Angewandte
Communications
Chemie
ingly ascribed to room-temperature phosphorescence (RTP).
The RTP spectra for all four molecules (Figure 1d) are almost
identical and show vibration progressions of the aromatic
ring, that is, the emission peaks at 541 nm, 588 nm, and
ꢀ1
6
42 nm are separated by approximately 1500 cm , which
indicates that the lowest excited triplet states are p-localized
3
excited states p-p* for all molecules. This is also consistent
with the observed sub-second lifetimes. All four molecules
share the same electron-deficient, p-conjugated NNI ring.
However, the N substituents exhibit increasing electron-
donating capability, and thus the DEST values of the ICT
states in the four compounds are expected to be in the order
DE (4) < DE (3) < DE (2) < DE (1). To confirm this
ST
ST
ST
ST
hypothesis, using the S1-optimized geometry computed with
[
21]
wB97XD/cc-pVDZ and the Tomasi model for the CH Cl2
2
[
22]
solvent,
GAMESS.
we conducted TDDFT calculations in
The lambda statistic (L) employed by Peach
which measures the locality extent excitations,
[
23,24]
[
25]
et al.,
enables the determination of their CT character. The
lambda value can range from 0 to 1, where the limits indicate
disjoint (L!0) and common (L!1) spaces occupied by the
origin and destination states of an excitation. The calculations
show that the fluorescence is an emission from an allowed
state at about 3.3 eV for each N-aryl species (see the
Supporting Information, Table S1). These are well-localized
excitations with L > 0.8 in each case. Orbital representations
confirm that these transitions are local to the NI frame
Figure 2. a) Steady-state emission spectra of 5a in PMMA under air
black) and in vacuo (red). b) Photographs of 5a in PMMA film under
(
air and in vacuo (l =365 nm). c) Steady-state emission spectra of 5b
ex
in a pH 6.8 buffer before (black) and after deoxygenation (red).
d) Photographs of 5b buffer solutions in air and after purging with
nitrogen gas (l =365 nm).
ex
potential of NNI-based phosphorescent molecules in biolog-
ical imaging, we also synthesized a series of NNI-based ROP
initiators I1–I6 (Scheme S1), from which the luminescent
PLAs 6–11 (Scheme 1) were made. NMR spectra and gel
permeation chromatography (GPC) data (see the Supporting
Information) indicate that polymers 6–11 are well-defined
(PDI < 1.15), and their molecular weights are given in
Table 2.
(
Figure 1e). The lowest singlet transitions from 3 and 4 are
unambiguously CT states (Figure 1 f, L = 0.099 and 0.083)
with a forbidden nature (dark state). This explains the lack of
fluorescence for 3 and 4 even in the solid-state films.
It is well-known that heavy atoms can help populate
triplet excitons by spin–orbit coupling. As shown in Figure 2a,
the steady-state emission spectra of compound 5a in PMMA
film (0.5% by weight) in air exhibit two major bands with
their maxima at l = 420 nm (4.2 ns) and l = 611 nm
F
RTP
Table 2: Molecular weight data of polymers 6–11.
(
(
5.1 ms). Compared to the ultralong “afterglow” of 3 and 4
tRTP ꢁ 0.23 s), the RTP of 5a has a much shorter lifetime and
6
7
8
9
10
11
[
a]
could not be entirely quenched even when exposed to air (see
Figure 2a,b). This is likely due to the enhanced triplet to
M (NMR)
5200
11000
1.09
4900
9300
1.06
7000
10400
1.04
6500
17400
1.14
10000
14500
1.11
11000
12800
1.10
n
[
b]
M
n
(GPC)
[
c]
PDI (GPC)
(
singlet) ground state decay rate from the heavy atom effect,
which outcompetes the relatively slow oxygen diffusion in
PMMA film (ca. 0.2 mm). When a hydrophilic group is used
to replace the butyl group (5b), RTP could also be detected in
deoxygenated buffer solution (Figure 2c,d) but with a shorter
lifetime (0.5 ms) and lower relative intensity, which is
presumably due to activated collisional quenching.
[a] Number-average molecular weight calculated from the integration
ratio of the CH (PLA) proton and residual initiator protons in NMR
spectra recorded in CDCl . [b] Number-average molecular weight
3
determined by GPC using THF as the eluent at 258C; the system was
calibrated with linear polystyrene standards. A correction factor of 0.58
[
26]
for PLA was applied to all data. [c] Polydispersity index (PDI=M /M ).
w
n
The experimental luminescence results for 5a and 5b are
[
20]
consistent with a recent report by Taddei and co-workers,
As shown in Figure 3a, films of 6 (PLA-Br), 7 (PLA-OPh-
Br), 9 (PLA-OPh-Cl), and 11 (1,2-OPh-PLA) exhibit barely
detectable steady-state emission under 365 nm excitation in
air, whereas emissions centered at 412 nm and 385 nm were
observed for 8 (PLA-Cl) and 10 (1,2-PLA), respectively.
After sealing the PLA films under vacuum, a dramatic
increase in the emission intensity of the lower-energy band
(500–700 nm) was observed for all polymers except for 10,
whose emission hardly changed. The luminescence lifetimes
of the newly added emissions for 6, 7, 8, 9, and 11 were
where RTP in both organic solution and the crystalline state
was observed for a Br-substituted NNI derivative. However,
we show in the following experiments that the RTP/fluores-
cence ratio can be further increased by the ICT enhancement
strategy (6–11). Introducing a heavy halogen atom (6–9) or
extending the p-conjugation (10 and 11) can affect the
miscibility between the NNI dye and the polymer matrix; we
thus resorted to the method of covalent dye conjugation with
a polymer by ring-opening polymerization (ROP), which was
[
26]
developed by Fraser and co-workers.
To extend the
measured to be t = 5.6 ms (lem = 611 nm), 3.6 ms (lem
=
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!