Enhanced Fluorescence of Remote Functionalized Diaminodicyanoquinodimethanes in the Solid State and Fluorescence Switching in a Doped Polymer by Solvent Vapors

Article abstract of DOI:10.1002/chem.200305123

Remote functionalized zwitterionic diaminodicyanoquinodimethanes are found to exhibit a dramatic enhancement of light emission in the solid state and when doped in polymer films, as compared to the solution state. Crystal structure analysis of prototypica

Full text of DOI:10.1002/chem.200305123

FULL PAPER
Enhanced Fluorescence of Remote Functionalized
Diaminodicyanoquinodimethanes in the Solid State and Fluorescence
Switching in a Doped Polymer by Solvent Vapors
S. Jayanty and T. P. Radhakrishnan*[a]
Abstract: Remote functionalized zwit-
terionic diaminodicyanoquinodime-
state molecular organization. Semiem-
pirical quantum chemical computations
provide a viable model to explain the
interesting phenomenon of fluores-
cence enhancement as arising from the
inhibition of geometry relaxation of
the vertical excited state to a nonemit-
ting state. The reversible switching of a
doped polymer film fluorescence trig-
gered by solvent vapors is demonstrat-
ed.
thanes are found to exhibit a dramatic
enhancement of light emission in the
solid state and when doped in polymer
films, as compared to the solution
state. Crystal structure analysis of pro-
totypical molecules reveals the role of
the remote functionality in the solid
Keywords: diaminodicyanoquinodi-
methane
¥
fluorescence
enhancement
¥
semiempirical
calculations ¥ solvent switch
Introduction
interesting and useful application potential in areas such as
electroluminescence and sensors. However, detailed pictures
of the physical origin of many of these observations are yet
to emerge.
The strong fluorescence exhibited by most organic dye mol-
ecules in the monomeric state is quenched in the aggregated
state. The current focus on the photoluminescence and elec-
troluminescence of molecular materials provides impetus to
explore novel chromophores which not only do not suffer
fluorescence quenching in the aggregated state, but display
enhanced light emission in the solid state and in polymer
films. Only a few such cases have been reported. The super-
radiance observed in some thiacarbocyanine molecules has
been attributed to dynamic energy transfer from the mono-
meric species to aggregate structures.[1] The highly emissive
phase of a poly(p-phenyleneethynylene) was proposed to
either arise from the metastable skewed chain alignments
that allowed efficient interchain interactions or result from
planarization effects.[2,3] Nano-sized aggregates of a sterically
crowded silole[4] and a biphenyl substituted ethylene[5] have
been reported to show strong light emission, presumably
due to significant ground state conformational changes ac-
companying the aggregate formation. These reports suggest
During our investigations of optical second harmonic gen-
eration in zwitterionic diaminodicyanoquinodimethanes
11]
(DADQ)[6
we have found that several molecules, espe-
cially the ones bearing remote functionalities[9
show en-
11]
hanced fluorescence in the solid state compared to the solu-
tions; similar effect was observed with these molecules
doped in polymer films. Remote functionality refers to
groups which are not in p-conjugation with the DADQ
chromophore unit and therefore do not disturb the low
lying electronic states of the chromophore. The role of vis-
cous solvents and polymer matrices in enhancing the light
emission of two related molecules have been investigated by
Bloor et al.[12] Even though similar effects in solids were
briefly noted, no detailed studies were presented or the
impact of the molecular assembly in crystals analyzed. Fur-
ther, these molecules do not possess remote functionalities
involved in intermolecular noncovalent interactions, which
as we will demonstrate below is a crucial feature in our sys-
tems. Our detailed investigation of the solution and solid
state of a family of molecules and doped polymer films, es-
tablishes the generality of this interesting phenomenon and
the molecular design principle involved. Zwitterionic
DADQ×s are prone to aggregation and the remote function-
alities such as amino groups facilitate specific and directed
intermolecular interactions promoting extended assembly.
The remote groups appear to contribute positively to the
[a] S. Jayanty, Prof. T. P. Radhakrishnan
School of Chemistry, University of Hyderabad
Hyderabad, 500 046 (India)
Fax : (+91)40-2301-2460
E-mail: tprsc@uohyd.ernet.in
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author and contains the
electronic absorption/emission of solutions of 1 8, details of the crys-
tal structures of 2 and 3, summary of semiempirical computations (18
pages).
791
Chem. Eur. J. 2004, 10, 791 797
DOI: 10.1002/chem.200305123
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPER
fluorescence enhancement since all the remote functional-
ized DADQ×s we have examined exhibit this effect whereas
many of the DADQ×s without such functionality do not.
More significantly, the presence of the remote groups im-
parts appreciable solubility to these molecules in water in
addition to organic solvents. This is an important advantage
since it facilitates convenient doping of these molecules in
polymers which are soluble only in water enabling potential
application in instances such as organic vapor sensing.
