Complementarity in bimolecular photochromism

Article abstract of DOI:10.1039/b502183a

Irradiating 2,3,6,7-tetraphenylanthracene in the presence of 9,10-dimethylanthracene leads to exclusive formation of the cross-dimer. No photochemical reaction is observed when either of these chromophores is irradiated in the absence of the other. The Roval Society of Chemistry 2005.

Full text of DOI:10.1039/b502183a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Complementarity in bimolecular photochromism{
David Bailey and Vance E. Williams*
Received (in Bloomington, IN, USA) 14th February 2005, Accepted 17th March 2005
First published as an Advance Article on the web 12th April 2005
DOI: 10.1039/b502183a
Sterically-hindered anthracene derivatives, such as 9,10-dimethy-
Irradiating 2,3,6,7-tetraphenylanthracene in the presence of
9,10-dimethylanthracene leads to exclusive formation of the
cross-dimer. No photochemical reaction is observed when
either of these chromophores is irradiated in the absence of the
other.
lanthracene (1), exhibit only a limited tendency to homodimerize.15
However, this compound reacts quite readily with anthracene
derivatives that lack functional groups at the 9- and/or
10-positions.12 Unfortunately, since anthracene and its
9-substituted derivatives tend to form homodimers, the reactions
involving these compounds still exhibit poor specificity.16 In an
effort to design a suitable partner for 1, we decided to target
an anthracene derivative that lacked groups at the 9- and
10-positions, but substituted bulky groups at the peripheral 2-,
3-, 6- and 7-positions. It was reasoned that a substitution pattern
of this type would inhibit homodimer formation while permitting
reaction with 9,10-disubsituted derivatives such as 1. A similar
strategy has been successfully employed to attenuate the
deleterious effects of self-association on the photophysical proper-
ties of conjugated polymers while still allowing interactions with
small molecule analytes.17 Since phenyl groups have been used
extensively as ‘‘insulation’’ for both reactive functionalities18,19
and nanostructured materials,20–23 it was thought that 2,3,6,7-
tetraphenylanthracene (2) would possess sufficient lateral bulk to
inhibit homodimerization.
Photochromism is perhaps one of the most powerful methods for
reversibly altering molecular and macroscopic properties by the
application of an external stimulus.1 Indeed the ability to
photochemically toggle molecules between two or more states
has been exploited in contexts ranging from information storage2,3
to enzyme activity modulation.4,5 In the vast majority of cases,
switching is brought about by the unimolecular isomerization of
molecules such as azobenzene or diarylethene derivatives.6
Molecular switches based on bimolecular photochromic systems
have received far less attention, despite them offering the distinct
advantage of being able to bring together two molecules in a
reversible fashion. Thus, for example, the [4 + 4] cycloaddition of
anthracene has been used in the assembly7 and cross-linking8 of
polymers, in the creation of dynamic receptors,9 and as a means
of attenuating magnetic interactions.10,11
Unfortunately, these photocycloaddition reactions suffer from
poor specificity. For example, irradiation of a mixture of two
anthracene derivatives A and B generally leads not only to the AB
product, but also to the homodimers AA and BB.12 This tendency
to form complex mixtures severely restricts the usefulness of
existing bimolecular photochromic systems. For example, recent
attempts to create reversible diblock copolymers from anthracene-
terminated segments were hampered by the accompanying
formation of the unwanted AA and BB photoproducts.13
Identifying pairs of chromophores that exclusively form AB
cross-dimers would therefore greatly extend the range of materials
that could be reliably and reversibly assembled using light. To
this end, we have set out to design pairs of complementary
anthracene derivatives that selectively form cross-dimers rather
than homodimers.14
Compound 2 was synthesized by LiAlH4–AlCl3 reduction24 of
the corresponding tetraphenylanthraquinone 3, which itself was
recently synthesized in our laboratory from 3,4-diphenylthio-
phene.25 The conversion of 3 to 2 under these conditions was
accompanied by the formation of an approximately equimolar
quantity of the over-reduced product 2,3,6,7-tetraphenyl-9,10-
dihydroanthracene. We found that this impurity could be cleanly
rearomatized to 2 by treating it with oxygen in the presence of
catalytic palladium.26 Rather than separating the products of the
LiAlH4–AlCl3 reduction, the crude product mixture was simply
heated in the presence of 10% Pd–C under aerobic conditions to
afford 2 as the major product (Scheme 1). This extra step not only
improved the overall yield of 2, but also obviated the need to
separate it from the over-reduced dihydroanthracene.
In order to investigate the photochemical behavior of this
compound, a benzene solution containing 4 mmol of 2 was
exposed to near-UV light.27 Gratifyingly, no reaction was
observed, even after 140 min irradiation. This compound
{ Electronic Supplementary Information (ESI) available: Full synthetic
and analytical details of 2 and 4; H NMR spectra of 4 before and after
1
heating at 150 uC; absorption spectra of compounds 1, 2 and 4. See http://
www.rsc.org/suppdata/cc/b5/b502183a/
*vancew@sfu.ca
Scheme 1 Reagents and conditions: (a) 10 equiv. LiAlH4, 4 equiv. AlCl3,
THF, reflux 21 h; (b) 10% Pd–C, mixed xylenes, reflux 5 d.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 2569–2571 | 2569

