Just as with ferrocene, 1 the formula first written for diborane B2H6 was orthodox, in conformity with the usual paradigmatic rules regarding molecular structure. And it was wrong. Correcting the mistake showed an extremely unusual bonding picture for the molecule. Just as in the ferrocene case, it opened up a whole new chapter of chemistry, later recognized with the award of a Nobel prize to William N. Lipscomb in 1976 for "his studies on the structure of boranes illuminating problems of chemical bonding."
I shall highlight here only a few episodes from the diborane story. In 1937, a former student of Linus Pauling, Simon H. Bauer - incidentally he is still active in research at Cornell, at the age of 88 - applied the technique of electron diffraction, a tool which he had learned to use at Caltech with Pauling, to diborane. He found and he reported a structure analogous to that of ethane, which therefore he wrote as H3B-BH3. 2 In 1942, Bauer reiterated his contention of the ethane-like structure for diborane. 3 At the same time of the early 1940s, H.I. Schlesinger, a chemistry professor at the University of Chicago, was also working on boron compounds. One of the reasons was his involvement in the Manhattan Project: it was thought that an isotope separation could be devised for uranium using such derivatives. In any case, Schlesinger was very much interested in the structure of diborane.
Professor Pierre Laszlo
Thus Schlesinger wrote a letter to Linus Pauling at Caltech on January 3 1941. 4 Let me quote:
As a result of our work on the metallo borohydrides I definitely feel
The structure I have in mind is a bridge structure, in which the two boron atoms are joined to each other through an unusual type of hydrogen bond, perhaps best represented by the following formula
From hindsight we know this Schlesinger formula to be the correct one. Also, he ought to have been given priority for proposing such a bridged structure. End of Act II.
I do not feel very friendly toward the structure which you mention in
In other words, at this point Pauling with his considerable insider knowledge of structural chemistry not only shot down the correct structure which Schlesinger had come up with, but he also indirectly convinced him not to bring it into print. What a shame! When Schlesinger did publish his review, he had summarized the chemical and physical properties of boron hydrides in terms of an ethane-like structure presenting electron-deficient boron-hydrogen bonds. 5 Act IV occurred soon afterwards. An undergraduate student by the name of H. Christopher Longuet-Higgins was at Balliol College, in Oxford. He was only in his second year of study. He got himself interested in this question of the diborane structure. He too came up with the same bridged structure, after examining all the physical evidence and after thinking out an adequate theoretical description. His paper, signed jointly with one of the professors of chemistry from Balliol, R.P. Bell, a reputed physical chemist, appeared in 1943 in the Journal of the Chemical Society. 6 Longuet-Higgins would go on to become one of the world leading chemical theoreticians. Act V saw the consolidation of the bridged structure of Schlesinger and Longuet-Higgins. Price confirmed in 1947-48 the earlier (1940, 1941) infrared spectroscopic evidence by Stitt 7, 8 in favor of the bridged structure. 9, 10 Shoolery brought to bear clinching nmr evidence. 11 Three different theorists (Mulliken, Pitzer and Walsh) gave molecular orbital and valence bond descriptions of the bridged structure in 1946 and 1947. 12, 13, 14 And in 1951, Hedberg and Schomaker did another electron diffraction study, showing that the data ruled out the open structure and was consistent with the bridged structure only. 15
There was even a precursor found (as is almost invariably the rule with science discoveries 16): Dilthey had anticipated the bridged structure as early as 1921 17 and this was revived by Soviet scientists in the early 1940s 18, 19 (Schlesinger was referring to them in his letter to Pauling). Yet, in the second edition of The Nature of the Chemical Bond, which came out in 1945, Linus Pauling devoted a full 3.5 pages to presenting diborane as an ethane-like structure on the strength of both G.N. Lewis's conceptualization 20 of this molecule with "six electron pair bonds resonating amongst seven positions" and of S.H. Bauer electron diffraction work. 21 True, his influence on structural chemistry was on the wane: in the felicitous wording of Mary Jo Nye's, his valence bond approach to molecular structure "earned most chemists' allegiance, at least up until 1940." 22
Both stories, that of ferrocene 1 and that of diborane, have a strong resemblance. They both bear out an account à la Kuhn of scientific revolutions. Both compounds were anomalies within the existing paradigms. Their representations disobeyed "conventional wisdom" to make use of the expression pioneered by Galbraith. 23 Accordingly, they suffered from a birth defect. Both were incorporated into existing knowledge, at a high cost. Their formulas were inconsistent with some of the experimental evidence, some of the data had been inadvertently misread because of the paradigmatic bias.
Both stories are inconsistent with the notion, central to a sociologist of science such as Bloor, of a Symmetry Principle. 24 According to this methodological edict, the historian or the sociologist of science should give equal footing to the two types of theories of nature, whether they are winners or losers. There is, to the contrary, blatant dissymmetry in both the ferrocene 1 and the diborane case. Their first investigators could not bring themselves to see, let alone accept the correct structure because the existing knowledge did not allow it. In order to make room for the correct formulation, two moves had to occur: provide a theoretical description for the novel entity; and retrieve from the waste basket those parts of the experimental evidence earlier glossed over, because they did not jibe with the postulated structure. As an empirical natural philosopher, I must credit Thomas Kuhn with an operationally-valid description; and I must beg to disagree with Bloor's symmetry principle: in both cases, the difference between the winning side and the losing side was not historically contingent and socially determined.
I thank the Ava Helen and Linus Pauling Archive at Oregon State University in Corvallis, Oregon and its director Clifford Mead for permission to quote from Pauling's correspondence.
1 P. Laszlo, R. Hoffmann, Angew. Chem. Int. Ed. Engl. (2000) .