Conformational and Electronic Engineering of Twisted Diphenylacetylenes

Article abstract of DOI:10.1021/ol035532+

(Formula presented). Three tethered diphenylacetylene derivatives were prepared by alkyne metathesis. In these cycles, the twist angle between the two benzene rings is variable and determined by the nature of the linker. The engineering of the twist angle

Full text of DOI:10.1021/ol035532+

ORGANIC
LETTERS
2003
Vol. 5, No. 21
3951-3954
Conformational and Electronic
Engineering of Twisted
Diphenylacetylenes
,†
Glen Brizius,†,‡ Kelvin Billingsley,‡ Mark D. Smith,‡ and Uwe H. F. Bunz*
School of Chemistry and Biochemistry, Georgia Institute of Technology,
Atlanta, Georgia 30332-0400, and Department of Chemistry and Biochemistry,
UniVersity of South Carolina, Columbia, South Carolina 29208
uwe.bunz@chemistry.gatech.edu
Received August 13, 2003
ABSTRACT
Three tethered diphenylacetylene derivatives were prepared by alkyne metathesis. In these cycles, the twist angle between the two benzene
rings is variable and determined by the nature of the linker. The engineering of the twist angle leads to a change of the UV−vis spectra of the
cycles. The larger the twist angle in the macrocycles, the more blue shifted their λmax (UV−vis), the lower their fluorescence quantum yield,
and the lower field shifted their 13C NMR signals of the alkyne carbons are.
This contribution presents the synthesis of diphenylacetylene
derivatives 12, 13, and 15 by ring-closing alkyne metathesis.
The optical properties of 12, 13, and 15 are dependent upon
the size of the twist angle R. Increasing the angle R leads to
larger HOMO-LUMO gaps and lower emissive efficiencies.
Diphenylacetylenes (DPA) are fascinating in their own
right and as building blocks for larger structures, oligo- and
poly(phenyleneethynylene)s. DPAs are rigid, and their only
degree of freedom is the rotation of the arene units around
the central alkyne. This rotational freedom has an impact
upon the optical and electronic properties of DPAs 1-6 as
well upon that of the poly(p-phenyleneethynylene)s (PPEs).1-6
The conformational change has been invoked to explain the
large solvatochromic and thermochromic effects that the
PPEs show and the change in fluorescence observed between
the pair 1a/2 and 2/6 upon binding to metal cations.1 While
the PPEs do show large effects in absorption, the DPAs seem
to show only a change in emission intensity, but not in their
absorption spectra. We found that curious and were interested
to “dial in” the twist angle R between the two arene rings in
DPAs to examine their electronic properties. There was only
sparse information regarding how to change the twist angle
in DPAs.7
(3) (a) Kim, J.; Levitsky, I. A.; McQuade, D. T.; Swager, T. M. J. Am.
Chem. Soc. 2002, 124, 7710-7718. (b) Kim, J.; Swager, T. M. Nature
2001, 411, 1030-1034; Nature 2001, 413, 548 (correction) (c) Zhou, Q.;
Swager, T. M. J. Am. Chem. Soc. 1995, 117, 12593-12603.
(4) Beeby, A.; Findlay, K.; Low, P. J.; Marder, T. B. J. Am. Chem. Soc.
2002, 124, 8280-8284.
† Georgia Institute of Technology.
‡ Univeristy of South Carolina.
(1) McFarland, S. A.; Finney, N. S. J. Am. Chem. Soc. 2002, 124, 1178-
1179.
(2) (a) Bunz, U. H. F. Acc. Chem. Res. 2001, 34, 998-1010. (b) Sluch,
M. I.; Godt, A.; Bunz, U. H. F.; Berg, M. A. J. Am. Chem. Soc. 2001, 123,
6447-6448. (c) Levitus, M.; Schmieder, K.; Ricks, H.; Shimizu, K. D.;
Bunz, U. H. F.; Garcia-Garibay, M. A. J. Am. Chem. Soc. 2001, 123, 4259-
4265. (d) Miteva, T.; Palmer, L.; Kloppenburg, L.; Neher, D.; Bunz, U. H.
F. Macromolecules 2000, 33, 652-654. (e) Halkyard, C. E.; Rampey, M.
E.; Kloppenburg, L.; Studer-Martinez, S. L.; Bunz, U. H. F. Macromolecules
1998, 31, 8655-8659.
(5) The rotational barrier of tolane is calculated to 0.86 kcal/mol utilizing
a B3PW91/6-311G** basis set: Seminario, J. M.; Zacarias, A. G.; Tour, J.
M. J. Am. Chem. Soc. 1998, 120, 3970-3974.
(6) Moore, J. S. Acc. Chem. Res. 1997, 30, 402-413. Nelson, J. C.;
Saven, J. G.; Moore, J. S.; Wolynes, P. G. Science 1997, 277, 1793-1796.
(7) (a) Crisp, G. T.; Bubner, T. P. Tetrahedron 1997, 53, 11881-11898.
(b) Crisp, G. T.; Bubner, T. P. Tetrahedron 1997, 53, 11899-11912.
10.1021/ol035532+ CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/16/2003

