is single; the others pair up as \(k\) and \(N-k\). Filling one single level and \(n\) degenerate pairs takes \(2+4n\) electrons.

For benzene (\(N=6\)) the levels are \(\alpha+2\beta\), \(\alpha+\beta\) (twice), \(\alpha-\beta\) (twice), \(\alpha-2\beta\). Six electrons fill the three bonding orbitals: \(E_\pi=2(\alpha+2\beta)+4(\alpha+\beta)=6\alpha+8\beta\). Three isolated ethylene π bonds give \(3\cdot2(\alpha+\beta)=6\alpha+6\beta\). The delocalization energy is \(2\beta\), a negative number, so delocalization stabilizes the molecule.

Constant valence, multiple bonds, models

For the general reader. Kekulé’s theory rested on a simple rule: each element has a fixed number of bonds. The rule worked beautifully for carbon, but other elements, such as phosphorus and nitrogen, seemed to change their combining power from compound to compound. Kekulé defended the fixed rule all his life. He also built some of the first molecular models, with carbon’s four bonds pointing to the corners of a tetrahedron, though he gave that geometry no physical meaning.

The doctrine had support: \(\mathrm{PCl_5}\) vapour dissociates into \(\mathrm{PCl_3}\) and \(\mathrm{Cl_2}\) on heating, and ammonium chloride vapour into ammonia and hydrogen chloride. But many chemists, including Edward Frankland and Wurtz, accepted variable valence, and it prevailed. Sulfur shows valence 2 in \(\mathrm{H_2S}\), 4 in \(\mathrm{SO_2}\) and 6 in \(\mathrm{SF_6}\), a stable gas that does not dissociate. Modern theory distinguishes oxidation state, coordination number and the number of electron-pair bonds; none of these is constant across all compounds of an element.

Multiple bonds

If carbon is always tetravalent, ethylene \(\mathrm{C_2H_4}\) leaves each carbon with one unused affinity. Kekulé’s answer, which became standard, was that the two carbons share two affinities with each other, a double bond. Acetylene \(\mathrm{C_2H_2}\) has a triple bond. Unsaturated compounds add two atoms of bromine or hydrogen per double bond, restoring single bonds: the reaction that benzene notably resists (Lesson 09).

Models

For lectures Kekulé used ball-and-wire models in which carbon’s four valences point toward the corners of a tetrahedron, an arrangement that allows single, double and triple links between carbons. Like most chemists of the time, he treated the geometry as a teaching device, not a claim about real space. Jacobus Henricus van ’t Hoff, who worked in his Bonn laboratory, took the tetrahedron literally in 1874, and drawings based on Kekulé-style models illustrated the new stereochemistry.

Ghent and Bonn: a school of chemists

At Ghent from 1858 Kekulé taught general, inorganic and organic chemistry in French and set up what has been described as the first chemical teaching and research laboratory in Belgium. Jean-Servais Stas had persuaded him to take the chair. Researchers came to him from abroad: Adolf von Baeyer, Wilhelm Körner, Albert Ladenburg and James Dewar all worked in his laboratory, and Théodore Swarts, who succeeded him at Ghent, was among his pupils. He wrote the first volumes of his Lehrbuch der organischen Chemie there, the first textbook built throughout on tetravalent carbon, and there he worked out the benzene theory.

In 1862 he married Stephanie Drory in Ghent. Their son Stephan was born in 1863, and Stephanie died two days later of puerperal fever. He was elected an associate member of the Royal Academy of Belgium on 15 December 1864.

Bonn

In 1867 Kekulé accepted the chair at Bonn left vacant by August Wilhelm Hofmann’s move to Berlin. The Belgian government accepted his resignation on 21 September 1867. The new chemical institute near Poppelsdorf Palace, planned for Hofmann, was occupied on 11 May 1868, and it became one of the great teaching laboratories of Germany. Kekulé declined a call to succeed Liebig at Munich, and the Prussian government extended his building in return. He served as rector of the university in 1877–78. In 1876 he married Luise Högel, with whom he had three more children.

His students and junior colleagues at Bonn included Emil Fischer, who began there before following Baeyer to Strasbourg; Jacobus Henricus van ’t Hoff, who spent 1872–73 in his laboratory; Otto Wallach, Ludwig Claisen, Theodor Zincke and Richard Anschütz, who later succeeded to his chair and wrote his biography. Three of the first five Nobel Prizes in Chemistry went to men who had worked under him: van ’t Hoff (1901), Fischer (1902) and Baeyer (1905). Wallach won in 1910. After an attack of measles weakened his health, Kekulé did less laboratory work and turned his attention to the later volumes of the Lehrbuch.

Dreams, priority and legacy

On 11 March 1890 the German Chemical Society held a celebration in Berlin for the 25th anniversary of the benzene theory, the Benzolfest. In his speech, printed in the Berichte, Kekulé described two reveries. One was the dance of atoms on the London omnibus (Lesson 02). The other came in Ghent, as he dozed by the fire while writing his textbook: chains of atoms twisting like snakes, until one seized its own tail. He woke, he said, and spent the night working out the consequences. He closed with advice to learn to dream, and then to test the dreams before the waking mind.

