Intramolecular charge transfer photoemission of a silicon-based copolymer containing carbazole and divinylbenzene chromophores. Electron transfer across silicon bridges

Article abstract of DOI:10.1021/jp504649p

A new copolymer consisting of N-isopropylcarbazole/dimethylsilylene bridge/divinylbenzene units was synthesized and characterized. Dual fluorescence was observed in this copolymer in polar solvents. The absence of the second band at the lower transition energy of the two emission maxima in nonpolar solvents and the quantitative correlation of the lower-energy emission band maxima with solvent polarity indicate that the lower-energy emission band arises from an intramolecular charge transfer (ICT) state. A series of model compounds was synthesized to investigate the source of the charge transfer. It was found that the Si-bridged dyad with a single N-isopropylcarbazole and a single divinylbenzene was the minimum structure necessary to observe dual luminescence. The lack of dual luminescence in low-temperature glasses indicates that the ICT requires a conformation change in the copolymer. Analogous behavior in the Si-bridged dyad suggests that the ICT in the copolymer is across the silicon bridge. Results from time-resolved luminescence measurements with picosecond and subnanosecond excitation were used to support the thesis that twisted charge-transfer states are the likely source of the observed dual luminescence.

Full text of DOI:10.1021/jp504649p

Article
pubs.acs.org/JPCA
Intramolecular Charge Transfer Photoemission of a Silicon-Based
Copolymer Containing Carbazole and Divinylbenzene
Chromophores. Electron Transfer Across Silicon Bridges
Malgorzata Bayda,*,†,‡ Monika Ludwiczak,‡ Gordon L. Hug,‡,§ Mariusz Majchrzak,‡ Bogdan Marciniec,†,‡
and Bronislaw Marciniak‡
†Center of Advanced Technologies, Adam Mickiewicz University, Grunwaldzka 6, 60-780 Poznan, Poland
‡Faculty of Chemistry, Adam Mickiewicz University, Umultowska 89 b, 61-614 Poznan, Poland
§Radiation Laboratory, University of Notre Dame, Notre Dame, Indiana 46556, United States
ABSTRACT: A new copolymer consisting of N-isopropylcar-
bazole/dimethylsilylene bridge/divinylbenzene units was
synthesized and characterized. Dual fluorescence was observed
in this copolymer in polar solvents. The absence of the second
band at the lower transition energy of the two emission
maxima in nonpolar solvents and the quantitative correlation
of the lower-energy emission band maxima with solvent
polarity indicate that the lower-energy emission band arises
from an intramolecular charge transfer (ICT) state. A series of
model compounds was synthesized to investigate the source of
the charge transfer. It was found that the Si-bridged dyad with
a single N-isopropylcarbazole and a single divinylbenzene was
the minimum structure necessary to observe dual lumines-
cence. The lack of dual luminescence in low-temperature glasses indicates that the ICT requires a conformation change in the
copolymer. Analogous behavior in the Si-bridged dyad suggests that the ICT in the copolymer is across the silicon bridge. Results
from time-resolved luminescence measurements with picosecond and subnanosecond excitation were used to support the thesis
that twisted charge-transfer states are the likely source of the observed dual luminescence.
of these fluorescence studies, charge-transfer character has been
attributed to the excited states responsible for the red-shifted
spectra in some of the donor−acceptor copolymers.11 These
spectral assignments are consistent with earlier work on silicon-
bridged donor−acceptor dyads.13 In addition, there has been
some indication that dual luminescence was involved with these
broad emission spectra (i.e., in copolymers and dyads).11,13 So
far there has been no discussion of whether either these charge-
transfer excited states or their dual luminescences were due to
twisted intramolecular charge transfer (TICT) states.14
However, systematic studies have not been done that are
aimed at the time-dependent excited-state processes involved
with this class of copolymers. Understanding these processes
has the potential for gaining insight into two physical features
that could effect charge transport in such polymers and could
also have wider theoretical implications. One issue is whether
the charge can be transported through or around the silicon
bridges (through space versus through bridge). Another
question is whether significant conformational changes are
necessary to populate charge-transfer excited states (implica-
tions for the TICT literature).14 In order to address both of
INTRODUCTION
■
Copolymers with donor and acceptor groups separated by
silicon atom spacers display a wide array of photophysical and
photochemical phenomena.1,2 Synthesis of these polymers has
been motivated by the realization that electroluminescence
from them is of interest in producing polymer-based, light-
emitting diodes (LED).3 The silicon spacers were inserted to
break the electronic conjugation1 in the archetypical conducting
polymer, polyphenylene-vinylene. If the conjugation in this
polymer were not broken, polyphenylene-vinylene would emit
at longer wavelengths only. However, even with the silicon
spacers breaking the electronic conjugation,4 electrolumines-
cence can still be seen in a variety of homopolymers and
copolymers.5 This indicates that the transport of holes and
electrons is still possible even though electronic conjugation has
been broken.
Silicon-bridged donor/acceptor copolymers with carbazoles
as the donors show both large fluorescence yields6 and large
hole mobilities.7 This makes it possible to use photophysical
techniques to understand the nature of the excited states8 that
could be precursors of electroluminescence and charge
transport. To date, most of the photoluminescence work on
these polymers has been steady-state luminescence spectra that
are routinely taken along with new syntheses.1,9−12 On the basis
Received: May 12, 2014
Published: June 5, 2014
© 2014 American Chemical Society
4750
dx.doi.org/10.1021/jp504649p | J. Phys. Chem. A 2014, 118, 4750−4758

The Journal of Physical Chemistry A
Article
these two general topics via photophysical techniques, a
copolymer was synthesized containing an electron donor (N-
isopropylcarbazole) connected to an electron acceptor
(divinylbenzene) by a silicon bridge. In addition, supplemen-
tary syntheses of small model compounds were used to address
these same two issues by providing contrasting structural
constraints relative to the copolymer.
The spectroscopic characteristics of synthesized compound
1a are described as follows:
EXPERIMENTAL METHODS
■
General Considerations. 1H NMR (400 MHz), 13C NMR
(100 MHz), 29Si NMR (80 MHz), and DEPT spectra were
recorded on a Bruker Avance II 400 MHz spectrometer in
CDCl3 solution. Chemical shifts are reported below in δ (ppm)
with reference to the residue solvent (CDCl3) peak for 1H and
13C and to TMS for 29Si. Mass spectra of the organic and
organometallic molecular products were obtained by GCMS
analysis (Varian Saturn 2100T, equipped with a BD-5 capillary
column (30 m), and an ion trap detector). Gel permeation
chromatography (GPC) analyses were performed using an
Agilent HPLC system equipped with a UV absorbance detector
and a RI detector (analysis conditions: mobile phase, THF;
flow rate 0.80 mL/min; temperature 20 °C; injection volume,
17 μL). The number-average molecular weight (Mn), weight-
average molecular weight (Mw), and polydispersity index (PDI)
were determined by a polystyrene standard calibration curve.