Results and Discussion
The family of molecules 1 8 exhibited fluorescence en-
hancement to varying extents in the crystalline state and in
polymer films, compared with the solutions. We present the
detailed results for 3 as an illustrative case and subsequently
provide a summary of the observations for the whole family.
The absorption and emission spectra of 3 in solvents of dif-
ferent polarity are shown in Figure 2a (see Experimental
Section for details). The negative solvatochromism of the
absorption spectra indicates the highly polar nature of the
ground state of the molecule consistent with dipole moment
We have carried out a systematic investigation on a family
of DADQ molecules with remote amino functionality in-
cluding some in the form of salts (Figure 1); we present
measurements on similar molecules reported earlier.[13
15]
The fluorescence quantum yield of 3 in acetonitrile solution
estimated by comparison with quinine sulfate is only about
0.11%; this is typical of such molecules.[12] However, the
solid sample as well as 3 doped in polymer films, polyvinyl
alcohol (PVA) and poly(sodium 4-styrenesulfonate) (PSS)
show bright green light emission when excited near their ab-
sorption peaks. The polymer films show clearly visible fluo-
rescence even under ambient lighting. The spectra of 3 in
the solid state and in polymer films are presented in Fig-
ure 2b. The strong enhancement of the emission compared
with the solutions is evident from the normalized intensities.
The absorption spectra in the solid state and in polymer
films show similar linewidth and shape as the solution spec-
tra, which indicates that the electronic states involved are
molecular. Unlike the systems reported earlier,[2,3,5] molecu-
lar aggregation in the solid state appears to have little
impact on the electronic absorption spectrum. This also im-
plies that the extent of aggregation in the polymer films
cannot be clearly established from the spectral data. The ex-
citation spectra of the various cases, solution, solid and poly-
mer films, show close resemblance to the absorption spectra
suggesting that the emission is from the vertically excited
state. The Stokes shifts observed are also within the range
expected for vibrational relaxation alone of the excited
state. The absorption and emission maxima of 1 8 in aceto-
nitrile solution, solid state and polymer films are summar-
ized in Table 1. The emission intensity of the various mole-
cules in different solvents have the same order of magni-
tude, but as in the case of 3 (Figure 2a) the emission from
low polar solvents are found to be slightly stronger than that
from more polar solvents. The large enhancement of the
normalized emission intensity in the solid state and in poly-
mer films over the acetonitrile solution for the various com-
Figure 1. Molecules studied; the remote functionalities are in bold.
herein the salient findings which establish their enhanced
fluorescence in the solid state and in polymer films. We also
present crystal structures of prototypical systems and semi-
empirical computational analysis of the molecular structure
and energy in the ground and excited states leading to a
simple model to explain this interesting phenomenon. A
novel application potential of these materials is demonstrat-
ed through experiments on a chromophore doped in poly-
mer film, showing reversible cycles of solvent vapor trig-
gered fluorescence switching.
em
Table 1. Visible absorption (lambasx) and emission (l ) peaks of 1 8 in acetonitrile solution, solid state and in PVA and PSS films and the ratio, R of the
max
maximum intensity of the normalized emission in the solid and polymer films to that in acetonitrile solution.
Molecule
Acetonitrile solution
Solid
[lmemax] /nm
PVA film
[lmemax] /nm
PSS film
[lmemax] /nm
abs
abs
max
abs
max
abs
max
l
[lmemax] /nm
l
R
l
R
l
R
max
1
2
3
4
5
6
7
8
418 [539]
420 [543]
407 [538]
407 [533]
408 [539]
449 [549]
422 [540]
408 [536]
418 [525]
422 [492]
422 [495]
419 [534]
410 [553]
423 [526]
418 [518]
403 [531]
13
408 [526]
414 [514]
405 [505]
391 [514]
390 [510]
411 [509]
398 [518]
399 [523]
18
38
100
14
218
60
431 [543]
431 [536]
420 [527]
415 [526]
413 [524]
447 [547]
422 [528]
421 [525]
29
27
132
69
31
34
36
35
171
216
63
27
227
23
25
33
30
792
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 791 797

Enhanced Fluorescence
791 797
fact that in spite of such effects,
most organic dye molecules ex-
hibit stronger fluorescence in
solution than in the solid state,
one has to seek an explanation
for the enhanced fluorescence
observed in the present materi-
als. Before embarking on an in-
vestigation of the origin of the
fluorescence enhancement we
have carried out single crystal
X-ray structure analysis of 2
and 3 to gain insight into the
molecular structure of these
compounds and to explore the
role of the remote functionality.
The crystals are found to
belong to the monoclinic space
group, P21/n and orthorhombic
space group, Pbca, respective-
ly.[16] The basic crystallographic
data are presented in Table 2
and the molecular structures
are shown in Figure 3.
