Exciton and charge-transfer interactions in nonconjugated merocyanine dye dimers: Novel solvatochromic behavior for tethered bichromophores and excimers

Article abstract of DOI:10.1021/ja983778h

A series of merocyanine dye dimers tethered at different sites with spacers ranging from 0 to 5 methylene groups and the corresponding monomer have been synthesized. The formation, structure, and excited-state properties of the consequent merocyanine dye "aggregates" in solutions and in rigid glass have been studied by UV/visible absorption spectra, steady-state fluorescence, and fluorescence lifetime measurements. The dimers with 0 and 1 methylene spacers must exist in more-or-less extended conformations; consequently, they show a very weak and distance-dependent "J"-type exciton coupling, evident in both absorption and fluorescence spectra. For the dimers with 2, 3, and 5 methylene spacers, absorption spectra in nonpolar solvents are consistent with a largely extended configuration and little or no evident exciton coupling. However, for more polar solvents, a blue-shifted absorption spectrum is observed, suggesting a folded configuration resulting in an "H" dimer. For the latter dimers, fluorescence spectra in a variety of solvents show a pronounced red-shift, which is attributed to a folded "excimer". As anticipated from its structure, the merocyanine monomer shows a weak positive solvatochromic effect that may be correlated using the Taft-Kamlet π* parameter. Remarkably, both the "J" coupled dimer (n = 0) and the dimers having 2-5 methylene groups show a much stronger solvatochromic behavior than the monomer. The strong solvatochromic effects in these tethered dimers may be attributed to interchromophoric charge-transfer interactions in both ground and excited states.

Full text of DOI:10.1021/ja983778h

8146
J. Am. Chem. Soc. 1999, 121, 8146-8156
Exciton and Charge-Transfer Interactions in Nonconjugated
Merocyanine Dye Dimers: Novel Solvatochromic Behavior for
Tethered Bichromophores and Excimers
Liangde Lu,1 Rene J. Lachicotte, Thomas L. Penner,2 Jerry Perlstein, and
David G. Whitten*,3
Contribution from the Center for Photoinduced Charge Transfer, Department of Chemistry, UniVersity of
Rochester, Rochester, New York 14627, and Chemical Science and Technology DiVision, Los Alamos
National Laboratory, Los Alamos, New Mexico 87545
ReceiVed October 29, 1998
Abstract: A series of merocyanine dye dimers tethered at different sites with spacers ranging from 0 to 5
methylene groups and the corresponding monomer have been synthesized. The formation, structure, and excited-
state properties of the consequent merocyanine dye “aggregates” in solutions and in rigid glass have been
studied by UV/visible absorption spectra, steady-state fluorescence, and fluorescence lifetime measurements.
The dimers with 0 and 1 methylene spacers must exist in more-or-less extended conformations; consequently,
they show a very weak and distance-dependent “J”-type exciton coupling, evident in both absorption and
fluorescence spectra. For the dimers with 2, 3, and 5 methylene spacers, absorption spectra in nonpolar solvents
are consistent with a largely extended configuration and little or no evident exciton coupling. However, for
more polar solvents, a blue-shifted absorption spectrum is observed, suggesting a folded configuration resulting
in an “H” dimer. For the latter dimers, fluorescence spectra in a variety of solvents show a pronounced red-
shift, which is attributed to a folded “excimer”. As anticipated from its structure, the merocyanine monomer
shows a weak positive solvatochromic effect that may be correlated using the Taft-Kamlet π* parameter.
Remarkably, both the “J” coupled dimer (n ) 0) and the dimers having 2-5 methylene groups show a much
stronger solvatochromic behavior than the monomer. The strong solvatochromic effects in these tethered dimers
may be attributed to interchromophoric charge-transfer interactions in both ground and excited states.
Introduction
which has been attributed to an extended configuration. This
coupling falls off with an increase in tether length but is
perceptible up to six methylene groups. In contrast, in polar
solvents, a predominantly blue-shifted spectrum is observed for
the “dimers”, which is attributed to an “H”-type or folded
configuration. Comparison of the fluorescence of the “J” dimers
in fluid nonpolar solvents with that in rigid sucrose octaacetate
(SOA) glass shows that the extended form rapidly converts to
a folded form upon excitation to the excited singlet of the “J”
dimer. As an extension of these studies, we have synthesized a
series of tethered merocyanine dye dimers to examine excitonic
interactions and photophysical behavior and their solvent
sensitivity in a system in which the corresponding monomer
would be expected to exhibit moderate to strong solvatochromic
behavior. The results obtained with this series of dimers show
both some similarities and interesting differences from the
previously investigated squaraines and other tethered dimers.
As might be anticipated from their lower oscillator strengths,
the merocyanines in extended configurations show a less
pronounced “J”-type exciton coupling, which quickly falls off
as the tether length is increased. However, quite analogously
to the flexible squaraine dimers, we find strong evidence
indicating that an extended configuration is preferred for
nonpolar organic solvents, while in polar solvents a folded form
may be favored. Also, in accord with the behavior of the
squaraines, we find evidence that the extended form may rapidly
undergo a change in geometry following photoexcitation, even
in nonpolar solvents, resulting in very short fluorescence
Due to the wide applications of cyanine, squaraines, and
related dyes in color photography, photovoltaics, and other
imaging processes and their use in solid-state or interfacial
systems, their aggregation and their photophysical and photo-
chemical behavior have been the subject of considerable research
interest. Most of the previous work on these dyes involved
investigation of moderate-to-large intermolecular aggregates,
including both red-shifted “J” aggregates and blue-shifted “H”
aggregates, and the relationship between aggregate structure,
size, and associated spectral features.4-6 Relatively few experi-
mental studies have dealt with tethered dimers and their
photophysics;7,8 however, recently some interesting results have
been obtained for tethered squaraine dimers.9 In these systems,
strong excitonic interactions are observed which are ascribed
to different configurations. In nonpolar solvents, a prominent
red-shifted spectral transition is observed (“J”-type coupling),
(1) Present address: Research Laboratories, Eastman Kodak Co., Roch-
ester, NY 14650.
(2) Research Laboratories, Eastman Kodak Co., Rochester, NY 14650.
(3) Los Alamos National Laboratory.
(4) West, W.; Pearce, S. J. Phys. Chem. 1965, 69, 1894.
(5) Paddy, J. F. J. Phys. Chem. 1968, 72, 1259.
(6) Chen, H.; Farahat, M. S.; Perlstein, J.; Law, K.-Y.; Whitten, D. G.
J. Am. Chem. Soc. 1996, 118, 2584.
(7) Sundstrˆım, V.; Gillbro, T. J. Chem. Phys. 1985, 83, 2733.
(8) Chibisov, A. K.; Zakharova, G. V.; Gˆırner, H.; Sogulyaev, Yu. A.;
Mushkalo, I. L.; Tolmachev, A. I. J. Phys. Chem. 1995, 99, 886.
(9) Liang, K.; Farahat, M. S.; Perlstein, J.; Law, K.-Y.; Whitten, D. G.
J. Am. Chem. Soc. 1997, 119, 830.
10.1021/ja983778h CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/31/1999

SolVatochromic BehaVior of Merocyanine Dye Dimers
Chart 1. Merocanine Dyes Used in This Study
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8147
Scheme 2. Syntheses of Dimer-0, Dimer-1, and Dimer-0
Scheme 1. Syntheses of Dimer-5 and Dimer-3
Table 1. Crystallographic Data and Structure Refinement for the
Monomer Dye
empirical formula/weight
C17H17N3O4/327.34
crystal system/space group monoclinic/P2(1)
cell constant
a ) 7.0506(4) Å
b ) 28.390(2) Å, â ) 109.2780(10)°
c ) 8.0611(5) Å
volume/Z
crystal size
θ range
index ranges
reflections collected
independent reflections
refinement method
data/restraints/parameters
goodness-of-fit (F2)
final R indices
1523.1(2) Å3/4
0.06 × 0.08 × 0.32 mm3
1.43-23.26°
-6e he 7, -31e k e 30, -8 e l e 6
6184
2129 [R(int) ) 0.0379]
full-matrix least-squares on F2
5876/1/371
1.105
R1 ) 0.0539 [I > 2σ(I)]
wR2 ) 0.1224 [all data]
lifetimes. In contrast to the squaraines, we observe prominent
fluorescence from the folded forms of the merocyanine dimers
that closely resembles the excimer fluorescence observed from
tethered dipyrenyl alkanes and related compounds.10,11 Perhaps
the most striking results of the present study are the findings of
strong solvatochromic effects on both absorption and fluores-
cence spectra of the folded and extended exciton coupled states.