The chemical reagents were obtained from the following
sources: hexane, toluene, diethyl ether, tetrahydrofuran (THF),
dimethylformamide (DMF), and acetone were purchased from
Avantor, carbazole from Alfa Aesar, styrene, 1,2-dibromo-
ethane, ethanol, methanol, isopropyl bromide, tetrabutylammo-
nium bromide, N-bromosuccinimide (NBS), magnesium
sulfate, calcium hydride, sodium hydride, potassium hydroxide,
and CDCl3 from Aldrich, and chlorodimethylvinylsilane from
Gelest. The solvents used for photophysical measurements
(chloroform, cyclohexane, dichloromethane, diethyl ether, and
dimethyl sulfoxide) were of spectroscopic grade from Merck
and were used as received. Acetonitrile (gradient grade for
liquid chromatography, Merck) and sulfuric acid (POCH) were
used as received. THF (inhibitor free, for HPLC, Sigma-
Aldrich) was distilled over sodium, under argon. 1,4-
Divinylbenzene15 as well as chlorodimethylvinylsilane were
purified by “bulb-to-bulb” distillation and stored under argon.
All the syntheses of monomers, macromolecular compounds,
and catalytic tests were carried out under an inert argon
atmosphere. The ruthenium complex [RuHCl(CO)(PCy3)2]
was prepared according to a literature procedure.16
1H NMR (400 MHz, CDCl3): δ 0.46 (s, 6H, Si−CH3), 1.73
(d, JHH = 7.0 Hz, 6H, H2′), 5.01 (septet, 1H, H1′), 5.83 (dd, JHH
= 3.9, 20.4 Hz, 1H, H2″), 6.11 (dd, JHH = 3.9, 14.6 Hz, 1H, H2″),
6.41 (dd, JHH = 14.6, 20.4 Hz, 1H, H1″), 7.24 (t, 1H, H7), 7.45
(t, 1H, H6), 7.54 (m, 1H, H8 and 1H, H2), 7.60 (d, JHH = 8.2
Hz, 1H, H1), 8.15 (m, 1H, H5), 8.30 (s, 1H, H4). 13C NMR
(100 MHz, CDCl3): δ −2.40 (Si−CH3), 20.81 (C2′), 46.65
(C1′), 109.70 (C1), 109.94 (C5), 118.5 (C8), 120.31 (C6),
123.28 (Ci−Si), 125.33 (C7), 126.16 (C4), 126.35 (Ci from N−
CC−), 130.67 (C2), 132.48 (C2″), 138.86 (C1″), 139.41 (N−
Ci), 140.13 (N−Ci). 29Si NMR (80 MHz, CDCl3): δ −10.60.
MS (EI): m/z (relative intensity %): 294•+ (9), 293 (36), 278
(100), 252 (16), 85 (23), 59 (17).
A General Procedure for Silylative Coupling (SC).
Compounds 1b, 2, 3, and 4 (Chart 1) were synthesized
Chart 1. Structures of Compounds 1−4
Syntheses of Organosilicon Compounds. 3-(Dimethyl-
vinylsilyl)-N-isopropylcarbazole (1a). Compound 1a was
obtained by Grignard’s reaction of chlorodimethylvinylsilane
and 3-bromo-N-isopropylcarbazole. A solution of 3-bromo-N-
isopropylcarbazole (1.58 g, 5.48 mmol) in 5 mL of THF was
added slowly dropwise to a suspension of Mg (0.2 g, 8.23
mmol; surface activated by 1,2-dibromoethane, 0.1 mL) and
chlorodimethylvinylsilane (0.9 mL, 6.57 mmol) in 5 mL of
THF. After the addition was completed, the reaction mixture
was refluxed, and the process of the reaction was controlled by
GC analysis. The mixture was cooled to room temperature, the
solvent was evaporated, and the product was extracted with
hexane/water solvents. The organic layer was dried under
magnesium sulfate and filtered. The solvent was evaporated,
and the residue was separated with a silica gel column (hexane/
CH2Cl2 = 1:1, Rf = 0.5). 3-(Dimethylvinylsilyl)-N-isopropyl-
carbazole was obtained as a colorless oil (yield 81%).
following a general procedure for silylative coupling reactions.17
The silylative coupling syntheses were carried out in 5 mL glass
reactors equipped with a magnetic stirring bar under an argon
atmosphere. The reaction mixtures contained toluene (0.5 M),
the vinylsilane derivative, olefin (DVB or styrene), and a
ruthenium−hydride complex [RuHCl(CO)(PCy3)2] (1−2 mol
%). The molar ratios of the reagents were stoichiometric. The
4751
dx.doi.org/10.1021/jp504649p | J. Phys. Chem. A 2014, 118, 4750−4758

The Journal of Physical Chemistry A
Article
system was heated in an oil bath at 100 °C for 24 h (molecular
compounds) or 72 h (copolycondensation). After the reactions
were complete, the solvents were evaporated under vacuum.
The molecular products were separated with a silica gel column
(hexane/CH2Cl2 = 1:1). The macromolecular product was
dissolved in THF, purified by repeated precipitation from
methanol, filtered, and dried under vacuum.
H1′), 5.25 (d, JHH = 10.8 Hz, 1H, −CHCH2), 5.76 (d, JHH
=17.6 Hz, 1H, −CHCH2), 6.69 (d, JHH = 11.0 Hz, 1H,
−CHCH2), 6.71 (d, JHH = 19.1 Hz, 1H, H2″), 6.98 (d, JHH
19.1 Hz, 1H, H1″), 7.20−7.25 (t, 1H, H7), 7.37−7.40 (d, JHH
=
=
8.3 Hz, 2H, H5″), 7.44 (d, JHH = 8.0 Hz, 2H, H4″), 7.45 (t, 1H,
H6), 7.53 (d, JHH = 8.3 Hz, 1H, H8), 7.55 (d, JHH = 8.1 Hz, 1H,
H2), 7.64 (d, JHH = 8.4 Hz, 1H, H1), 8.15 (d, 7.8, 1H, H5), 8.32
(s, 1H, H4). 13C NMR (100 MHz, CDCl3): δ −2.27 (Si−CH3),
20.82 (C2′), 46.68 (C1′), 109.75 (C1), 113.79 (−CHCH2),
118.5 (C8), 123.28 (Ci−Si), 125.36 (C7), 126.24 (C4), 126.38
(C5″), 126.71 (C4″), 126.45 (Ci from NCC−), 130.80
(C2), 136.45 (−CHCH2), 137.29 (>CiCHCH2), 137.88
(C3″), 139.42 (C2″), 140.17 (N−Ci), 144.50 (C1″). 29Si NMR
(80 MHz, CDCl3): δ −10.02. MS (EI): m/z (relative intensity
%): 395•+ (100), 380 (40), 338 (6), 276 (8), 171 (8), 59 (5).
