Date: Wed, 6 Oct 1999 05:14:43 -0500 (CDT)
Precedence: bulk
From: Zhengyu Wu 
Subject: Paper of the week
MIME-Version: 1.0

Hello, everyone,

Here you will find the introduction of the paper for this week. The
attachment is the Word file. I would like to hear the comments from you
and answer your questions as much as I can.



                           Paper of the week 4

     The paper is from J.Phys.Chem. B 1999, 103, 7145-7150. The name of
the paper is "Competition between Wolff Rearrangement and 1,2-Hydrogen
Shift in b-Oxy-a-ketocarbenes: Electrostatic and Specific Solvent
Effects". The first author is Calvo-Losada from the Departamento de
Quimica Fisica.

     Stereoselective Wolff rearrangement of b-Oxy-a-ketocarbenes gives
1-alkyl-2-oxy synthons. This is a useful pathway for the synthesis of some
antibiotic macrolides. 1,2-hydrogen shift reaction is a competitive
pathway in this reaction process and it becomes dominant in the case of
b-hydroxyl-a-ketocarbenes. One can get 100% yield of the vinyl-ketone
product, which means the 1,2-hydrogen shift is 100% selective in this
case. The authors are interested in understanding the reason for this fact
and thus pursued a theoretical analysis to find the factors governing the
competition between Wolff rearrangement and 1,2-hydrogen shift processes.
The analysis is concentrated on the calculation of the activation barriers
of the two reaction processes (W-rearrangement and 1,2-H shift) because
the reactions and percent yields are kinetic-controlled. The quantum
mechanical calculations were carried out using the Gaussian 94 program
(which is what we will use in this course! :-) at B3LYP and HF levels
with 6-31G* and 6-31G** basis sets.

     Firstly, the activation Gibbs energies of the two reaction processes
in gas-phase were calculated. The structures of RKC, TSH and TSW (
ketocarbene, transition state of 1,2-H shift, transition state of Wolff
rearrangement respectively) were optimized, too. From the optimized
structures, one can tell easily that the reason that 1,2-H shift is
preferred is because of the H-bond between the hydroxyl group hydrogen and
the carbonyl oxygen which keep the rotation of  carbonyl group which is
required in Wolff rearrangement process. A 96% yield of vinyl-ketone
product was calculated from the Gibbs energy difference, which is

     Secondly, the authors tried to reproduce the experimental result
(100% yield for 1,2-H shift) in solution by adding the pure electrostatic
effect of water as solvent. The result is 60% yield, which means this
model failed.

     Then they came up with the association complexes model, which means
at this time they did not consider only the pure electrostatic effect of
water, but the association process of the each solute molecule with
several (here it is 3) water molecules as well. To find the reasonable
association complex, they first calculate the energies of various
complexes between one water molecule and the ketocarben intermediate to
find the most important so-called "specific water-ketocarbene interaction
sites". Then they built the association complex model with one ketocarbene
molecule and three water molecules because they found three important
interaction sites. Following this step is are the optimizations of the
RKC, TSH and TSW structures and  their energy calculations. The most
interesting result is an extra H-bonding stabilization in TSH.  The
percent selectivity of 1,2-H shift is 100% according to the energy
calculation of this model. By comparing the structures and energy values
between the association model and SCRF calculation, the authors came up
with their conclusion that specific H-bond interactions among the
keocarbene and the solvent molecules.

     Finally, the contribution of different energy components to the
relative solvent-solute ineraction energies in the association complexes
was examined by means of the generalized molecular interaction potential
with polarization (GMIPp) method. From the calculation results, we can
find that he electrostatic effect plays the most important role in
lowering the activation barrier of 1,2-H shift, while the van der Waals
interaction favors Wolff rearrangement. One can certainly expect a 100%
selectivity of the Wolff rearrangement in a non-polar solvent.

     The reason I chose this paper is that the idea of the author is clear
and straightforward. The calculation process is systematic and they used
the potential energy surface knowledge that we learned recently. It is
easy to understand why they performed optimizations and energy
calculations. The data analysis also makes sense. It is a good example on
how to solve a problem from the very beginning. We may not be very clear
on the exact meaning of different calculation models. But it is not the
point of the paper and should not affect our understanding of this paper.
We still do not know how to choose proper calculation methods for specific
cases. I believe we will learn it later this semester.