View Article Online
was also photochemically unreactive in acetonitrile, indicating that
its failure to dimerize can be traced to the effectiveness of its phenyl
groups at inhibiting self-association rather than an effect relating
to solvent polarity.12 Similarly, compound 1 did not react under
these conditions, whereas anthracene was converted to dianthra-
cene in 78% yield.
Although compounds 1 and 2 failed to react in isolation,
irradiation of an equimolar mixture of them for 140 min under the
same conditions afforded a single product in 65% yield (Scheme 2).
This product showed a high degree of thermal stability and could
be isolated from the starting materials by either column
chromatography or recrystallization.
1
Both the H and 13C NMR spectra of this photoproduct are
consistent with cross-dimer 4. Of particular note is the presence of
two singlets in the 1H NMR spectrum at 4.1 and 2.2 ppm,
corresponding to the dimer’s bridgehead protons and methyl
groups respectively. These chemical shifts are almost identical to
those of the head-to-tail dimer of 9-methylanthracene,28 as one
would expect given their similar chemical environments (i.e. a
bridgehead proton adjacent to a methyl group). This assignment
was further confirmed by the observation of a strong nuclear
Overhauser effect between these two sets of protons, indicating
that they are in close proximity to each another. Characterization
of this compound by mass spectrometry posed a greater challenge,
since dianthracenes tend to fragment into their component
monomers when ionized.29 Thus, the MALDI–TOF spectrum of
4 exhibited only peaks corresponding to the two monomers, while
the FAB spectrum showed a weak molecular ion peak for the
dimer, in addition to a much more intense peak for compound 2.
Having demonstrated that 1 and 2 selectively form the cross-
dimer, we next examined the reversibility of this reaction—since
anthracene dimers are known to revert either thermally or
photochemically to their component monomers.29 As is typical
for dianthracenes, the UV–visible absorption spectrum of
compound 4 is blue-shifted relative to both monomers 1 and 2,
with a lmax of 259 nm. Exposing a sample of 4 to 252 nm light30
caused the characteristic anthracene peaks of 1 and 2 to appear in
the UV–visible spectrum. These peaks became progressively more
intense with longer exposure times (Fig. 1), and photoreversion
was 56% complete after 90 min.
Fig. 1 The UV–visible absorption spectra after 0, 5, 15, 40, 60 and 90 min
of a 0.04 mM solution of 4 in CH3CN irradiated with 252 nm light.
This suggests that it is the number of bridgehead methyl groups
that limits the thermal stability, and that the peripheral phenyl
groups do not appreciably destabilize the photodimer. The slow
decomposition of this compound at low temperatures indicates
that dimers of this type could be used to construct relatively robust
structures.
In conclusion, we have demonstrated the viability of a new
strategy for enforcing cross-dimer formation in [4 + 4] photo-
cycloaddition reactions by a judicious choice of substituents on the
two anthracene derivatives. The resulting photoproduct is
thermally stable at moderate temperatures and can be reconverted
to its constituent monomers either photochemically or at elevated
temperatures. We are currently investigating the scope and
limitations of this strategy for enforcing selective cross-dimeriza-
tion, as well as examining its potential for controlling the
reversible, light driven assembly of materials.
The authors gratefully acknowledge the Natural Sciences and
Engineering Council of Canada (NSERC) and Simon Fraser
University for funding, Mr. Mikki Yang for carrying out
microanalyses and Professor Neil Branda for helpful discussions.
David Bailey and Vance E. Williams*
Department of Chemistry, Simon Fraser University, 8888 University
Dr., Burnaby, British Columbia, Canada V5A 1S6.
E-mail: vancew@sfu.ca; Fax: 1 604 291 3765; Tel: 1 604 415 3481
Compound 4 was found to possess considerable thermal
stability at room temperature; indeed, no decomposition of it
was observed at temperatures below 100 uC. The thermal reversion
of this compound was more thoroughly investigated by 1H NMR
(Fig. S2{). After heating a sample of 4 in DMSO-d6 at 135 uC for
15.5 h, only 15% of the dimer had been converted to monomers 1
and 2. At 150 uC, approximately 50% of 4 had been consumed
after 70 min. The reaction was approximately 93% complete after
150 min at the same temperature. This rate of fragmentation is
similar to that of the head-to-tail dimer of 9-methylanthracene,28
which also has two methyl groups at the bridgehead positions.
Notes and references
1 Molecular Switches, ed. B. L. Feringa, Wiley–VCH, New York, 2001.
2 R. van Delden, M. van Gelder, N. Huck and B. Feringa, Adv. Funct.
Mater., 2003, 13, 319–324.
3 M. Maesri, F. Pina and V. Balzani, in Molecular Switches, ed. B. L.
Feringa, Wiley–VCH, New York, 2001, pp. 309–337.
4 C. Dugave and L. Demange, Chem. Rev., 2003, 103, 2475–532.
5 I. Willner, Acc. Chem. Res., 1997, 30, 347–356.
6 H. Tian and S. Yang, Chem. Soc. Rev., 2004, 33, 85–97.
7 F. C. de Schryver, L. Anand, G. Smets and J. Switten, J. Polym. Sci.,
Part B: Polym. Lett., 1971, 9, 777–780.
8 J. R. Jones, C. L. Liotta, D. M. Collard and D. A. Schiraldi,
Macromolecules, 2000, 33, 1640–1645.
9 G. McSkimming, J. H. R. Tucker, H. Bouas-Laurent, J. P. Desvergne,
S. J. Coles, M. B. Hursthouse and M. E. Light, Chem.–Eur. J., 2002, 8,
3331–3342.
10 S. Nakatsuji, T. Ojima, H. Akutsu and J. Yamada, J. Org. Chem., 2002,
67, 916–921.
Scheme 2 The reversible formation of 4 from 1 and 2.
2570 | Chem. Commun., 2005, 2569–2571
This journal is ß The Royal Society of Chemistry 2005