The compound was isolated as a white powder, yield 60%.
Poly[dimethylsilylene-(3,6-N-isopropylcarbazolylene)-di-
methylsilylene-(E)-vinylene-(1,4-phenylene)-(E)-vinylene] (4).
The spectroscopic characteristics of the compounds
synthesized by SC are described as follows:
[(E,E)-1,4-Bis(trimethylsilyl)ethenyl]benzene (1b). H NMR
1
(400 MHz, CDCl3): δ 0.09 (s, 18H, Si−CH3), 6.41 (d, JHH
=
19.1 Hz, 2H, H2″), 6.79 (d, JHH = 19.1 Hz, 2H, H1″), 7.33 (s,
4H, H4′). 13C NMR (100 MHz, CDCl3): δ −1.26 (Si−CH3),
126.53 (C4″), 129.52 (C2″), 137.94 (Ci), 143.16 (C1″). 29Si
NMR (80 MHz, CDCl3): δ −6.23. MS (EI): m/z (relative
intensity %): 274•+ (60), 259 (80), 201 (37), 184 (25), 171
(37), 144 (15), 73 (100), 59 (30), 45 (20). The compound was
isolated as a yellow powder, yield 88%.
{(E)-[(3-N-Isopropylcarbazolyl)dimethylsilyl]ethenyl}-
benzene (2). 1H NMR (400 MHz, CDCl3): δ 0.54 (s, 6H, Si−
1H NMR (400 MHz, CDCl3): δ 0.52 (s, 12H, −CH3), 1.67 (d,
6H, H2′), 4.99 (m, 1H, H1′), 5.22 (trace H2CCH−), 5.73
(trace H2CCH−), 6.42 (trace H2CCH−), 6.67 (d, 1H,
H2′), 6.95 (d, 1H, H1′), 7.41 (s, 4H, H4″, and H5″), 7.52 (d, 2H,
H2), 7.61 (d, 2H, H1), 8.14 (H5), 8.27 (HIV), 8.33 (d, 2H, H4).
13C NMR (100 MHz, CDCl3): δ −2.0 (Si−CH3), 20.8 (C2′),
46.7 (C1′), 109.7 (C1), 110.0 (Ci-Si), 118.7 (CVIII), 120.3 (CVI),
123.1 (CV), 125.3 (CVII), 126.1 (C4), 126.5 (Ci from NC
C−), 126.7 (C4″ and C5″), 130.9 (C2), 138.0 (C2″), 140.15 (N−
Ci), 144.5 (C1″). 29Si NMR (80 MHz, CDCl3): δ −10.02,
−9.96. GPC analysis: Mw = 4663 [g·mol−1], Mn = 3768 [g·
mol−1], PDI (Mw/Mn) = 1.24, n = 9. The compound was
isolated as a white powder, yield 58%.
Photophysical Measurements. UV−vis spectra were
recorded at room temperature using a Cary 5000 UV−vis−
NIR. Fluorescence spectra, at room temperature and at low
temperature (77 K), were measured on a PerkinElmer LS 50B
and were corrected for instrumental response.18 Both
absorption and emission spectra at RT were recorded using 1
cm × 1 cm rectangular cells.
CH3), 1.72 (d, JHH = 7.0 Hz, 6H, H2′), 5.01 (septet, 1H, H1′),
6.70 (d, JHH = 18.9 Hz, 1H, H2″), 7.01 (d, JHH = 18.9 Hz, 1H,
H1″), 7.20−7.28 (m, 2H, H7 and H6″), 7.34 (t, 2H, H5″), 7.44 (t,
1H, H6), 7.48 (d, JHH = 7.3 Hz, 2H, H4″), 7.53 (d, JHH = 8.2 Hz,
1H, H8), 7.56 (d, JHH = 8.2 Hz, 1H, H2), 7.64 (d, JHH = 8.2 Hz,
1H, H1), 8.15 (d, JHH = 7.6, 1H, H5), 8.33 (s, 1H, H4). 13C
NMR (100 MHz, CDCl3): δ −1.98 (Si−CH3), 20.82 (C2′),
46.67 (C1′), 109.74 (C1), 109.93 (C5), 118.71 (C8), 120.35
(C6), 123.22 (Ci−Si), 125.35 (C7), 126.24 (C4), 126.52 (Ci
from NCC−), 128.04 (C6″), 128.10 (C5″), 125.51 (C4″),
130.81 (C2), 138.33 (C2″), 139.42 (N−Ci), 140.17 (N−Ci),
144.98 (C1″). 29Si NMR (80 MHz, CDCl3): δ −10.01. MS
(EI): m/z (relative intensity %): 369•+ (40), 354 (100), 312
(10), 252 (20), 194 (10), 145 (28). The compound was
isolated as a white powder, yield 87%.
{(E)-4-[(3-N-Isopropylcarbazolyl)dimethylsilyl]ethenyl}-
1
styrene (3). H NMR (400 MHz, CDCl3): δ 0.53 (s, 6H, Si−
Low-temperature measurements were carried out in NMR
tubes (5 mm o.d.) immersed in liquid nitrogen in a quartz
CH3), 1.72 (d, JHH = 7.09 Hz, 6H, H2′), 4.95−5.06 (septet, 1H,
4752
dx.doi.org/10.1021/jp504649p | J. Phys. Chem. A 2014, 118, 4750−4758

The Journal of Physical Chemistry A
Article
Dewar. A neutral-density filter was employed to decrease the
very high fluorescence intensity of compounds 3 and 4 at 77 K.
