Communications
that for 2 the dimer 2a remains an important constituent (K =
An important question at this point pertains to the actual
structure of the dimer. Both 1a and 2a have no vacant
coordination site and should be unable to complex 4. We
therefore propose the half-open structure 1’a as the species
responsible for epoxide binding to 1 (Scheme 1),with a
1000mÀ1),but to a noticeably lesser extent. The presence of
the bulkier ligand in 3 makes the monomer 3b the only
species detectable by cyclic voltammetry (K < 100mÀ1,as
deduced from digital simulations; see Supporting Informa-
tion). For 2-mm Zn-reduced solutions of 1, 2,and 3,this
implies that the dimer/monomer ratio is 1.5,0.8,and < 0.15,
respectively.
Kinetic analysis of the opening of 4[2a] with Zn-reduced
solutions of 1, 2,and 3 was then carried out (Table 1). The rate
Scheme 1. Dimeric (1a or 1’a) and monomeric species (1b) present in
Zn-reduced THF solutions of 1.
Table 1: Rate constants k [mÀ1 sÀ1] of the reductive ring opening of 4 with
Zn-reduced THF solutions of 1–3.
similar structure in the case of 2. Complex 1’a is a highly Lewis
acidic intermediate in the formation of 1a according to the
principle of activation of electrophiles through dimeric
association.[6]
catalyst
k(monomer)
k(dimer)
1
2
3
0.5
1.3
0.8
1.4
3.9
–
On the basis of the above experimental results,we turned
our attention towards calculations to gain further insight into
the reaction mechanism. Activation and reaction energies
were determined along with the structures of all pertinent
intermediates and transition states by density functional
theory (DFT) calculations[7] with the BP functional and a
TZVP basis set. The [Cp2TiCl]-derived complexes,which are
valuable models for higher-substituted titanocenes,were
investigated first. As model compounds for the epoxide we
focused on the simpler propene oxide (5) and isobutene oxide
(6) rather than 4.
The results concerning the binding of 5 and 6 with
[Cp2TiCl] to afford the corresponding complexes 7 and 8 (see
Table 2) proved to be interesting as there is a binding-energy
difference in favor of 7 by about 4 kcalmolÀ1. The calculations
show that this is because the hydrogen atom at the substituted
carbon atom of the epoxide part in 7 cannot be replaced by a
second methyl group (giving 8) without causing geometrical
changes: For steric reasons the epoxide has to be rotated by
908 in 8 compared to 7 in order to enable complexation by
constants, k,were extracted by monitoring the disappearance
of the TiIII species with a UV dip-probe in the presence of an
excess of 4 (see Supporting Information). As in the synthetic
work, 1,4-cyclohexadiene (1,4-C6H8) was added in order to
chemically reduce the radical intermediate formed upon
electron-transfer-mediated ring opening. The kinetics were
unaffected by the concentration of 1,4-C6H8. Thus,electron
transfer is the rate-controlling step of the overall reduction.
This renders any mechanism with a quick and reversible ring
opening before radical trapping unlikely.
To establish the nature of the reductant,with specific
emphasis on the influence of the monomer/dimer distribu-
tion,kinetic traces were recorded at different Ti
III
concen-
trations in the range 1–8 mm. In the case of 3 no effect was
observed on the extracted pseudo-first-order rate constants,
thus indicating that the monomer is the only reducing agent.
This is consistent with the cyclic voltammetry studies,where
no dimer constituent could be detected. Also,the reaction
rate is relatively low in this case (k = 0.8 mÀ1 sÀ1),which can be
attributed to larger steric constraints during substrate binding
or in the transition state (TS). In contrast,in the case of 1 and
2 the Zn-reduced solutions contain appreciable amounts of
the dimer,and the reaction rate is strongly affected by the Ti III
concentration. This was interpreted as though both monomer
and dimer have a measurable reactivity towards 4 under these
conditions,in agreement with the kinetic analysis of the
reaction between Zn-reduced 1 and benzaldehyde or benzyl
chloride.[5]
À
titanium. Despite this structural change the Ti O bond length
in 8 is slightly shorter than in 7. Illustrations of complexes 7
and 8 are given in Figure 1.
All activation energies, DE°,are in the range 7.0–
9.4 kcalmolÀ1,thus indicating that radical generation should
be facile at room temperature,in agreement with the
experimental results. It should also be noted that DE° for
the formation of 7a and 8a is lower than for 7b and 8b,and
that the higher-substituted radicals are the thermodynami-
By numerically fitting the decay curves pertaining to the
concentration of TiIII (see Supporting Information) we
established that the dimers (k = 1.4 and 3.9 mÀ1 sÀ1 for 1 and
2,respectively) open 4 faster than the corresponding mono-
mers (k = 0.5 and 1.3 mÀ1 sÀ1). For an employed concentration
of TiIII of 10 mm for instance,this implies that 84% and 75%
of 4 will be opened by the dimers,respectively.
Figure 1. DFT-calculated structures of 7 and 8.
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2041 –2044