Optical properties and photo-oxidation of tetraphenylethene-based fluorophores

Article abstract of DOI:10.1002/chem.201202715

We report the optical properties of tetraphenylethene (TPE) and other TPE derivatives functionalised with an octyl group (TPE-OCT) and polyethyleneglycol group (TPE-PEG) in the side chain. We compared TPE-OCT and TPE-PEG with TPE in both organic solvents and under aqueous conditions. All materials exhibit aggregation-induced emission, however, uncommonly, TPE-PEG seems to aggregate in aqueous solution with enhanced photoluminescence quantum efficiency (PLQE) relative to that in organic solvents. All three materials can be photo-oxidised in solution to their diphenylphenanthrene derivative by irradiation with UV light (at both ≈1 and ≈5 mW cm^-2), with a subsequent enhancement in PL efficiency. The electron-donating ether group increases the rate of oxidation relative to bare TPE and also photo-oxidation was shown to be solvent and concentration dependent. Finally, photo-oxidation was also demonstrated in the aggregate state.

Full text of DOI:10.1002/chem.201202715

FULL PAPER
DOI: 10.1002/chem.201202715
Optical Properties and Photo-Oxidation of Tetraphenylethene-Based
Fluorophores
Matthew P. Aldred, Chong Li, and Ming-Qiang Zhu*[a]
Abstract: We report the optical proper-
ties of tetraphenylethene (TPE) and
other TPE derivatives functionalised
with an octyl group (TPE-OCT) and
polyethyleneglycol group (TPE-PEG)
in the side chain. We compared TPE-
OCT and TPE-PEG with TPE in both
organic solvents and under aqueous
conditions. All materials exhibit aggre-
gation-induced emission, however, un-
commonly, TPE-PEG seems to aggre-
gate in aqueous solution with enhanced
photoluminescence quantum efficiency
(PLQE) relative to that in organic sol-
vents. All three materials can be
photo-oxidised in solution to their di-
phenylphenanthrene derivative by irra-
diation with UV light (at both ꢀ1 and
ꢀ5 mWcmÀ2), with a subsequent en-
hancement in PL efficiency. The elec-
tron-donating ether group increases the
rate of oxidation relative to bare TPE
and also photo-oxidation was shown to
be solvent and concentration depend-
ent. Finally, photo-oxidation was also
demonstrated in the aggregate state.
Keywords: aggregation
·
biosen-
sors · luminescence · optical phys-
ics · oxidation
Introduction
toluminescence quantum efficiencies (PLQE) can be used.
Although the term “aggregation” implies that the mecha-
nism for emission enhancement is intermolecular, in fact the
AIE of TPE and TPE-based molecules is due to intramolec-
ular effects.[4] It is no surprise that, owing to their high
PLQE in the solid state, TPE-based materials have found
great promise in organic light-emitting diodes (OLEDs)[5]
like the analogous phenylenevinylene-based fluorophores.[6]
Also, another area in which TPE has found increased atten-
tion over recent years and has been intensively researched is
sensors, that is, bio-/chemosensors[7] and ion sensors.[8] The
experiments performed in biosensing research are typically
in aqueous solutions, in which the TPE molecule has been
derivatised to enable dissolution in water. The environmen-
tal surroundings of TPE are crucial in determining its fluo-
rescence behaviour.
The photophysics of TPE have been investigated previ-
ously,[9] however, despite its newly discovered AIE proper-
ties and recent rise in popularity, the susceptibility of TPE-
based materials to photo-oxidation has received little atten-
tion.[10] TPE and cis-stilbene are similar in terms of structure
and photochemistry, however, the photochromism[11] and
photo-oxidation[12] of stilbene and stilbene-based materials
have been widely researched. Stilbene undergoes trans–cis
isomerisation following irradiation with UV light, in which
subsequent photocyclisation of cis-stilbene forms trans-
4a,4b-dihydrophenanthrene that is unstable and, if not trap-
ped, reverts back to cis-stilbene. However, if trapped by
means of oxidation, dihydrophenanthrene can be oxidised
to yield phenanthrene, which is irreversible and detrimental
to the photochromism of cis-stilbene (Scheme 1). Owing to
the intense research of TPE in recent years and a subse-
quent plethora of published papers, we thought it worthy to
further investigate its photostability. Therefore, herein we
Mainly as a result of the pioneering research by Tangꢀs
group, in recent years tetraphenylethene (TPE) has become
the focus of much attention, primarily due to its aggrega-
tion-induced emission (AIE) properties.[1] Although there
have been a number of aryl–ethene-based molecules report-
ed to exhibit AIE properties,[2] TPE-based materials have
received the most attention[3] owing to their well-defined
AIE properties, high solid-state fluorescence quantum yields
and ease of synthesis. TPE and TPE-based materials are
highly emissive in the “aggregate” state compared with their
weak emission in solution. The aggregate state has been
shown to form in the solid state (thin film or powder), nano-
particle state and in aqueous solutions upon binding with
“guests” such as biomolecules or ions. Therefore, the AIE
phenomenon can be investigated by comparing the fluores-
cence changes in solution and the aggregate state. The rep-
recipitation method using a solvent and anti-solvent, in
which nanoparticles are formed at low fluorophore concen-
tration and high water content, has been widely used in the
literature. Here, the addition of an anti-solvent (most com-
monly water) at a critical concentration causes nano-precipi-
tation that enhances the photoluminescence intensity sharp-
ly owing to restriction of internal motions. In addition, a
comparison of the solution and nanoparticle/solid-state pho-
[a] Dr. M. P. Aldred, C. Li, Prof. M.-Q. Zhu
Wuhan National Laboratory of Optoelectronics (WNLO)
Huazhong University of Science and Technology (HUST)
Wuhan, Hubei 430074 (P.R. China)
Fax : (+86)27-87793419
E-mail: mqzhu@hust.edu.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201202715.
Chem. Eur. J. 2012, 18, 16037 – 16045
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
16037