A series of bridged DPAs was checked by us via fast AM1
calculations. Xylylene linkers between the diphenylacetylenes
led to structures 12-14 with varying R when going from
12 to 14. Ab initio calculations of these topologies (RHF
6-31G**) refined the picture and showed that 13 is the least
twisted cycle (R ) 40°) and that both 12 and 14 are
considerably more distorted (Figure 1). While these calcula-
Figure 1. Computationally obtained (RHF 6-31G**, Spartan 2002
on a Windows platform) geometries for 12 (left), 13 (middle), and
14 (right). The R values are 58.82°, 39.12°, and 88.02°.
tions give gas phase structures, the 6-31G** basis set is
generally reliable for ground-state geometries. Noticeable was
the significant shift in the HOMO and LUMO energies upon
the twisting action in the three isomeric cycles (Figure 1).
While the HOMO-LUMO values do not directly correspond
to a physical property they qualitatively match changes in
the absorption spectra. An increasing angle R led to an
increased HOMO-LUMO gap according to the calculations.
Because these cycles are quite flexible, we performed an
energy profile calculation with respect to the rotation around
the CC triple bond. In 13 and 14, only one minimum is
observed (see Figure 1), but the conformational behavior of
12 is more complex. Two energy minima, one at 0° and one
at approximately 60°, are found (Figure 2a). According to
different levels of theory either one is somewhat more stable.
AM1 and RHF 6-31G** predict the twisted form of 12 to
be slightly more stable while B3LYP 6-31G** predicts the
planar form of 12 to be more stable. The two conformers
are separated by a barrier that was calculated to be between
4 and 7 kcal mol-1. As a comparison, the rotational barrier
of 1b was calculated (Figure 2b). All levels of theory show
that the planar syn form in which both methoxy groups are
Figure 2. (a) Rotational profile of 12 calculated on three levels of
theory. (b) Rotational profile of 1b calculated on three levels of
theory. (c) Rotational profile of diphenylacetylene calculated on
three levels of theory. In a-c, the geometry optimization was
performed on the AM1 level and the ab initio calculations were
single point calculations utilizing optimized AM1 geometries.
Abbreviations: HF, restricted Hartree-Fock; BLP, Becke-Lee-
Yang-Parr functional (B3LYP). x-axis: torsion angle (deg).
located at the same side of the molecule is 0.7-0.8 kcal
mol-1 less stable than the planar anti form, where both
methoxy groups avoid each other. The difference between
the calculations is in the height of the barrier, which is
nonexistent in the AM1 calculation and approximately 1 kcal
mol-1 in the DFT calculation.
All of the calculations were single-point energy calcula-
tions performed on optimized AM1 geometries. To asses the
3952
Org. Lett., Vol. 5, No. 21, 2003

