Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 34, 2001 8427
formation depends on curvature at the interface, we reasoned that
if we could add some volume to the core of the palladium
complexes, then perhaps we could control the curvature and
thereby induce cubic phase formation. We attempted this by
making palladium alkanoate complexes of the stilbazoles (5) and
found to our surprise that whether we employed mono-, di-, or
even trialkoxystilbazoles and whatever the chain length, only
nematic phases were seen!15 Thus, addition of these chains to
the smectogenic complexes did not modify the interface, rather
leading to the loss of microphase separation giving a nematic
phase. Similarly, addition of the alkanoate chains to columnar-
forming complexes might have been expected to reduce curvature
and lead to the observation of a cubic phase, rather than the
nematic phase found.
However, consideration of the structures of 2 and 5 shows that
while the alkanoate chains in the latter may have contributed to
the volume of the core, there is no proper structural comparison
between the two as one contains two lateral chains while the other
contains one. We therefore considered structural types which we
might use to realize complexes with only one lateral chain and
we decided upon monoalkyne complexes using Pt as the central
metal (6). We needed to use Pt as unsymmetric complexes of
this type would be much less likely to disproportionate, and in
any case, the mesomorphic behavior of the Pt complexes (4) is
very similar to that of the palladium congeners (3).
columnar phase; curiously the cubic phase is absent. However,
when we add a single lateral chain and a charged moiety (2), we
lose the SC phase, retain the columnar phase, and add the cubic
phase; loss of the smectic phase is almost certainly due to the
presence of the lateral alkyl sulfate chain. If we add two chains
to 3 to produce 5, then we find only nematic phases. However,
in 7 we retain the ionic function (not present in 5), but lose the
lateral chain (although there is little doubt that the triflate anion
will be associated closely with the silver cation). Finally, we
consider 6 where we have retained a single, lateral chain, but
lost the possibility of intermolecular electrostatic interactions and
here we find a nematic and a SC phase.
It has been speculated previously that intermolecular interac-
tions6,10,19 are important for cubic phase formation in calamitic
materials and this is consistent with the structures of the materials
which show these phases.7 However, it is only now with the
systematic structural variations which have been possible on
account of the presence of a metal center that we can say with
confidence that this is the case. Thus, while complexes 6 have a
somewhat more restricted geometry than the silver complexes
(1, 2, and 7), we believe that they provide a good structural
analogy minus the intermolecular electrostatic interactions, and
the discovery of a nematic phase rather than a cubic phase
supports the importance of these interactions. Of course, we also
find that while electrostatic interactions are clearly pivotal in these
systems, they are not the only factor of importance, although it
does seem that cubic phases are absent without them.10
The coupling of the acetylene to the Pt center was carried out
with CuI and an amine base, following procedures outlined by
Raithby.16 While, in principle, the synthesis of the complexes is
straightforward, we found that care was needed to ensure
monosubstitution rather than disubstitution. Short reaction times
seemed to offer the necessary control, although it was not possible
to define a single, optimum reaction time. Curiously, however,
we also found that the monosubstituted complexes were best
formed using a 3-fold excess of the acetylene. The compounds
For some time, the factors influencing the formation of this
rare phase in calamitic mesogens have been the subject of
discussion and we believe that we have produced important
evidence for a key factor, namely specific intermolecular interac-
tions. These interactions imply once more that aggregate formation
may be significant, suggesting a common factor with the purely
polycatenar systems described above.
1
were characterized by H and 13C NMR spectroscopy and by
Acknowledgment. C.M. and B.D. thank the EU for a Marie Curie
Fellowship and a Category 20 Fellowship, respectively.
elemental analysis, all of which were in agreement with the
proposed formulation.17 We were also able to obtain the disub-
stituted complexes by similar procedures using longer reaction
times and 6 equiv of acetylene.
JA015944I
(17) Complex 413 (n ) 4) (1 g, 1.09 mmol) was dissolved in a mixture
dichloromethane/diisopropylamine (10/1; 120 mL/12 mL) at room temperature.
