.
Angewandte
Communications
the high torsion angle within the
biphosphole unit, the p–p* transi-
tions observed in the UV/Vis spec-
tra of the PdII complexes 8a–c are
considerably blue-shifted com-
pared to those of the AuI com-
plexes 6a–c (for example: Dlmax
=
95 nm for 6c versus 8c; see
Table S1 in the Supporting Inform-
tion). These results clearly show
how it is possible to fully exploit
the biphosphole coordination
chemistry to control their torsion
Figure 3. Views of the X-ray crystal structures of a 2,2’-biphosphole derivative series (see the
Supporting Information for the other series).[11]
torsion angles are very high because of the tetrasubstituted
P atoms which impose a twisted conformation to minimize
steric congestion. These twisted conformations probably also
exist in solution because of an additional steric repulsion of
the methyl groups at the 3,4-positions of the P rings (4b,
Figure 3). This structural feature can account for the rather
high HOMO–LUMO gap of the thiooxobiphosphole deriv-
atives 4a–c. Complexes 6a–c are also isolated as a racemic
mixture, but the P-methyl substituents now occupy the axial
positions (6b, Figure 3). This conformation, which contrasts
to that of the thiooxophosphole derivative 4b is due to the
fact that two coordinated AuI ions are engaged in aurophilic
interactions[7] (6a: 2.9889(2) ꢀ, 6b: 3.050(4) ꢀ; 6c:
3.077(3) ꢀ). This metallophilic interaction results in the
formation of a C2P2Au2 six-membered rings, which display a
highly distorted chair conformation as a result of the long Au–
Au distance and the presence of bulky methyl substituents
(Figure 3). It is very likely that these six-membered rings exist
in solution since 1) the energy associated with aurophilic
interaction is about 10 kcalmolÀ1,[7] and 2) this conformation
minimizes the steric repulsion between the ring substituents.
Therefore, the metallophilic interactions lock the 2,2’-
diphosphole subunits in place and the torsion angles between
the two P rings are now fixed (6a: 51.38, 6b: 55.08; 6c: 54.28).
These torsion angles are lower than those recorded for their
thioxophosphole analogues and can explain why these two
series of derivatives exhibit different HOMO–LUMO gaps.
This use of metal coordination to control the torsion angle of
conjugated building blocks through metallophilic interactions
nicely illustrates the impact of the biphosphole moiety for the
molecular engineering of p-conjugated systems.
To further tune the torsion angle between the two central
P rings of the biphosphole subunit, its bidentate behavior
towards a PdII metal ion was exploited (Scheme 1, 8a–c). The
X-ray diffraction study of the complex 8a reveals a square-
planar metal center as part of a five-membered ring
(Figure 3). Note that here the racemic mixture is also
obtained and, as a result of the chelating behavior of the
biphosphole fragment, the methyl groups occupy the axial
positions. The formation of the five-membered ring upon
coordination on the PdII ion imposes a large torsion angle
within the biphosphole unit (87.78). Furthermore, the phosp-
holes rings are bent with respect to each other to help the
phosphorus atoms to accommodate the square-planar coor-
dination sphere of the PdII center (see Figure S2 in the
Supporting Information). Interestingly, in accordance with
angle using transition-metal atoms with specific properties,
and to impact the HOMO–LUMO gap of the corresponding
p-conjugated systems.
The versatility of this approach can be additionally
extended to using 2,2’-biphosphole building blocks with a
restricted rotational ability that results from a C3 alkyl chain
that is covalently bonded to the P atoms (derivatives 5, 7, and
9; Scheme 1). Replacing the methyl substituent on P with the
bridging C3 alkyl chain has almost no influence on the UV/vis
spectra of the corresponding p-conjugated systems that
incorporate either the s3-biphosphole or s4-thioxobiphosp-
hole building blocks (see Table S1 in the Supporting Infor-
mation). Indeed, the solid-state structure of the dithioxo-
phosphole derivative 6c reveals a torsion angle of 678
between the two P rings, a value that is within the range of
those recorded for the nonbridged compounds 4a–c. These
data show that the C3 chain does not lock the biphosphole
unit with a precise torsion angle and that a certain rotation
À
about the inter-ring C C is possible. In contrast to what was
I
À
observed within the P CH3 series, the Au complexes 7b,c
exhibit absorption maxima that are similar to the correspond-
ing dithioxophospholes 5b,c (see Table S1 in the Supporting
Information). The origin of this difference in behavior
between the bridged and nonbridged series (Scheme 1) was
rationalized by considering the solid-state crystal structure of
complex 7b. In this racemic mixture, which contains the
bridging C3 chain, the two AuI ions occupy the axial positions
and therefore cannot be engaged in aurophilic interactions
(Figure 3). The torsion angle between the two P rings (64.98)
is controlled by the C3 chain only and the P rings are not
locked. As a consequence, the HOMO–LUMO gaps of these
AuI complexes 7b,c are similar to those of the thioxobiphosp-
holes 5b,c, as observed for monophosphole-based p-conju-
gated systems. Indeed, the presence or absence of a covalent
C3 bridge between the P rings allows to change the relative
position of the AuI metal ions in the resulting complexes in
order to control the torsion angle within the biphospholes.
The trick to obtaining a biphosphole unit doubly locked
through both covalent bonding and metal coordination, was
to prepare the PdII complex 9b using the precursor 3b
(Scheme 1). Indeed, the X-ray diffraction study of 9b
revealed a torsion angle of 25.88, by far the smallest value
observed within the biphosphole series. For example, in the
related PdII complex 8a having P-CH3 moieties, this angle
reaches 86.78 (Figure 3). The consequence of this double-lock
approach (C3 chain and metal coordination) is a large red-
216
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 214 –217