Fluorescence lifetime measurements were carried out using a
time-correlated single-photon-counting technique (Fluores-
cence Lifetime Spectrofluorimeter FluoTime 300 from
PicoQuant, equipped with a 300 5 nm diode as its excitation
source whose time profile, measured by the detection system
following scattering through Ludox, was approximately 750 ps
full width at half-height). Time-resolved fluorescence experi-
ments were also performed in the Centre of Ultrafast Laser
Spectroscopy, Adam Mickiewicz University. The Centre’s
Ti:sapphire Tsunami laser, tunable in the 720−1000 nm
range, was pumped with a 10 W (532 nm) Millennia X Pro
laser. The output of the Tsunami laser consisted of 2 ps pulses
at a repetition rate of about 82 MHz and a mean power of over
1 W. A pulse selector reduced the repetition rate from 82 to 4
MHz. A harmonic generator was used to generate 285 nm as
the excitation wavelength by tripling the fundamental beam
frequency (855 nm).19
The additivity of the absorption spectra of the two
chromophores to give the copolymer’s spectrum is taken to
show that there is little communication between the donor
(carbazole, modeled by compound 1a) and acceptor
(divinylbenzene, modeled by 1b) moieties across the silicon
atom bridge in the ground state of the copolymer 4. Figure 1
shows much less interaction between the monomers 3 and the
copolymer 4 with the small shift and broadening probably due
to exciton interactions.8 Furthermore, the absorption spectra of
the copolymer do not change significantly with solvent, i.e.,
acetonitrile (ACN), butyronitrile (BUN), chloroform (CHL),
cyclohexane (CY), dichloromethane (DCM), diethyl ether
(DEE), dimethyl sulfoxide (DMSO), or tetrahydrofuran
(THF).
In contrast, the emission spectra of the copolymer 4 were
significantly dependent on the solvent. In cyclohexane, the
emission spectrum (Figure 2) of the copolymer 4 was
Electrochemical Measurements. The cyclic voltammetry
experiments were carried out with a potentiostat VersaSTAT3-
400 from Princeton Applied Research. The measurements were
made with a standard three-component cell that contained a
PTE platinum working electrode (6 mm o.d. × 1.6 mm i.d.), a
platinum-wire counter electrode and an SCE (saturated calomel
electrode) as the reference electrode. To determine the
oxidation potential, Eox, of compound 1a (0.01 M) and the
reduction potential, Ered, of compound 1b (0.01 M),
voltametric measurements were performed on deoxygenated
ACN solutions (purged with argon) with the addition of 0.1 M
tert-butylammonium perchlorate as the supporting electrolyte.
Figure 2. Fluorescence spectra of compound 4 in various solvents, λexc
= 310 nm.
RESULTS AND DISCUSSION
■
From Chart 1, it can be seen that copolymer 4 is made up of
units of model compound 3, but alternately copolymer 4 can be
seen as being made of overlapping units of model compounds
1a and 1b. This latter parsing of the copolymer into units is
convenient for analyzing the chromophoric content of the
copolymer because the silicon atoms can shift the spectra of the
aromatic moieties. With this in mind, the absorption spectra of
these two chromophores 1a and 1b are shown along with the
absorption spectrum of the copolymer 4 in Figure 1.
reminiscent of the fluorescence spectrum of carbazole20 and,
in particular, of compound 1a (Figure 3). However, the
Figure 3. Normalized fluorescence spectra of compounds 1a, 1b, 3,
and 4 in ACN.
copolymer 4 emission spectrum in acetonitrile (Figures 2 and
3) showed that a second band appeared at long-wavelengths
(low energies) in addition to the 370 nm band system in its
emission spectrum in cyclohexane (Figure 2). The emission
spectrum of 1a in acetonitrile (Figure 3) showed no such band
in the green part of its emission spectrum. The emission
spectrum of 4 in cyclohexane was similar to the emission
Figure 1. Absorption spectrum of 4 is displayed as a least-squares fit to
the sum of spectra 1a and 1b, and the absorption spectrum of 3 is
displayed as a least-squares fit to the absorption spectrum of 4. The
solvent was ACN in all spectra.
4753
dx.doi.org/10.1021/jp504649p | J. Phys. Chem. A 2014, 118, 4750−4758

The Journal of Physical Chemistry A
Article
Table 1. Absorption and Emission Properties of Compounds 1a, 1b, 3, and 4 in ACN
λf,max (nm)
a
b
0−0
compd
λabs,max (nm)
LE
ICT
λS (nm)
E(S1)(kJ mol−1
)
Φf
τf (ns)
15.3
1a
1b
3
346
297
346
347
353, 369
344
−
−
488
349
322
349
350
343
372
343
342
0.37
b
c
0.003
<0.5
d
e
353, 370
355, 370
0. 17
20.4
d
e
f
4
494
0.17
22.0 (15)
a
b
Calculated as total fluorescence quantum yields. Carbazole in cyclohexane used as a reference compound to determine fluorescence quantum
yields (0.38).20 For a similar system (E,E)-{1,4-bis(2-dimethylphenylsilyl)ethenyl}benzene, its fluorescence lifetime was determined to be 13.2
c
2.0 ps in acetonitrile.21 Quinine sulfate in 1 N H2SO4 used as a reference compound to determine fluorescence quantum yields (0.54).22
d
e
f
Fluorescence lifetimes determined for ICT band as single-exponential decays. Longest fluorescence lifetime determined for ICT band in THF.
spectra of 1a in both cyclohexane and acetonitrile (see Figures
2 and 3).
In Figure 4 the wavenumbers of the maxima of the low-
energy transitions are plotted against the solvent parameter Δf.
In contrast to the absorption spectrum (Figure 1) of
compound 1b, which appeared prominently as a component in
the absorption spectrum of copolymer 4, there was no evidence
that the fluorescence (Figure 3) of 1b makes any appearance in
the fluorescence spectra of copolymer 4. This could be due to
the very low fluorescence quantum yield of 1b relative to 1a
(see Table 1), which could be masking the fluorescence of the
fragment of 1b in compound 4’s luminescence.
However, it can also be seen from Figure 3 (and also Table
1) that the first excited singlet state of 1b is higher in energy
than the first excited singlet state of 1a. This could mean that
radiationless transitions23 or Forster singlet energy transfer24 is
̈
rapid from the divinylbenzene component 1b to the carbazole
fragments 1a in the copolymer. For example, Forster singlet
̈
energy transfer goes as the reciprocal of the sixth power of the
distance between energy donor−acceptor centers and could
account for the lack of a contribution to the luminescence of 4
from its 1b components. The silicon spacers disrupt the
conjugation, but they do not disrupt the communication
between energy (or electron) donors and acceptors, see also
below.