Subsequent addition of copper(I) iodide (2%, 22 mg) was then followed by
addition of 1-pentadecyne (3 equiv, 853 µL). The solution was then allowed
to stir at room temperature for 10 min after which the solvents were evaporated
and the crude reaction mixture analyzed by 1H NMR spectroscopy to determine
the yield (94.5%); traces of the corresponding disubstituted complex were
detected. The monosubstituted complex was then purified by column chro-
matography. A 1/3 ethyl acetate/hexane mixture eluted remaining alkyne and
traces of disubstituted complex, while the monosubstituted complex was
recovered by eluting with pure ethyl acetate. It was then crystallized from
diethyl ether and finally recovered as a yellow solid by centrifugation.
We made a series of these complexes, but for the purposes of
this paper we need discuss only two, namely those for which
n ) 4 and 10 with, in each case, p ) 13. Pentadecyne was chosen
as the acetylene as we wished to make a close comparison with
the silver complexes where there is a 15-atom lateral chain
(O-S-O-C12H25). For complex 2 with n ) 4, only a cubic phase
is observed, while when n ) 10 both a cubic and a columnar
phase are seen.13 However, for 6 with n ) 4 and p ) 13, we
found a smectic C and a nematic phase (Cr‚130‚SC‚147‚N‚153‚
dec); the same mesomorphism was found for n ) 10 and p ) 13
(Cr‚88‚SC‚96‚N‚138‚I). For comparison and for the purposes of
discussion, it is also relevant to compare the properties of 7 where
no mesophase is found for n ) 4, while a cubic phase only is
found for n ) 10.18
δ
H (400 MHz; CDCl3) 6(n ) 4, p ) 15): 0.83 (t, 3J ) 6.7 Hz, 3H), 0.95 (2t,
3
3J ) 7.4 Hz, 12H), 1.47 (m, 32H), 1.78 (m, 8H), 2.34 (t, J ) 6.8 Hz, 2H),
3.98 (t, 3J ) 6.6 Hz, 4H), 4.01 (t, 3J ) 6.6 Hz, 4H), (t, 3J ) 6.7 Hz, 3H), 6.79
4
(l, AB, J ) 16.1 Hz, 2H), 6.82 (j, d, J ) 6.8 Hz, 2H), 7.05 (i, dd, Jig ) 1.7
3
4
Hz, Jij ) 6.8 Hz, 2H), 7.07 (g, d, Jgi ) 1.7 Hz, 2H), 7.21 (n, AA′XX′,
3
|Jno + Jno′| ) 6.9 Hz, 4H), 7.27 (k, AB, Jkl ) 16.1 Hz, 2H), 8.91 (o, |Jnm
+
Jon′| ) 6.9 Hz, 4H). δC (100 MHz; CDCl3): 13.8, 14.0, 20.9, 28.9, 29.3, 29.4,
29.6, 29.7, 30.6, 31.2, 31.3, 31.9, 22.6, 68.8, 69.1, 69.0, 91.2, 112.0, 113.4,
121.5, 121.6, 122.0, 128.6, 135.8, 146.8, 149.3, 150.7, 153.8. Elemental
analysis Calcd (Found): 6(n ) 4, p ) 15) C 62.9 (62.9); H 7.5 (7.6); N 2.5
(2.3); 6(n ) 10, p ) 15) C 72.3 (72.1); H 9.6 (9.8); N 1.7 (1.5).
(18) Donnio, B.; Bruce, D. W. New J. Chem. 1999, 23, 275.
(19) Donnio, B.; Rowe, K. E.; Roll, C. P.; Bruce, D. W. Mol. Cryst., Liq.
Cryst. 1999, 332, 2893.
Let us begin by recalling that the complexes 3 and 4 behave
as classical polycatenar mesogens showing both a SC and a
(15) Roll, C. P.; Donnio, B.; Weigand, W.; Bruce, D. W. Chem. Commun.
2000, 709.
(16) Adams, C. J.; James, S. L.; Raithby, P. R. Chem. Commun. 1997,
2155.