SCIENCE & TECHNOLOGY
EXPLAINING C-H BOND STRENGTHS Alternative to hyperconjugation invokes steric strain, but not everyone is buying the idea ELIZABETH K. WILSON, C&EN WEST COAST NEWS BUREAU
F
tance of repulsive interactions in molecular geometries during their careers. Gronert has devised an empirical model, based on the number and types ofgeminal in teractions, to calculate the heats of formation of alkanes and their radicals. The calculated values, and the corresponding bond disso ciation energies, closely match experimental values (J. Org. Chem. 2006,71,1209). It's a somewhat heterodox idea that Gronert hopes will stir up discussion among his theoretical organic chemist colleagues. And create a stir it has. "The work is a breath of fresh air in that it forces us to think about reasons for fun damental chemical facts," says theoretical organic chemist Andreas A. Zavitsas at Long Island University, in Brooklyn. "I have, for a long time, been convinced that hyperconjugation was not the expla nation for the differences in alkyl radical stabilities," Wiberg says.
ing hyperconjugation, he argues that as ROM THEIR FIRST ORGANIC more substituents are added onto a central chemistry course, students learn carbon, and the bulkier they become, the that the more substituted an alkane, the weaker the remaining more these groups will want to get out of C-H bonds, and the more stable each other's way. the molecule's corresponding radical. The More precisely, Gronert says, C - H bonds reason for these variations in bond strengths is are weakened by the steric strain between so well-accepted that few chemical educators two atoms or alkyl groups that are attached give it much thought. to the same carbon atom, a phenomenon he That explanation is hyperconjugation, in calls "geminal" repulsion. And when a C - H which hybrid orbitals on a neighboring C-H bond is broken, the release of the strain is or C-C bond overlap with a singly occupied OTHERS, HOWEVER, aren't convinced what leads to lower bond dissociation en ρ orbital on the central carbon, thereby deat all. "The fact that Gronert's set of pa ergies, he believes. Gronert says he doesn't localizing electrons and thus increasing the rameters can reproduce the energies of a dismiss hyperconjugation altogether, but radical's stability. limited number of hydrocarbons does not he doesn't believe it's responsible for the prove the validity of his assumptions," says But how do chemists know that that's effects in this case. Paul v. R. Schleyer, chemistry professor at really what's going on? Though they can Though hyperconjugation has largely the University of Georgia, Athens. "He measure bond dissociation energies, hy reigned supreme since it was proposed by does not consider the vast modern theo perconjugation itself can't be directly Nobel Laureate Robert S. Mulliken in the retical and computational support for observed. I hyperconjugation." It can be inferred from some experi Weston Τ Borden, chair of the mental results, such as electron spin | chemistry department at the University resonance spectra, and many chem of North Texas in Denton, is also wary. ists point to a great deal of theoretical I "The danger in drawing a conclusion evidence that supports it. In addition to I about physics from any empirical fitting bond strengths, hyperconjugation neatly scheme is that a fortuitous cancellation explains a host of other properties, from of errors can lead to a successful fit while the gauche effect to ethane rotation. So ignoring physical effects that may actu why mess with it? ally be present," he says. For a handful of chemists, hyper conjugation has some shortcomings "Replacing an explanation based they just can't reconcile. For example, on sound quantum mechanics with a although C - H bond dissociation ener numerical correlation seems like a step gies decrease as alkane systems become backward," agrees Kendall N. Houk, a more substituted, systems involving chemistry professor at UCLA. REPULSIVE Gronert's explanation for bonds between carbon and other at But chemistry professor Barry K. the variation in stabilities of C-H bonds in oms, such as halogens, do not. "This Carpenter of Cornell University ac alkanes, shown in this cartoon, invokes is something we've known for a long knowledges some of the limitations repulsive forces (represented by the red time, but it is literally ignored in most of hyperconjugation. "I think Scott is springs) between atoms attached to the same textbooks," says Robert G. Bergman, a making some important points," he atom. When a C-H bond breaks, the springs chemistry professor at the University of says. "But finding a definitive answer to and associated strain energy are released. California, Berkeley. what is responsible for bond energies is likely to be tough." These different trends imply that some 1930s, the idea that steric strain affects bond other effect—involving interactions in the Andrew Streitwieser, emeritus chemis stability isn't new. Gronert himself discov parent molecule and not just hyperconjuga ered a 1932 Journal'of'the Amencan Chemical try professor at UC Berkeley, under whom tion in the radical—must be at work. Society paper by noted theoretical chemist Gronert got his Ph.D., comments that Scott Gronert, a chemistry professor at chemists who think about this problem Henry Eyring that suggests that very notion. San Francisco State University, has now will be enriched by considering both in Others, such as emeriti chemistry professors proposed a different scenario to explain the terpretations. "We'll end up with a better Lawrence S. Bartell, at the University of variations in alkane C - H bond strengths. understanding that we'll be teaching to un Michigan, and Kenneth B. Wiberg, at Yale In a simple fashion, and without invok dergraduates in the future," he says. • University, also have advocated the impor WWW.CEN-0NLINE.ORG
C&EN
/ MARCH
6,2006
65