Both molecules show
a
strong twist between the diami-
nomethylene unit and the ben-
zenoid ring plane,
a feature
common among these mole-
11,17]
cules;[6
the dihedral angles,
C2-C1-C7-N9 and C6-C1-C7-
N10 are 52.6 and 53.58 in 2 and
50.6 and 53.08 in 3, respectively.
It should be noted that the sit-
uation is quite different from
that in the case of the phenyle-
neethynylenes,[2,3] silole[4] and
cyano-biphenylethylene[5] where
the planar ground state confor-
mation in the aggregates is con-
sidered to be an important
factor contributing to the strong
fluorescence. The bond lengths
Figure 2. Absorption (c) and emission (a) spectra of 3 a) in different solvents and b) in the solid state
and polymer films (acetonitrile solution spectra with similar absorption lmax are shown for comparison). The
emission spectra are normalized by dividing by the optical density at lmax; note the different scales in a) and
b).
pounds are listed in Table 1 (the absorption lmax of solids
and doped polymer films are closest to that in acetonitrile
than in other solvents); the enhancement factor, R ranges
from 13 to 227. We note that the fluorescence intensities of
the solids and doped polymer films measured may be less
than the actual values due to factors such as self-absorption
and wave-guiding. The latter is particularly relevant for pol-
ymer films where the light emission from the edges is not
fully accounted for in the normal sample geometry em-
ployed. While these effects are difficult to quantify, the qual-
itative considerations indicate that the fluorescence en-
hancements we report for these materials are really their
lower limits.
in the aromatic ring of 2 and 3 are typical of benzenoid
structures and point to the zwitterionic nature of the mole-
cules.[17] Examination of intermolecular contacts in these
crystals reveals moderately close ones between the carbon
atoms connected to the remote N atom of the piperazine
units and the negatively polarized dicyanomethylene end of
the zwitterionic diaminodicyanoquinodimethane moiety.
The extended structure which results from such contacts in 2
and 3 are shown in Figure 4. Even though these noncovalent
interactions are likely to be weak, their cooperative influ-
ence would steer the organization of the zwitterionic
DADQ molecules and contribute to the enhanced tendency
of these remote functionalized molecules towards aggregate
formation. Strong intermolecular hydrogen-bond interac-
tions mediated by the remote groups have been observed in
6[11] and a derivative of 4 and 5[7] in earlier studies. All these
Deactivation of the chromophore excited state due to col-
lisions with solvent molecules may contribute to the lower
fluorescence observed in solutions. However, in view of the
793
Chem. Eur. J. 2004, 10, 791 797
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

S. Jayanty and T. P. Radhakrishnan
FULL PAPER
Table 2. Basic crystallographic data for 2 and 3.
2
3
formula
Fw
C20H26N6
350.47
C19H23N5
321.42
crystal system
space group
color/appearance
dimensions [mm3]
a [ä]
monoclinic
P21/n (No. 14)
yellow plate
0.64î0.64î0.36
8.581(3)
20.157(3)
11.4040(13)
90.085(18)
4
orthorhombic
Pbca (No. 61)
yellow plate
0.52î0.32î0.24
16.32(7)
12.400(17)
18.13(2)
90.0
b [ä]
c [ä]
b [8]
Z
8
0.72
293 (2)
m [cmÀ1
T [K]
]
0.74
293 (2)
l [ä]
0.71703
0.71703
no. refls
with Iꢀ2sI
no. parameters
GOF
3080
4597
235
0.992
0.0429
217
0.983
0.0550
R
[for Iꢀ2sI]
wR2
0.1085
0.2374
[for Iꢀ2sI]
molecular associations are likely to play a significant role in
obstructing the excited state conformational relaxation dis-
cussed below.
In order to investigate the likely molecular structure in
solution, we have optimized the geometry of 3 using the
AM1 method.[18] The molecular structure from crystal analy-
sis served as a convenient starting point for the optimization
Figure 4. Intermolecular contacts (broken lines) leading to tape structures
in a) 2 (extending along the crystallographic a axis; R1=3.399, R2=
3.564, R3=3.711, R4=3.787 ä) and b) 3 (extending along the crystallo-
graphic b axis; R1=3.698, R2=3.760, R3=3.897 ä).
and the solvent environment was modeled by invoking the
value of about 668 at e>40. The q observed in the crystal is
obtained when e is in the range 2 4. CI calculations on the
optimized ground state geometry provided the excitation en-
ergies. The enthalpy of formation of the ground state and
the lowest vertical excited state at different e are shown in
Figure 5.
22]
COSMO subroutine.[19] As demonstrated earlier,[11,20
the
same approach can be used to model the impact of the mo-
lecular environment in the solid state as well. Similar to our
observation on related molecules,[11] the molecular twist, q
of the optimized ground state geometry of 3 is found to in-
crease with the dielectric constant, e employed in the com-
putation; it increases from 39.88 at e=1 and saturates at a
The computed lmax agree very well with those observed
for low polar solvents like dichloromethane and chloroform
Figure 3. Molecular structure of a) 2 and b) 3 from single crystal X-ray analysis. 10% probability thermal ellipsoids are indicated and hydrogen atoms
are omitted for clarity.
