It was again pleasant to take a trip to the archives stacks containing the Wenner Collection at AIP’s Niels Bohr Library and Archives back in early 2020. And again, it took little time to find a box containing papers written by two giants of the early 20th century, concerned with one of the basic problems of both physics and chemistry. The papers are:
- Lewis, G.N. 1916. The Atom and the Molecule. Journal of the American Chemical Society 38:4, 762-785.
- Langmuir, I. 1919. The Arrangement of Electrons in Atoms and Molecules. Journal of the American Chemical Society 41:3, 868-934.
- Langmuir, I. 1920. The Octet Theory of Valence and its Applications with Special Reference to Organic Nitrogen Compounds. Journal of the American Chemical Society 42:2, 274-292.
These three papers partly capture one of the most fertile and contentious eras that brought physics and chemistry together in the persons of the chemist Gilbert N. Lewis (1875-1946) and the chemist-physicist-electrical engineer Irving Langmuir (1881-1957). Lewis studied with Walther Nernst and Wilhelm Ostwald in Germany, but after a brief return to Harvard he spent the rest of his career as a physical chemist at University of California, Berkeley. While Lewis never won a Nobel prize, he mentored many chemists who did. He is known as the person who coined the name “photon.” Langmuir also studied in Germany with Nernst, but returned and eventually joined the General Electric Research Laboratory at Schenectady, N.Y. Langmuir built on his work with the Nernst lamp and contributed major insights into a number of problems including with light bulbs and filaments. His Nobel was awarded for surface chemistry. These papers from Lewis and Langmuir represent one part of the continuous strand of scientific inquiry attempting to figure out the mysteries of bonding in chemical compounds as related to atomic structure.
Linus Pauling (1901-1994) later picked up this strand of research, as evidenced by his ownership of Lewis’ paper of 1916:
Front cover of “The Atom and the Molecule” with Linus Pauling’s signature. Screenshot Courtesy of Special Collections & Archives Research Center, Oregon State University Libraries website.
Caption: Front cover of “The Atom and the Molecule” with Linus Pauling’s signature. Screenshot Courtesy of Special Collections & Archives Research Center, Oregon State University Libraries website.
Pauling summed and brought valence bond work up to date in 1939 with his masterful “The Nature of the Chemical Bond.”
Gilbert Lewis outdoors. Credit: AIP Emilio Segrè Visual Archives, Uhlenbeck Collection, image information: Lewis Gilbert B2
There is a well-known succession of advances in the understanding of the mysteries of chemical combination and behavior. Chemical reactions have been used since antiquity for a multitude of practical, predictable products such as metals, soap, glass, dyes, fertilizers, and medicines. How combinations occur between what entities was a constant question, as exemplified in the startling appearance of a bright yellow precipitant upon combining two colorless solutions of lead (II) nitrate and potassium iodide. A favorite beginning for the concept of atoms is often noted in the poem of Lucretius (c.99 B.C.E.-c.55 B.C.E.), “On the Nature of Things,” which cites the atomism of Democritus, including the world view of Epicurus. As the nineteenth century progressed, the regular forms of crystalline minerals, plus knowledge of their constituents led to various theories of combination. After John Dalton, and through the 19th century, the conception of matter as small, indivisible particles, or atoms, continued (see Ex Libris Universum blog on Dalton of July 2, 2020).
Then evidence emerged that atoms were not indivisible but instead composed of parts. To initial disbelief, J.J. Thomson (1856-1940) from 1897 into 1907 established that “cathode rays” were actually small particles—electrons—that were only a part of atoms. Then in 1914 Thomson published a paper investigating the import of electron sharing in bonding. Lewis noted that Thomson’s findings corresponded to his own conclusions about polar and nonpolar compounds. Lewis preferred that designation to “inorganic” and “organic,” although he did point out that there are inorganic substances that are predominately nonpolar as well as organic substances that are polar in at least part of the molecule (Lewis, 1916, 763). Lewis provided lists of the characteristics of polar and nonpolar molecules, the difference between the two types consisting of the strength with which the electrons were held by the nucleus. After some discussion, he came to the conclusion that “the distinction of the most extreme polar and nonpolar types is only one of degree, and that a single molecule, or even part of a molecule, may pass from one extreme type to another, not by a sudden and discontinuous change, but by imperceptible gradations” (Lewis, 1916, 767).
Lewis laid the foundation for Lewis structures, a basic concept in university chemistry classes. He noted that since the “the total difference between the maximum negative and positive valences or polar numbers of an element is frequently eight, and is in no case more than eight, I designed what may be called the theory of the cubical atom” (Lewis, 1916. 767).
Lewis’ cubical atoms, as seen on p. 767 of “The Atom and the Molecule.” Screenshot Courtesy of Special Collections & Archives Research Center, Oregon State University Libraries website. Also available in print from the Niels Bohr Library & Archives.