Scheme 1. Photocyclisation and oxidation of bis-stilbene.
report some optical properties and photo-oxidation experi-
ments of TPE. We have found that TPE and analogous
ether-functionalised TPE compounds are dramatically
photo-oxidised in organic solvents, aqueous solutions and in
the nanoparticle state. This photo-oxidation should provide
an incentive for improving the photostability of TPE-based
materials, especially in the areas of sensory research, and
encourage the design and synthesis of novel TPE structures
with improved photostability.
Table 1. Optical properties of TPE, DPP, TPE-OCT and TPE-PEG.
labs [nm]
lem [nm]
FF [%]
solution aggregate solid solution[a] aggregate[b]
TPE
DPP
238, 308
258, 301, 355, 373,
n.d.[c]
468
474
448
474
0.5
13
1.2
5.3[b]
351
317
392
TPE-
OCT
TPE-
PEG
n.d.[c]
484
485
474
492
0.4
0.3
1.7
1.5
316
n.d.[c]
[a] FF estimated by the optical dilution method. The standard is anthra-
cene (0.27 in ethanol) unless stated otherwise. [b] The standard is quinine
sulfate (0.54 in 0.1m H2SO4). PL in solution using THF as the solvent
(lex =310); TPE and DPP in the solid state (lex =310); TPE-OCT and
TPE-PEG in the solid state (lex =330); aggregate is measured in a 99%
water/THF solvent mixture. The solid-state emission was recorded from
the powder or oil (TPE-PEG). [c] n.d. =not determined.
Results and Discussion
Steady-state absorption and emission: The TPE derivatives
were synthesised using standard synthetic procedures, ac-
cording to Scheme 2.
periments, in which a low concentration of fluorophore in
THF (10À5 m) is added to water with a final water/THF (v/v)
ratio of 99:1. After vigorous agitation of the solution, the
nanoparticle solution appears transparent and homogeneous
with no sign of precipitation. Owing to nanoprecipitation,
the molecules aggregate enough to restrict internal motions
and consequently the PL intensity (Figure 1) and PLQY
values are enhanced. TPE exhibits the highest fluorescence
efficiency (13%) of the fluorophores. The low PLQE values
for TPE-OCT (1.7%) and TPE-PEG (1.5%) reflect the det-
rimental effect of the long alkyl side chains, which probably
decrease the aggregation between the aromatic cores and
consequently the internal motions are not restricted effi-
ciently. 9,10-Diphenylphenanthrene (DPP) shows classical
aggregation-caused quenching (ACQ) behaviour and the
PLQE values in THF (5.3%) are higher than those in the
nanoparticle state (1.2%). However, the absolute PLQE
values are expected to be much higher, owing to the over es-
timation of the UV absorbance caused by the light scattering
of the nanoparticles.[13] The AIE phenomenon that exists in
TPE-based materials is due to the reduction of torsional and
rotational motions and out-of-plane bending vibrations;
these internal motions around the ethylenic bond and the
bonds connecting the phenyl rings to the ethylenic double
bond have both been shown to contribute significantly.[8c,f]
In fact, the effect of internal motions on the fluorescence
properties of TPE-based materials, such as TPE and stil-
bene, has been known for a long time.[8] However, taking
Scheme 2. Synthetic routes to TPE and its derivatives TPE-OCT, TPE-
PEG and DPP.
The emission from solutions of the TPE materials in THF
was not detectable owing to considerable quenching. From
the PLQE data shown in Table 1, it is clear that all the TPE
derivatives—including TPE functionalised with an octyl
group (TPE-OCT) or a polyethyleneglycol group (TPE-
PEG) in the side chain—exhibit AIE behaviour. This can be
further confirmed by solvent (THF)–anti solvent (water) ex-
16038
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16037 – 16045