794
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 791 797

Enhanced Fluorescence
791 797
Figure 6. Schematic diagram illustrating the mechanism of fluorescence
enhancement of diaminodicyanoquinodimethanes in the solid state and
doped polymer films.
the structure relaxation would be hindered either by the ri-
gidity of the polymer matrix or the aggregate formation, de-
pending on the extent of aggregation. The energy gap be-
tween the vertical and relaxed excited states increases with
e (Figure 5) suggesting that, as the polarity of the solvent en-
vironment increases, the relative population of molecules in
the vertical excited state would diminish progressively lead-
ing to reduction of fluorescence intensity; this is in agree-
ment with the emission spectra observed in solution (Fig-
ure 2a). Future investigations would target the excited state
dynamics in solution, solid and doped polymer states and
their temperature dependence to gain further insight into
the mechanism we have presented.
In order to examine the application potential of the fluo-
rescence enhancement phenomenon, we have carried out an
experiment on PVA films containing 3, by alternately expos-
ing it to chloroform vapors and drying. The fluorescence in-
tensity of the film showed a clear reduction when it was en-
veloped by chloroform vapors emanating from the solvent
placed at the bottom of the spectrometer cuvette. When the
chloroform was removed and the film dried free of the
vapors, the original fluorescence intensity was nearly re-
gained (Figure 7). Two subsequent cycles showed stronger
reduction of the fluorescence in presence of the vapors and
Figure 5. AM1/CI calculated enthalpy of formation at different e, of 3 in
the optimized ground state, the lowest vertical excited state, the opti-
&
mized excited state and the ground state of the optimized excited state:
~
optimized ground state,
&
vertical excited state,
optimized excited
~
state, ground state of optimized excited state.
as well as the highly polar solvent, water. The lmax at e=4 is
in good agreement with the experimental value in the solid
state. As noted earlier, the observed emission is likely to be
from the vertical excited state alone. Even though geometry
relaxation of the excited state is unlikely in the rigid solid
state or polymer matrix, it is quite feasible in the solution
state. In order to investigate the likely geometry relaxation,
we have optimized the excited state structure under differ-
ent e values. The conspicuous change is the increase of q to
~908 in all cases. The enthalpy of formation of these opti-
mized geometries are plotted in Figure 5. The Figure also
shows the enthalpy of formation of the ground states of
these geometries from which we can estimate the upper
bound for the emission energies of these relaxed geometries.
The values are found to be in the range, 1588 673 nm (for e
ranging from 1 80), very much smaller than the observed
emission energies suggesting that no visible light emission
from the relaxed excited state geometries is observed in the
solution, solid or polymer film. These computational results
combined with the experimental facts noted earlier lead us
to the conclusion that the enhancement of the fluorescence
in the solid state and polymer films arise as a result of ar-
resting the geometry relaxation of the excited fluorescent
state to a nonfluorescent state. Conversely, geometry relaxa-
tion of the excited state leads to diminished fluorescence in
solutions (Figure 6). Both excited state geometries are twist-
ed and the difference is the extent of twisting; this situation
may be contrasted with TICT systems[23] wherein planar and
twisted geometries are considered. The current model is
consistent with the qualitative picture proposed earlier for
the matrix dependence of light emission from similar mole-
cules.[12] It is important to note that in the model we develop
for the DADQ×s, the role of molecular aggregation is to fa-
cilitate further, the inhibition of excited state structure re-
laxation, rather than to force any kind of ground state ge-
ometry changes as implied in the models put forward in the
Figure 7. The variation of emission maximum (at 505 nm) of 3 in PVA
film on exposure to chloroform vapors and subsequent drying, through
5]
case of the systems reported earlier.[3 In the polymer films,
consecutive steps: on drying, on exposure to chloroform.
&
*
795
Chem. Eur. J. 2004, 10, 791 797
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

S. Jayanty and T. P. Radhakrishnan
FULL PAPER
the solution was stirred at 708C for 0.5 h and then at 308C for 1 h. Yel-
lowish green fluorescent precipitate of 3 which separated out was filtered
and dried (0.21 g, 81%).
return to higher fluorescence in the dry state. Some change
in the dry state fluorescence was observed during the initial
cycles. The values stabilized subsequently and good reversi-
bility was attained. At this stage, the reduction of fluores-
cence due to the vapors is quite substantial (~33%). Since
the PVA matrix does not dissolve in chloroform and abso-
lutely no leaching of the compound is observed throughout,
we believe that the solvent vapors diffuse into the film and
possibly cause local dissolution of part of the compound
leading to fluorescence reduction. On drying, solid aggre-
gates may be reformed recovering the enhanced fluores-
cence. Alternatively, a mild softening of the PVA matrix in-
duced by the solvent vapor may facilitate structure relaxa-
tion and fluorescence quenching, which is reversed on
drying. The present study provides a quantitative demon-
stration of the reversibility and repeatability of solvent
vapor induced switching of the enhanced fluorescence in a
chromophore doped polymer film.
