|
One of the great tenets of organic chemistry is that its central element, carbon, can form a maximum of four covalent bonds. Methane, the archetypal organic molecule is essentially a carbon atom at the centre of a tetrahedron of four hydrogen atoms, with each hydrogen bonded to the carbon through a single covalent bond, filling carbon's valency. Other organic molecules may contain carbon double bonds, and even carbon triple bonds. But, the principle holds that the maximum valency is 4. However, a novel structure studied using X-ray crystallography hints at the possibility of a carbon atom that, at first sight seems to be a little different. Could the oldest rule of organic chemistry have been broken at last, or is low atomic separation being equated too keenly with the presence of a bond? Torahiko Yamaguchi and Yohsuke Yamamoto of Hiroshima University, Daisuke Kinoshita and Kin-ya Akiba of Waseda University, Yun Zhang and Christopher Reed of the University of California, Riverside, Daisuke Hashizume of RIKEN, and Fujiko Iwasaki of the Rigaku Corporation, in some sense, set out to break the rules. They hoped to explore the possibility of hypervalency in carbon compounds. They anticipated that steric constraints that prevent atoms sitting where they ought to in a structure might provide the answer. Adding large, bulky chemical groups to a central carbon might make it behave badly. The team used the bulk of an anthracene derivative containing a central sulfur atom where normally there would be a carbon and hooked two of these molecules up to each other through a central carbon atom, an allenic bridge. In other words, a central carbon atom with a double bond connected to one sulfur-containing anthracene group and another double bond diametrically oppososite linked to a second anthracene. So far, so conventional. Four-coordinated carbon, breaking no valency rules. However, by bringing four ether oxygen atoms into close proximity with the allenic carbon they could simulate what appears to be a carbon atom surrounding by six close neighbours. This results in a carbon with an anthracene, doubly bonded top and bottom and an arrangement with four oxygen atoms one at each corner of a square. The carbon thus sits at the centre of an octahedron. The X-ray structure reveals the details of this formation and experimental charge density analysis and density functional theory (DFT) calculations confirm that the carbon is surrounded by six atoms. Carbon is usually considered unique, it seemingly creates limitless molecular diversity among millions of molecules, but, as Yamamoto and colleagues explain, its behaviour is taught to new chemistry students as being understood in relatively simple bonding concepts. Heavier elements, such as phosphorus and sulfur have none of the molecular diversity and yet their bonding patterns seem so much more impenetrable, with their ability to commonly form hypervalent species, with more than four bonds. In contrast, examples of hypervalent carbon, outside the theoretical transition states of organic substitution reactions and their ilk have remained not only highly improbable, but the organic dogma suggested for a long time that isolating hypervalent carbon species would be impossible. Computational chemist Steven Bachrach of Trinity University in San Antonio, Texas, highlighted the Yamamoto work in his Computational Organic Chemistry blog, which accompanies his Wiley book of the same name. Bachrach points out that while the distances between the allenic carbon and the oxygen atoms are short, it is not a trivial question to ask whether there are bonds between the carbon and those oxygen atoms, or whether the atoms have simply been forced together like to like poles of a pair of bar magnets. Yamamoto and colleagues actually propose that other C...O intermolecular interactions with similar separations (about 2.6 to 2.8 Å) may also be examples of hypervalent carbon; there are some 2000 such structures in the Cambridge crystallographic database that fit this criterion. Perhaps there is a weak stabilizing interaction between the carbon and oxygen. "I look at the structure in this paper as more than just 'atoms jammed together'," says Bachrach, "The allene is bent to try to move things apart - yet the C...O distance is shorter than van der Waals distances, and there is a bond path in the experimental and theoretical densities." "So," he asks, "is this a 'bond'? Probably not. Is the carbon 'hexavalent'? Probably not. Is it 'hexacoordinate'? Probably yes." "I don't think most chemists would want to call the stabilizing interaction between the central C and O in this molecule a 'bond'," Bachrach told SpectroscopyNOW, "what we will need to develop, if there are other molecules like this, is a new term to describe this kind of weak interaction - analogous to the 'hydrogen bond', which is just a very short ion-dipolar interaction with greater strength than a typical interaction of this type. The ubiquity of the hydrogen bond demanded a name - perhaps the same will develop for something like what's going on in this molecule." Related links:
|
A hex on carbon
|