The dual luminescence of copolymer 4 in acetonitrile was
seen in other solvents, particularly in polar solvents. Figure 2
shows the emission spectra of 4 in the different solvents. From
this figure it can be seen that the carbazole-like emission bands
of copolymer 4 did not shift significantly but that the low-
energy transition shifted significantly with solvent polarity.
These shifts to lower energy with solvent polarity are
reminiscent of the Lippert−Mataga theory describing the shift
in luminescence with solvent polarity as arising from larger
dipole moments in the excited states of the molecules
compared to the dipole moments in their ground states.24−26
The Lippert−Mataga equation is given by
Figure 4. Fluorescence solvatochromism of compound 4.
Figure 4 shows a roughly linear plot as predicted by the
Lippert−Mataga equation for emission from an excited state
with a large dipole moment. In donor−acceptor compounds
with the moieties separated by silicon bridges, van Walee et
al.13 saw similar linear Lippert−Mataga plots with slopes of
−3.651 × 104 cm−1 for NH2−C6H4−Si(CH3)2−C6H4CN
compared to −3.2 × 104 cm−1 for the copolymer 4 in Figure
4. They interpreted the behavior of their compounds as arising
from intramolecular charge transfer from the donor part of their
molecules to the acceptor part of their molecules. Given the
similarity of the slopes of their Lippert−Mataga plots and that
in Figure 4, this is suggestive that the solvent shifts in the low-
energy transition in copolymer 4 are also due to emission from
a charge-transfer excited state.
The most likely charge transfer would be the full transfer of a
charge from the carbazole moiety 1a to the divinylbenzene
moiety 1b. The energy of this excited state can be estimated
from the Rehm−Weller equation:27
2μ2
e
νC = νC (0) −
Δf
EICT = Eox(carbazole) − Ered(divinylbenzene) + C
̅
T
̅
T
(3)
hcρ3
(1)
The Eox(1a) = 1.18 V vs SCE in acetonitrile and the Ered(1b) =
−2.32 V vs SCE in acetonitrile. Taking C, the Coulombic
energy for an oppositely charged ion pair in a polar solvent as
−0.056 eV as is commonly used for exciplex calculations in
acetonitrile,27 the energy of the ICT state relative to the
equilibrium ground state is 27 780 cm−1. This is below the local
excited state (LE) of the carbazole moiety 1a, 28 650 cm−1 (λ =
349 nm, Table 1). The energy of the transition from this ICT
state would be expected to shift with solvent because of the
large dipole moment of this state. The Coulombic energy will
also change with dielectric constant of the solvent. This
calculated energy is consistent with the scenario of the LE state
where
Δf =
n2 − 1
εS − 1
2εS + 1
−
4n2 + 2
(2)
νCT is the maximum of the emission band in wavenumbers,
̅
̅
νCT(0) is the hypothetical maximum of emission in vacuum, μe
is the dipole moment of the excited state, h is Planck’s constant,
c is the speed of light, ρ is Onsager’s solute cavity radius within
the solvent, εS is the dielectric constant of the solvent, and n is
the solvent’s refractive index.
4754
dx.doi.org/10.1021/jp504649p | J. Phys. Chem. A 2014, 118, 4750−4758

The Journal of Physical Chemistry A
Article
being able to populate the closely lying ICT state as the solvent
becomes more polar.
There are several possible means by which the ICT could
occur. Intermolecular transfer seems unlikely because of the
low concentration of the copolymer and lack of evidence for
aggregation in the absorption spectrum in Figure 1. However,
being a copolymer, the formation of the ICT excited state could
be formed by an intramolecular electron transfer from the
carbazole moiety to the divinylbenzene in two distinct ways.
First, there could be transfer in the excited state from the
carbazole to the first adjacent divinylbenzene moieties on either
side of the carbazole, or, second, this excited-state electron
transfer could happen between nonadjacent carbazole and
divinylbenzene units. Electron transfer would occur through the
overlap between occupied orbitals of the donor and unoccupied
orbitals of the acceptor. Since this overlap decreases
exponentially with distance,28 at first glance, one might expect
that electron transfer between adjacent donor and acceptor
units would not be particularly favorable. It might seem that
nonadjacent transfer of an electron between two intrachain
moieties that have diffused together might afford a better
overlap of donor and acceptor orbitals than the overlap
between adjacent donor and acceptor orbital across a silicon
bridge. (For details see eqs 4 and 5 below.)
A synthetic strategy was initiated to put some limits on the
answer to these questions. First of all, the absence of dual
luminescence in compound 1a indicates that just having a
dimethylsilylvinylene group as the acceptor in the ICT is not
sufficient to form an excited intramolecular CT state. Kim et al.
interpreted their data as being due to ICT to a negative ionic
site on the silicon atom that, in their case, was stabilized by
phenyl groups in their diphenylsilylene bridges.9 In the current
work, the bridges are dimethylsilylene groups, and these bridges
do not apparently provide adequate stabilization to form a
stable negative ionic site for an ICT state. Even with the
terminal vinyl group for stabilization, it was not enough. The
next degree of complexity was to synthesize 2, i.e., a carbazole
linked to a styrene through the dimethylsilylene bridge. This
compound also showed only carbazole-like fluorescence.
However, when divinylbenzene was linked to the carbazole
moiety through the dimethylsilylene bridge in compound 3,
dual luminescence was observed; see Figure 3.
This is consistent with the van Walree et al. observations
where they saw intramolecular CT states forming across
dimethylsilylene bridges.13 Through their electrochemical
measurements they got redox potentials for the substituted
aromatic donors (typically p-aminophenyl-) and acceptors
(typically p-cyanophenyl-) that showed that the silicon atom
was involved in more favorable redox potentials for the electron
transfer. We saw no difference between the Eox of N-
isopropylcarbazole and 1a. Furthermore, in the current work,
the overlap between the donor and acceptor orbitals is not as
favorable due to the more extended nature of the donor and
acceptor moieties. So it is likely that the silicon atom plays a
kinetic role in the electron-transfer process such as through a
superexchange mechanism.28,29
Thus, the ICT state can be formed on adjacent moieties in
the copolymer 4. Whether or not there are more complicated
kinetics was investigated by performing time-resolved single-
photon-counting experiments. The luminescent decay traces
are shown in Figure 5 for copolymer 4 in THF.
