Solvent effects on the triplet-triplet annihilation upconversion of diiodo-Bodipy and perylene

Article abstract of DOI:10.1039/c6cp06897a

Solvent effects play a very important role in photochemical reactions and energy transfer processes in solution; however, these effects are rarely mentioned in the triplet-triplet annihilation (TTA) upconversion fluorescence experiments. In a typical TTA upconversion system of a photosensitizer of diiodo-Bodipy (I₂-Bodipy) and a triplet acceptor of perylene, five common inert solvents, hexane, heptane, toluene, 1,4-dioxane, and dimethyl sulfoxide (DMSO), were used to investigate the solvent effects on the overall quantum yield of upconversion fluorescence. Femtosecond and nanosecond time-resolved transient difference absorption spectra were obtained to study the efficiencies of intersystem crossing (ISC) and triplet-triplet energy transfer (TTET). From the obtained upconversion fluorescence emission spectra, the overall TTA upconversion fluorescence quantum yield was derived. Among the five solvents, the upconversion quantum yield in dioxane is the highest at 19.16%, more than twice that that in toluene (8.75%). For the solvents hexane, heptane, toluene, and dioxane, the yields generally follow the sequences of polarity and viscosity. However, a very low upconversion quantum yield (1.51%) was observed in DMSO although the TTET process and fluorescence quantum yield of perylene in DMSO were almost as efficient as in dioxane. Based on density functional theory calculations, a reasonable explanation for these solvent effects was proposed.

Full text of DOI:10.1039/c6cp06897a

View Article Online
View Journal
PCCP
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: Q. ZHOU, M.
Zhou, Y. Wei, X. ZHOU, S. Liu, S. Zhang and B. Zhang, Phys. Chem. Chem. Phys., 2016, DOI:
1
0.1039/C6CP06897A.
Volume 18 Number
1
7
January 2016 Pages 1–636
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
PCCP
Physical Chemistry Chemical Physics
www.rsc.org/pccp
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9076
PERSPECTIVE
Darya Radziuk and Helmuth Möhwald
Ultrasonically treated liquid interfaces for progress in cleaning and
separation processes
rsc.li/pccp

Page 1 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
Solvent Effect on Triplet-Triplet Annihilation Upconversion of
diiodo-Bodipy and Perylene
1
2
1
1,3,‡
1,3
Qiaohui Zhou, Miaomiao Zhou, Yaxiong Wei, Xiaoguo Zhou,
Shilin Liu ,
2
,‡
2
Song Zhang , Bing Zhang
1
Hefei National Laboratory for Physical Sciences at the Microscale and Department of Chemical
Physics, University of Science and Technology of China, Hefei, Anhui 230026, China
2
State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, Wuhan Institute of
Physics and Mathematics, Chinese Academy of Sciences, Wuhan 430071, China
3
Synergetic Innovation Center of Quantum Information & Quantum Physics, University of Science and
Technology of China, Hefei, Anhui 230026, China
ABSTRACT
Solvent effect plays a very important role in photochemical reaction and energy transfer
process in solution, but it is rarely mentioned in tripletꢀtriplet annihilation (TTA) upconversion
fluorescence experiments. In a typical TTA upconversion system of a photosensitizer of
diiodoꢀBodipy (I ꢀBodipy) and a triplet acceptor of perylene, five common inertia solvents,
2
hexane, heptane, toluene, 1,4ꢀdioxane and dimethyl sulfoxide (DMSO), were used to investigate
the solvent effect on the overall quantum yield of upconversion fluorescence. Femtosecond and
nanosecond timeꢀresolved transient difference absorption spectra were measured to study
efficiencies of intersystem crossing (ISC) and tripletꢀtriplet energy transfer (TTET). From the
recorded upconversion fluorescence emission spectra, the overall TTA upconversion
fluorescence quantum yield was derived. Among the five solvents, the upconversion quantum
yield in dioxane is the highest as 19.16%, and more than twice value in toluene, 8.75%. For the
solvents of hexane, heptane, toluene and dioxane, the yields generally follow the sequence of
polarity and viscosity. However, a very low upconversion quantum yield (1.51%) was observed
in DMSO, although the TTET process and fluorescence quantum yield of perylene in DMSO
were almost as efficient as it in dioxane. Based on the calculation of density functional theory, a
reasonable explanation of solvent effect was proposed.
KEYWORDS: tripletꢀtriplet annihilation, energy transfer, upconversion fluorescence, quantum
yield, solvent effect
_________________________________
*
Authors to whom correspondence should be addressed. Electronic mail: xzhou@ustc.edu.cn (X. Zhou),
zhangsong@wipm.ac.cn (S. Zhang)

Physical Chemistry Chemical Physics
Page 2 of 20
View Article Online
DOI: 10.1039/C6CP06897A
1. INTRODUCTION
1
Since upconversion fluorescence was observed over 50 years ago, it has attracted extensive
attention due to the potential applications in photoꢀdynamic therapy, photovoltaics and
2
–7
photocatalysis. A few traditional methods were developed for this aim including upconversion
with twoꢀphoton absorption dyes, upconversion with inorganic crystals and rare earth
8
,9
materials. However, the drawbacks of high excitation power, too weak absorption of visible
light and low upconversion quantum yield, restricted application of these methods. Recently, the
tripletꢀtriplet annihilation (TTA) upconversion has received wide interest due to its
overwhelming advantages of an intense absorption of solar light, as well as high upconversion
1
0,11
quantum yields.
Moreover, the TTA upconversion fluorescence creates incomparable
1
2–14
application in bioimaging when the prompt fluorescence technique cannot be applied.
An overall TTA upconversion process is shown in Scheme 1, and four steps are mainly
involved upon excitation: intersystem crossing (ISC) from the singlet excited state to triplet state
of photosensitizer, tripletꢀtriplet energy transfer (TTET) from triplet sensitizer to triplet acceptor,
TTA of triplet acceptor and delayed fluorescence (upconversion fluorescence) emission of the
excited acceptor. To synthesize a new triplet photosensitizer with more intense absorption ability
of visible light and the higher ISC efficiency, is most attractive. Some organic compounds, e.g.
RuBpy, Pt (II) and Ir (III) porphyrin complexes, have been often used as the triplet
1
5–21
photosensitizer since they have a very high ISC efficiency due to “heavyꢀatom effect”.
Among these sensitizers, boronꢀdipyrromethene (Bodipy) derivatives show a couple of major
merits: the strong absorption in visible light range and feasibly modified structure. A lot of
efforts were made to synthesize the new Bodipyꢀbased sensitizers for improving antiꢀStokes shift
2
2–25
and the upconversion efficiency.
Recently, Zhao and his coꢀworkers have synthesized a
series of metal complex and pure organic triplet photosensitizers, especially Bodipy derivative
2
6–29
dayds.
These sensitizers had not only a strong absorption ability in near infrared wavelength
range, but also a high ISC efficiency which benefited from the long lifetime of the triplet state.
The upconversion quantum yield was reported up to 39.9% for naphthalimide (NI)
3
0
acetylideꢀcontaining Pt(II) complex.
2