7-(3-R-(+)-Hydroxypyrrolidino)-7-(piperazino)-8,8-dicyanoquinodime-
thane p-toluenesulfonate (8): Compound 5 was synthesized following
similar procedure as for 3. A solution of p-toluenesulfonic (PTS) acid
(0.175 g, 1 mmol) in acetonitrile (5 mL) was added to a solution of 5
(0.25 g, 0.73 mmol) in acetonitrile (50 mL). The yellow microcrystalline
powder of the PTS salt, 8 which precipitated out immediately was filtered
and dried (0.307 g, 77%).
Characterization including single crystal structure of 1[25] and 6[11] have
been reported earlier. The detailed characterization of the new com-
pounds are provided below.
7,7-Bis(N-methylpiperazino)-8,8-dicyanoquinodimethane (2): Yield=
82%, recrystallized from acetonitrile; m.p. 278 2808C (decomp); FTIR
(KBr): n˜ =2172.0, 2135.4, 1595.3 cmÀ1; UV/Vis (acetonitrile): lmax =269,
420 nm; 1H NMR (CDCl3): d=1.7 (s, 6H), 2.6 2.7 (m, 8H), 3.5 3.6 (m,
8H), 7.1 ppm (s, 4H); 13C NMR ([D6]DMSO): d=34.5, 45.5, 50.9, 54.6,
114.1, 118.1, 123.2, 131.8, 149.8, 168.8 ppm; elemental analysis (%) calcd
for C20H26N6: C 68.57, H 7.43, N 24.00; found: C 68.31, H 7.11, N 24.61.
7-(N-Methylpiperazino)-7-pyrrolidino-8,8-dicyanoquinodimethane
(3):
Yield=75%; recrystallized from acetonitrile; m.p. 260 2628C (decomp);
FTIR (KBr): n˜ =2941.7, 2170.1, 2133.5, 1595.3 cmÀ1, UV/Vis (acetoni-
trile): lmax =267, 407 nm; 1H NMR (CDCl3): d=1.6 (s, 3H), 2.0 2.1 (t,
4H), 2.15 2.25 (t, 4H), 3.5 3.6 (t, 4H), 3.60 3.75 (t, 4H), 7.0 ppm (s,
4H); 13C NMR ([D6]DMSO): d=24.1, 25.7, 33.1, 45.5, 50.1, 52.7, 54.6,
115.2, 118.1, 123.7, 130.5, 148.3, 165.9 ppm; elemental analysis (%) calcd
for C19H23N5: C 71.03, H 7.17, N 21.80; found: C 70.93, H 7.22, N 21.72.
Conclusion
We have explored a family of zwitterionic diaminodicyano-
quinodimethanes and demonstrated enhanced light emission
from them in the solid state and polymer films. The utility
of the remote functionality in promoting molecular assembly
is revealed by the crystallographic investigation of two of
the molecules. A viable mechanism for the fluorescence en-
hancement is proposed through semiempirical computation-
al analysis of the ground and excited state geometries and
energies. We believe that such analysis may prove useful in
understanding the basis for similar observations in other
molecules reported previously. Experimental demonstration
of reversible fluorescence switching triggered by solvent
vapors is presented. These materials provide an ideal testing
ground for the fascinating and relatively rare phenomenon
of light emission enhancement in aggregates and are poten-
tial candidates for the development of novel photonic devi-
ces based on such phenomena.
7-(3-Hydroxypyrrolidino)-7-piperazino-8,8-dicyanoquinodimethane (4):
Yield=75%; recrystallized from acetonitrile/methanol; m.p. 245 2468C
(decomp); FTIR (KBr): n˜ =3390.0, 2170.1, 2131.5, 1597.2 cmÀ1; UV/Vis
(acetonitrile): lmax =273, 407 nm; 1H NMR ([D6]DMSO): d=1.9 2.1 (m,
3H), 2.75 2.90 (m, 4H), 3.7 3.8 (m, 4H), 4.3 4.4 (m, 4H), 5.3 (s, 1H),
6.8 6.9 (d, 2H), 7.15 7.20 ppm (d, 2H), hydroxy proton was not ob-
served; 13C NMR ([D6]DMSO): d=32.5, 33.1, 34.2, 46.1, 50.6, 51.8, 60.3,
67.6, 69.1, 115.1, 118.1, 123.8, 130.6, 148.3, 166.5 ppm; elemental analysis
(%) calcd for C18H21N5O: C 66.87, H 6.50, N 21.67; found: C 66.86, H
6.49, N 21.68.