The 370 nm decay gave a very poor quality fit unless three
exponentials were used. In contrast, the ICT decay trace at 500
Figure 5. Fluorescence decays of compound 4 in THF, λmonit = 370
and 500 nm (λexc = 300 nm).
nm could be fit very well with a double exponential. The 370
nm fluorescence trace is characterized initially, at least, by very
short fluorescence lifetime, <500 ps, followed by longer
lifetimes (3.8 and 14 ns). The ICT emission shows an apparent
very rapid growth following closely the pulse shape. This rapid
growth is then followed by slow decays of 4.4 and 15 ns.
The source of the multiple decays of the 370 and 500 nm
traces is most likely related to multiple conformations of the
polymer, but there are several possibilities for the nature of this
decay. If the small perturbations on the absorption spectrum of
3 compared to 4 are due to excitons, then exciton dynamics
could be responsible for the complex decay. Such variations in
the exciton dynamics would then also be related to the differing
polymer conformations. These conformations would have
differing resonance interactions between excited carbazole
sites because of varying distance between these sites in the
various polymer strands. If this is the source of the multiple
lifetimes, the divinylbenzene units would have to be involved
since only a single time constant was seen in the homopolymers
involving silicon-bridged N-isopropyl carbazoles.12
Intramolecular electron transfer is often associated with
conformal changes. For instance, there is a large body of
literature on twisted intramolecular charge transfer (TICT)
states.14,23 The issues of whether or how conformational
changes affect the ICT state and its formation were pursued by
performing experiments in low-temperature glasses. Both BUN
and DCM:MeOH (1:1) were used to form 77 K glasses
containing 4 in solution. Fluorescence spectra were measured
in these systems. The fluorescence spectra of 4 in BUN and
DCM:MeOH showed similar behavior. See Figure 6 for
DCM:MeOH. Both of these solvents are polar enough such
that red-shifted fluorescences (500 nm) were seen at RT along
with the carbazole-like fluorescence (370 nm). However, the
significant feature of Figure 6 is that there was no dual
luminescence of 4 in DCM:MeOH at 77 K. The same was seen
for 4 in BUN at 77 K.
The contrasting fluorescence behavior of 4 in DCM:MeOH
and BUN at the two temperatures is suggestive that there are
critical conformational changes that are responsible for the low-
energy fluorescence band at RT. There are three likely
candidates for this behavior. The first is that the ICT is due
to a twisted intramolecular charge transfer (TICT) state. In the
TICT literature14 there are many examples of TICT states
resulting from conformational changes. If such a rotation is
necessary for the ICT and if the low-temperature glass hinders
the rotation, this would account for the difference in the
4755
dx.doi.org/10.1021/jp504649p | J. Phys. Chem. A 2014, 118, 4750−4758

The Journal of Physical Chemistry A
Article
4 is due to electron transfer across the silicon bridge between
adjacent carbazole and divinylbenzene moieties.
The third possibility for dual luminescence at RT and none
at LT is that separate conformations are present at RT in which
ICT is possible in one such conformation, but that this
conformation is not present at LT. The absence of the ICT
enabled conformation at LT could be that the conformation is
inhibited by the structure of the LT glasses. But the more
plausible scenario is that the ICT-enabled conformation is
somewhat more unstable than 3 or 4 in their LE-enabled
conformation. The low temperatures would then favor the
more stable conformation leading to the lack of ICT
fluorescence. Such ground-state conformers of varying stability
have proven to be critical in understanding photophysical and
photochemical phenomena, for instance, s-cis and s-trans in
polyenes. Work on vitamin D precursors is a recent example.30
The difference between these two competing hypotheses can
be summarized by the TICT phenomena being a conforma-
tional change in the excited-state manifold whereas the two-
conformer model relates to ground-state conformations. There
is an experimental way to distinguish between these possible
scenarios by looking at the formation kinetics of the
fluorescence following a short excitation pulse. In the two-
ground-state-conformer scenario, one would expect a minimum
of difference in the rise times of the two fluorescence traces.
Any shifts in the rising part of the trace would be related to
normal shifts, dependent on the decay time. In any event, there
would be little correlation between the two traces. In the case
of the TICT scenario, it would be expected that the LE state
would show a rapid decay that would match a rise time of the
ICT state’s fluorescence trace. Then the two traces might decay
with the same lifetime if they are in thermal equilibrium.
The luminescence traces in Figure 5 do show the hint of a
fast decay of the 370 nm LE fluorescence trace, but the growth
of the two traces (370 and 500 nm) appeared to convolute
adequately with the immediate response function (IRF) of the
detection system. However, the full width at half-height of this
detection system and exciting diode is about 750 ps. In order to
put more stringent limits on the rise times, solutions of 4 were
excited with a picosecond excitation pulse.
Figure 6. Fluorescence spectra of compound 4 in DCM:MeOH (1:1),
λexc = 310 nm.
fluorescence spectra at RT (room temperature) and LT (low
temperature, 77 K) as seen in Figure 6.
The second possible explanation for this difference in
behavior is that there are conformations having nonadjacent
carbazole and divinylbenzene in close contact that are not
favored in low-temperature glasses. This was mentioned above
as a possibility that could account for the low-energy
fluorescence band that would involve ICT without having to
use the silicon bridges and hence have better overlap between
initial and final electron orbitals.
Orbital overlap plays a critical role because electron transfer
involves the electronic exchange interaction matrix element
VET = Ψ (r )Ψ (r2)|1/(r − r2)|Ψ (r2)Ψ (r )
(4)
D
1
A
1
D
A 1
where ΨD and ΨA are one-electron molecular orbitals on the
donor and acceptor, respectively. Because the two molecular
orbitals involve overlaps with themselves at the two different
sites, r1 and r2, the square of the electronic exchange interaction
varies approximately exponentially with the distance (r1 − r2)
between donor and acceptor.28
|VET|2 = |VET,0|2exp{−β(r − r2)}
(5)
1
Room temperature fluorescence traces of 4 in THF at 370
and 500 nm following excitation by a 2 ps laser pulse at 285 nm
are shown in Figure 7. Also shown is a fluorescence trace of
xanthione (XT) in CY at 460 nm. A fluorescence trace of
xanthione was used to reconstruct the shape of the IRF to be
used in deconvoluting the 500 nm fluorescence trace of 4.31
where the time dependence has been factored out of VET and β
is the reciprocal of the distance between donor and acceptor
that the interaction goes to 1/e. Because of the flexibility of the
copolymer chain nonadjacent carbazole and divinylbenzene
sites could involve smaller intersite distances than adjacent sites
which include the extra distance of the silicon linker.
One final low-temperature experiment was performed that
helped to distinguish between these two possible explanations.
