| Barry
K. Carpenter- Cornell University
Horace White Professor
Educational Background:
Ph.D., University College,
London, 1973
B.Sc., Warwick University, 1970
Awards:
- Alfred P. Sloan Research Foundation
Fellow
- John Simon Guggenheim Memorial Foundation
Fellow
- Alexander von Humboldt Senior Scientist
Award
- American Association for Advancement
of Science Fellow
- ACS Cope Scholar Award
- ACS James Flack Norris Award
Research Description:
Our research uses a combination
of theory and experiment to study fundamental questions of
reaction mechanism. On the theoretical side we use ab
initio electronic-structure calculations and quasiclassical
molecular dynamics simulations. The experimental work frequently
involves organic synthesis, followed typically by kinetics
studies, some of which are carried out in supercritical fluids.
The Influence of Nonstatistical Dynamics
on the Interpretation of Reaction Mechanisms
Figure 1
For many decades now the study of mechanism in organic chemistry
has, implicitly or explicitly, relied on models of reaction
kinetics that employ the so-called statistical approximation.
This approximation says that the redistribution of excess
vibrational energy (i.e. vibrational energy above the zero-point
level) in polyatomic molecules is so fast that for a collection
of molecules of a given kind at a given total energy, the
distribution of the vibrational energy can be treated as always
statistically distributed among the available vibrational
modes. Among the common theories that rely on this approximation
is Transition State Theory.
Application of the statistical approximation
made calculation of reaction rates feasible at a time when
fast electronic computers were not available. However, it
has also had a subtle but nonetheless profound effect on the
way organic chemists have thought about their reactions. Some
of the fundamental ideas of mechanistic organic chemistry,
for example that achiral intermediates can give only achiral
or racemic products, or that the product distribution from
an intermediate is independent of the way in which you make
it, carry as a hidden assumption the validity of the statistical
approximation.
Our work over the last several years
has caused us to question the general applicability of the
statistical approximation, particularly as it is applied to
reactive intermediates, and hence has led us also to question
some of these fundamental claims of mechanistic chemistry.
Our molecular-dynamics simulations have led to predictions
of behavior that would be indescribable by Transition State
Theory, such as the trajectory bifurcation illustrated in
Figure 1. Several different phenomena of this kind have been
discovered. The challenge, in each case, has been to design
an experiment that can clearly distinguish these unconventional
models of chemical reactivity from the more standard ways
of thinking. Such experiments frequently involve the synthesis
and kinetic investigation of isotopically labeled molecules,
such as those shown in Figure 2.
Figure 2
Chemistry Related to Hydrocarbon Combustion
Figure 3
Although the combustion of hydrocarbon fuels is conceptually
simple in its overall plan – the conversion of a hydrocarbon
to carbon dioxide and water with concomitant release of energy
– the details of how this occurs are enormously complex.
The combustion of even a relatively simple hydrocarbon may
involve thousands of elementary steps. Detailed knowledge
of the energetics of each of these steps would permit fuels
to be used with maximum efficiency and minimum pollution,
and so the Department of Energy sponsors research that is
designed to acquire that information.
Since almost all of combustion chemistry
involves free radicals, our approach has been to design, synthesize
and study precursors to key free radicals that are known or
believed to be important in combustion. In addition, we use
ab initio electronic structure theory to study the chemistry
of these radicals.
Sometimes the research that is initiated
with these goals in mind leads us into unexpected areas. A
case in point is illustrated in Figure 3 above. This is a
representation of the potential energy surfaces involved in
cyclopropyl iodide photolysis. The relevance to the project
is that a common way of generating free radicals is by photolysis
of organic halides, especially iodides. In a collaborative
effort with Professor Paul Houston and his students, we studied
the translational kinetic energy and angular scattering of
the fragments from cyclopropyl iodide photolysis, and were
able to show that the reaction must involve opening of the
three-membered ring in concert with C–I bond dissociation.
Calculations revealed that this probably occurred by a double
crossing from a radical-pair potential energy surface to an
ion-pair surface and then back again. Such nonadiabatic events
are of importance because they can cause unanticipated structural
rearrangements in the hydrocarbon fragment, leading to the
formation of radicals of unexpected structure.
Investigations of Nonadiabatic Thermal
Reactions
The kind of surface crossing discussed in the section on combustion-related
chemistry is common in photochemistry, but rarely invoked
for thermal reactions. However, we have come to believe that
the thermal generation of electronic excited states of molecules,
especially reactive intermediates, may be more common than
has been recognized. One case that we have studied in detail
concerns the cyclization of (Z)-1,2,4-heptatrien-6-yne. This
reaction is known to generate the a,3-didehydrotoluene biradical.
In methanol, this intermediate reacts by abstracting a hydrogen
atom from the methyl group. However, it has long been known
that in the same reaction some other intermediate is formed
that abstracts the O–H hydrogen from methanol, to generate
benzyl methyl ether. The identity of this second intermediate
has been tough to pin down. Now, after combined experimental
and computational work, we have come to the conclusion that
it is probably a zwitterionic excited state of the biradical,
as shown in Figure 4.
Figure 4
As this diagram implies, the zwitterionic state of the intermediate
is far above the biradical state when both are at their equilibrium
geometry. However, the activation energy for the cyclization
and the high potential energy of the reactant combine to raise
the lower surface to a point where it can touch the upper
one, in the vicinity of the transition state for the cyclization.
This touching of surfaces gives thermal access to the excited
state during the cyclization reaction. We believe that a similar
phenomenon may occur in several other quite different reactions,
and we are now conducting experiments and calculations to
find out whether we are right.
Selected Publications:
Reyes, M.B.; Lobkovski, E.B.; Carpenter,
B.K. Interplay of Orbital Symmetry and Nonstatistical Dynamics
in the Thermal Rearrangements of Bicyclo[n.1.0]polyenes. J.
Am. Chem. Soc. 2002, 124, 641.
Debbert, S.L.; Carpenter, B.K.; Hrovat,
D.A.; Borden, W.T. The Iconoclastic Dynamics of the 1,2,6-Heptatriene
Rearrangement. J. Am. Chem. Soc. 2002, 124, 7896.
Nummela, J.A.; Carpenter, B.K. Nonstatistical
Dynamics in Deep Potential Wells: A Quasiclassical Trajectory
Study of Methyl Loss From the Acetone Radical Cation. J. Am.
Chem. Soc. 2002, 124, 8512.
Carpenter, B. K.Nonexponential Decay
of Reactive Intermediates: New Challenges for Spectroscopic
Observation, Kinetic Modeling, and Mechanistic Interpretation.
J. Phys. Org. Chem. 2003, 16, 858.
Cremeens, M. E.; Carpenter, B. K. Access
to an Excited State via the Thermal Ring-opening of a Cyclopropylidene.
Org. Lett. 2004, 6, 2349.
Research:
Our research uses a combination
of theory and experiment to study fundamental questions of
reaction mechanism. On the theoretical side we use ab initio
electronic-structure calculations and quasiclassical molecular
dynamics simulations. The experimental work frequently involves
organic synthesis, followed typically by kinetics studies,
some of which are carried out in supercritical fluids.
Contact:
Barry K. Carpenter-
Cornell University
Office: 328 Baker Laboratory
Phone: (outside the University preceded by 1-607-25) 5-3826
Email: bkc1@cornell.edu
Back
to Top
|