7-(3-R(+)-Hydroxypyrrolidino)-7-piperazino-8,8-dicyanoquinodimethane
(5): Yield=77%; recrystallized from acetonitrile/methanol; m.p. 213
2158C (decomp); FTIR (KBr): n˜ =3308.0, 2170.0, 2131.0, 1597.0 cmÀ1
UV/Vis (acetonitrile): lmax =270, 408 nm; H NMR ([D6]DMSO): d=1.8
;
1
2.1 (m, 3H), 2.7 2.9 (m, 4H), 3.6 3.9 (m, 4H), 4.2 4.5 (t, 4H), 5.3 (s,
1H), 6.8 6.9 (d, 2H), 7.2 7.35 ppm (d, 2H), hydroxy proton was not ob-
served; 13C NMR ([D6]DMSO): d=32.2, 33.0, 34.3, 46.0, 50.6, 51.7, 60.3,
68.0, 69.1, 115.1, 118.1, 123.8, 130.6, 148.3, 166.5 ppm; elemental analysis
(%) calcd for C18H21N5O: C 66.87, H 6.50, N 21.67; found: C 66.22, H
6.85, N 21.68.
7-(Hydroxypyrrolidino)-7-piperazino-8,8-dicyanoquinodimethane p-tolu-
enesulfonate (7): Yield=75%; recrystallized from acetonitrile/methanol;
m.p. 250 2558C (decomp); FTIR (KBr): n˜ =3398.2, 3022.8, 2177.8,
2137.0, 1597.0, 1005.0, 887.0, 814.0 cmÀ1; UV/Vis (acetonitrile): lmax =267,
Experimental and Computational Section
Synthesis and characterization
1
422 nm; H NMR ([D6]DMSO): d=1.9 2.0 (m, 3H), 2.1 2.2 (m, 4H), 2.3
Compounds 1 8 were prepared using minor modifications of procedures
(s, 3H), 3.8 4.0 (m, 4H), 4.30 4.45 (t, 2H), 5.25 5.40 (m, 2H), 6.8 6.9
(m, 2H), 7.1 7.2 (d, 2H), 7.3 7.4 (m, 2H), 7.5 7.6 (d, 2H), 8.85 9.15 ppm
(s, 2H), hydroxy proton was not observed; 13C NMR ([D6]DMSO): d=
21.0, 32.5, 33.3, 34.0, 43.1, 44.7, 49.7, 50.8, 60.1, 60.6, 67.5, 69.1, 114.7,
118.1, 123.7, 125.7, 128.4, 130.7, 138.2, 145.5, 148.5, 167.0 ppm, the extra
peaks additional to the number of symmetry nonequivalent atoms ap-
pears to result from clustering of these zwitterionic molecules, since we
have verified their reproducibility and by repeated purification ensured
that they are not due to any impurities which is further confirmed by the
satisfactory elemental analysis; elemental analysis (%) calcd for
C25H29N5SO4: C 60.61, H 5.86, N 14.14; found: C 60.62, H 5.89, N 14.24.
8,24]
reported[6
for similar molecules. Three typical cases are described below.
7,7-Bis(N-methylpiperazino)-8,8-dicyanoquinodimethane (2): N-Methyl-
piperazine (0.34 g, 3.4 mmol) was added to a warm solution of TCNQ
(0.30 g, 1.47 mmol) in acetonitrile (30 mL) (CAUTION: HCN is a by-
product). The solution turned dark green immediately and changed to
yellow subsequently. Yellow crystalline product formed in about 1 h. The
reaction mixture was stirred for 1.5 h more at 758C. The precipitate was
filtered and dried (0.39 g, 82%).
7-(N-Methylpiperazino)-7-pyrrolidino-8,8-dicyanoquinodimethane
(3):
Pyrrolidine (0.094 g, 1.32 mmol) was added to a warm solution of TCNQ
(0.30 g, 1.47 mmol) in acetonitrile (30 mL). The solution turned purple
immediately and a purple crystalline compound precipitated in 1 h. The
reaction mixture was stirred for 1.5 h more at 758C. The precipitate of 7-
pyrrolidino-7,8,8-tricyanoquinodimethane, PTCNQ was filtered and dried
(0.305 g, 83%). N-Methylpiperazine (0.12 g, 1.2 mmol) was added to a
warm solution of PTCNQ (0.20 g, 0.98 mmol) in acetonitrile (20 mL) and
7-(3-R(+)-Hydroxypyrrolidino)-7-(piperazino)-8,8-dicyanoquinodime-
thane p-toluenesulfonate (8): Yield=77%; recrystallized from acetoni-
trile/methanol; m.p. 239 2448C (decomp); FTIR (KBr): n˜ =3362.2,
3015.0, 2177.8, 2137.3, 1597.2, 1005.0, 887.3, 814.0 cmÀ1; UV/Vis (acetoni-
trile): lmax =265, 408 nm; 1H NMR ([D6]DMSO): d=1.9 2.0 (m, 3H),
2.1 2.2 (m, 4H), 2.3 (s, 3H), 3.6 3.9 (m, 4H), 4.25 4.45 (m, 2H), 5.25
796
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 791 797

Enhanced Fluorescence
791 797
5.40 (m, 2H), 6.8 6.9 (m, 2H), 7.1 7.2 (d, 2H), 7.25 7.4 (m, 2H), 7.5 7.6
(d, 2H), 8.8 9.2 ppm (s, 2H), hydroxy proton was not observed;
13C NMR ([D6]DMSO): d=21.0, 32.6, 33.7, 34.1, 43.0, 47.1, 50.7, 51.3,
60.0, 61.0, 67.6, 69.3, 114.5, 118.1, 123.6, 125.7, 128.4, 130.9, 138.4, 145.2,
148.8, 166.7 ppm, the extra peaks additional to the number of symmetry
nonequivalent atoms appears to result from clustering of these zwitter-
ionic molecules, since we have verified their reproducibility and by re-
peated purification ensured that they are not due to any impurities,
which is further confirmed by the satisfactory elemental analysis; elemen-
tal analysis (%) calcd for C25H29N5SO4: C 60.61, H 5.86, N 14.14; found:
C 60.55, H 5.93, N 14.21.
