Chemists have found a chemical reaction that sidesteps its reaction coordinate and entirely bypasses transition state theory to display two distinct “roaming” mechanisms. The discovery could have implications for understanding the atmospheric photochemistry of nitrogen(VI) oxide as it dissociates into nitrogen(II) oxide and dioxygen. You can read my news report on the research in Chemistry World today.
I suppose the point is is that the difference between roaming and transition states is really just one of metaphor, it’s not as if molecules really do have hills to climb it’s simply about finding a way to picture the changes in their energies as atoms shuffle around from reactants to products…
I also asked Larrry Harding of Argonne National Laboratory in Chicago, Illinois, whose team made an early discovery regarding roaming reactions to comment. He sent me a rather detailed reply, which I think is worth publishing:
The relatively recently discovered phenomena of roaming radical reactions are proving to be very diverse in nature. Roaming radical mechanisms were first reported in the photodissociation of formaldehyde by van Zee et al(1) in 1993. They noted an anomaly in the product state distributions of CO produced in the photodissociation and suggested that this could be explained by two distinct mechanisms for the molecular dissociation. One mechanism passes through the well-known, tight saddle point while the other passes through regions in which one CH bond is greatly extended, allowing a nascent hydrogen atom to orbit around the HCO fragment and eventually abstract the remaining hydrogen atom to form molecular hydrogen. Townsend et al(2) confirmed this hypothesis in 2004 with a joint experimental/theoretical study. Shortly thereafter Kable and Houston(3) reported a second example of a roaming mechanism, the photodissociation of acetaldehyde. In this case a nascent methyl radical orbits around an HCO fragment and eventually abstracts the remaining hydrogen atom to form methane.
The roaming radical mechanism was initially thought of as defying transition state theory(4) since the roaming trajectories do not pass anywhere near the well known and well defined tight saddle points used in transition state theory. Later it was discovered(5) that these roaming mechanisms are associated with their own saddle points, which occur in parts of the potential surface where theorists had not previously thought to look for saddle points. The roaming saddle points connect two distinct regions of the potential surface, one being a bond cleavage region and the second a radical-radical, abstraction region. As a molecule begins to dissociate into two radicals it will pass through a region where the covalent bond is essentially broken but the radical fragments have not completely separated. This typically corresponds to a separation between the two radical fragments greater than ~3.5 Å. In this region the two radical fragments are free to reorient themselves with respect to each other and it is this reorientational motion that allows the radical fragments to pass through a roaming saddle point from the bond cleavage region of the potential surface to the abstraction region. It has recently been demonstrated(6) that one can use transition state theory to accurately model these roaming radical reactions.
In the last few years a number of other examples of roaming radical reactions have been reported. These include the dissociation of MgH2 (for which a full quantum dynamics treatment has been done(7), dimethyl ether(8) and most alkanes(9). We are indeed coming to the conclusion that these roaming mechanisms are quite common, having been overlooked only because their associated saddle points occur in parts of the potential surfaces that had not been well explored. Now in a recent Science article, Michael Grubb and colleagues have reported yet another new facet to these roaming reactions. In the dissociation of NO3 they find two distinct roaming mechanisms occurring on two different electronic surfaces. In most cases when two radicals, such as CH3 and HCO, add or abstract, there is only one reactive potential surface (asymptotically there are two degenerate potential surfaces but one of these is non-reactive). For NO3 the roaming fragments are an NO2 radical and an oxygen atom. It is well known that when an oxygen atom adds to a radical there are usually two reactive potential surfaces(10) that differ in the orientation of the doubly occupied oxygen lone pair with respect to the radical. Similarly when oxygen atoms abstract, there are again usually two reactive surfaces. What Grubb et al have shown is that when oxygen atoms roam, they can roam on either of these two reactive potential surfaces. This is an important discovery, not only for the case of NO3, which will have significant implications in atmospheric chemistry, but it also suggests that whenever oxygen atoms roam they will do so on two distinct potential surfaces.
References:
- van Zee, R. D.; Foltz, M. F.; Moore, C. B. J. Chem. Phys. 1993, 99, 1664.
- Townsend, D.; Lahankar, S. A.; Lee, S. K.; Chambreau, S. D.; Suits, A. G.; Zhang, X.; Rheinecker, J. L.; Harding, L. B.; Bowman, J. M. Science, 2004, 306, 1158.
- Houston, P. L.; Kable, S. H. Proc. Natl. Acad. Sci. 2006, 103, 16079.
- Bowman, J. M. Proc. Natl. Acad. Sci. 2006, 103, 16061.
- Harding, L. B.; Klippenstein, S. J.; Jasper, A. W. Phys. Chem. Chem. Phys. 2007, 9, 4055.
- Klippenstein, S. J.; Georgievskii, Y.; Harding, L. B. J. Phys. Chem. A. 2011, 115, 14370.
- Takayanagi, T.; Tanaka, T. Chem. Phys. Lett. 2011, 504,130.
- Sivaramakrishnan, R.; Michael, J. V.; Wagner, A. F.; Dawes, R.; Jasper, A. W.; Harding, L. B.; Georgievskii, Y.; Klippenstein, J. J. Combustion and Flame, 2011, 158, 618.
- Harding, L. B.; Klippenstein, S. J. J. Phys. Chem. Lett. 2010, 1, 3016.
- Harding, L. B.; Klippenstein, S. J.; Georgievskii, Y. 30th Symposium (International)
on Combustion 2005, 30, 985.