The snake story has been endlessly retold and sometimes doubted. Historians note that it was first told a quarter of a century after the event, in a speech partly designed to answer rival priority claims. A parody of 1886, in a mock journal of the “Thirsty Chemical Society”, had already drawn benzene as six monkeys holding each other’s tails. Whether or not the dream happened, the ring hypothesis rested on years of work on formulas, isomer counts and the type theory.

Priority

The structure theory had several authors. Couper published the self-linking of carbon a month after Kekulé and drew bonds as lines. Butlerov made structure the organizing principle of a textbook in the 1860s. Loschmidt drew cyclic benzenes in 1861. Kekulé was sensitive about these claims. His lasting contribution was to turn the scattered ideas into a programme: a consistent valence theory, a textbook built on it, and predictions of isomer numbers that experiments confirmed.

Honours and death

The Royal Society awarded him the Copley Medal in 1885. In 1893 he received the Prussian order Pour le Mérite for Sciences and Arts. In 1895 Kaiser Wilhelm II recognized an old Bohemian title in his family, and he took the name Kekule von Stradonitz, dropping the accent. He died in Bonn on 13 July 1896, aged 66, and was buried in an honorary grave in the Poppelsdorf cemetery.

Legacy

• Structural formulas: the tetravalent carbon and the carbon chain underlie every structure drawn in organic chemistry.
• Aromatic chemistry: the benzene ring organized the chemistry of dyes, drugs and explosives, and the German chemical industry grew on it.
• Teaching: his laboratories in Ghent and Bonn trained a generation, including several Nobel laureates.
• Theory: the problem of benzene’s equivalent bonds led directly to resonance and molecular-orbital theory, and his ball-and-wire lecture models influenced van ’t Hoff’s tetrahedral carbon.

Connections

People linked to August Kekulé across this collection. A name without a link has no page here yet. See August Kekulé in the network of all 337 lives →

Teachers and advisors

• Justus von Liebig 1847-1852

Students

• Jacobus Henricus van 't Hoff 1872-1873
• Adolf von Baeyer 1856-1860
• Albert Ladenburg
• Emil Fischer 1871-1872
• James Dewar
• Ludwig Claisen
• Otto Wallach
• Richard Anschütz
• Wilhelm Körner

Influenced by

• Alexander Williamson 1853-1855
• Charles Gerhardt 1851-1852
• Jean-Baptiste Dumas 1851-1852

Colleagues and collaborators

• Robert Bunsen 1856-1858
• Stanislao Cannizzaro 1860
• Adolphe Wurtz
• August Wilhelm Hofmann

Rivals

• Aleksandr Butlerov
• Archibald Scott Couper
• Josef Loschmidt

Family

• Stephan Kekule von Stradonitz

Also mentioned in this life

• Amedeo Avogadro
• Charles-Augustin de Coulomb
• Michael Faraday

Fields

• organic chemistry 4 lives

Sources and further reading

Biography

1. August Kekulé, Wikipedia; and the German article August Kekulé.
2. Science History Institute, August Kekulé and Archibald Scott Couper.
3. Université de Gand, Biographie nationale entry, Kekulé (Friedrich-August).
4. Kekulé, Friedrich August, Encyclopedia.com.
5. R. Anschütz, August Kekulé, 2 vols., Verlag Chemie, Berlin, 1929.
6. A. J. Rocke, Image and Reality: Kekulé, Kopp, and the Scientific Imagination, University of Chicago Press, 2010.
7. O. T. Benfey, August Kekulé and the birth of the structural theory of organic chemistry in 1858, J. Chem. Educ. 35 (1958) 21.
8. Karlsruhe Congress, Wikipedia.

Chemistry

1. A. Kekulé, Ueber die Constitution und die Metamorphosen der chemischen Verbindungen und über die chemische Natur des Kohlenstoffs, Annalen der Chemie und Pharmacie 106 (1858) 129–159.
2. A. S. Couper, Sur une nouvelle théorie chimique, Annales de chimie et de physique 53 (1858).
3. A. Kekulé, Sur la constitution des substances aromatiques, Bulletin de la Société Chimique de Paris 3 (1865) 98–110.
4. A. Kekulé, Untersuchungen über aromatische Verbindungen, Annalen der Chemie und Pharmacie 137 (1866) 129–196.
5. A. Kekulé, Ueber einige Condensationsproducte des Aldehyds, Annalen der Chemie und Pharmacie 162 (1872) 309–320.
6. A. Kekulé, Lehrbuch der organischen Chemie, Enke, Erlangen and Stuttgart, 1859–1887.
7. A. Kekulé, Benzolfest: Rede, Berichte der Deutschen Chemischen Gesellschaft 23 (1890) 1302–1311.
8. P. J. Ramberg, Pragmatism, belief, and reduction: stereoformulas and atomic models in early stereochemistry, HYLE 6 (2000).
9. E. Hückel, Quantentheoretische Beiträge zum Benzolproblem, Z. Physik 70 (1931) 204–286.
10. L. Pauling and G. W. Wheland, The nature of the chemical bond. V, J. Chem. Phys. 1 (1933) 362–374.