G. A. Chass, D.-C. Fang, M. G. Organ et al.
additional steric interactions that may aid the expulsion of
the cross-coupling product, contrary to the accepted mecha-
the suspension was stirred until the solids dissolved. After this time, n-bu-
tylzinc bromide (0.8 mL, 1.0m in DMI, 0.8 mmol) and 3-phenyl-1-bromo-
propane (3) (0.5 mmol) were added. The septum was replaced with a
Teflon-lined screw cap under an inert atmosphere and the reaction was
stirred for 2 h at room temperature. After this time, GC/MS analysis was
conducted by removing a 2 mL aliquot of the reaction mixture and dilut-
ing with distilled hexane (1 mL). The resulting solution was then filtered
through a plug of celite and analyzed. GCMS analysis of product ratios
was conducted (in triplicate) on a Varian Series GC/MS/MS 4000 System
using undecane as an internal standard. The above experiment (for each
catalyst system) was conducted twice and the averaged yield is reported
in Table 1. The spectra obtained for products 4–7 were consistent with
nism (Scheme 1), which requires ZnBr to dissociate prior to
2
commencement of reductive elimination, and suggests that
these processes may not be discrete steps. The Pd–Zn inter-
action (1 =0.059 au, 2.426 ꢀ and 1 =0.058 au, 2.438 ꢀ, for
b
b
2
and 1, respectively) persists even after the departure of
the cross-coupling product. The adduct between the highly
0
electron-rich NHC–Pd and the Lewis acidic ZnBr is ob-
2
[31]
served as an additional intermediate in the cycle (CPL).
[
22]
that reported in the literature.
In conclusion, the first computational study of the com-
plete catalytic cycle for the Pd-mediated alkyl–alkyl Negishi
reaction revealed that the bulky NHC ligands introduce im-
portant differences into the traditionally accepted mecha-
nism. First, transmetalation and not oxidative addition,
which is considered to be slow in comparison with that for
[
33]
Computational methods: The Gaussian 03 (G03) program package was
used for all computations in this work. Analytical frequencies were com-
puted on the geometry-optimized structures to ensure the identity of
each structure as residing at minima or first-order saddle points on the
PEHSs of the systems. Zero-point energy (ZPE) and the thermodynamic
parameters required to quantify the free energy (DG) of each structure
were also extracted from these frequency evaluations.
[18]
aryl halides, was found to be the rate-determining step for
the whole cycle. Second, the inorganic salt by-product
An established numeric and modular description of all molecular struc-
[
34]
tures was used in the generation of inputs for geometry optimizations.
(
ZnBr ) was found not to dissociate prior to reductive elimi-
2
This allowed for novel Perl and shell-based UNIX/Linux scripting to be
continuously built-up and refined, to facilitate file construction, job sub-
mission, data storage/extraction, tabulation and automation. Accurate in
silico determinations are highly dependent upon accurate initial starting
structures.
nation, as suggested in the traditional mechanism. This in-
creased steric crowding during transmetalation helps to re-
lease the product (n-butane). The complex between the
0
highly electron-rich IPr–Pd and the Lewis acidic ZnBr per-
2
Geometry optimizations were carried out with density functional theory
sisted even after the cross-coupled product was formed.
Third, four (iPr)H atoms from the two N-(2,6-diisopropyl-
phenyl) substituents in 1 were found to form fleeting weak
interactions with Pd that appeared and disappeared during
the various stages of the cycle, depending on the steric
crowding around Pd. We propose that these interactions fa-
cilitate the molecular mechanism that results in the high ac-
tivity of IPr-ligated 1. They change the free energy land-
[23a,c]
[24]
(
DFT) using the B3LYP
method, employing the DZVP basis set
with a self-consistent reaction field polar continuum (solvent) model
[
28a–c]
(SCRF-PCM=THF).
In regard to the atoms-in-molecules (AIM) approach, critical points
CPs) of rank 3 were identified in the electron densities; these include
(
bond critical points (BCPs), ring critical points (RCPs), and cage critical
points (CCPs). The existence of a BCP between two atoms in an equilib-
rium molecular geometry is the necessary condition for two atoms to be
defined as being bound to one another. The pairs of gradient paths that
originate at a BCP and terminate at neighboring neclei, define a line
through which electron distribution, 1(r), is a maximum with respect to
any lateral displacement.
0
scape through increasing the entropy term (ꢁTDS ) in the
free energy, destabilizing intermediates more than transition
states, leading to higher turnover frequency and coordina-
tively saturate Pd throughout the cycle. Although the exact
molecular mechanism by which the structure of the IPr
ligand helps steer the course of the reaction to cross-cou-
pling or b-hydride elimination remains unsolved at this time,
this study has shed light on some of the ligand effects gov-
erning the product ratio resulting from the two competing
pathways. Finally, it is important to note that the use of
truncated computational models, for example, using the di-
methylimidazolium analogue of diarylimidazolium-based 1,
failed to reveal any of the aforementioned subtle interac-
tions between the isopropyl substituents of the NHC ligand
and the reacting metal center that are responsible, and in
fact critical for directing 1 down a productive path in the
B3LYP/DZVP wavefunctions, generated with G03, were used to obtain
[35,36]
molecular graphs and BCP properties using the AIM2000 program.
Acknowledgedments
We thank GIOCOMMS and Project 985 (China) for computational sup-
port, ORDCF (Ontario, Canada), NSERC (Canada), CAFMaD (Wales,
UK), and the National Natural Sciences Foundation of China (20773016)
for funding.
[
1] Metal-catalyzed cross-coupling reactions, 2nd ed. (Eds.: A. De Mei-
jere, F. Diederich), Wiley, New York, 2004.
[32]
catalytic cycle.
3] Handbook of Organopalladium Chemistry for Organic Synthesis
[
(
Experimental Section
[
[
Cross-coupling experiments: The following cross-coupling protocol was
used for both catalyst 1 (Pd–PEPPSI–IPr) and 2 (Pd–PEPPSI–IXy). In
air, a vial was charged with Pd–PEPPSI–IPr (3.4 mg) or Pd–PEPPSI–IXy
[6] a) N-Heterocyclic Carbenes in Transition Metal Catalysis (Ed.: F.
Synthesis (Ed.: S. P. Nolan), Wiley-VCH, Weinheim, 2006; d) W. A.
(
1
3.0 mg) (1 mol%) and under an inert atmosphere, LiBr (139.0 mg,
.6 mmol), and a stir bar were added. The vial was then sealed with a
septum and purged with argon after which THF (1.6 mL) was added and
4286
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 4281 – 4288