Angewandte
Communications
Chemie
Thus, a Finkelstein-type reaction involving 5 and excess LiBr
tris(chlorido) counterpart 5 can be rationalized by the
in acetone was carried out to access the tris(bromido)
complex 7 (Scheme 3). This reaction was found to be quite
retarded and complex 5 was almost completely recovered
increased steric bulk of Br versus Cl, which is expected to
induce a fast reductive elimination in the former case.
IV
The decent thermo stability of Pd complexes 5 and 7
1
after stirring for 12 h, as indicated by H NMR spectroscopy.
allowed us to grow their single crystals under ambient
conditions and attain their solid-state structures. The molec-
ular structures of 4, 5, and 7 are shown in Figure 2. A
II
In sharp contrast, Pd complex 4 can undergo a clean ligand
substitution yielding 6 under the same conditions. This is no
6
surprise since octahedral d complexes are expected to be
8
kinetically more inert than square-planar d counterparts.
IV
However, the target tris(bromido) Pd complex 7 can be
II
accessed through oxidative addition of Pd mono-bromido
complex 6 with Br or CuBr (Scheme 3 and Table 1). Both
2
2
reactions are clean and fast, and the product can be obtained
in very good yields. This reactivity is in sharp contrast to
Kraftꢀs bis(NHC) compounds (II, see above), for which
IV
accessing a Pd bromido complex was unsuccessful. To the
best of our knowledge, complex 7 is the first carbene-
IV
supported Pd bromide.
The resonance signal for the carbene carbon atom of 7 was
found at 152.3 ppm, which shows a marked upfield shift (Dd =
5
.8 ppm) compared to that for tris(chlorido) analogue 5
(
Table 2). This upfield shift can be rationalized by the
Table 2: Comparison of the resonances for the carbene carbon atoms of
–7.
4
Complex
13Ccarbene NMR signal [ppm][a]
Figure 2. Solid-state structures of 4, 5, and 7, ellipsoids set at 50%
probability. Hydrogen atoms and solvent molecules have been omitted
for clarity. Selected bond lengths [ꢀ] and angles (deg): For 4: C7ꢀPd1
4
6
(X=Cl)
(X=Br)
164.4
163.8
2
.053(4), C13ꢀPd1 1.938(4), C14ꢀPd1 2.062(4), Cl1ꢀPd1 2.3996(9);
C13ꢀPd1ꢀC7 78.94(14), C13ꢀPd1ꢀC14 79.04(15). For 5: C1ꢀPd1
1
.967(5), C5ꢀPd1 2.051(4), Cl1ꢀPd1 2.3146(13), Cl2ꢀPd1 2.4199(12),
5
7
(X=Cl)
(X=Br)
158.1
152.3
Cl3ꢀPd1 2.3185 (13), C5ꢀPd1 2.051(4); C1ꢀPd1-C5 79.59(12). For 7:
Pd1ꢀBr1 2.4564(5), Pd1ꢀBr2 2.5472(5), Pd1ꢀBr3 2.4569(5), Pd1ꢀC1
[
b]
1
.965(4), Pd1ꢀC6 2.059(3), Pd1ꢀC6 2.059(3); C1ꢀPd1ꢀC6 79.70(9).
[
[
a] Measured in CDCl and referenced to the solvent signal at 77.16 ppm.
3
comparison of the key bond parameters revealed an elonga-
b] Measured in [D ]DMSO due to the decreased solubility of complex 7
6
II
tion of the PdꢀC and PdꢀCl (trans to aryl) bonds from Pd
in CDCl , and referenced to the solvent signal at 39.52 ppm.
aryl
3
IV
complex 4 to Pd counterpart 5. For example, PdꢀC bond
aryl
distance increases from 1.938(4) ꢁ (for 4) to 1.967(5) ꢁ (for
5). This can be rationalized by the steric effect since the
binding of two additional Cl ligands from 4 to 5 makes the
enhanced electron density (that is shielding effect) of Br
versus Cl, and a similar trend was previously reported for
III
[15]
Au carbene complexes. When complex 7 was kept in the
solid state under ambient conditions for weeks, only slight
decomposition back to 6 was observed. Heating the solution
of 7 in CDCl3 at 508C for 12 h gave 6 as the major
decomposed product in a yield of about 25% (Supporting
Information, Figure S5). This implies the reversibility of the
reaction of 6!7 and similar reactivity was observed for Foutꢀs
coordination sphere more sterically demanding. Finally, the
IV
Pd ꢀC
distances of 5 and 7 are longer compared to those
MIC
IV
[4a,b]
in a handful of other Pd –NHC complexes.
Previously, Kraft et al. have used Pd –bis(NHC) com-
plexes (see above) to study their stoichiometric and inter-
IV
[4b,c]
molecular chlorine transfer reactions to olefins.
To com-
IV
pare, the pincer-type Pd –MIC complexes 5 and 7 reported
herein were treated with styrene to test their reactivity in such
transformations (Scheme 4). Tris(chlorido) complex 5 was
found to be inert, while vicinal dibromide 10 was obtained in
a yield of 52% in the case of tris(bromido) complex 7
(Supporting Information, Figure S6). Furthermore, no Br-
substituted product was observed, while such halogenated
alkene species was found in Kraftꢀs case. Interestingly, when
trans-stilbene was used as the olefin substrate, the reaction
gave an intractable product mixture including 6, but no
corresponding vicinal dibromide was observed as indicated by
IV
[16]
[13a,17]
Ni tris(halido) complex.
Notably, van Koten
and
[18]
others
molecules to [NCN]-type pincer complexes [Pt X(NCN)]
X = halide). It was found that for X = Cl or Br, the resulting
have studied the oxidative additions of halogen
II
(
IV
Pt species is stable enough and no reverse reactions back to
Pt precursors were observed. In contrast, in the case of X = I,
the reaction stops at an “intermediate” stage, giving an
interesting complex [Pt I(NCN)(h -I )]. Similar “intermedi-
ate”-type complex was not observed in our case. Finally, the
decreased stability of tris(bromido) Pd complex 7 versus
II
II
1
2
IV
Angew. Chem. Int. Ed. 2019, 58, 1 – 5
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!