precision of these calculations, the rotational barrier of
diphenylacetylene was computationally evaluated. This
system had been calculated5 earlier in its planar as well in
its 90° twisted conformation. The barrier of rotation was
calculated to 0.86 kcal mol-1.5 When utilizing AM1 opti-
mized geometries and the B3LYP 6-31G** method this
barrier is well reproduced with a value of 0.97 kcal mol-1.
The relative error utilizing AM1 geometries in combination
with single-point ab initio calculations is <0.1 kcal mol-1.
The use of AM1-calculated geometries is therefore reason-
able. The control calculation backs up the calculational results
obtained for 1b and 12-14.
matography on silica gel. Major byproducts are scarcely
soluble oligomers and polymers that could not be purified/
separated. Repeated attempts to close 10 to 14 under different
conditions failed with a variety of catalytic in situ systems.
To gain an understanding of the structures of the cycles,
we attempted to grow single-crystalline specimens. Cycle
13 is a colorless glass that resisted single-crystal growth.
The ortho cycles 12 and 15 formed single crystals from
hexafluorobenzene (12) and hexane/dichloromethane (15),
respectively. The use of hexafluorobenzene as a crystalliza-
tion solvent seems critical because it favors aromatic face-
to-face interactions.12-14 The structures of 12 and 15 are
shown in Figure 3a,b. There are three independent molecules
of 12 in the unit cell. Each of them has a different angle R
(22.4-26.6°). The energy potential for rotation is quite soft
for 12, but that is not too unexpected. In the case of 15,
only one twist angle (29.9°) is observed. The twist angle of
12 and 15 in the solid state is significantly smaller than
calculated, but this is a packing effect in combination with
the soft rotational profile in 12 and 15. While the X-ray data
are an important proof of the topology of the cycles, and
show a fascinating solid state ordering, they probably do not
correctly represent the conformation of the cycles in solution
(Figure 3).
The synthesis of 12, 13, and 15 starts (Scheme 1) with
the benzylation of 7 followed by propynylation utilizing a
Scheme 1. Synthesis of Twisted Tolanes
Figure 3. (a) ORTEP representation of 12; R ) 26.6°. (b) Ball
and stick representation of 15; R ) 29.9°.
The UV-vis spectra of 12, 13, 15, and 1b are shown in
Figure 4. While 1b has its λmax at 332 nm, the ortho cycles
12 and 15 show their λmax at 313-315 nm, almost 20 nm
blue-shifted. All of these spectra display a fine structure. The
meta cycle 13 features a broad and undistinguished UV-
vis spectrum that is intermediate between that of 1b and the
Pd-catalyzed coupling of the Sonogashira type.8 The bis-
ethers 8-11 are isolated in 28-56% yield. Alkyne metathesis
with either the Grela9 system or our own preactivated
variant10 of the Mortreux11 catalyst furnished the cycles 12,
13, and 15 in 18-19% isolated yield after repeated chro-
(11) Mortreux, A.; Blanchard, M. J. Chem. Soc., Chem. Commun. 1974,
786-787. (b) Kaneta, N.; Hikichi, K.; Mori, M. Chem. Lett. 1995, 1055-
1066. (c) Kaneta, N.; Hirai, T.; Mori, M. Chem. Lett. 1995, 627-628.
(12) (a) Weck, M.; Dunn, A. R.; Matsumoto, K.; Coates, G. W.;
Lobkovsky E. B.; Grubbs, R. H. Angew. Chem., Int. Ed. 1999, 38, 2741-
2744. (b) Coates, G. W.; Dunn, A. R.; Henling, L. M.; Dougherty, D. A.;
Grubbs, R. H. Angew. Chem., Int. Ed. Engl. 1997, 36, 248-251.
(13) Bunz, U. H. F.; Enkelmann, V. Chem. Eur. J. 1999, 5, 263-266.
(14) (a) Dai, C.; Nguyen, P.; Marder T. B.; Scott, A. J.; Clegg, W.; Viney,
C. Chem. Commun. 1999, 2493-2494. (b) Collings, J. C.; Roscoe, K. P.;
Robins, E. G.; Batsanov, A. S.; Stimson, L. N.; Howard, J. A. K.; Clark, S.
J.; Marder, T. B. New J. Chem. 2002, 26, 1740-1746.
(8) Bunz, U. H. F. Chem ReV. 2000, 100, 1605-1644. Sonogashira, K.
J. Organomet. Chem. 2002, 653, 46-49.
(9) Grela, K.; Ignatowska, J. Org. Lett. 2002, 4, 3747-3749.
(10) (a) Brizius, G.; Bunz, U. H. F. Org. Lett. 2002, 4, 2829-2831. (b)
Kloppenburg, L.; Song, D.; Bunz, U. H. F. J. Am. Chem. Soc. 1998, 120,
7973-7974.
Org. Lett., Vol. 5, No. 21, 2003
3953