View Article Online
11 T. Ojima, H. Akutsu, J. Yamada and S. Nakatsuji, Polyhedron, 2001,
20, 1335–1338.
12 H. Bouas-Laurent and R. Lapouyade, Chem. Commun., 1969, 817–818.
13 J. T. Goldbach, T. P. Russell and J. Penelle, Macromolecules, 2002, 35,
4271–4276.
14 One promising approach to this problem was reported by Bouas-
Laurent and Lapouyade (ref. 12), who showed that irradiation of a
mixture of 9-cyanoanthracene and 9,10-dimethylanthracene at rt affords
the cross-dimer in high yields. Unfortunately, the outcome of this
reaction is highly sensitive to the conditions employed. In polar sovents
or at elevated temperatures, the more thermally stable homodimer of
9-cyanoanthracene is formed as the major product.
15 Although 9,10-dimethylanthracene has been reported to homodimerize,
it does so only under much more stringent conditions than were
employed in the present study, and at no time did we observe the
formation of this photoproduct. See: H. Bouas-Laurent and
A. Castellan, Chem. Commun., 1970, 23, 1648–1649.
16 Interestingly, while neither 9,10-dimethylanthracene or 9,10-dimethox-
yanthracene tend to form homodimers, these two compounds do form
the cross-dimer quite readily. Unfortunately, this dimer exhibits
relatively poor thermal stability. See: R. Lapouyade, A. Castellan and
H. Bouas-Laurent, C. R. Seances Acad. Sci., Ser. C, 1969, 268, 217–219.
17 J. S. Yang and T. M. Swager, J. Am. Chem. Soc., 1998, 120,
11864–11873.
18 J. Clyburne and N. McMullen, Coord. Chem. Rev., 2000, 210, 73–99.
19 Y. Miura, S. Kurokawa and M. Nakatsuji, J. Org. Chem., 1998, 63,
8295–8303.
20 M. Theander, O. Ingana¨s, W. Mammo, T. Olinga, M. Svensson and
M. R. Andersson, J. Phys. Chem. B, 1999, 103, 7771–7780.
21 J. Lu, J. Zhang, X. Shen, D. M. Ho and R. A. Pascal, Jr., J. Am. Chem.
Soc., 2002, 124, 8035–8041.
22 M. D. Watson, A. Fechtenko¨tter and K. Mu¨llen, Chem. Rev., 2001, 101,
1267–1300.
23 A. J. Berresheim, M. Mu¨ller and K. Mu¨llen, Chem. Rev., 1999, 99,
1747–1785.
24 K. Shahlai and H. Hart, J. Org. Chem., 1991, 56, 6905–6912.
25 D. Bailey and V. E. Williams, Tetrahedron Lett., 2004, 45, 2511–
2513.
26 J. Luo and H. Hart, J. Org. Chem., 1987, 52, 4833–4836.
27 All photodimerization reactions were carried out under an N2
atmosphere in glass Schlenk tubes. Since anthracene derivatives
commonly form endoperoxides when irradiated in the presence of
oxygen, all solutions were subjected to three freeze–pump–thaw cycles in
order to ensure oxygen was excluded from the reaction mixtures.
Photodimerization experiments were carried out in a Rayonet reactor
fitted with ten RPR3000A lamps.
28 D. Y. Wu, B. Chen, X. G. Fu, L. Z. Wu, L. P. Zhang and C. H. Tung,
Org. Lett., 2003, 5, 1075–1077.
29 H. Bouas-Laurent, A. Castellan, J. P. Desvergne and R. Lapouyade,
Chem. Soc. Rev., 2000, 29, 43–55.
30 Photoreversion experiments were carried out by irradiating deoxyge-
nated acetonitrile solutions in quartz cuvettes using a PTI C60
photon-counting spectrofluorimeter as the light source (slit width 5
2 nm).
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 2569–2571 | 2571