Results
The structures of the monomer merocyanine dye and the
tethered dimers synthesized in this study are shown in Chart 1.
The syntheses of the different dyes are outlined in Schemes 1
and 2. Although it is usually difficult to grow single crystals of
merocyanine dyes suitable for structure analysis, we were able
to grow crystals of the monomer and obtain structural data,
summarized in Table 1 (a more detailed table is included as
Supporting Information). The structure of the monomer is shown
Figure 1. X-ray structure of the merocyanine monomer.
in Figure 1. The monomer has one molecule in an asymmetric
unit, and the whole molecule is planar, with the benzoxazole
and barbiturate subunits trans to one another. The ethyl attached
at the nitrogen atom of the benzoxazole and the directly attached
alkenic proton point in the same direction. Bond lengths are
(10) Hirayama, F. J. Chem. Phys. 1965, 42, 3163.
(11) Reynders, P.; Kuhnle, W.; Zachariasse, K. A. J. Am. Chem. Soc.
1990, 112, 3940.

8148 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Lu et al.
Table 2. Absorption Maxima of Monomer and Dimer-0 in
Different Solvents
monomer λmax
dimer-0 λmax
solvent
Tafta π*
nm
kcal
nm
kcal
cyclohexane
triethylamine
diethyl ether
toluene
ethanol
tetrahydrofuran
methanol
CHCl3
CH2Cl2
dimethylformamide
dimethyl sulfoxide
0
0.14
437
439
438
441
440
440
65.4
65.1
65.2
64.8
65.0
65.0
0.273
0.535
0.54
0.576
0.586
0.76
0.802
0.875
1.000
Figure 2. Extended configuration of dimer-0 from molecular modeling.
470
467
472
471
473
475
60.9
61.2
60.4
60.7
60.4
60.2
443
442
442
444
64.6
64.7
64.7
64.4
a From ref 9.
Figure 3. Folded configuration of dimer-2 from molecular modeling.
included as Supporting Information. Remarkably, the diene unit
connecting the benzoxazole and barbiturate rings shows nearly
identical bond lengths for all three C-C bonds (ca. 1.38 Å),
and these are only slightly longer than those for an isolated
carbon-carbon double bond. If we assume that the merocyanine
should exist in the ground state as a largely uncharged species,
we would expect that the central C-C bond might be consider-
ably longer due to its largely single bond character. Using the
X-ray structural information for the monomer, a Monte Carlo
(MC)-simulated annealing using a modified MM2 force field
has been carried out for dimer-0, dimer-1, dimer-2, and dimer-3
(in the absence of solvent). The configurations corresponding
to the global energy minimum from these calculations are
extended (Figure 2), V-shaped, folded (Figure 3), and edge-to-
face, respectively.
UV/Visible Absorption and Solvatochromic Effect. Table
2 compares the absorption maxima of the monomer and dimer-
0. The monomer shows a maximum at ca. 440 nm and a
shoulder at ca. 417 nm (Figure 4), which are independent of
concentration over the range 10-5-10-7 M. When the solvent
is changed from nonpolar cyclohexane, to tetrahydrofuran
(THF), to ethanol (protic), to methylene chloride, and to very
polar dimethyl sulfoxide (DMSO), the absorption maximum
undergoes a modest red shift (437-444 nm). The merocyanine
dyes are generally expected to have a π,π* transition with weak
charge-transfer character in the ground state but considerably
stronger charge separation in the excited state; consequently, a
positive solvatochromic effect is anticipated,12-15 as is observed.
Figure 4. Absorption spectra of the monomer in (a) cyclohexane, (b)
ethanol, (c) CH2Cl2, and (d)DMSO.
The transition energies of several merocyanine dyes show
reasonable quantitative correlations with a number of empirical
solvent polarity parameters, such as ET(30) and the Kamlet-
Taft π* parameter.12-15 In the case of the latter, a plot of the
transition energy vs the empirical solvent polarity parameter
π* gives a linear plot with a negative slope (-s) in the case
where there is more charge separation in the excited state than
in the ground state.15 The Kamlet-Taft treatment is attractive
since it allows the separation of “polarity” from other solvent-
solute interactions, such as hydrogen bonding. When the
absorption maxima of the monomer dye used in this study are
plotted versus the Kamlet-Taft π* solvent polarity parameter,12
a reasonable linear plot with a relatively small slope (for a
merocyanine) is obtained by fitting eq 1 using a least-squares
regression method:
Emax(monomer) ) 65.4 kcal - 0.87π*
R ) 0.92, N ) 10 (1)
(1)
The absorption energies from Table 2 have also been plotted
against the Brooker øR values,14 and this plot gives only a fair
(R ) 0.86) fit, even though the øR parameter is based on a
merocyanine dye.
(a) “J”-Type Exciton Coupling.16,17 Dimer-0 is anticipated
to be an extended structure, with the two monomer units
connected by a single bond. The predicted dihedral angle
(12) Reichardt, C. SolVents and SolVent Effect in Organic Chemistry;
Verlag Chemie: New York, 1988; Chapter 6.
(13) Shin, D. M.; Whitten, D. G. J. Phys. Chem. 1988, 92, 2945. Schanze,
K. S.; Mattox, T. F.; D. M.; Whitten, D. G. J. Org. Chem. 1983, 48, 2808.
(14) Brooker, L. G. S.; Craig, A. C.; Heseltine, D. W.; Jenkins, P. W.;
Lincoln, L. L. J. Am. Chem. Soc. 1965, 87, 2443.
(15) Kamlet, M. J.; Abboud, J. L.; Taft, R. W. J. Am. Chem. Soc. 1977,
99, 6027.

SolVatochromic BehaVior of Merocyanine Dye Dimers
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8149
Figure 6. Absorption spectra of dimer-2 in different solvents: (a)
CHCl3, (b) CH2Cl2, (c) C2H5OH, and (d) DMSO.
Figure 5. Absorption spectra of dimer-0 in different solvents: (a) CH3-
OH, (b) 50 vol % CH3CN/H2O, (c) CH2Cl2, and (d) DMSO.
between the two planes of the monomer units of dimer-0 in its
lowest energy configuration is 28.2° based on the Monte Carlo
calculations. On the basis of the anticipated extended config-
uration of dimer-0, the anticipated exciton coupling is expected
to be a “J”-type splitting, characterized by a more intense long-
wavelength transition and a less intense short-wavelength band.
Compared to the monomer, dimer-0 shows a splitting with two
bands in its absorption spectrum (Figure 5) in a series of organic
solvents having an intensity ratio of ca. 2 (A470/A420).18 These
bands both show a red shift as the solvent polarity increases in
the series THF, methanol, methylene chloride, dimethyl sul-
foxide (Figure 5). As in the case of the monomer, the
solvatochromic effect of the dimer-0 may be correlated quan-
titatively with the Kamlet-Taft π* parameter, and the relation-
ship shown in eq 2 is obtained.
Emax(dimer-0) ) 62.2 kcal - 2.02π*
Figure 7. Absorption spectra of dimer-3 in different solvents: (a) SOA
glass, (b) CHCl3, (c) CH2Cl2, (d) CH3CN, (e) CH3OH, and (f) 80 vol
% H2O/CH3CN.
R ) 0.90, N ) 6
(2)
Due to the low solubility of dimer-0, we were only able to obtain
a correlation with solvents having π* values larger than 0.5.