Page 3 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
Scheme 1. Jablonski diagram of the overall TTA upconversion process. I ꢀBodipy is the triplet
2
donor (photosensitizer) and perylene is the triplet acceptor in the present experiment.
In TTET and TTA processes, a triplet donor needs to collide with the triplet acceptor in a
same solvent cage to generate an encounter. Thus solvent effect plays a very important role in
the processes as well as characteristics of the triplet sensitizer. However, it is rarely mentioned in
1
1
TTA upconversion fluorescence experiments, although the solvent effect has been wellꢀknown
3
1–34
in photochemical reaction and energy transfer process.
Generally, solvents with different
polarity and viscosity not only affect the reaction rate, but also are able to change the resonance
2
9,35,36
wavelength and absorption intensity of molecules.
Since dipole moments of singlet and
triplet states are usually different, the energy gap between them can be influenced by solvents,
3
7–39
and consequently the different dynamic behaviors are probably visible in different solvents.
Very recently, Yokoyama et al. observed the solvent viscosity effect on TTA upconversion of
4
0
platinum octaethylporphrin and DPA. By changing temperature and applying magnetic field,
the effect of the solvent viscosity dependence of the delayed fluorescence of DPA was found.
However, the viscosity (or polarity) dependence of absorption and ISC efficiency of triplet
sensitizer was artificially ignored, and the solvent effect on the prompt fluorescence quantum
yield of annihilator was not discussed ether.
In the present work, a typical TTA upconversion system of I ꢀBodipy and perylene has been
2
investigated in five solvents, hexane, heptane, toluene, 1,4ꢀdioxane and DMSO, respectively. All
these solvents are chemical inert and less volatile. Through comparing the dynamic behavior and
upconversion efficiency, the solvent effect can be clearly observed. As the previous experiments
2
7
shown, I ꢀBodipy in toluene absorbs the visible light in a wavelength range of 500ꢀ550 nm, and
2
produces the triplet state with a high ISC efficiency due to the heavy atom effect of iodine. The
triplet state of a ~50 ꢁs long lifetime supports its role of the triplet donor, and the triplet perylene
can be formed as the triplet acceptor in TTET process. The TTA upconversion fluorescence has
3

Physical Chemistry Chemical Physics
Page 4 of 20
View Article Online
DOI: 10.1039/C6CP06897A
been finally observed in a range of 430ꢀ530 nm with a moderate upconversion efficiency of 7.5%.
Although the upconversion spectra and efficiency have been already obtained, dynamic details in
the overall process, e.g. ISC efficiency and so on, are still unknown. More importantly, solvent
effect is never paid attention in this TTA upconversion system. Therefore, femtosecond and
nanosecond transient difference absorption spectra and upconversion spectra will be applied to
measure these useful dynamic information. Consequently, solvent effect on all the involved
dynamic processes will be discussed respectively.
2. MATERIALS AND METHODS
2.1 Triplet photosensitizer and triplet acceptors of TTA upconversion.
Bodipy 1 has a strong absorption in visible light region, and its fluorescence quantum yield
is also quite high. Based on its modifiable structure, iodine atoms were attached on the core of
fluorophore as shown in Scheme 2. The details of synthetic process were very similar to those in
1
13
reference 27. The structure of I ꢀBodipy was verified by mass spectrometry, H NMR and C
2
NMR. All the precursors of analytically pure were purchased from SigmaꢀAldrich Co. without
any purification. Solvents were dried and distilled prior to use.
Scheme 2. Synthesis of 2,6ꢀDiiodoꢀ1,3,5,7ꢀtetramethylꢀ8ꢀphenylꢀ4,4ꢀdifluoroboradiazaindacene
2
7
(
I ꢀBodipy). (i) N atmosphere, CH Cl , Et N, and BF ·Et O; (ii) NIS and CH Cl , 1 h.
2 2 2 2 3 3 2 2 2
In the present experiment, I ꢀBodipy was used as the lightꢀharvesting antenna and triplet
2
donor at the same while. Usually a triplet acceptor in TTA upconversion should have a high
fluorescence quantum yield and the lower triplet energy than that of donor. Another essential
limitation of TTA for the triplet acceptor was that its energy of S state should be just lower than
1
the twice of the T energy. Thus the most triplet acceptors were polycyclic aromatic hydrocarbon
1
compounds, and here perylene was chosen as the triplet acceptor in the present TTA
upconversion system.
2.2 Experimental setups.
4

Page 5 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
The steadyꢀstate UVꢀVis absorption and fluorescence spectra were recorded from 400ꢀ650
nm (UVꢀ3600, Shimadzu) and 500ꢀ680 nm (Fꢀ4600, Hitachi) respectively. Structures of all the
1
synthesized chemicals were identified using the standard NMR and mass spectra. H NMR
spectra were recorded with a 400 MHz spectrophotometer (AVANCE III 400, Bruker), where
CDCl was used as solvent and TMS was the standard for which δ = 0.00 ppm. Highꢀresolution
3
mass spectra (HRꢀMS) were measured with a MALDIꢀTOFMS spectrometer (GCT, Micromass
UK).
The photochemical behaviors of the present chemical systems were studied using the
femtosecond and nanosecond timeꢀresolved transient absorption spectra. The details of
4
1
femtosecond transient absorption spectra have been described elsewhere, and thus only a brief
has been introduced here. Ti:Sapphire in femtosecond laser system was pumped by CW second
harmonic of a Nd:YVO4 laser and amplified by a Nd:YLF to generate a 35 fs pulse width
centered at 800 nm with a maximum energy of 1 mJ /pulse. A fraction of the laser was frequency
doubled in a 1 mm thick BBO crystal, and a pulse at 400 nm with an energy of 100 µJ was
yielded to pump the NOPA. The output NOPA was set at 532 nm and used as the excitation
pulse in the present experiment. The energy was about ~4.5 µJ by an attenuation. The NOPA
pulse was temporally compressed to obtain the minimum pulse width compatible with their
bandwidth. A white light continuum generated by focusing 800 nm on a CaF plate was reflected
2
from the front and back surfaces of a quartz plate to obtain the probe and reference beams. The
relative polarization of the pump and probe pulses was kept as 54.7° for all the measurements.
The absorption spectra were detected with a CCD camera equipped with a spectrometer
(Princeton, SpectraPro 2500i). The instrumental response function of the system was typically
better than 150 fs.
Nanosecond timeꢀresolved transient absorption spectra and dynamics curves were measured
4
2,43
based on the homeꢀbuilt laser flash photolysis system.
The second harmonic (532 nm) of a
QꢀSwitched Nd:YAG laser (Labꢀ170, Spectra Physics, repetition rate of 10 Hz) was used as the
excitation light (pulse duration 8 ns, pulse energy ~7 mJ/pulse). An analyzing light from a 500 W
Xenon lamp and the pulsed excitation laser passed through a flow quartz cuvette perpendicularly.
The transient absorption spectra and dynamic curves were measured with a monochromator
equipped with a photomultiplier (CR131, Hamamatsu) and the data were recorded with an
oscilloscope (TDS3052B, Tektronix).
5

