It took crystallography to prove that the audacious proposal of a sandwich structure was correct
Sometimes what science needs to move forward is a hearty dose of disbelief. Such was the case with ferrocene, the beloved sandwich-shaped molecule that’s been credited with ushering in the modern age of organometallic chemistry.
On Dec. 15, 1951, Peter L. Pauson and Thomas J. Kealy, two chemists at Duquesne University, in Pittsburgh, reported a new type of organo-iron compound (Nature, DOI: 10.1038/1681039b0). They suggested their remarkable yellow crystals consisted of two cyclopentadiene moieties each making a single bond to a central iron atom. It was a modest report, taking up less than a page, but it certainly raised some eyebrows.
Unbeknown to Pauson and Kealy, British Oxygen Co. chemists Samuel A. Miller, John A. Tebboth, and John F. Tremaine had made the exact same compound through an entirely different synthetic route. In a paper that appeared shortly after Pauson and Kealy’s, they too proposed a C5H5–Fe–C5H5 structure (J. Chem. Soc. 1952, DOI: 10.1039/jr9520000632).
Harvard University chemists Robert Burns Woodward and Geoffrey Wilkinson read Pauson and Kealy’s paper and believed their proposed structure to be incorrect. In short order, they proposed an alternative structure in which the iron sat sandwiched between two parallel cyclopentadiene rings (J. Am. Chem. Soc. 1952, DOI: 10.1021/ja01128a527). Ernst Otto Fischer, of the Technische Hochschule, in Munich, made a similar proposal (Z. Naturforsch. B: Chem. Sci.
1952, 7, 377).
“I think it is difficult today to appreciate just how surprising, unorthodox, even revolutionary, this structure must have appeared to most chemists,” writes crystallographer Jack D. Dunitz in an essay on ferrocene he wrote in the 1993 book “Organic Chemistry: Its Language and Its State of the Art.” Dunitz, who in 1951 was a research fellow at the University of Oxford, recalls that his own reaction was one of extreme skepticism. “I thought: What a nerve these Harvard chemists have! To publicly put forward such a structure on such scanty evidence!”
He was not alone. Marshall D. Gates, an assistant editor for the Journal of the American Chemical Society, wrote to Woodward: “We have dispatched your communication to the printers but I cannot help feeling that you have been at the hashish again.”
Dunitz shared his disbelief with theoretical chemist Leslie E. Orgel, who was also at Oxford. The two decided to persuade an organic chemist they knew to prepare crystals of ferrocene, which Dunitz used to solve the structure. “Extraordinary as it seemed to me, the Harvard proposal was correct,” Dunitz recalls. “There was no doubt about it.”
Dunitz published the structure along with Orgel’s explanation of ferrocene’s stability based on molecular orbital theory (Nature 1953, DOI: 10.1038/171121a0). Independently, Pennsylvania State University researchers Philip F. Eiland and Raymond Pepinsky also used crystallography to confirm the sandwich structure (J. Am. Chem. Soc. 1952, DOI: 10.1021/ja01139a527). “That was the marvelous thing about crystal structure analysis,” Dunitz remarks. “When it worked, the result had a satisfying definiteness about it.”—Bethany Halford