How solid are the theoretical foundations for computational nanotechnology? While there can be no doubt that computational chemistry is a mature discipline, there are occasional surprises. ScienceDaily points to this news release from the University of Georgia “UGA researcher leads discovery of a new driving force for chemical reactions“:
New research just published in the journal Science [abstract] by a team of chemists at the University of Georgia and colleagues in Germany shows for the first time that a mechanism called tunneling control may drive chemical reactions in directions unexpected from traditional theories.
The finding has the potential to change how scientists understand and devise reactions in everything from materials science to biochemistry.
The discovery was a complete surprise and came following the first successful isolation of a long-elusive molecule called methylhydroxycarbene by the research team. While the team was pleased that it had “trapped” the prized compound in solid argon through an extremely low-temperature experiment, they were surprised when it vanished within a few hours. That prompted UGA theoretical chemistry professor Wesley Allen to conduct large scale, state-of-the-art computations to solve the mystery.
“What we found was that the change was being controlled by a process called quantum mechanical tunneling,” said Allen, “and we found that tunneling can supersede the traditional chemical reactivity processes of kinetic and thermodynamic control. We weren’t expecting this at all.”
What had happened? Clearly, a chemical reaction had taken place, but only inert argon atoms surrounded the compound, and essentially no thermal energy was available to create new molecular arrangements. Moreover, said Allen, “the observed product of the reaction, acetaldehyde, is the least likely outcome among conceivable possibilities.” …
“We knew that the rate of a reaction can be significantly affected by quantum mechanical tunneling,” said Allen. “It becomes especially important at low temperatures and for reactions involving light atoms. What we discovered here is that tunneling can dominate a reaction mechanism sufficiently to redirect the outcome away from traditional kinetic control. Tunneling can cause a reaction that does not have the lowest activation barriers to occur exclusively.”
Does this surprising result require a change in the approach to computational chemistry used for computational nanotechnology? Because quantum tunneling can cause an isolated molecule at 11 K to undergo an unexpected reaction, is it a necessary factor to consider in positional control and diamond mechanosynthesis (DMS)? I asked Robert A. Freitas Jr. (2009 Feynman Prize in Nanotechnology for Theory) for his reaction. He graciously forwarded this opinion from the Nanofactory Collaboration:
In the collective opinion of the Nanofactory Collaboration: This paper reports that in special cases, H tunneling can occur preferentially through a higher energy barrier that is narrower (in the direction of a relevant reaction coordinate) than through a lower energy barrier that is wider. For this particular molecule, it is observed that an H bonded to an O will tunnel at noticeable rates to an adjacent carbene but not from a C to an adjacent carbene. We expect that hydrocarbon structures containing carbenes should be at relatively low risk for this sort of transformation, and for the molecule shown we strongly suspect that the H tunneling wouldn’t happen if the O was linked to a methyl group. Furthermore, in DMS we generally try to avoid the use of carbenes in either tools or product structures because they’re so annoyingly reactive, and OH groups are rarely employed, so this result does not seem to add a significant complication to theoretical calculations involving positional control in diamond mechanosynthesis at cryogenic temperatures.
So quantum tunneling is an important phenomenon to keep in mind, but does not seem to complicate current theoretical work on diamond mechanosynthesis.