Physical Chemistry Chemical Physics
Page 6 of 20
View Article Online
DOI: 10.1039/C6CP06897A
The upconversion fluorescence spectra were measured by a modified commercial
spectrometer. A pump laser of ~20 mW from a stable CW laser (VerdiꢀV5, 532 nm, Coherent)
was used as the excitation light in TTA upconversion. The fluorescence were dispersed and
detected with a triple monochromator system (TriplePro, Acton Research) and a CR131
4
4,45
photomultiplier. In order to reduce selfꢀabsorption, the backscattering geometry was applied.
ꢀ1
The normal spectral resolution was ~1.0 cm . In the present experiments, the concentrations of
ꢀ6
ꢀ4
I ꢀBodipy and perylene were fixed respectively at 5.0×10 M and 3.0×10 M in the five selected
2
solvents. All the samples in nanosecond transient and upconversion experiments were
deoxygenated by purging with highꢀpurity argon (99.99%) for at least 20 minutes in advance.
And the degassing was kept during the measurements to be sure an anaerobic environment.
2.3 Theoretical Calculations.
The geometries of I ꢀBodipy and perylene were optimized with density functional theory
2
(DFT) at B3LYP/6ꢀ31G(d,p) level of theory. Due to the heavyꢀatom of iodine, LanL2DZ basic
set was introduced into the DFT calculations of the photosensitizer, I ꢀBodipy. The spin density
2
surface of I ꢀBodipy was analyzed at the same level of theory. Based on the optimized structure
2
of the ground state, the timeꢀdependent DFT (TDDFT) with PCM model was used to calculate
excitation energies of the lowest singlet S and triplet T states of I ꢀBodipy and perylene in
1
1
2
4
6
different solvents. All the calculations were performed with Gaussian 09W program package.
3. RESULTS AND DISCUSSIONS
3.1 Steady-state UV-Visible absorption and fluorescence Spectra.
Figure 1(a) shows the steadyꢀstate UVꢀVisible absorption spectra of I ꢀBodipy in the five
2
solvents. The maximum absorption is located at 537 nm along with a shoulder band at ~504 nm.
Besides a slight redꢀshift is recognized, the intensities of the maximum absorption in different
5
solvents are obviously changed. The molar extinction coefficients ε varies from 1.24×10
ꢀ1
ꢀ1
4
ꢀ1
ꢀ1
M ·cm in heptane to 6.97×10 M ·cm in DMSO. Upon photoexcitation at 532 nm, the
fluorescence emission spectra of I ꢀBodipy are shown in Figure 1(b). Similar to the UVꢀVis
2
absorption spectra in Figure 1(a), a slight redꢀshift can be observed in a range of 552ꢀ561 nm,
and the maximum intensity shows a 4.7ꢀfold enhancement from DMSO to heptane.
6

Page 7 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
Figure 1. UVꢀVis absorption (a) and fluorescence emission (b) spectra of I ꢀBodipy in different
2
ꢀ6
solvents (c=5×10 M).
Table 1 summarizes the absorption and emission wavelengths, molar absorption coefficients
and fluorescence quantum yields. Without substitution of iodine atom, Φ of Bodipy 1 was
F
0
.712, while Φ of I ꢀBodipy dramatically decreased to a few percent. As the direct
F 2
photoexcitation to triplet state (T ) is transitionꢀforbidden, the unique way to form T state is via
l
1
ISC process to compete with the fluorescence emission. As shown in the Jablonski diagram of
Scheme 1, the decreased Φ of I ꢀBodipy indicates more population of the triplet state.
F
2
Table 1. Photophysical parameters of I ꢀBodipy
2
4
ꢀ1
ꢀ1
a
b
Solvent
Hexane
Heptane
Toluene
Dioxane
DMSO
λ (nm)
ε (×10 M ·cm )
λ (nm)
Φ (%)
F
abs
em
534
535
537
534
536
8.18
12.4
8.6
552
554
558
555
561
4.86±0.05
3.14±0.05
3.01±0.08
3.27±0.08
1.12±0.09
8.0
7.0
a
molar absorption coefficient.
prompt fluorescence quantum yield with the photoexcitation at 532 nm.
b
3.2 Femtosecond transient difference absorption spectra.
Femtosecond transient difference absorption spectra were measured to characterize the
initial photophysical processes of photosensitizer, e.g. lifetime of the singlet excited state, ISC
rate from S to T , and so on. Figure 2(aꢀe) show the recorded femtosecond transient difference
1
1
absorption spectra of I ꢀBodipy in the five solvents upon photoexcitation at 532 nm. The delay
2
time was gradually changed from 40 fs to 900 ps. There are a negative (~540 nm) and three
positive bands (346, 471 and 716 nm) in all spectra.
7

Physical Chemistry Chemical Physics
Page 8 of 20
View Article Online
DOI: 10.1039/C6CP06897A
Figure 2. Femtosecond timeꢀresolved transient difference absorption spectra of I ꢀBodipy in (a)
2
hexane, (b) heptane, (c) toluene, (d) dioxane, (e) DMSO. (f) The experimental (dotted lines)
and fitted (solid lines) normalized decay curves of the absorption band at 346 nm (λ = 532
ex
ꢀ5
nm, c=1×10 M).
Taking the spectra in hexane of Figure 2(a) as an example, the strongest negative band at
40 nm is assigned as the stimulated emission and/or the ground bleaching. From the initial 200
5
fs to 47 ps, its position and intensity do not essentially change. The slight blueꢀshift is due to the
overlapping of the adjacent broad band of 600ꢀ700 nm (positive peak). The band at 346 nm is a
typical positive peak and its intensity monotonously decays with the delay time. Thus the band
can be attributed to absorption of the S state and its decay mainly corresponds to the S ꢀS
0
1
1
process. Interestingly, the band at 471 nm with moderate intensity is quickly increased to the
maximum in the initial one picosecond, then slowly decays and gradually blueꢀshifts to 450 nm.
The isoabsorptive point is located at 453 nm and its intensity is kept strong all the while in the
present delay time range. Therefore, an ISC process to produce the triplet state is expected and
the absorption of the latter contributes the band. With the delay time increasing, the weak peak at
~700 nm shows an overall trend of growing and broadening. From 1 ps on, the intensity at 716
8