The fluorescence spectra of compound 3 was also measured in
DCM:MeOH at RT and 77 K. The preliminary spectra were
perfectly analogous to the fluorescence spectra of 4 in Figure 6.
In particular, a fluorescence spectrum of 3 in DCM:MeOH at
RT showed both low- and high-energy bands, but at 77 K only
the high-energy band was present. In compound 3, there is only
adjacent carbazole and divinylbenzene, so the ICT has to be
across the silicon bridge. Furthermore, there is likely a
conformational change that is being inhibited by the LT
glass. So even though this does not totally eliminate the
possibility of an ICT between nonadjacent carbazole and
divinylbenzene moieties in 4, the existence of dual
luminescence in 3 at RT and the lack of it at LT makes it
plausible that the low-energy band of the dual luminescence in
Figure 7. Fluorescence traces of compound 4 in THF, λmonit = 370 and
500 nm, in the picosecond range (λexc = 285 nm).
4756
dx.doi.org/10.1021/jp504649p | J. Phys. Chem. A 2014, 118, 4750−4758

The Journal of Physical Chemistry A
Article
Notes
The decay of the LE fluorescence trace at 370 nm and the
growth of the ICT fluorescence trace at 500 nm were
determined taking into account deconvolution with the IRFs
with respect to the respective wavelengths.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Following deconvolutions, the decay of the 370 nm LE
fluorescence trace matches the rise of the 500 nm ICT
fluorescence trace. Both time constants are approximately 25
ps. This agreement between these two time constants is
consistent with the notion that the ICT state is being formed
from the LE state which, in turn, is consistent with the TICT
hypothesis. Furthermore, the long decay times of the LE state
at 370 nm and the ICT state at 500 nm, shown in Figure 5, are
consistent with the two states being in thermal equilibrium.
Combining the results in Figures 5 and 7, the time constant of
the initial decay of LE and the ICT growth is associated with
their approach to thermal equilibrium, and, after achieving
equilibrium, they decay together.
Financial support from The National Science Centre Grant
(Poland) (Grant MAESTRO 2011/02/A/ST5/00472) is
gratefully acknowledged. This is document no. NDRL-4988
from the Notre Dame Radiation Laboratory which is supported
by the Office of Basic Energy Sciences of the US Department of
Energy. The authors thank Dr. Jerzy Karolczak from the Centre
of Ultrafast Laser Spectroscopy Adam Mickiewicz University
for performing and analyzing the picosecond experiments.
DEDICATION
■
This paper is dedicated to the memory of Professor Klaus-
Dieter Asmus.
Finally, it was also necessary to show that the low-energy
emission was a fluorescence from a singlet ICT state and not a
phosphorescence from a triplet state of the copolymer. To the
latter issue, it is noted that the emission lifetimes were of the
same order of magnitude as the fluorescence of carbazole, and
the triplet yield of a related silicon-bridged compound had a
very low triplet yield.21 Lack of the lower-energy luminescence
at 77 K also can be counted as evidence against this emission
being phosphorescence. These observations suggest that the
emission is fluorescence and not phosphorescence.
REFERENCES
■
(1) Wang, F.; Kaafarani, B. R.; Neckers, D. C. Synthesis of Silicon-
Containing Unsaturated Polymers by Hydrosilylation Reactions.
Photophysical Studies. Macromolecules 2003, 36, 8225−8230.
(2) Luh, T.-Y.; Cheng, Y.-J. Alternating Divinylarene-Silylene
Copolymers. Chem. Commun. 2006, 4669−4678.
(3) Grimsdale, A. C.; Chan, K. L.; Martin, R. E.; Jokisz, P. G.;
Holmes, A. B. Synthesis of Light-Emitting Conjugated Polymers for
Applications in Electroluminescent Devices. Chem. Rev. 2009, 109,
897−1091.
(4) Luh, T.-Y.; Hu, Z. Thorpe−Ingold Effect in Organosilicon
Chemistry. Dalton Trans. 2010, 39, 9185−9192.
(5) Jung, S.-H.; Kim, H. K.; Kim, S.-H.; Kim, Y. H.; Jeoung, S. C.;
Kim, D. Palladium-Catalyzed Direct Synthesis, Photophysical Proper-
ties, and Tunable Electroluminescence of Novel Silicon-Based
Alternating Copolymers. Macromolecules 2000, 33, 9277−9288.
(6) Sasano, T.; Sogawa, H.; Tamura, K.; Shiotsuki, M.; Masuda, T.;
Sanda, F. Synthesis and Properties of Conjugated Polymers
Containing 3,9-Carbazolylene and Silylenevinylene Moieties in the
Main Chain. J. Polym. Sci., Part A: Polym. Chem. 2010, 48, 1815−1821.
(7) Zou, Y.; Gendron, D.; Badrou-Aïch, R.; Najari, A.; Tao, Y.;
Leclerc, M. A High-Mobility Low-Bandgap Poly(2,7-Carbazole)
Derivative for Photovoltaic Applications. Macromolecules 2009, 42,
2891−2894.
(8) Barford, W. Excitons in Conjugated Polymers: A Tale of Two
Particles. J. Phys. Chem. A 2013, 117, 2665−2671.
(9) Paik, K. L.; Baek, N. S.; Kim, H. K.; Lee, J.-H.; Lee, Y. White
Light-Emitting Diodes from Novel Silicon-Based Copolymers
Containing Both Electron-Transport Oxadiazole and Hole-Transport
Carbazole Moieties in the Main Chain. Macromolecules 2002, 35,
6782−6791.
CONCLUSIONS
■
Dual emission was observed upon photoexcitation of the newly
synthesized copolymer 4. The dependence of the lower energy
emission on solvent polarity showed that this excited state had
a large dipole moment. Lack of this low-energy emission at low
temperature indicated that both emissions were from singlet
excited states. Using a synthetic strategy for systematically
building up the copolymer until dual luminescence occurred, it
was shown that dyad 3 is the minimum structure for observing
ICT in the copolymer 4.
Additional detail about the nature of the ICT state was
obtained in the low-temperature (77 K) experiments on 3 and
4 that showed no ICT fluorescence at LT in polar glasses. This
observation was interpreted as being due to the LT glasses
restricting conformational changes necessary for ICT or to the
temperature changing the equilibrium populations of con-
formers associated with LE and ICT states. Both possibilities
would involve electron transfer through the dimethylsilylene
bridge. These conformational changes were shown to be
associated with excited-state processes (likely involving a TICT
state) using time-resolved fluorescence with picosecond and
subnanosecond excitations.