Here we see his conjectures of how the outer electrons are arranged in cubical atoms, followed by reference to his earlier work on possible arrangements in a variety of compounds and speculation about rules that govern them. Each row of the periodic chart added a concentric cube of electrons to the atom. He quickly learned that this was insufficient, but retained it for explanatory purposes. He couldn’t accept Planck’s or Bohr’s conjectures about electron motion, but also noted “it will not be the first time in the history of science that an increase in the range of observational material has required a modification of generalizations based on a smaller field of observation” (Lewis, 1916, 773).
Lewis structures, as seen on p. 778 of “The Atom and the Molecule.” Screenshot Courtesy of Special Collections & Archives Research Center, Oregon State University Libraries website. Also available in print from the Niels Bohr Library & Archives.
After more discussion, Lewis continued on to theorize about molecular structures, using combinations of his cubical atoms. As a shorthand, he used what are now called Lewis dot diagrams (Lewis structures), for example as in the ammonium ion.
There is a great deal more of interest in this seminal paper, in which a very creative mind grapples with both regularities and contradictions to expand knowledge of the world. Servos (1990, 279-285+) has an excellent discussion of the interactions of Lewis, Planck, Bohr, de Broglie, Heisenberg and Pauling, as they attempted to resolve the quantum/matrix approaches to account for bonding.
Irving Langmuir. Credit: AIP Emilio Segrè Visual Archives, Physics Today Collection, Image information: Langmuir Irving B1
Irving Langmuir is known less for his original work in bonding theory, than for his popularization and explanation of Lewis’ ideas. He worked to great effect in a variety of fields as evidenced by his activity in chemistry, physics, and electrical engineering. His Nobel Prize in 1932 was in surface chemistry. His accomplishments are stunning in their variety and effectiveness, but his 1919 paper “The Arrangement of Electrons in Atoms and Molecules” has been noted as possibly his most influential. Langmuir began with:
"The problem of the structure of atoms has been attacked mainly by physicists who have given little consideration to the chemical properties which must ultimately be explained by a theory of atomic structure. The vast store of knowledge of chemical properties and relationships, such as is summarized by the Periodic Table, should serve as a better foundation for a theory of atomic structure than the relatively meager experimental data along purely physical lines" (Langmuir, 1919, 868).
He stated that Lewis’ explanations were satisfactory for the alkaline and alkaline earth elements and a number of others (35 of the then known 88) but that they did not apply to those with more electrons. For those, it’s very hard to position multiple cubic structures around them to continue with Lewis’ model. In the same year, 1916, Walther L.J. Kossel (1888-1956) had published a theory similar to Lewis’ but was most successful for the first 23 elements. Langmuir critiqued it, saying that he felt Lewis gave a more detailed explanation (Langmuir, 1919, 869). Langmuir listed seven detailed postulates which he used to derive the properties of the elements, those concerned with the tendencies of elements as atoms to add or subtract electrons. He continued with many examples applying the octet rule in agreement with Lewis’ use and summarized his conjectures and conclusions with many examples.
Langmuir continued his research on valence and atomic structure. He began his 1920 paper, “The Octet Theory of Valence and its Applications with Special Reference to Organic Nitrogen Compounds,” with:
"The octet theory of valence (reference to Lewis) leads to structural formulas for organic compounds which are identical with those given by the ordinary valence theory whenever we can assume a valence of 4 for carbon, 3 for nitrogen, 2 for oxygen, and one for hydrogen and chlorine" (Langmuir, 1920, 274).
Langmuir evolved a mathematical expression for his octet theory in organic nitrogen compounds and gave many examples. However, he noted that it still did not account for all observed physical properties, such as melting points. There is a great deal of detail in this paper that gives a close look at two excellent minds puzzling through explanations for the invisible.
These three papers are a microcosm of a fertile time of explanation and exploration of the mysteries of atomic structures. The dot structure model is still useful for visualization, while nestled cubes were, and are still, awkward. Unlike Thomson, who considered all bonds as essentially polar, with transfer of electrons from one to another atom, Lewis advocated shared electrons. Brush (1983, 220) stated “The Lewis-Langmuir theory successfully accounted for most of the (then) known facts about valence. The emphasis of shared pairs of electrons was justified by the fact that almost all molecules have an even number of electrons.”  These papers contributed to understanding of organic and quantum chemistry, but also to classical physical chemistry and to ionization in general (Servos, 1990, 135). The papers are conveniently available in the Wenner Collection.
 Thomson, J.J. 1897. Cathode Rays. Philosophical Magazine, series 5, 44:293-316; 1907, The Corpuscular Theory of Matter. London: Constable
 Servos, John W. 1990. Physical Chemistry from Ostwald to Pauling: The Making of a Science in America. Princeton: Princeton, New Jersey.
 Brush, S.G. 1983. Statistical Physics and the Atomic Theory of Matter: from Boyle and Newton to Landau and Onsager. Princeton: Princeton University Press.