Page 9 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
nm keeps monotonous enhancing, and the absorption is gradually blueꢀshifted. Such a long delay
timescale indicates that a triplet excited state is generated, which is consistent with the above
conclusion. Thus the absorption of the triplet state T is believed to contribute the bands at 450
1
and 716 nm. Actually, the following nanosecond transient difference spectra also verify the
spectral assignment.
As shown in Figure 2(aꢀe), the peak patterns keep almost the same in different solvents,
however lifetimes of the excited states of I ꢀBodipy obviously change from hexane to DMSO. As
2
shown in Figure 2(f), the intensity at 346 nm was normalized to compare and obtain the lifetime
of the S state. With the solvent polarity and viscosity increasing, the decay rate gradually
1
increases. By fitting all the curves with a tripleꢀexponential function, all dynamic parameters in
decay process of the S state were obtained. Three dynamic decay processes were taken into
1
account with three typical time components, τ , τ and τ , respectively. The first (τ ) was the
1
2
3
1
natural lifetime of S state (here it mainly corresponds to internal conversion of S ꢀS , the
1
1
0
second (τ ) was the ISC time to form the triplet state T , and the third (τ ) was the other slow
2
1
3
decays involving collision and unknown processes. Since we do not care about the slowest
process of τ in the present ultrafast experiment, Table 2 only summarizes the data of τ and τ .
3
1
2
ꢀ12
Usually, internal conversion is a very fast decay (~10 s), and τ decreases indeed from 1.9 ps to
1
0
.36 ps with a polarity order from hexane to DMSO. With the same sequence, τ of ISC process
2
is reduced from 362 ps to 83 ps.
1
*
Table 2. The decay dynamic data of I ꢀBodipy in five solvents.
2
a
Solvent
Polarity
0.06
0.2
τ1(ps)
1.90
1.05
0.62
0.45
0.36
τ2(ps)
362
375
235
234
83
Hexane
Heptane
Toluene
Dioxane
2.4
4.8
DMSO
7.2
a
The relative dielectric constant ε.
The energy gap ꢂEST between the S and T states is the essential element of ISC rates.
1
1
However, the excitation energy of the T state of I ꢀBodipy could not be easily obtained due to
1
2
none phosphorescent at either room temperature or lowꢀtemperature, and thus the TDꢀDFT
calculations were performed to estimate the excitation energies of the lowꢀlying electronic states.
Figure 3 shows the calculated excitation energies of S ꢀS and S ꢀT of I ꢀBodipy, where the
0
1
0
1
2
9

Physical Chemistry Chemical Physics
Page 10 of 20
View Article Online
DOI: 10.1039/C6CP06897A
energy of the ground state in DMSO is set as zero. As noted in Fig.3, both the S ꢀS and S ꢀT
1
0
1
0
excitation energies are almost unchanged (2.65ꢀ2.67 eV and 1.52ꢀ1.53 eV) from hexane to
dioxane. It is interesting that the S energy dramatically decreases to 2.53 eV in DMSO of the
1
highest polarity, but the S ꢀT energy only increases a little bit from 1.52 eV in hexane to 1.55
0
1
eV in DMSO. Thus ISC from S to T of I ꢀBodipy in DMSO should be the most favorable due
1
1
2
to its lowest energy gap ꢂE , compared with the other solvents. This conclusion is in excellent
ST
agreement with the present experimental data in Table 2.
Figure 3. The excitation energies of the lowꢀlying electronic states of I ꢀBodipy and perylene in
2
different solvents, calculated at TDꢀB3LYP/6ꢀ31G(d,p) level of theory.
3
.3 Quenching process of triplet I -Bodipy by perylene.
2
3
*
Upon photoexcitation at 532 nm, the triplet photosensitizer, I ꢀBodipy , is produced and
2
quenched by triplet acceptors and collision in solutions. Nanosecond transient difference
absorption spectra were used to study its quenching process in the five solvents. Figure 4 shows
the transient difference absorption spectra of I ꢀBodipy in toluene with photoexcitation at 532
2
nm, and the spectra in the other solvents are summarized in the supplementary information. All
the spectra have very similar profile in the five solvents. The bleaching band at 537 nm is due to
the depletion of the ground state of I ꢀBodipy. Another two positive bands at 443 nm and 710 nm
2
are contributed to absorption of the triplet state T of I ꢀBodipy. The assignment was confirmed
1
2
2
7
by an additional aerated experiment. The intensities of the three bands show almost the same
variation with the delay time.
1
0

Page 11 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
Figure 4. Nanosecond transient difference absorption spectra of I ꢀBodipy in toluene under
2
ꢀ5
photoexcitation at 532 nm (c=1×10 M), and the experimental and fitted decay curves at 710
nm are plotted in the insert panel.
Concentration of photosensitizers always plays a key role in the lifetime of the triplet state,
and too high concentration probably leads to the serious selfꢀquenching. Thus the concentration
−
5
of I ꢀBodipy in the present experiment was kept as 1.0×10 M to balance the better
2
signalꢀtoꢀnoise ratio and measurement of lifetime. Since the three bands have the very close
decay rates, only the dynamic curve of 710 nm is shown in the insert panel of Figure 4. In
3
*
toluene, a lifetime (τ ) of 50.12 ꢁs is obtained for the triplet state T of I ꢀBodipy by fitting the
0
1
2
3
*
curve. With the viscosity of solvent increasing, the T lifetime of I ꢀBodipy is prolonged from
1
2
13.0 ꢁs in hexane to 95.8 ꢁs in DMSO, as shown in Figure 5(a). The trend is consistent with the
calculated energies of the T state in different solvents of Figure 3.
1
3
*
Figure 5. (a) Lifetime of the triplet state T of I ꢀBodipy in five solvents; (b) SternꢀVolmer
1
2
plots generated from the T quenching of I ꢀBodipy in the present of perylene in five solvents
1
2
−
6
(
[I ꢀBodipy] = 5.0×10 M, λ = 532 nm).
2 ex
3
*
Such a long lifetime of I ꢀBodipy provides a high possibility of TTET between it and the
2
triplet acceptor, when the latter is added into the solution. The TTET process between
3
*
I ꢀBodipy and perylene can be briefly introduced as the following formula (1). In the present
2
experiment, all the absorption bands in Figure 4 were dramatically quenched in the presence of
3
*
perylene (as triplet acceptor) indeed. The obvious quenching rate k =1/τ of I ꢀBodipy was
obs
2
obtained by fitting its decay curve at 710 nm. Figure 5(b) shows relationships between 1/τ and
the concentration of perylene in different solvents. Using the SternꢀVolmer equation (2), the
bimolecular quenching rate constant, k , was obtained by linearꢀfitting the relationship of Fig.
q
5(b).
3
*
1
3
*
I ꢀBodipy (T ) + perylene (S ) ꢀ I ꢀBodipy (S ) + perylene (T )
(1)
2
1
0
2
0
1
k = 1/τ = 1/τ + k ·[perylene]
(2)
obs
0
q
1
1