(10) Liu, X.-M.; Xu, J.; Lu, X.; He, C. A Comparative Study on
Luminescent Copolymers of Fluorene and Carbazole with Conjugated
or δ-Si Interrupted Structures: Steric Effects. Macromolecules 2006, 39,
1397−1402.
(11) Wang, H.-W.; Cheng, Y.-J.; Chen, C.-H.; Lim, T.-S.; Fann, W.;
Lin, C.-L.; Chang, Y.-P.; Lin, K.-C.; Luh, T.-Y. Photoinduced Electron
Transfer in Silylene-Spaced Copolymers Having Alternating Donor-
Acceptor Chromophores. Macromolecules 2007, 40, 2666−2671.
(12) Ludwiczak, M.; Majchrzak, M.; Bayda, M.; Marciniak, B.;
Kubicki, M.; Marciniec, B. Stereoselective Synthesis and Luminescence
Properties of Novel Trans-Regular N-Alkylcarbazolylene-Silylene-
Vinylene Polymers. J. Organomet. Chem. 2014, 750, 150−161.
(13) van Walree, C. A.; Roest, M. R.; Schuddeboom, W.; Jenneskens,
L. W.; Verhoeven, J. W.; Warman, J. M.; Kooijman, H.; Spek, A. L.
Comparison between SiMe2 and CMe2 Spacers and σ-Bridges for
Photoinduced Charge-Transfer. J. Am. Chem. Soc. 1996, 118, 8395−
8407.
Experiments and calculations are underway on dyad 3 to
determine what is the precise nature of the time-dependent,
excited-state processes.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: mbayda@amu.edu.pl.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given the approval to the final version
of the manuscript.
(14) Grabowski, Z. R.; Rotkiewicz, K.; Rettig, W. Structural Changes
Accompanying Intramolecular Electron Transfer: Focus on Twisted
4757
dx.doi.org/10.1021/jp504649p | J. Phys. Chem. A 2014, 118, 4750−4758

The Journal of Physical Chemistry A
Article
Intramolecular Charge-Transfer States and Structures. Chem. Rev.
2003, 103, 3899−4031.
(15) Ludwiczak, M.; Majchrzak, M.; Marciniec, B.; Kubicki, M.
Highly Stereoselective Synthesis of Novel Trans-Thiophenylene-
Silylene-Vinylene-Arylene Molecular and Macromolecular Com-
pounds. J. Organomet. Chem. 2011, 696, 1456−1464.
(16) Yi, C. S.; Lee, D. W.; Chen, Y. Hydrovinylation and [2 + 2]
Cycloaddition Reactions of Alkynes and Alkenes Catalyzed by a Well-
Defined Cationic Ruthenium−Alkylidene Complex. Organometallics
1999, 18, 2043−2045.
(17) Marciniec, B. Catalysis by Transition Metal Complexes of
Alkene Silylation - Recent Progress and Mechanistic Implications.
Coord. Chem. Rev. 2005, 249, 2374−2390.
(18) Gardecki, J. A.; Maroncelli, M. Set of Secondary Emission
Standards for Calibration of the Spectral Responsitivity in Emission
Spectroscopy. Appl. Spectrosc. 1998, 52, 1179−1189.
́
(19) Karolczak, J.; Komar, D.; Kubicki, J.; Wrozowa, T.; Dobek, K.;
̇
Ciesielska, B.; Maciejewski, A. The Measurements of Picosecond
Fluorescence Lifetimes with High Accuracy and Subpicosecond
Precision. Chem. Phys. Lett. 2001, 344, 154−164.
(20) Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic
Molecules, 2nd ed.; Academic Press: New York, 1971.
(21) Burdzinski, G.; Bayda, M.; Hug, G. L.; Majchrzak, M.;
Marciniec, B.; Marciniak, B. Time-Resolved Studies on the Photo-
isomerization of a Phenylene-Silylene-Vinylene Type Compound in Its
First Singlet Excited State. J. Lumin. 2011, 131, 577−580.
(22) Meech, S. R.; Phillips, D. Photophysics of Some Common
Fluorescence Standards. J. Photochem. 1983, 23, 193−217.
(23) Turro, N. J.; Ramamurthy, V.; Scaiano, J. C. Modern Molecular
Photochemistry of Organic Molecules; University Science Books:
Sausalito, CA, 2010.
(24) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Springer
Science+Business Media: New York, 2010.
(25) Lippert, E. Habilitationsschrift Zur Erlangung Der Lehrber-
ichtigung (Venia Legendi) Fur Das Fach Physikalische Chemie Ander
Techischen-Hochschule-Stuttgart - Spektroskopische Bestimmung Des
Dipolmomentes Aromatischer Verbindungen Im Ersten Angeregten
Singulettzustand. Z. Elektrochem. 1957, 61, 962−975.
(26) Mataga, N.; Kaifu, Y.; Koizumi, M. Solvent Effects Upon
Fluorescence Spectra and the Dipolemoments of Excited Molecules.
Bull. Chem. Soc. Jpn. 1956, 29, 465−470.
(27) Rehm, D.; Weller, A. Kinetics of Fluorescence Quenching by
Electron and H-Atom Transfer. Isr. J. Chem. 1970, 8, 259−271.
(28) Newton, M. D. Quantum Probes of Electron-Transfer Kinetics:
The Nature of Donor-Acceptor Interactions. Chem. Rev. 1991, 91,
767−792.
(29) McConnell, H. M. Intramolecular Charge Transfer in Aromatic
Free Radicals. J. Chem. Phys. 1961, 35, 508−515.
(30) Redwood, C.; Bayda, M.; Saltiel, J. Photoisomerization of Pre-
and Provitamin D3 in EPA at 77 K: One-Bond-Twist, Not Hula-Twist.
J. Phys. Chem. Lett. 2013, 4, 716−721.
́
(31) Wrozowa, T.; Ciesielska, B.; Komar, D.; Karolczak, J.;
̇
Maciejewski, A. Measurements of Picosecond Lifetimes by Time
Correlated Single Photon Counting Method: The Effects of the
Refraction Index of the Solvent on the Instrument Response Function.
Rev. Sci. Instrum. 2004, 75, 3107−3121.
NOTE ADDED IN PROOF
■
We have been made aware that an alternate explanation is that
the conformation that takes place in the excited state of
compound 4 might be the conformational relaxation of the
solvent. Further experiments are planned.
4758
dx.doi.org/10.1021/jp504649p | J. Phys. Chem. A 2014, 118, 4750−4758