Remarkably, dimer-0 shows a greater sensitivity to solvent
polarity than the monomer.
spectrum of dimer-3 shows a pronounced variation with solvent
which is independent of concentration at low concentrations (ca.
10-6 M) (Figure 7). In rigid SOA glass, chloroform, and
methylene chloride, the spectral shape and position for dimer-3
are similar to those of the monomer, except that a slightly wider
and more pronounced shoulder is observed. This implies that
any intramolecular exciton interaction is very weak in these
nonpolar solvents and that the chromophores are probably quite
far apart. As the solvent polarity increases, a blue-shifted band
appears (ca. 415 nm) in addition to the band at 440 nm
associated with “monomer” or extended dimer, and in aceto-
nitrile its intensity is almost the same as the monomer in
intensity. In methanol and ethanol, the blue-shifted band
becomes stronger than the monomeric peak (ca. 440 nm). Based
on similar spectral shifts for the tethered squaraines,9 the short-
wavelength band may be attributed to a folded “H”-type dimer.19
These shifts are also similar to those observed for “H”
aggregates for other merocyanine dyes.17
For dimer-1 and dimer-2 (Figure 6), the two monomers are
connected at C13 by one and two methylene groups, respectively,
and their absorption spectra in chloroform or rigid sucrose
octaacetate (SOA) glass are very similar to that of the monomer,
except for small red shifts for the dimers of ca. 10 and 8 nm,
respectively. These may be attributed to a progressively weaker
“J”-type exciton-coupling through longer distances; the direction
of the observed shifts is consistent with the predominant
configuration in these solvents being an extended one. The
behavior observed here is similar to that observed for squaraine
dimers with flexible tethers;9 however, the fall-off in splitting
with distance (length of tether) is much more rapid for the
merocyanines.
(b) “H”-Type Exciton Coupling. Dimers-3 and -5 have the
merocyanines connected at the nitrogens in the benzoxazole ring
by three and five CH2 groups, respectively. The absorption
In water-acetonitrile mixtures, the relative intensity (blue-
shifted “exciton” band vs monomeric band) increases as the
percentage of water increases. In the extreme case, when the
(16) Hochstrasser, R. M.; Kasha, M. Photochem. Photobiol. 1964, 3,
317.
(17) Herz, A. H. Photogr. Sci. Eng. 1974, 18, 323.
(18) There is also a shoulder on the short-wavelength side of the more
intense “red” band which resembles the shoulder observed for dilute
solutions of the monomer. None of the spectra shown in Figures 5-7 are
concentration-dependent.
(19) We expect, in a “well folded” cofacial structure, that the excitonic
splitting (allowed-forbidden character) may be more pronounced than in
the somewhat bent structures.

8150 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Lu et al.
Figure 8. Excitation spectra: a (λem ) 530 nm) and b (λem ) 475 nm) of dimer-3 in CHCl3; and its emission spectra c (λex ) 445 nm) and d (λex
) 420 nm). Inset: The fluorescence spectra of the monomer in CHCl3 (λem ) 475 nm and λex ) 420 nm).
Table 3. Absorption Data of Dimer-3 and Dimer-5 (10-6 M)
λ
max1, λmax2 absorbance ratio
dye molecule
dimer-3
solvent
CH2Cl2
CH3CN
(nm)
(Amax1/Amax2)
442, 422
440, 416
CH3OH H2O/CH3CN 440, 414
1.41
0.99
0.74
0.57
1.37
1.11
0.93
0.87
(1/1 vol)
CH2Cl2
CH3CN
CH3OH
H2O/CH3CN
(1/1 vol)
441, 411
442, 424
438, 418
440, 416
439, 415
dimer-5
percentage of H2O is g80%, the dominant band is the exciton
band, with a low absorbance due to the low solubility of dimer-3
in the mixture (when the solution was heated, the absorbance
increases to close to that of other aqueous mixtures).
Dimer-5, with five CH2 groups as the spacer, shows spectra
Figure 9. Emission spectra of dimer-3 (λex ) 420 nm) in (a) SOA
glass, (b) CHCl3, (c) CH2Cl2, (d) CH3CN, (e) CH3OH, and (f) 80 vol
% H2O/CH3CN.
similar to those of dimer-3 in the same solvents at ca. 10-6
M
and shows a spectral change pattern similar to that of dimer-3
as solvent properties change (Table 3). The “H”-type exciton
splitting shows a very small decrease as the chain length
increases from three to five carbon units, and there is no
evidence for “J”-type exciton coupling for either dimer-3 or
dimer-5 in any solvent.
Fluorescence. The emission spectra and excitation spectra
of the monomer in chloroform are shown in Figure 8 (ca. 10-6
M). The emission spectra are independent of excitation wave-
lengths, and the excitation spectra do not change with emission
wavelengths. The emission maxima show small red shifts with
increasing solvent polarity, corresponding to the shifts in the
absorption spectra. Similar fluorescence spectra are obtained
in the same solvents for dimer-0 and dimer-1; here again, there
is no wavelength dependence of either fluorescence or excitation.
Although the same general trend of red shifts in emission
maximum is observed for monomer, dimer-0, and dimer-1, due
to the breadth of the fluorescence maxima for all three
compounds, no sensible quantitative correlation of solvatochro-
mic effects could be made.
spectrum obtained by monitoring the fluorescence at 460 nm
resembles that for the monomer (inset, Figure 8) and also tracks
quite closely the absorption spectrum (Figure 7). The intensity
of the broad band at 530 nm is much higher when the sample
is excited at 420 nm, which corresponds quite closely to the
maximum of the blue-shifted “H” exciton band observed for
dimer-3 in more polar solvents. The excitation spectrum
monitoring the fluorescence at 530 nm resembles the absorption
spectrum of dimer-3 in more polar solvents (Figure 7), such as
methanol. The broad, structureless nature of the long-wavelength
emission and the excitation corresponding to the folded “H”-
type configuration suggest that this emission may be attributed
to a folded excimer-like state, similar to that associated with
dipyrenyl propane.20
Figure 9 compares the fluorescence spectra of dimer-3 in a
series of different solvents. In SOA, which may be regarded as
a rigid, nonpolar solvent, the only emission that is observed is
the sharp band at ca. 455 nm which corresponds quite closely
to monomer. As mentioned above, dimer-3 in chloroform shows
absorption very similar to that for monomer but also both the
sharp emission, similar to that of monomer and dimer-3 in SOA,
Excimer Formation. The emission and excitation spectra of
dimer-3 in chloroform (ca. 10-6 M) are compared in Figure 8.
The fluorescence spectrum consists of a sharp band at 460 nm
and a broader band at 530 nm; the relative intensity of the two
bands is dependent on the excitation wavelength. The excitation
(20) Birks, J. B. Nature 1967, 214, 1187.

SolVatochromic BehaVior of Merocyanine Dye Dimers
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8151
Table 5. Quantum Yields of the Monomer and Dimer Dyes in
Different Degassed Solvents
excitation
wavelength
dye
solvent
(nm)
φta
φmon
φex
monomer
CHCl3
CH3CN
SOAb
420 or 430
420 or 430
420 or 430
420
430
420 or 430
420
430
420
430
420 or 430
420
0.0064
0.0024
0.23
0.0064
0.0024
0.23
dimer-5
dimer-3
CHCl3
0.019
0.016
0.028
0.012
0.010
0.026
0.024
0.011
0.083
0.071
0.013
0.014
0.14
DMSO
CHCl3
0.0041
0.0043
0.0029
0.0039
0.011
0.010
0.011
0.0079
0.0059
0.023
DMSO
0.020
dimer-2
CHCl3
DMSO
0.073
0.060
430
dimer-1
dimer-0
CHCl3
DMSO
CHCl3
420 or 430
420 or 430
420
Figure 10. Emission spectra (λex ) 420 nm) of the dimers in DMSO:
(a) dimer-1, (b) dimer-2, (c) dimer-3, (d) dimer-5, and (e) emission
spectra of dimer-2 in a solid state.