Table 1. Spectroscopic Data of Cycles 12, 13, and 15
1b
12
13
15
315
UV-vis (CHCl3, nm)
ꢀ at λmax
fluorescence (nm)
quantum yield (CHCl3)
CH2Cl2
hexanes
ethyl ether
13C NMR alkyne shift,
4 ppm (1b)a
330
313
309
20 000
352
0.12
0.33
0.16
0.15
+1.20
17 300
350
0.69
0.77
0.66
0.69
+0.71
21 700
350
0.10
0.21
0.08
0.07
+1.64
348
0.45
0.75
0.54
0.55
0
a For the influence of mechanical strain on the 13CNMR spectral shifts
of alkynes, see: Lee, D. C.; Sahoo, S. K.; Cholli, A. L.; Sandman, D. J.
Macromolecules 2002, 35, 4347-4355. Cholli, A. L.; Sandman, D. J.; Maas,
W. Macromolecules 1999, 32, 4444-4446.
compound 1b has a fluorescence quantum yield of 45%. The
quantum yields differ considerably upon the solvent utilized,
but the trend is unchanged by the choice of solvents.
Coinciding with the quantum yields and the blue shift in the
absorption are the 13C NMR chemical shifts of the alkyne
carbons that deshield in the same way, reaching a maximum
of 1.64 ppm in 15. Again, the meta cycle 13 is the one that
resembles 1b most.
In conclusion, we have shown that the optical gap and
the fluorescence quantum yield in DPAs 1b, 12, 13, and 15
vary in dependence upon the twist angle of the two
constituting phenyl rings. A self-consistent picture emerges
from experimental and calculational results, in which an
increased torsion angle leads to a decreased fluorescence
quantum yield and an enlarged optical gap.
Figure 4. (a) UV-vis spectra (chloroform) of 1b and the cycles
12 and 13. (b) Emission spectra (chloroform, height normalized)
of 1b and the cycles 12 and 13. Emission of 15 is superimposable
to that of 12 and therefore not shown.
Acknowledgment. We thank the National Science Foun-
dation (CHE 0138659) and the USC NanoCenter for gener-
ous support. U.B. is a Camille Dreyfus Teacher-Scholar
(2000-2004).
other cycles. The fine structure in the spectra of 12 and 13
cannot be assigned to any specific vibration(s), and the
numeric value of the fine structure in 1b (2059 cm-1) might
be assigned to a CC triple-bond stretch. The fluorescence
spectra of 1b and 12, 13, and 15 are similar but their quantum
yields are significantly different (Table 1). The fluorescence
quantum yield (chloroform) of 13 is 69%, and 12 and 15
feature quantum yields in the range of 10-22%. The parent
Supporting Information Available: Experimental pro-
cedures and characterization for all compounds and CIFs for
12 and 15. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL035532+
3954
Org. Lett., Vol. 5, No. 21, 2003