Acknowledgement
Financial support from the Department of Science and Technology, New
Delhi (Swarnajayanti Fellowship), the use of the National Single Crystal
Diffractometer Facility funded by the DST at the School of Chemistry,
University of Hyderabad and the ™University with Potential for Excel-
lence∫ program of the University Grants Commission, New Delhi are ac-
knowledged. S.J. thanks the CSIR, New Delhi for a senior research fel-
lowship. We thank Profs. M. Durga Prasad and A. Samanta for insightful
discussions.
Preparation of doped polymer films
[1] S. ÷zcelik, D. L. Akins, J. Phys. Chem. B 1999, 103, 8926.
[2] R. Deans, J. Kim, M. R. Machacek, T. M. Swager, J. Am. Chem. Soc.
2000, 122, 8565.
[3] M. Levitus, K. Schmieder, H. Ricks, K. D. Shimizu, U. H. F. Bunz,
M. A. Garcia-Garibay, J. Am. Chem. Soc. 2001, 123, 4259.
[4] J. Luo, Z. Xie, J. W. Y. Lam, L. Cheng, H. Chen, C. Qiu, H. S.
Kwok, X. Zhan, Y. Liu, D. Zhu, B. Z. Tang, Chem. Commun. 2001,
1740.
[5] B. An, S. Kwon, S. Jung, S. Y. Park, J. Am. Chem. Soc. 2002, 124,
14410.
[6] M. Ravi, D. N. Rao, S. Cohen, I. Agranat, T. P. Radhakrishnan,
Chem. Mater. 1997, 9, 830.
In the present study we have considered only water soluble polymers,
PVA (MW =15000) and PSS (MW =70000), since we were interested in
exploring the fluorescence response of the doped polymer films in pres-
ence of organic solvent vapors; however, these compounds can be incor-
porated in other polymers as well. Equal volumes of aqueous solutions of
the compound (5 mM) and the polymer (~0.2 gmLÀ1) were mixed and
subjected to ultrasonication for 5 min. The films were prepared by spin
casting on glass plates followed by drying in a vacuum desiccator and hot
air oven at 808C for 1 h each.
Electronic spectroscopy
The electronic absorption spectra of solutions and doped polymer films
were recorded on a Shimadzu UV/Vis spectrometer (Model UV-3101).
The concentrations were adjusted to achieve an optical density of ~0.3 in
all cases. The specular reflectance (88 incidence) spectrum of the solid
compounds in the form of KBr pellets were recorded using the integrat-
ing sphere (ISR 3100) attachment of the spectrometer. The reflectance
spectra were converted to absorption profiles using the Kubelka Munk
function. The weight ratio of the solid compound to KBr was adjusted in
each case to achieve a final optical density of ~0.3. The emission spectra
were recorded on a Jobin Yvon Horiba spectrofluorimeter (Model Fluo-
[7] M. Ravi, P. Gangopadhyay, D. N. Rao, S. Cohen, I. Agranat, T. P.
Radhakrishnan, Chem. Mater. 1998, 10, 2371.
[8] P. Gangopadhyay, S. Sharma, A. J. Rao, D. N. Rao, S. Cohen, I.
Agranat, T. P. Radhakrishnan, Chem. Mater. 1999, 11, 466.
[9] S. Jayanty, T. P. Radhakrishnan, Chem. Mater. 2001, 13, 2072.
[10] S. Jayanty, P. Gangopadhyay, T. P. Radhakrishnan, J. Mater. Chem.
2002, 12, 2792.
[11] S. Jayanty, T. P. Radhakrishnan, Chem. Mater. 2001, 13, 2460.
[12] D. Bloor, Y. Kagawa, M. Szablewski, M. Ravi, S. J. Clark, G. H.
Cross, L. PÂlsson, A. Beeby, C. Parmer, G. Rumbles, J. Mater.
Chem. 2001, 11, 3053.