Physical Chemistry Chemical Physics
Page 12 of 20
View Article Online
DOI: 10.1039/C6CP06897A
whereτ represents the lifetime in the absence of perylene. Table 3 lists the fitted 1/τ and
0
0
bimolecular quenching rate constants k in the five solvents, as well as the diffusion rate kdiffuse
.
q
kdiffuse can be determined as 8k T/(3η) as the upper limit of k , where η is the viscosity of solvent.
B
q
3
*
Obviously, all k values are close to kdiffuse, indicating that the TTET between I ꢀBodipy and
q
2
perylene is mainly a diffusionꢀcontrolled dynamic process. In addition, both k and kdiffuse follow
q
the same sequence as 1/τ in the five solvents.
0
As suggested in Dexter mechanism, the exchanged electrons should occupy the orbital of
the other party, and the energy transfer rate is proportional to the spectral integral overlap at the
short collision distance. Thus diffusion rate in the solvent of various viscosity may be the
1
1
rateꢀcontrolling factor in the overall TTA upconversion process. In addition, a suitable triplet
acceptor should have the lower energy than that of triplet donor. Similar to that the triplet energy
3
*
of I ꢀBodipy (T ) varies with the solvents, the T energy of perylene needs to be checked
2
1
1
3
*
carefully in different solvents. At B3LYP/6ꢀ31G(d,p) level of theory, the energy of perylene (T )
1
is determined as 1.51 eV and almost unchanged in all the five solvents (in Figure 3), which
2
7
agrees with the previous data (1.53 eV in toluene). Compared with the energy of
3
*
I ꢀBodipy (T ), perylene is really a great triplet acceptor due to its slightly lower energy than the
2
1
triplet donor. As shown in Table 3, ꢂETTET are slightly increased from 0.01 eV in hexane to 0.03
eV in DMSO, and all of them are so small. Therefore, the viscosity should play a larger role in
the TTET process, and the TTET rates are expected to decrease with the solvent viscosity
increasing. The present experimental results are consistent with the theoretical prediction indeed.
3
*
Table 3. Bimolecular reaction rate constants k of I ꢀBodipy quenched by perylene, as well as
q
2
3
*
3
*
ꢂETTET between I ꢀBodipy and perylene in five solvents.
2
4
ꢀ1
9
-1 ꢀ1
9
-1 ꢀ1
5
-1
Solvent
Hexane
Heptane
Toluene
Dioxane
DMSO
η(cp)
0.33
0.41
0.59
1.54
2.24
1/τ (×10 s ) k (×10 M ·s ) kdiffuse (×10 M ·s ) ꢂETTET(eV) K (×10 M )
0
q
SV
7.64
4.77
1.99
1.10
1.04
13.30
10.33
5.57
19.67
15.83
11.0
0.014
0.014
0.017
0.016
0.032
1.74
2.17
2.79
2.70
1.94
2.98
4.21
2.03
2.89
Generally, KSV is a reliable criterion to evaluate the quenching ability of a triplet sensitizer
by an acceptor, with a definition of K =k ·τ As shown in Table 3, K does not show a straight
SV
q
0.
SV
5
-1
order of viscosity and polarity of solvents. The maximal value of KSV is 2.79×10 M in toluene,
1
2

Page 13 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
2
7
and nearly ~1.4ꢀfold of KSV in DMSO. This is a potential reason for the previous investigation
to invariably use toluene as solvent in TTA upconversion. Additionally, the TTET quantum
yields, ΦTTET = k ·[perylene]/(1/τ + k ·[perylene]), are expected as the same sequence as K .
q
0
q
SV
ꢀ5
When the concentration of perylene is increased to 7×10 M, ΦTTET is close to unity
(
k ·[perylene] = 20/τ ) in dioxane. In the following TTA upconversion measurement, the
q
0
ꢀ4
concentration of perylene was 3×10 M, and thus the TTET quantum yield was near unity in all
the solvents.
3
*
3
.4 TTA upconversion fluorescence spectra of perylene .
Upon the triplet perylene is produced, an upconversion fluorescence can be observed due to
3
* 48
TTA process of perylene . In the present experiment, the white upconversion fluorescence (in
Figure 6(a)) could be observed under photoexcitation of a CW 532 nm laser (~20 mW, power
ꢀ2
density 630 mW·cm ), and the intensities of fluorescence were remarkably different in the five
solvents. Since such the upconversion fluorescence cannot appear upon photoexcitation of
I2ꢀBodipy or perylene alone, it can be attributed to the upconverted delayed fluorescence instead
of the prompt fluorescence. Figure 6(b) shows the dispersed upconversion fluorescence spectra
in the five solvents. Obviously, the observed fluorescence is a mixture of the yellowishꢀgreen
fluorescence of I ꢀBodipy itself (~552 nm) and the blue upconversion fluorescence of perylene
2
(400ꢀ500 nm).
Figure 6. (a) Photographs of the upconversion fluorescence in different solvents; (b) TTA
upconversion fluorescence spectra of I ꢀBodipy and perylene, where the curve of hollow
2
circles is the prompt fluorescence of I ꢀBodipy alone in heptane under excitation at 532 nm.
2
ꢀ6
ꢀ4
[
I ꢀBodipy]=5×10 M, and [perylene]=3×10 M.
2
As shown in Figure 6(a), the brightest delay fluorescence was emitted in dioxane, while it
was the weakest in DMSO. In hexane and heptane, the upconversion fluorescence is weaker than
the prompt fluorescence of the photosensitizer (I ꢀBodipy) although it can be clearly
2
1
3

Physical Chemistry Chemical Physics
Page 14 of 20
View Article Online
DOI: 10.1039/C6CP06897A
2
7
distinguished in Figure 6(a). Similar to the previous experiment in toluene, the upconversion
fluorescence was observed as well in a wavelength range of 440ꢀ480 nm. Using the fluorescence
of I ꢀBodipy as the standard (Φ = 2.7% in ACN), the upconversion quantum yields (Φ ) could
2
std
UC
be determined with the equation (3),
ꢅꢆꢇꢈ
ꢋꢆꢉꢊ
ꢌꢆꢉꢊ
Φꢀꢁ = 2 ∙ Φꢂꢃꢄ ∙ (ꢅ ) ∙ (ꢋ ) ∙ (
)ꢍ
(3)
ꢆꢇꢈ
ꢌꢆꢇꢈ
ꢆꢉꢊ
where A, I, and η are the absorbance intensity, integrated luminescence intensity, and the
refractive index of the solvents used for standard and samples. In the equation, a factor of 2 is
1
2
multiplied in order to make the maximum quantum yield to be unity. Thus the upconversion
quantum yields ΦUC of the present system are determined as 5.77% in hexane, 5.59% in heptane,
8.44% in toluene, 19.16% in dioxane and 1.51% in DMSO, respectively. The present yield in
2
7
toluene is slightly higher than the previous value (7.5%). Table 4 summarizes the quantum
yields of the TTA upconversion fluorescence under irradiation at 532 nm and the prompt
fluorescence of perylene alone.
3
*
Table 4. Quantum yields of the TTA upconversion fluorescence of perylene in the five solvents
under photoexcitation at 532 nm, as well as the prompt fluorescence yield of perylene alone.
a
b
c
Solvent
Hexane
Heptane
Toluene
Dioxane
DMSO
ꢂETTA(eV)
0.31
ꢂE2T1ꢀT2(eV)
0.001
Φ (%)
Φ (%)
Py
UC
44.06±0.04
53.32±0.02
84.81±0.01
96.78±0.02
78.78±0.02
5.77±0.03
5.59±0.02
8.44±0.05
19.16±0.02
1.51±0.06
0.32
0.001
0.34
0.002
0.33
0.002
0.47
0.011
a
Energy difference of perylene calculated as ꢂETTA =2·E(T ) – E(S ).
1
1
b
Energy difference ꢂE2T1ꢀT2 =2·E(T ) – E(T ), where E(T ) and E(T ) are the energies of the first and second
1
2
1
2
triplet states of perylene.
Prompt fluorescence quantum yield of perylene.
c
It is wellꢀknown that the delayed fluorescence intensity shows a quadratic dependence on
1
1
the excitation power at low intensities, and shifts to a linear dependence at higher intensities. In
4
8
the previous experiment of Wu et al., a linear relationship between the upconverted
fluorescence intensity and the excitation power density was observed within 14.2~237.8
ꢀ2
mW·cm . Thus the maximum Φ for the present TTA upconversion system was not reached
UC
1
1
yet, since ΦUC may linearly increase with the excitation power under the threshold I . An
th
additional experiment was performed to figure out the threshold power density as shown in
Figure 7, where the delayed fluorescence intensity was plotted with the excitation power density.
1
4

Page 15 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
A change of dependence was observed from the quadratic function to the linear relationship. The
ꢀ2
threshold I was determined as ~259 mW·cm as the turning point in Figure 7. In the present
th
TTA upconversion experiment, the excitation power density is higher than I , and hence the ΦUC
th
2
7
value in Table 4 is slightly higher than the previous data.
Figure 7. The normalized integrated delayed fluorescence intensity plotted as a function of
ꢀ6
ꢀ4
the light power, where [I ꢀBodipy]=5×10 M, and [perylene]=3×10 M.
2
It is very interesting that a high quantum yield (ΦUC = 19.16%) was obtained in dioxane,
even higher than the maximal upconversion quantum yield (11.1%) predicted from the
4
7
spinꢀstatistic law in TTA process. The exceptional yield is not unique, and the higher Φ of
UC
24.3% and 39.9% have been observed for naphthalimide(NI) acetylideꢀcontaining Pt(II)
30,35
complex.
A probable explanation is provided according to the reference 12 and 47 that both
the quintet and triplet encounter complexes contribute to the formation of singlets of annihilator.
It is especially worthy to note that among the five solvents, ΦUC in DMSO is exceptionally low
as 1.51%, although both the quantum yields, ΦISC and ΦTTET, in DMSO are close to unity as well
as the other solvents.
As indicated in Scheme 1, ΦUC can be determined as the following equation (4).
Φꢀꢁ = Φꢋꢎꢁ ∙ Φꢏꢏꢐꢏ ∙ Φꢏꢏꢅ ∙ Φꢑꢒ
(4)
1
*
3
*
where ΦISC is the quantum yield of ISC from I ꢀBodipy (S ) to I ꢀBodipy (T ), and Φ is the
2
1
2
1
Py
fluorescence quantum yields of perylene upon excitation. Using the fluorescence yield Φ of
F
1
*
I ꢀBodipy (S ) in Table 1, Φ can be calculated by Φ =1–Φ as 0.95 in hexane to 0.99 in
2
1
ISC
ISC
F
DMSO. Moreover, Φ shows moderately dependent on polarity of solvent as shown in Table 4.
Py
Figure 8 shows the fluorescence emission spectra of perylene alone under photoexcitation at 390
nm. The maximal Φ value is obtained in dioxane (96.78%), while it is moderate in DMSO
Py
(78.78%).
1
5

Physical Chemistry Chemical Physics
Page 16 of 20
View Article Online
DOI: 10.1039/C6CP06897A
Figure 8. Fluorescence emission spectra of perylene alone in the five solvents under
ꢀ4
photoexcitation at 390 nm. c=3×10 M.
In general, the bottlenecks limiting the overall TTA upconversion efficiency are the TTET
and TTA processes. As mentioned above, the quantum yields ΦTTET are close to unity in all the
solvent here. Moreover, a little difference is observed for the fluorescence quantum yields of
perylene in dioxane and DMSO, as shown in Table 4. Thus such a dramatic difference of ΦUC in
dioxane and DMSO is original from the TTA process. In general, the TTA process proceeds as
the Dexter exchange mechanism, the solvent viscosity play a key role. The present experimental
conclusions indicates that the energy gap ꢂETTA between the singlet state and double of the
triplet state, ꢂETTA=2·E(T )–E(S ) is another important role besides the diffusion rate in the TTA
1
1
4
9
process. As V. Gray et al. suggested, “ꢂE
ideally is positive but close to zero for a good
TTA
annihilator in a TTA upconversion system.” At TDꢀB3LYP/6ꢀ31G(d,p) level of theory, ꢂETTA of
perylene was calculated and listed in Table 4 as well. In DMSO, ꢂETTA is apparently larger than
its value in the other solvents. Thus the remarkable increase of ꢂETTA in DMSO is thought as a
key reason for such a low TTA upconversion quantum yield in DMSO. In addition, as Schmidt
5
0
and Castellano recently pointed out, the TTA yield of singlets can be reduced when the energy
of the second triplet state, T , is lower than the twice of the T energy. As shown in Table 4, the
2
1
energy differences, ꢂE = 2×T – T , are apparently increased from near zero to 0.011 eV in
1
2
DMSO. This is another important reason to cause such a low TTA yield in DMSO. Furthermore,
the distance between two triplets in the annihilator encounter complex can also be certainly
affected by solvent polarity and viscosity. Therefore, the solvent effect on the TTA yield is
considerable, and more accurate theory and experiments are ongoing.
4. SUMMARY AND CONCLUSIONS
1
6

Page 17 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
It is wellꢀknown that solvent effect plays a very important role in photochemical reaction
and energy transfer process in solution, but it is rarely mentioned in TTA upconversion
fluorescence experiments. In a TTA upconversion system of I ꢀBodipy (as the triplet
2
photosensitizer) and perylene (as the triplet acceptor), five common inertia solvents, hexane,
heptane, toluene, 1,4ꢀdioxane and dimethyl sulfoxide (DMSO), were used to investigate solvent
effect on the overall quantum yield of upconversion fluorescence.
In femtosecond timeꢀresolved transient difference absorption spectra of I ꢀBodipy, three
2
positive peaks were observed. The bands at 450 and 716 nm are attributed to absorption of the
triplet state T , and the absorption of S state contributes the peak at 346 nm. Through fitting the
1
1
decay curves at 346 nm, the lifetime of the S state and the ISC rate from S to T were obtained
1
1
1
respectively. Among the five solvents, the lifetime of S state decreases from 1.9 ps to 0.36 ps
1
with a polarity order from hexane to DMSO. Similarly, ISC rate is increased as well with the
same sequence, which agrees very well with the calculated energy gap ꢂE between S and T
ST
1
1
states.
3
*
Upon the triplet I ꢀBodipy is produced, it can be quickly quenched by triplet acceptors and
2
collision in solutions. From nanosecond transient difference absorption spectra, the lifetime of
3
*
I ꢀBodipy was derived with the absence of perylene. With the viscosity of solvent increasing,
2
the lifetime is prolonged from 13.0 ꢁs in hexane to 95.8 ꢁs in DMSO. Such a long lifetime of
3
*
I ꢀBodipy provides a high TTET possibility between it and perylene, when the latter exists in
2
the solution. The bimolecular quenching rate constants k were obtained by linearꢀfitting the
q
SternꢀVolmer relationship in the five solvents, and gradually decreased with the viscosity of
solvent increasing. Additionally, all of them were close to the diffusion rate kdiffuse, indicating
that the TTET process is a diffusionꢀcontrolled dynamic process indeed.
From the TTET process then on, the triplet perylene was formed and then upconversion
fluorescence could be emitted via TTA process. Through recording the upconversion
fluorescence emission spectra, the overall TTA upconversion fluorescence quantum yield was
obtained. Among the five solvents, the upconversion quantum yield in dioxane is the highest as
19.16%. The moderate yields are 8.75% in toluene, 5.77% in hexane and 5.59% in heptane. For
the solvents of hexane, heptane, toluene and dioxane, the yields generally follow the sequence of
viscosity. To our surprise, a very low upconversion quantum yield of 1.51% was observed in
DMSO. Although the key role in TTA process is generally thought as the diffusion as the Dexter
1
7

Physical Chemistry Chemical Physics
Page 18 of 20
View Article Online
DOI: 10.1039/C6CP06897A
mechanism, the present experimental conclusions indicates that the energy gap ꢂETTA between
the singlet state and double of the triplet state, ꢂETTA=2·E(T )–E(S ) is another important role. In
1
1
addition, as the energy differences, ꢂE = 2×T – T , are apparently increased from near zero to
1
2
0.011 eV in DMSO, the TTA yield of singlets is naturally reduced to result in a low TTA yield.
In summary, an obvious solvent effect has been observed in the TTA upconversion system
of I ꢀBodipy and perylene under photoexcitation at 532 nm. Although toluene is the most
2
popular solvent in the similar TTA system, it is not the best in the present system from the view
of TTA upconversion efficiency. Therefore, to choose a more suitable solvent is also able to
efficiently improve upconversion quantum yield, and solvent effect is another important factor
although it has been ignored before.
ACKNOWLEDGMENTS
This work was supported by the National Key Basic Research Foundation (Grant
Nos.2013CB834602 and 2013CB922202) and the National Natural Science Foundation of China
(Grant Nos. 21373194, 21573210 and 11674355). The quantum chemical calculations in this
study had been done on the supercomputing system in the Supercomputing Center of University
of Science and Technology of China. A few experimental instruments were supported by the
Ministry of Science and Technology of China (No. 2012YQ220113) as well.
REFERENCES
1
2
T. A. Joyce, Analyst, 1965, 90, 1–8.
B. C. O’Regan, K. Walley, M. Juozapavicius, A. Anderson, F. Matar, T. Ghaddar, S. M. Zakeeruddin, C.
Klein and J. R. Durrant, J. Am. Chem. Soc., 2009, 131, 3541–3548.
T. Lazarides, T. M. McCormick, K. C. Wilson, S. Lee, D. W. McCamant and R. Eisenberg, J. Am. Chem.
Soc., 2011, 133, 350–364.
3
4
5
D. P. Hari, T. Hering and B. König, Org. Lett., 2012, 14, 5334–5337.
A. Gorman, J. Killoran, C. O’Shea, T. Kenna, W. M. Gallagher and D. F. O’Shea, J. Am. Chem. Soc., 2004,
1
26, 10619–10631.
6
7
S. Duman, Y. Cakmak, S. Kolemen, E. U. Akkaya and Y. Dede, J. Org. Chem., 2012, 77, 4516–4527.
M. A. Baldo, C. Adachi and S. R. Forrest, Phys. Rev. B - Condens. Matter Mater. Phys., 2000, 62, 10967–
1
0977.
8
9
P. Zhang, C. Shao, Z. Zhang, M. Zhang, J. Mu, Z. Guo and Y. Liu, Nanoscale, 2011, 3, 3357–3363.
M. Haase and H. Schäfer, Angew. Chemie - Int. Ed., 2011, 50, 5808–5829.
1
8

Page 19 of 20
Physical Chemistry Chemical Physics
View Article Online
DOI: 10.1039/C6CP06897A
1
1
0
1
V. Gray, D. Dzebo, M. Abrahamsson, B. Albinsson and K. MothꢀPoulsen, Phys. Chem. Chem. Phys., 2014,
6, 10345–10352.
1
A. Monguzzi, J. Mézyk, F. Scotognella, R. Tubino and F. Meinardi, Phys. Rev. B - Condens. Matter Mater.
Phys., 2008, 78, 2–6.
1
1
2
3
T. N. SinghꢀRachford and F. N. Castellano, Coord. Chem. Rev., 2010, 254, 2560–2573.
V. Gray, K. Börjesson, D. Dzebo, M. Abrahamsson, B. Albinsson and K. MothꢀPoulsen, J. Phys. Chem. C,
2
016, 120, 19018–19026.
1
1
1
1
1
1
2
4
5
6
7
8
9
0
P. Ceroni, Chem. - A Eur. J., 2011, 17, 9560–9564.
D. V Kozlov and F. N. Castellano, Chem. Commun., 2004, 1, 2860–2861.
W. Wu, S. Ji, W. Wu, J. Shao, H. Guo, T. D. James and J. Zhao, Chem. - A Eur. J., 2012, 18, 4953–4964.
T. N. SinghꢀRachford and F. N. Castellano, J. Phys. Chem. A, 2008, 112, 3550–3556.
W. Zhao and F. N. Castellano, J. Phys. Chem. A, 2006, 110, 11440–11445.
T. N. SinghꢀRachford and F. N. Castellano, J. Phys. Chem. A, 2009, 113, 5912–5917.
H. C. Chen, C. Y. Hung, K. H. Wang, H. L. Chen, W. S. Fann, F. C. Chien, P. Chen, T. J. Chow, C. P. Hsu
and S.ꢀS. Sun, Chem. Commun., 2009, 4064–4066.
2
1
Y. Y. Cheng, B. Fückel, T. Khoury, R. G. C. R. Clady, N. J. EkinsꢀDaukes, M. J. Crossley and T. W.
Schmidt, J. Phys. Chem. A, 2011, 115, 1047–1053.
2
2
2
2
2
3
4
5
T. Yogo, Y. Urano, Y. Ishitsuka, F. Maniwa and T. Nagano, J. Am. Chem. Soc., 2005, 127, 12162–12163.
G. Ulrich, R. Ziessel and A. Harriman, Angew. Chemie - Int. Ed., 2008, 47, 1184–1201.
A. Loudet and K. Burgess, Chem. Rev., 2007, 107, 4891–4932.
O. Altan Bozdemir, S. ErbasꢀCakmak, O. O. Ekiz, A. Dana and E. U. Akkaya, Angew. Chemie - Int. Ed.,
2
011, 50, 10907–10912.
2
2
6
7
K. Xu, Y. Xie, X. Cui, J. Zhao and K. D. Glusac, J. Phys. Chem. B, 2015, 119, 4175–4187.
C. Zhang, J. Zhao, S. Wu, Z. Wang, W. Wu, J. Ma, S. Guo and L. Huang, J. Am. Chem. Soc., 2013, 135,
1
0566–10578.
2
2
3
3
8
9
0
1
J. Zhao, W. Wu, J. Sun and S. Guo, Chem. Soc. Rev., 2013, 42, 5323–51.
C. Zhang and J. Zhao, J. Mater. Chem. C, 2016, 4, 1623–1632.
J. Zhao, S. Ji and H. Guo, RSC Adv., 2011, 1, 937–950.
H. Imahori, M. E. Elꢀkhouly, M. Fujitsuka, O. Ito, Y. Sakata and S. Fukuzumi, J. Phys. Chem. A, 2001, 105，
3
25–332.
W. Qin, M. Baruah, M. Van Der Auweraer, F. C. De Schryver and N. Boens, J. Phys. Chem. A, 2005, 109,
371–7384.
3
2
7
3
3
3
4
F. Santoro, V. Barone, T. Gustavsson and R. Improta, J. Am. Chem. Soc., 2006, 128, 16312–16322.
T. Asahi, M. Ohkohchi, R. Matsusaka, N. Mataga, R. P. Zhang, A. Osuka and K. Maruyama,
J.Am.Chem.Soc., 1993, 115, 5665–5674.
3
3
5
6
H. Guo, Q. Li, L. Ma and J. Zhao, J. Mater. Chem., 2012, 22, 15757.
S. Vyas and P. K. Sharma, J. Chem. Sci., 2002, 114, 137–148.
1
9

Physical Chemistry Chemical Physics
Page 20 of 20
View Article Online
DOI: 10.1039/C6CP06897A
3
3
3
4
7
8
9
0
S. Feskov, V. Gladkikh and A. I. Burshtein, Chem. Phys. Lett., 2007, 447, 162–167.
O. Holtomo, M. Nsangou, J. J. ules Fifen and O. Motapon, J. Mol. Model., 2014, 20, 2509.
C. Fehling and G. Friedrichs, J. Am. Chem. Soc., 2011, 133, 17912–17922.
K. Yokoyama, Y. Wakikawa, T. Miura, J.I. Fujimori, F. Ito and T. Ikoma, J. Phys. Chem. B, 2015, 119,
5901–15908.
1
4
4
4
4
4
1
2
3
4
5
Y. Wang, S. Zhang, S. M. Sun, K. Liu and B. Zhang, Chinese J. Chem. Phys., 2013, 26, 651–655.
X. Liu, L. Chen, Q. Zhou, X. Zhou and S. Liu, J. Photochem. Photobiol. A Chem., 2013, 269, 42–48.
L. Chen, Q. Zhou, X. Liu, X. Zhou and S. Liu, Chinese J. Chem. Phys., 2015, 28, 493–500.
A. Yu, K. Lin, X. Zhou, H. Wang, S. Liu and X. Ma, J. Phys. Chem. C, 2007, 111, 8971–8978.
L. Chen, W. Zhu, K. Lin, N. Hu, Y. Yu, X. Zhou, L. F. Yuan, S. M. Hu and Y. Luo, J. Phys. Chem. A, 2015,
1
19, 3209–3217.
4
4
6
7
M. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani,
V. Barone, B. Mennucci and G. A. Petersson, Gaussian 09, Gaussian. Inc., Wallingford, CT, 2009.
Y. Y. Cheng, T. Khoury, R. G. C. R. Clady, M. J. Y. Tayebjee, N. J. EkinsꢀDaukes, M. J. Crossley and T.
W. Schmidt, Phys. Chem. Chem. Phys., 2010, 12, 66–71.
4
4
8
9
W. Wu, H. Guo, W. Wu, S. Ji and J. Zhao, J. Org. Chem., 2011, 76, 7056–7064.
V. Gray, D. Dzebo, A. Lundin, J. Alborzpour, M. Abrahamsson, B. Albinsson and K. M. Poulsen, J. Mater.
Chem. C, 2015, 3, 11111ꢀ11121.
50
T. W. Schmidt and F. N. Castellano, J. Phys. Chem. Lett., 2014, 5, 4062–4072.
2
0