460
420
460
420
0.12
DMSO
SOA
0.050
0.039
0.65
Table 4. Stabilization Energy of Dimer-2, Dimer-3, and Dimer-5
460
0.39
stabilization energy (cm-1
)
a φt is the quantum yield with 5% error, cwhich could consist of
two parts: quantum yield of the monomer-like species (φmon) and that
of the excimer (φex). b The dimer-5, dimer-3, dimer-2, and dimer-1 show
similar quantum yields (0.15-0.20) similar to that of the monomer in
SOA glass.
dimer
in DMSO
5000
3200
900
5000 (minor)
in methanol
dimer-2
dimer-3
dimer-5
5300
5000
Fluorescence Quantum Yields and Lifetime Measure-
ments. The fluorescence quantum yields and lifetimes of the
monomer and the various dimers in different solvents are listed
in Tables 5 and 6, respectively. The fluorescence lifetimes of
the monomer in fluid solvents are in the range of 5-20 ps, and
the quantum yield (less than 1%) is very low, probably due to
isomerization4 and/or torsional motions which provide rapid
nonradiative decay.22 The emission quantum yield and lifetime
of the monomer are independent of excitation wavelength and
emission wavelength in all solvents. Both the fluorescence
lifetime and the quantum efficiency are enhanced dramatically
when the monomer is “solubilized” in rigid SOA glass,
reinforcing the idea that conformational changes are responsible
for the short lifetimes in fluid media.22 Compared with the
monomer, dimer-0 shows a higher fluorescence efficiency in
fluid solvents and a longer fluorescence lifetime (at least in the
one common solvent studied) as well as an increased fluores-
cence quantum efficiency. Similar to the monomer, there is a
big increase in both fluorescence lifetime (see below) and
efficiency when dimer-0 is studied in SOA. In fact, all of the
dimers synthesized (except dimer-0) show similar quantum
efficiencies (0.15-0.20), only slightly lower than that of the
monomer in SOA. In heterogeneous environments such as
bilayer vesicles or SOA glasses, where a number of different
solute states with differing emission properties may be popu-
lated, the use of the usual multiexponential decay model for
the decay data may not be justified.23 An appropriate model
describing such a system is one that attributes a range of
lifetimes to a single emitting species corresponding to a
nonequilibrium population of chromophores. Such a method was
and the broad, red-shifted band at 530 nm. In methylene
chloride, which shows only a slight broadening of the absorption
spectrum and a small blue shift compared to monomer or
dimer-3 in SOA and chloroform, the major emission upon
excitation in the “monomer” transition is the broad “excimer”
band. The “excimer” emission becomes predominant when
dimer-3 is excited in acetonitrile and shows a red-shift compared
to methylene chloride and chloroform. In protic solvents such
as methanol or 80% H2O/CH3CN, the excimer band becomes
even broader and is further red-shifted to 600-620 nm. The
red shift in the “excimer” emission maxima of dimer-3 with
increase in solvent polarity (excluding chloroform and methylene
chloride, as discussed below) may be correlated with the
Kamlet-Taft π* solvent polarity parameter;12,15 a linear plot is
obtained by fitting eq 3 with a least-squares regression method.
Emax(dimer) ) 55.0 kcal - 2.28π*, R ) 0.93, N ) 5 (3)
The excimer emission for the different dimers shows an
interesting structure dependence. The emission spectra of dimers
with different spacers in DMSO are shown in Figure 10. Not
surprisingly, dimer-1, which cannot fold, shows only monomer-
like emission. Dimer-2 shows a predominant excimer band at
600 nm in addition to a weak emission corresponding to
monomer, while only excimer emission is observed for dimer-2
in the solid state. For dimer-3, the excimer band is less red-
shifted, and both monomer and excimer emission are observed.
In the case of dimer-5, the emission is more complex, with
maxima at 483 nm, partially overlapped with the monomer-
like emission, and a less prominent red-shifted emission spread
over the range 530-700 nm. The stabilization energy of the
excimers in DMSO and in methanol (Table 4), which is the
spectroscopic energy difference between the monomer emission
and excimer emission maxima,21 varies from 5000 to 900 cm-1
for dimer-2 and dimer-5, respectively.
(22) As suggested by a reviewer, a “TICT” state could be involved in
such isomerization processes. Previous papers have reported evidence for
twisted excited states as intermediates in merocyanine isomerization (e.g.:
Harriman, A. J. Photochem. Photobiol. A: Chem. 1992, 65, 79).
(23) Whitten, D. G.; Farahat, M. S.; Gaillard E. R. J. Photochem.
Photobiol 1997, 65 (1), 23.
(21) Jenekhe, S. A.; Osaheni, J. A. Science 1994, 265 (5), 765.

8152 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Lu et al.
Table 6. Lifetimes of the Monomer, Dimer-5, Dimer-3, Dimer-2,
Dimer-1, and Dimer-0 in Degassed Solvent and in Solid SOA Glass
(Excitation Wavelength ) 420 nm)
Discussion
The merocyanine monomer and corresponding suite of
tethered dimers were selected for the anticipated contrast to the
squaraines previously studied that these might offer, and it is
relevant to compare the behaviors observed for the two series
of linked dye dimers. For the merocyanine series, some of the
important anticipated differences from the squaraines may be
seen by comparing differences between the two monomers.
Thus, the merocyanine has a lower exctinction coefficient and
a consequent lower transition moment, which would be expected
to lead to a lower excitonic coupling for dimers having the same
spacing. However, since the merocyanine was anticipated to
have a substantial dipole moment in the ground state and an
even greater charge separation in its lowest excited singlet state,
additional possibilities for interactions in both ground and
excited states of the tethered dimers might be expected. The
merocyanine dyes consist of an electron-donating group (D)
linked to an electron-accepting group (A) through conjugation.
The ground-state electronic structure of the dyes lies somewhere
between that of the polyene form and that of the polymethine
form, depending on solvent polarity.12
emission
wavelength
(nm)
τ1
τ2
B1/B2
(%)
dye
solvent
(ns)a (ns)b
ø2
monomer CHCl3
460 or 530 0.02
CH3CN 460 or 530 0.005
SOAc
460 or 530 0.16
0.90 35/65 1.4
4.6
dimer-3
CHCl3
460
530
460
530
460
530
460
530
460
530
0.03
0.03
0.02
0.02
0.10
0.11
0.03
0.03
0.02
0.02
6.5
19/81 1.1
DMSO
SOAc
5.8
1.1
1.2
2.1
1.7
2.3
2.1
1.7
13/87 2.8
40/60 1.0
48/52 0.99
85/15 1.9
22/78 0.97
88/12
dimer-5
dimer-2
CHCl3
DMSO
22/78 1.6
35/65 0.98
SOAc
CHCl3
DMSO
460 or 530 0.15
460 or 530 0.03
460
530
460
530
0.01
0.02
0.11
0.19
4.4
36/64 3.0
SOAc
0.96 40/60 1.1
1.1 45/55 1.1
dimer-1
dimer-0
CHCl3
DMSO
460 or 530 0.02
460
530
0.03
0.02
SOAc
CHCl3
DMSO
SOAc
460 or 530 0.20
460 or 530 0.3
460 or 530 0.03
0.65 90/10 1.7
3.5
460
530
0.11
0.27
1.2
1.5
38/62 2.1
19/81 1.2
a The short lifetime (5-30 ps), close to the instrument response (60
ps), fitting by A + B1 exp(-t/τ1) + B2 exp(-t/τ2), is not precise, the
measurement is subject to considerable error, and only the order of
magnitude can be approximated. b The long lifetime (0.10-7.5 ns) has
an error of (1 on the second significant number. c The lifetime is mean
lifetime from a distribution of lifetimes by simulation analysis of the
fluorescence decay.
As pointed out in the Results section, there is a general red
shift with increase in solvent polarity in the absorption spectra
of monomer and dimer-0, which is anticipated for positive
solvatochromic dyes in which electronic excitation is ac-
companied by a pronounced increase in charge separation. The
finding of very little bond alternation in the diene unit linking
the two halves of the merocyanine supports the idea of some
charge-transfer interactions already present in the ground state;
the small negative slope for the plot of λmax vs π* (-s ) 0.87)
indicates that there is increased charge transfer in the excited
state.
Turning to the dimeric merocyanines, it is useful to first
consider what might be anticipated in terms of possible excitonic
interactions. All of the “tethers” linking the merocyanine
chromophores provide some flexibility in terms of the arrange-
ment between chromophores; in terms of potential excitonic
interactions, according to the point dipole approximation, two
limiting arrangements are most interesting: a folded and an
extended configuration.24 Both of these configurations may not
used to fit the fluorescence decay of the monomer and various
dimers in SOA glass, and it was found that the decay curves of
the monomer and dimers in SOA glass can be fitted nicely with
two lifetime distributions: the shorter one is 0.1-0.3 ns, and
the longer one is 1-2 ns.
As mentioned above, the emission spectra of dimers-2, -3,
and -5 in fluid environments exhibit both monomer-like (460
nm) and excimer emission; therefore, two lifetimes and quantum
yields are expected. Due to the overlap of the monomer and
excimer bands of dimer-5, no attempt was made to evaluate
the individual quantum efficiencies, and only the overall
quantum yield is listed. The overall quantum yield is higher
when the sample is excited at 420 nm than that at 430 nm,
because the blue-shifted exciton band is situated around 420
nm. For dimer-3 in fluid media, a species of a short lifetime
(10-30 ps) close to that of monomer was found when the
sample was monitored at an emission wavelength of 460 nm,
while a much longer-lived species (ca. 6 ns) is found at an
emission wavelength of 530 nm; therefore, the short-lived
species may be assigned to a monomer-like or extended dimer
state, while the long-lived species is the excimer. The quantum
yield is higher in polar solvents such as acetonitrile due to the
predominance of excimer emission; the quantum yield is much
lower in solvents such as chloroform, where monomer-like
emission is dominant.
be possible for each dimer in the series, and they may or may
not be able to interconvert, depending upon the individual dimer
(24) For the merocyanine dimers, several folded configurations are
possible. MC simulations indicate an edge-to-face configuration is one
minimum. Two limiting face-to-face configurations (one with the cho-
mophores aligned parallel and one with them antiparallel) are possible and
reasonable on energetic grounds. It is not possible to determine which
configuration or configurations may exists for either the ground or the
excited (excimer) state. For purposes of illustration, we assume a “parallel”
folded arrangement.
Oxygen was found to have no effect on the fluorescence
intensity and lifetimes of both monomer and dimers. This is in
accord with their short fluorescence lifetimes of no longer than
a few nanoseconds.8

SolVatochromic BehaVior of Merocyanine Dye Dimers
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8153
structure. For those dimers that can assume an extended
configuration, this configuration may be anticipated to be more
stable in the ground state for solutions with nonpolar solvents
due to the lack of gauche or eclipsed conformations in the tether,
provided there are no or weak dipoles. The extended or nearly
extended configuration (open “V” shape, for example) should
give rise to a “J”-type excitonic coupling, characterized by a
split in the monomer long-wavelength transition (if polarized
along the long axis), leading to lower energy (more allowed)
and higher energy (less allowed) transitions compared to those
of the monomer. Since our calculations indicate that dimers such
as dimer-2 or dimer-3 are not likely fully extended and the
coupling may fall off, it is reasonable that both long- and short-
wavelength transitions may be of comparable intensity. The
folded configuration may be favored in cases where there is
attraction between the two chromophores (for example dipole-
dipole attraction) or in solvents where strong solvent-solvent
interactions favor a minimization of the solute area to be
solvated (solvophobic effects).24 A completely folded config-
uration should result in an “H”-type excitonic coupling, also
characterized by a split in the monomer transition, resulting in
a lower energy (forbidden) and a higher energy (more allowed)
transition.25 Although the cartoon above shows limiting extended
and folded configurations, a variety of additional configurations
may be possible when the tether is long and flexible and
depending upon the site of the tether and other properties of
the solute and solvent.
of 28° (Figure 2); while there should be little charge transfer
between the two merocyanine units in the ground state, it seems
quite likely that extensive charge delocalization may occur in
the excited state. This could lower the energy of the excited
state and partially account for the observed red shift. That this
may be important is also suggested by the relatively large
(compared to the monomer) positive solvatochromic sensitivity
of the absorption spectrum of dimer-0. The value of the slope
(-s) of 2 is more than twice that of the monomer.
The similarity of the absorption and fluorescence spectra of
dimer-1 and dimer-2 to those of the monomer in nonpolar
solvents and SOA is consistent with these dimers existing in a
more-or-less extended configuration in the ground state, in which
there is little or no excitonic coupling between the two
merocyanine units. The finding that dimer-2 exhibits dual
fluorescence in solvents of moderate to high polarity is
consistent with a folded form of the dimer playing an important
role in these situations, quite analogous to the behavior of the
tethered squaraines with two or more methylene units in the
tether. The absorption spectra of dimer-2 in different solvents
suggest that there may be relatively little ground-state population
of folded forms in chloroform or less polar solvents. In solvents
such as DMSO, where there is dual emission, the fluorescence
lifetime and the quantum efficiency corresponding to the
monomer are very similar to those measured in chloroform,
where only “monomer” fluorescence is observed. Given the
extremely short fluorescence lifetime, it seems unreasonable to
assume that interconversion from an extended to a folded form
occurs in competition with fluorescence and most reasonable
that the observed “excimer” emission may be attributed to a
small amount of folded dimer-2 present in the solution. The
much higher quantum efficiency for fluorescence observed for
dimer-2 is most reasonably attributed to the much longer lifetime
of the excimer.
If we compare the properties of dimers-0, -1, and -2 with
those of the monomer and the roughly corresponding series of
methylene-linked squaraines,9 we observe some similarities as
well as pronounced differences. In fluid nonpolar solvents and
in SOA, the most prominent effect for the series of merocyanines
is the splitting and red shift in the absorption spectum of dimer-0
compared to the monomer. Dimer-0 shows a “J”-type splitting
(ca. 30 nm, depending slightly on solvent) in the long-
wavelength transition and a corresponding red shift in its
fluorescence. Dimers-1 and -2 show almost no changes from
the monomer in absorption. This may be compared to the
squaraine series, where (while we have no n ) 0 dimer) we
observe a shift of 24 nm for n ) 2, and a perceptible splitting
persists to n ) 7. Since the extinction coefficient of the
merocyanine (∼1 × 105 in chloroform) is lower than that of
the squaraine monomer (∼3 × 105 in chloroform) and exciton
theory predicts that the coupling should be in proportion to the
cube of the interaction distance and the square of the transition
moment of the interacting chromophores,26 it is not surprising
that the “J” exciton splitting is much smaller for the merocyanine
dimers than for the squaraines. In fact, if we examine the
splitting observed for dimer-0 compared to that for the monomer,
it is not clear that it is really correct to ascribe it solely to
excitonic interactions. First, while the two “halves” of dimer-0
are surely in a somewhat extended configuration in all solvents,
it is reasonable to question how much conjugation there is
between the two halves. For the MC simulated global minimum
of dimer-0, we obtain a dihedral angle between the two halves
The behavior of dimers-3 and -5 is also consistent with
interconversion between extended and folded forms, with a
steady increase in the population of the folded form as solvent
polarity and/or solvent-solvent hydrogen bonding increase.
Here again, “excimer” fluorescence, which presumably arises
primarily from folded ground-state configurations, is the most
prominent emission in polar solvents, and the lifetime of the
excimer is much longer than that for the monomer. As indicated
in the Results section, the excimer emission from dimer-5 is
complex, and it is not possible, at this time, to offer a simple
analysis for its complexity. On the other hand, the excimer
emission from dimer-3 seems somewhat simpler and perhaps
easier to understand. Since the blue-shifted absorption of dimer-3
that is attributed to the folded “H” configuration shows relatively
little solvent sensitivity, it may be assumed that the folded
ground-state structure does not change appreciably as the solvent
is changed. On the other hand, the corresponding “excimer”
fluorescence shows a strong solvatochromic effect with a slope
(-s) of 2.3, the highest value observed in this study. The
observation that the excimer emission is nearly structureless in
all solvents (Figure 9) suggests that it is truly excimer-like and
involves emission to an unbound ground state. A reasonable
interpretation for the solvatochromic behavior of the excimer
fluorescence as well as the strong “Stokes” shift observed is
that the excimer may have a much smaller interchromophore
separation than does the folded ground state. Here again, as
was suggested for the extended configuration of dimer-0, there
exists the attractive possibility that charge transfer (or charge
(25) If the configuration results in strong face-face or edge-edge
interactions, we may anticipate that the difference between the “allowed”
(short wavelength) and “forbidden” components of the split bands may be
very great. Thus, for example, with the intermolecular “H” aggregates of
squaraines, we are able only to detect the short-wavelength band in
absorption, while for corresponding aggregates of stilbenes, only the short-
wavelength transition is readily observable in absorption but emission clearly
originates from a “hidden” lower energy “forbidden” state.
(26) McRae, E. G.; Kasha, M. Physical Process in Radiation Biology;
Academic Press: New York, 1964; p 17.

8154 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Lu et al.
experiment to search for the apparent global minimum and nearby local
minima. The method is similar to that used earlier in predicting the
geometries of tethered squaraine dyes9 and for determining packing
patterns for Langmuir monolayers.29 Starting geometries for the
simulation consisted of an energy-minimized “stretched out dimer”.
Bond lengths and bond angles for each half of the dimer were taken
from the X-ray structure. The simulations consisted of cooling the
starting geometry from 4000 to 300 K in 10% increments, with rotation
about all single bonds as variables. At each temperature, each single
bond was randomly rotated an angle <10°. The nonbonded and
electrostatic energy of the dimer was then computed using a modified
MM2 force field.30 Structures were saved or rejected on the basis of
the Boltzmann criteria until the final temperature was reached. The
lowest energy structure at the final temperature was then saved as a
local minimum. The cooling cycle was then repeated, starting with the
saved structure, until 700 structures were collected.
Crystal Growing, Data Collection, Structure Solution, and
Refinement. A bright yellow plate of the monomer dye, grown from
concentrated solution of CH2Cl2 by slow evaporation, of approximately
0.06 × 0.08 × 0.32 mm3 was mounted on a glass fiber. All aspects of
data collection were performed on a standard Siemens SMART CCD
area detector using Mo KR radiation (λ ) 0.710 73 Å) at 193 K. The
unit cell parameters were determined by a least-squares analysis of 2751
reflections. The intensity measurements were then performed using
ω-scans (maximum 2θ of 56.6). Instrument and crystal stability was
verified by monitoring three check reflections every 100 data points.
No measurable decay was detected. The space group was assigned P21/n
on the basis of intensity statistics. The structure was solved using direct
methods and refined against F2 using the SHELXTL/PC v5.04 software
suite.31 All non-hydrogen atoms were refined anisotropically, and the
hydrogen atoms were included in calculated positions, using a standard
riding model and thermal parameters proportional to the non-hydrogen
atom to which they were attached. The final refinement converged with
the residuals wR2 ) 0.1181 (based on F2) for all data and R1 ) 0.0539
(based on F) for data with I >2σ(I). Additional details of data collection
and refinement parameters can be found in Table 1. Full details
including atomic positions and anisotropic displacement parameters are
included in the Supporting Information.
delocalization) across the two chomophores occurs as shown
below:24
In this case, especially with the relatively small spacer present
in dimer-3, the ground state may have a relatively larger
separation due to dipole-dipole repulsions (and other steric
considerations), while interchromophore charge resonance can
result in increased attraction and hence reduced separation
(within 3-4 Å)11 in the excimer. It is interesting to note that
the strong solvatochromic behavior observed for the dimero-
cyanine propane (dimer-3) is in strong contrast with the behavior
observed for analogous propyl-tethered aromatic dimers and may
be attributed specifically to the fact that the merocyanine
chromophore has strong donor and acceptor sites, whose
reactivity is enhanced following photoexcitation. Thus, the
classical attribution of excimer stabilization20,27 to a combination
of dominant excitation resonance with very minor charge
resonance contributions may be reversed in this case, with the
latter playing a very important and perhaps dominant role.
Experimental Section
Instrumentation and Methods. 1H NMR spectra were recorded on
a General Electric QE 300-MHz spectrometer. FAB mass spectra were
obtained at the Center for Mass Spectrometry at the University of
Nebraska in Lincoln, NE. HPLC was performed on a Hewlett-Packard
liquid chromatograph series 1050 with a variable-wavelength absor-
bance detector and a Rannin 150-mm × 4.6-mm silica gel column.
UV/vis absorption spectra were measured on a Perkin-Elmer Lambda
19 spectrophotometer. Fluorescence emission and excitation spectra
were recorded on a Spex Fluorolog-2 spectrofluorimeter and were
corrected. Solid-state spectra were obtained as a fine dispersion in a
KBr pellet. Fluorescence quantum yields were determined by com-
parison with a standard, coumarin 6, in ethanol (0.78).28
Fluorescence Lifetime Measurements. Time-correlated single-
photon-counting experiments were carried out on an instrument
consisting of a mode-locked Nd:YLF laser (Quantronix) operating at
76 MHz as the primary laser source. The second harmonic (KTP crystal)
of the Nd:YLF laser was used to synchronously pump a dye laser
(Coherent 700), circulating Rhodamine 6G in ethylene glycol as the
gain medium. The pulse width of the dye laser was typically 8 ps, as
determined by autocorrelation, and was cavity-dumped at a rate of 1.9
MHz. The dye laser was tuned to the desired excitation wavelength
(420 nm). Emission from the sample was collected by two convex
lenses, focused at the entrance slit of a Spex 1681 monochromator (0.22
m), and detected by a multichannel plate detector. The single-photon
pulses from the detector were amplified and used as the start signal
for a time-to-amplitude converter (TAC), while the signal from a
photodiode, detecting a small fraction of the dye output, was used as
the stop signal for the TAC. The start and stop signals for the TAC
were conditioned before entering the TAC by passing through two
separate channels of a constant fraction discriminator (CFD). The output
of the TAC was connected to a multichannel analyzer (MCA) interface
board inside a PC. The MCA was controlled by software from
Edinburgh Instruments. The same software was used to deconvolute
the data and do exponential fitting by the nonlinear least-squares
method. Measurements were made on air-saturated or Ar-degassed
samples.
Synthesis. All starting materials and liquids were reagent grade or
better and were used as received except where indicated otherwise.
(a) Synthesis of Dimer-5 According to Scheme 1. (i) 2,2′-
Dimethyl-3,3′-pentamethylendibenzoxazolium Iodides. To a round-
bottom flask were added 8 mL (68 mmol) of 2-methylbenzoxazole and
1.5 mL (12 mmol) of 1,5-diiodopentane, stabilized by copper particles.
The mixture was heated at 108-110 °C for 20 h to give a yellowish
precipitate. Acetonitrile (20 mL) was added to the solid precipitate,
which was filtered and washed with CH3CN (3 × 10 mL). A yellowish
solid (4.0 g, 56%) was obtained. 1H NMR (300 MHz/CD3OD): δ 8.20-
7.77 (m, 8H, aromatic), 4.67-4.62 (t, J ) 7.2 Hz, 4H, N-CH2), 3.16
(s, 6H, CH3), 2.13-2.05 (q, J ) 7.2 Hz, 4H, N-C-CH2), 1.78 (m, 2H,
N-C-C-CH2).
(ii) 2,2′-(2-Anilidovinyl)-3,3-pentamethylenedibenzoxazolium Io-
dide. In a mortar, 1.0 g (1.7 mmol) of 2,2′-dimethyl-3,3′-pentameth-
ylenedibenzoxazolium iodides and 1.0 g (5.1 mmol) of N,N′-diphe-
nylformamidine were mixed well by pounding, and the mixed solid
was heated with stirring at 150-157 °C for 5 min under a dry
atmosphere. The solution was cooled to give a reddish solid. The solid
was dissolved in ethanol (50 mL), and diethyl ether (500 mL) was
added to precipitate the product, which was filtered. The crude product
was dissolved in hot ethanol (30 mL), and diethyl ether (3 mL) was
added the cold ethanol solution to induce a precipitate. A yellow solid
1
(0.8 g, 60%) was obtained. H NMR (300 MHz/CD3OD): δ 8.88-
8.70 (d, J ) 12 Hz, 2H, CHd), 7.70-7.23 (m, 8H, aromatic), 5.99-
5.95 (d, J ) 12.3 Hz, 2H, CHd), 4.36-4.31 (t, J ) 6.9 Hz, 4H, N-CH2),
2.05-1.95 (m, 2H, N-C-CH2), 1.66-1.60 (m, 2H, N-C-C-CH2).
Molecular Modeling. Possible geometries of the tethered dimers
were determined by carrying out a Monte Carlo-simulated cooling
(29) Perlstein, J. J. Am. Chem. Soc. 1994, 116, 11420.
(30) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.;
Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J. Comput. Chem.
1990, 11, 440.
(31) SHELXTL: Structure Amalysis Program, Version 5.04; Siemens
Industrial Automation Inc.: Madison, WI, 1995.
(27) Birks, J. B. Organic Molecular Photophysics; Academic Press: New
York, 1975; Vol. 2.
(28) Reynolds, G. A.; Drexhage, K. H. Opt. Commun. 1975 13 (3), 222.

SolVatochromic BehaVior of Merocyanine Dye Dimers
J. Am. Chem. Soc., Vol. 121, No. 36, 1999 8155
(iii) Dimer-5. Under nitrogen, 0.2 g (0.25 mmol) of 2,2′-(2-
anilidovinyl)-3,3-pentamethylenedibenzoxazolium iodide and 0.16 g
(0.82 mmol) of 1,3-dimethylbarbituric acid were dissolved in dried
methylene chloride (100 mL). Triethylamine (0.5 mL) was added to
the above solution.32 The reaction mixture was stirred for 3 days to
yield an orange precipitate, which was filtered and then washed with
methanol (3 × 5 mL) to give the crude product. The crude product
was chromatographed first with a mix of 5:40:55 methanol/chloroform/
ethyl acetate(0.5 L) and then with a mix of 10:75:15 methanol/
chloroform/ethyl acetate to yield 70 mg of yellow solid (42%, one peak
(iv) Bis(3-amino-4-hydroxyphenyl)ethane. A suspension of 4.0 g
(13.8 mmol) of bis(3-nitro-4-hydroxyphenyl)ethane with 0.21 g (0.92
mmol) of PtO2 in 200 mL of absolute ethanol was hydrogenated on a
Parr hydrogenator at 50 psi for 26 h. A total 1.5 g (45%) of greenish
solid was obtained after another 200 mL of methanol was added to the
reaction mixture, which was then sonicated, filtered, and washed with
700 mL of ethanol. H NMR (300 MHz/DMSO-d6): δ 8.65 (s, OH),
6.48-46 (d, J ) 8.2 Hz, 2H), 6.4 (d, J ) 1.5 Hz, 2H), 6.20-6.17 (dd,
J1 ) 8.2 Hz, J2 ) 1.5 Hz, 2H), 4.35 (s, NH2), 3.26 (s, 4H).
1
(v) 2,2′-Dimethyl-6,6′-ethylenedibenzoxazole. In 5 mL of acetic
anhydride, 1.7 g (7.0 mmol) of bis(3-amino-4-hydroxyphenyl)ethane
was first heated at 80-90 °C for 20 h under a dry atmosphere. The
mixture was filtered, and the filtrate was precipitated by addition of
H2O (100 mL). A total 1.2 g of white solid was obtained after the
solid was filtered and dried. The intermediate product was heated further
at 240-280 °C for 10 min. The reaction residue was taken up in 40
mL of CHCl3 with silica gel. The loaded silica gel was filtered, extracted
with 1:3:6 CH3CN/ethyl acetate/hexane (1 L), and condensed to a
yellowish crude product, which was purified by chromatography with
1:3:6 CH3CN/ethyl acetate/hexane (2 L) to afford 0.38 g (18%) of off-
1
by HPLC with 1:19 methanol/chloroform). H NMR (300 MHz/CD3-
CO2D): δ 8.61-8.57 (d, J ) 10.5 Hz, 2H, CHd), 7.68-7.40 (m, 10H),
4.35-4.20 (t, 4H, N-CH2), 3.40-3.20 (s, 12H, NCH3), 2.10-1.90 (m,
4H, N-C-CH2), 1.65-1.55 (m, 2H, N-C-C-CH2). MS (FAB): m/e 667.2
(M+ + 1).
(b) Dimer-3 and the Monomer Were Synthesized by a Method
Similar to That for Dimer-5. (i) Dimer-3. A yellow solid, in 65%
1
(one peak by HPLC with 1/19 methanol/chloroform). H NMR (300
MHz/CF3CO2D): δ 8.90-8.75(d, J ) 14.4 Hz, 2H, CHd), 8.4-7.5-
(m, 10H), 5.2-4.8 (t, J ) 7.2 Hz, 4H, N-CH2), 3.8-3.5 (s, 12H, NCH3),
2.90-2.75 (m, 2H, C-CH2-C); MS (FAB) m/e 639.2 (M+ + 1).
(ii) Monomer Dye. A yellow solid, in 89% (one peak by HPLC
with 1/19 methanol/chloroform). 1H NMR (300 MHz/CDCl3): δ 8.83-
8.70 (d, J ) 14.4 Hz, 1H, CHd), 7.60-7.20 (m, 5H), 4.20-4.10 (m,
2H, CH2-N), 3.41 (s, 3H, NCH3), 3.31 (s, 3H, NCH3), 1.56-1.50 (t, J
) 7.2 Hz, 3H, N-C-CH3). Elemental analysis: calculated for C17H17N3O4,
C 62.38, H 5.23, N 12.84; found, C 61.97, H 5.28, N 12.50.
1
white solid. H NMR (300 MHz/CDCl3): δ 7.39 (s, 2H), 7.35-7.32
(d, J ) 8.4 Hz, 2H), 7.08-7.05 (d, J ) 8.4 Hz, 2H), 3.05 (s, 2H), 2.62
(s, 6H).
(vi) 3,3′-Diethyl-2,2′-dimethyl-6,6′-ethylenedibenzoxazolium Tosyl-
ates. To a round-bottom flask were added 0.35 g (1.3 mmol) of 2,2′-
dimethyl-6,6′-ethylenedibenzoxazole and 2.0 g (10 mmol) of ethyl
tosylate. The solid mixture was heated at 100 °C under nitrogen for 20
h to give a yellowish precipitate. White solid (0.7 g, 78%) was obtained
after the precipitate was taken up in CH2Cl2 (8 mL), filtered, and washed
with CH2Cl2 (5 mL × 3) and diethyl ether (5 mL × 2). 1H NMR (300
MHz/CD3OD): δ 7.97 (s, 2H), 7.85-7.82 (d, J ) 8.7 Hz, 2H), 7.66-
7.63 (d, J ) 8.7 Hz, 2H), 7.63-7.60 (d, J ) 7.5 Hz, 4H), 7.20-7.18
(d, J ) 7.5 Hz, 4H), 4.60(q, J ) 7.2 Hz, 4H), 3.25 (s, 4H, CH2), 3.06
(s, 6H, CH3), 2.33 (s, 6H, CH3), 1.54 (t, J ) 7.2 Hz, 6H, CH3-C).