[13] Y. Kagawa, M. Szablewski, M. Ravi, N. Hackman, G. H. Cross, D.
Bloor, A. S. Batsanov, J. A. K. Howard, Nonlinear Opt. 1999, 22,
235.
[14] R. S. Gopalan, G. V. Kulkarni, M. Ravi, C. N. R. Rao, New J. Chem.
2001, 25, 1108.
[15] J. M. Cole, R. C. B. Copley, G. J. McIntyre, J. A. K. Howard, M. Sza-
blewski, G. H. Cross, Phys. Rev. B 2002, 65, 125107.
[16] CCDC-206354 and -206355 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or from the
Cambridge Crystallographic Data Centre, 12, Union Road, Cam-
bridge CB21EZ, UK; fax: (+44)1223-336-033; or deposit@ccdc.
cam.ac.uk).
[17] M. Ravi, T. P. Radhakrishnan, J. Phys. Chem. 1995, 99, 17624.
[18] M. J. S. Dewar, E. G. Zoebisch, E. F. Healy, J. J. P. Stewart, J. Am.
Chem. Soc. 1985, 107, 3902.
[19] A. Klamt, G. Sh¸¸rmann, J. Chem. Soc. Perkin Trans. 2 1993, 799.
[20] S. Sharma, T. P. Radhakrishnan, J. Phys. Chem. B 2000, 104, 10191.
[21] S. Sharma, T. P. Radhakrishnan, J. Phys. Chem. B 2003, 107, 147.
[22] S. Sharma, T. P. Radhakrishnan, ChemPhysChem 2003, 4, 67.
[23] a) W. Rettig, Angew. Chem. 1986, 98, 969; Angew. Chem. Int. Ed.
Engl. 1986, 25, 971; b) K. Bhattacharyya, M. Chowdhury, Chem.
Rev. 1993, 93, 507.
romax-3) by exciting approximately 10 nm below the absorption lmax
.
The solid and polymer film samples were mounted diagonally in the stan-
dard cuvette so that the signal is detected at 908 to the exciting beam in
the normal geometry used for fluorescence measurements in solution.
The spectra are normalized by dividing by the optical density at lmax of
the same sample, so that a fair comparison can be made between the in-
tensity of emission from solution, solid and polymer films.
Crystal structure determination
X-ray diffraction data were collected on an Enraf-Nonius MACH3 dif-
fractometer. MoKa radiation with a graphite crystal monochromator in
the incident beam was used. Data was reduced using Xtal3.4;[26] Lorentz
and polarization corrections were included. All non-hydrogen atoms
were found using the direct method analysis in SHELX-97[27] and refined
anisotropically. After several cycles of refinement the positions of the hy-
drogen atoms were calculated and added to the refinement process.
Graphics were handled using ORTEX6a.[28] In the case of 2, the data
were collected and structure solved under monoclinic P lattice; however
the b angle deviates from 908 by only 0.085(18). Therefore we have ex-
amined the lattice transformation to orthorhombic P; however, the set of
systematic absences found in the transformed reflection data is not com-
patible with any of the space groups in this Bravais lattice. Further, we
have checked the symmetry of the Friedel pairs of several reflections;
out of the four pairs in an (hkl) set, the intensity of reflection of two are
distinctly different from (ranging from 2 50% of) the other two, indicat-
ing the monoclinic symmetry of the lattice. Detailed crystallographic data
are submitted as Supporting Information.
[24] L. R. Hertler, H. D. Hartzler, D. S. Acker, R. E. Benson, J. Am.
Chem. Soc. 1962, 84, 3387.
[25] M. Ravi, S. Cohen, I. Agranat, T. P. Radhakrishnan, Struct. Chem.
1996, 7, 225.
[26] Xtal 3.4, (Eds.: S. R. Hall, G. S. D. King, J. M. Stewart), University
of Western Australia, Perth (Australia), 1995.
[27] G. M. Sheldrick, SHELX-97, University of Gˆttingen, Gˆttingen
(Germany), 1997.
Computations
Semiempirical quantum chemical computations were carried out using
the MOPAC93 program package[29] using the AM1 method.[18] The geom-
etry optimizations involved no constraints and the PRECISE option was
employed. The microenvironment of the molecule was modeled using the
COSMO subroutine[19] employing various dielectric constants in the
range 1 to 80. The parameter, NSPA (number of segments per atom) was
set equal to 60 in all calculations. Enthalpies of formation were computed
following full configuration interaction (CI) calculation within the 6 mo-
lecular orbitals bracketing the HOMO/LUMO, involving the 400 singlet
microstates. The computational results are presented in the Supporting
Information.
[28] P. McArdle, J. Appl. Crystallogr. 1995, 28, 65.
[29] MOPAC93 Fujitsu Inc., Japan.
Received: May 9, 2003
Revised: July 4, 2003 [F5123]
797
Chem. Eur. J. 2004, 10, 791 797
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim