and 3 in other studied solvents. Conversely, 1 in THF exhibited
Table 1 Photophysical properties of 1–3 in various solvents
exclusively a new emission band maximized at ~ 590 nm (a
2
1
PL lmax
/
Stokes shift of ~ 9000 cm ), while the normal emission
attributed to 2 or 3 individually was too weak to be resolved.
The excitation spectrum is effectively identical to the absorption
profile, excluding its origin from traces of impurity. Detailed
analyses revealed that the intensity of the 590-nm emission was
linearly proportional to the studied concentrations in the range
Solvent
l
max/nma nm
F
t/nsb
DGc
2
2
1
CYC
BEN
THF
DCM
ACN
THF
THF
382
384
382
383
378
382
368
500
557
588
609
642
392
373
4.49 3 10
3.57 3 10
1.08 3 10
18.25
42.09
37.77
20.63
20.68
20.94
20.96
21.03
22
22
5.20 3 1023 20.66
2
5
24
23
23
of 10 –10 mol dm . Thus, the possibility of the emission
resulting from the aggregation effect can also be eliminated.
Alternatively, the results can be more plausibly rationalized by
a mechanism incorporating a photoinduced electron transfer
1.08 3 10
6.45
1.05
0.91
2
3
a
0.58
~ 1.0
Absorption, data was taken from the first vibronic peak. b Samples for the
lifetime measurements were degassed via three freeze–pump–thaw cycles.
Free energies (DG) of the electron transfer for 1 in various solvents were
calculated according to eqn. (2). See text for the detailed description.
(ET) process. This viewpoint can first of all be supported by the
c
strong solvent-dependent spectral properties shown in Fig. 2.
The spectral shift of the fluorescence upon increasing solvent
polarity (Df) depends on the difference in permanent dipole
moments between ground and excited states, which can be
quantitatively expressed as:
fitted single exponential decay, the rise dynamics of the
emission for 1 was irresolvable (í15 ps), indicating the
occurrence of a fast ET process.
Attempts have also been made by synthesizing two analogues
a and 1b (Scheme 1, see ESI†), in which the molecular
symmetry is lowered by bearing only a pair of D(triarylamine)/
A(1,3,4-oxadizaole) chromophore rather than two D/A pairs in
. Our preliminary results revealed a similar highly efficient ET
process, resulting in an anomalous CT band (not shown here).
Thus, under a decent driving force (DG) the ET process can be
generalized among the spirobifluorene-conjugated D/A sys-
tems.
vac
2
3
(1)7
n˜
f
= n˜ f 2 (2¡ m˘
e
2 m˘
g
¡ /hca )Df
0
As shown in the inset of Fig. 2, the plot of n˜
f
versus Df is
1
2
1
sufficiently linear, in which the large slope of 14 200 cm is
obtained, supporting the proposed electron-transfer mechanism
in the excited state. As a result, a charge transfer (CT) emission
band was observed in 1. Secondly, the free-energy change (DG)
for photoinduced electron transfer between an excited donor
molecule (D*) and a ground-state acceptor (A) at a distance (d)
can be expressed as:
1
2
DG = Eox(D) 2 Ered(A) 2 E00(D) 2 (e /ed)
Assuming an ET efficiency of approximately unity, the
solvent-polarity dependent yields ranging from 10 to 0.05 for
2
+
2
(2)
23
2
(e /2)(1/r
D
+ 1/r
A
)(1/e
r
2 1/e)
where Eox(D) and Ered(A) are the oxidation and reduction
1 (see Table 1) are relatively smaller than those of parent
molecules 2 and 3 (Table 1). This can be rationalized by the
facts that back electron transfer is considered a forbidden
process and is normally dominated by the radiationless
deactivation. In addition, as the local excitation (LE)-ET zero-
order gap increases by increasing the solvent polarity, the
radiative decay rate of the ET band decreases due to the
reduction in LE/ET interaction. The combination of these two
factors rationalizes the decrease of the emission yield as the
solvent polarity increases.