(vii) 2,2′-Di(2-anilidovinyl)-3,3′-diethyl-6,6′-ethylenedibenzoxazo-
lium Tosylates. To a round-bottomed flask were added 350 mg (0.51
mmol) of 3,3′-diethyl-2,2′-dimethyl-6,6′-ethylenedibenzoxazolium to-
sylates, 350 mg (1.8 mmol) of N,N′-diphenylformamidine, and 5 mL
of acetic anhydride. The reaction mixture was heated with stirring at
124-128 °C for 7 min under a dry atmosphere and then cooled to
yield an orange precipitate. A total 110 mg (0.12, 24%) of yellow solid
was obtained after the mixture was filtered and washed with acetone
(5 mL × 3). 1H NMR (300 MHz/CD3OD): δ 8.80-8.76 (d, J ) 12.3
Hz, 2H, CHd), 7.72-7.20 (m, 16H, aromatic), 5.9-5.86 (d, J ) 12.3
Hz, 2H, CHd), 4.33-4.31 (q, J ) 7.5 Hz, 4H, N-CH2), 3.19 (s, 4H,
CH2), 2.36 (s, 6H, CH3), 1.51-1.46 (t, J ) 7.5 Hz, 6H, N-C-CH3).
(viii) Dimer-2. Under nitrogen, 100 mg (0.079 mmol) of 2,2′-di(2-
anilidovinyl)-3,3′-diethyl-6,6′-ethylenedibenzoxazolium tosylates and
100 mg (0.45 mmol) of 1,3-dimethylbarbituric acid were dissolved in
dried methylene chloride (100 mL). Triethylamine (0.4 mL) was added
to the above solution. The reaction mixture was condensed to a solid
residue after it was stirred for 4 d. The solid residue was chromato-
graphed with a mix of 5:45:50 methanol/chloroform/ethyl acetate (1.3
L), and with a mix of 5:95 methanol/chloroform, to give 48 mg of
yellow solid in 60% yield (99% pure by HPLC with 1:19 methanol/
(c) Synthesis of Dimer-2 According to Scheme 2. (i) Bis(4-
methoxyphenyl)ethane. To a round-bottom flask were added 10 g
(0.039 mmol) of 4′-methoxy-2-(4-methoxyphenyl)acetophenone, 100
mL of toluene, 30 g of amalgamated zinc,33 and 40 mL of 20% HCl.
The mixture was stirred at 40 °C for 4 h and then at 100 °C overnight.
The organic layer was separated and condensed to a solid residue, which
was dissolved in CHCl3 (250 mL) and combined with the extract from
the aqueous layer. The CHCl3 solution was washed with H2O (2 ×
100 mL), dried with MgSO4, and evaporated under vacuum to yield
9.0 g of yellowish solid. The crude product was taken up in diethyl
ether (50 mL) and filtered to give 6.5 g of an off-white solid. Finally,
5.5 g (0.023 mol, 59%) of white solid was obtained after the crude
was chromatographed on silica gel using a mix of 1:3:6 of ethyl acetate/
1
hexane/benzene (1.3 L). H NMR (300 MHz/CDCl3): δ 7.10-7.07
(d, J ) 8.2 Hz, 4H), 6.84-6.81 (d, J ) 8.2 Hz, 4H), 3.70 (s, 6H), 2.83
(s, 4H).
(ii) Bis(4-hydroxyphenyl)ethane. In a 150-mL Erlenmyer flask,
fitted with a septum, was dissolved 3.6 g (0.015 mol) of bis(4-
methoxyphenyl)ethane in 20 mL of dried CH2Cl2. The solution was
cooled to -70 °C in a dry ice/2-propanol bath, and boron tribromide
(1.6 mL, 0.011 mol) was added through septum. The reaction mixture
was stirred at room temperature for 2 h and then poured into icy water
(40 mL) in an icebath to give a suspension after the solution was
neutralized with sodium acetate. The suspension was filtered, washed
with H2O (2 × 20 mL), and air-dried overnight to give a white crude
product. Finally, 3.0 g (0.014 mol, 93%) of white solid was obtained
after the crude was taken up in methanol (100 mL), filtered, and
condensed. 1H NMR (300 MHz/CD3OD): δ 6.90 (d, J ) 8.2 Hz, 4H),
6.65 (d, J ) 8.2 Hz, 4H), 2.70 (s, 4H).
1
chloroform). H NMR (300 MHz/CF3CO2D): δ 8.75 (s, 2H, CHd),
7.75 (m, 2H, aromatic H), 7.54 (m, 6H, aromatic H), 4.58 (t, 4H,
N-CH2), 3.66 (s, 12H, NCH3), 3.33 (s, 4H, CH2), 1.68 (t, J ) 6.0 Hz,
6H, C-CH3). MS (FAB): m/e 681.3 (M+ + 1).
(iii) Bis(3-nitro-4-hydroxyphenyl)ethane. To a diluted nitric acid
solution at 5 °C was added 5.0 g (0.023 mol) of bis(4-hydroxyphenyl)-
ethane.33 The mixture was stirred at 25 °C for 20 h and then for another
2 d after 10 mL of acetonitrile was added to the above mixture. Finally,
4.0 g (57%) of yellow solid was obtained after the mixture was filtered
(ix) Dimer-1. Synthesis of dimer-1 is two steps shorter because the
intermediate, bis(4-hydroxyphenyl)methane, is commercially available.
A yellow solid is obtained in 42% yield (one peak by HPLC with 1:19
methanol/chloroform). 1H NMR (300 MHz/CF3CO2D): δ 8.754 (s, 2H,
CHd), 7.794-7.766 (d, J ) 8.4 Hz, 2H, aromatic H), 7.565 (m, 6H,
aromatic H), 4.558 (t, 4H, N-CH2), 4.487 (s, 2H, CH2), 3.643 (s, 12H,
NCH3), 1.666 (t, J ) 6.3 Hz, 6H, C-CH3). MS (FAB): m/e 667.2 (M+
+ 1).
1
and washed with water (10 mL × 5) and methanol. H NMR (300
MHz/CDCl3): δ 10.48 (s, OH), 7.88 (d, J ) 2.1 Hz, 2H), 7.37-7.33
(dd, J1 ) 8.4 Hz, J2 ) 2.1 Hz, 2H), 7.10-7.07 (d, J ) 2.1 Hz, 2H),
2.92 (s, 4H).
(32) Brooker, L. G. S.; Keyes, G. H.; Sprague, R. H.; Vandyke, R. H.;
Vanlare. E.; Vanzandt, G.; White, F. L. J. Am. Chem. Soc. 1951, 73, 5326.
(33) Furniss, B. S.; Hannaford, A. J.; Smith, P. W. G.; Tatchell, A. R.
In Vogel’s Textbook of Practical Organic Chemistry; Vogel, A. I., Ed.;
Longman: Essex, England, 1989.
(x) Dimer-0. Synthesis of dimer-0 is four steps shorter because the
intermediate, 3,3′-diamino-4,4′-dihydroxybiphenyl, is commercially
available. An orange solid is obtained in 52% yield (one peak by HPLC
1
with 1:19 methanol/chloroform). H NMR (300 MHz/CF3CO2D): δ

8156 J. Am. Chem. Soc., Vol. 121, No. 36, 1999
Lu et al.
8.838 (s, 2H, CHd), 8.088 (s, 2H, aromatic H), 8.000-7.972 (d, J )
8.4 Hz, 2H, aromatic H), 7823-7.795(d, J ) 8.4 Hz, 2H, aromatic H),
4.635 (q, J ) 7.2 Hz, 4H, N-CH2), 3.643 (s, 12H, NCH3), 1.752-
1.711 (t, J ) 7.2 Hz, 6H, N-C-CH3). MS (FAB): m/e 653.2 (M+ + 1).
Supporting Information Available: Complete crystal-
lographic information for the monomer dye, including atomic
coordinates for all atoms, bond lengths and angles, anisotropic
displacement parameter, and torsion angles (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
Acknowledgment. We are grateful to the National Science
Foundation for support of this research and Dr. Steve Atherton
for help in obtaining the fluorescence lifetimes.
JA983778H