Theoretically, it is feasible to adjust the D/A strength as well
as the number of spirobifluorene conjugation so that the flow of
electrons can be precisely regulated. Such a conceptual design
may be advantageously exploited in e.g. photovoltaic devices.
To achieve this goal, focus on poly-spirobifluorene conjugated
bipolar or triad systems is currently in progress.
potentials of triarylamine and 1,3,4-oxadiazole moieties, re-
+
spectively, E00(D) is the energy of zero–zero transition, r
r
D
and
are effective ionic radii, e is the dielectric constant of
solvent, and d is the center-to-center distance between D and A.
ox and Ered for 1, measured by cyclic voltammetry experiments
2
A
E
(
see ESI†), were measured to be 0.88 and 21.69 V, respectively
8
in THF. E00 of 3.25 eV ( ~ 382 nm) was obtained from the first
vibronic peak of the UV spectrum. According to the X-ray
structure, the center-to-center distance between D and A of 1
was estimated to be ~ 7.7 Å. A semi-empirical PM3 approxima-
+
–
tion was further made on r
D
A
~ r = 5.3 Å. With all values
substituted into eqn. (2), the free energies of ET (DG) were
calculated and shown in Table 1. As indicated by the DG values
of < 20.63 eV, the excited state ET reaction is apparently an
exergonic process in all solvents applied. This viewpoint can be
further supported by the corresponding ET dynamics for 1 in
various solvents. The fluorescence rise and decay dynamics
were measured by a time-correlated photon counting system,
rendering a temporal resolution of ~ 15 ps (see ESI†). The
results listed in Table 1 clearly showed that except for the well-
Notes and references
1
2
R. M. Metzger, J. Mater. Chem., 2000, 10, 55.
A. P. de Silva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley, C.
P. McCoy, J. T. Rademacher and T. E. Rice, Chem. Rev., 1997, 97,
1
515.
3
4
5
6
F. D. Lewis, T. Wu, Y. Zhang, R. L. Letsinger, S. R. Greenfield and M.
R. Wasielewski, Science, 1997, 277, 673.
D. Gust, T. A. Moore and A. L. Moore, Acc. Chem. Res., 1993, 26,
1
98.
(a) S. Y. Kim, M. Lee and B. H. Boo, J. Chem. Phys., 1998, 109, 2593;
b) P. Malask, Adv. Mater., 1994, 6, 405.
Crystal data for 1: C78.50 70Cl , M = 1306.66, monoclinic, space
group P2 /n, a = 13.7737(1), b = 27.0557(3), c = 20.0245(2) Å, b =
(
H
5 6 2
N O
1
3
23
1
0
05.0516(4)°, U = 7206.26(12) Å , Z = 4, D c = 1.204 Mg m , m =
21
.251 mm , F(000) = 2736, T = 150 K, wavelength 0.71073 Å, crystal
size 0.25 3 0.22 3 0.15 mm, qmax 25.00°, 41176 reflections measured,
suppdata/cc/b2/b208269a/ for electronic files in CIF or other electronic
format.
vac
f
7
n˜
f
and n˜
in eqn. (1) are the spectral position of the solvation
equilibrated fluorescence maxima and the value extrapolated to the
diluted gas-phase, respectively, a
0
denotes the radius of solute, m˘
g
and m˘
e
Fig. 2 The emission spectra of 1 in cyclohexane (5), benzene (-),
tetrahydrofuran (D), dichloromethane (8) and acetonitrile (1). The optical
density at the excitation wavelength (330 nm) was prepared to be 0.2 in all
studied solvents. Inset: the plot of emission peak frequencies as a function
of solvent polarities.
are the dipole moment vectors of the ground and excited states, and Df is
the Lippert solvent polarity parameter expressed as Df = (e 2 1)/(2e +
2
2
1) 2 (n 2 1)/(2n + 1).
8 T. J. Kang, M. A. Kahlow, M. D. Giser, S. Swallen, V. Nagarajan, W.
Jarzeba and P. F. Barbara, J. Phys. Chem., 1988, 92, 6800.
CHEM. COMMUN., 2002, 2874–2875
2875