Tetraphenylethene-Based Fluorophores
FULL PAPER
materials and the fluorophores here exhibit broad emission
spectra that are considerably redshifted in the solid state.
From these characteristics at first glance one would suspect
excimer formation, however, the fluorescence efficiency is
higher than that in the solution state, which is opposite to
excimer formation, and in most cases drastically quenches
the fluorescence. Also, compared to the solid state there is
negligible change in the emission band structure and wave-
length after dispersing TPE in polymethylmethacrylate
(PMMA) at low to medium concentrations (Figure 2), which
Figure 1. THF/water solvent mixture experiments to investigate the emis-
sion properties of the fluorophores in the “isolated” and “aggregated”
states. Excitation=310 nm, concentration=10À5 m.
advantage of the extremes in fluorescence properties, that is,
quenching in the “monomeric” state and emission enhance-
ment in the aggregate state, within applied science for useful
applications, has only recently been demonstrated.[14] Al-
though TPE-OCT and TPE-PEG differ by their insoluble
and soluble nature in water, respectively, interestingly their
emission profiles in 99% water–THF are similar, which indi-
cates a degree of “aggregation” of TPE-PEG in water owing
to phase segregation of the hydrophilic PEG side chains and
the hydrophobic TPE core. TPE-PEG in aqueous solution
also exhibits a long-wavelength emission peak that is similar
to its solid state and to TPE and TPE-OCT in the nanoparti-
cle state. The PLmax in the solid state is very similar to that
in the nanoparticle.
From previous studies we know that the TPE group in
solution exhibits unrestricted internal bond motions result-
ing in poor fluorescence efficiencies. When TPE is connect-
ed to other aromatic cores it generally has a strong effect
and consequently most reported TPE-based materials, even
with complex structural architectures, still exhibit low fluo-
rescence quantum efficiencies in solution.[4,15] This is in
sharp contrast to the “aggregate” or solid state, where the
fluorescence efficiencies are dramatically enhanced due to
restriction of intramolecular rotation (RIR). Therefore,
TPE-based materials in the “aggregate” and solid state are
usually efficient light emitters. The optical properties of
TPE-based materials show large Stokes shifts in the solid
state compared with small Stokes shifts in the solution state.
This has previously been explained by the extremes in the
structural conformations of the TPE molecule in solution
and in the solid state.[4] Accordingly, there is a big difference
in the solution and solid-state emission peak wavelengths.
The TPE-based fluorophores reported here also follow
these same characteristics. Most of the published TPE-based
Figure 2. PL spectra of pure TPE and TPE-PMMA at different wt% con-
centrations of TPE.
suggests that intermolecular affects are not governing the
emission properties to a considerable extent. The blueshift
in emission wavelength of TPE in the solid state relative to
TPE dispersed in a PMMA matrix is probably due to the
morphology, in which TPE in the solid state is crystalline
but has an amorphous morphology when dispersed in
PMMA. In this respect, TPE in the amorphous state exhib-
its a more planar conformation than the more twisted con-
formation in the crystalline state. This blueshift has been re-
ported by other research groups.[5d] To further confirm this
difference in emission of the amorphous and crystalline
states we prepared thin films of 80 wt% TPE in PMMA on
a quartz substrate and compared the emission spectra
before and after thermal treatment (Figure 3).
Before thermal treatment the film can be regarded as
amorphous owing to its high transparency and glassy tex-
ture, which is facilitated by the small amount of PMMA.
After thermal treatment of this same film by heating just
below the glass transition temperature (Tg) of PMMA, at
approximately 1008C, the resulting film crystallises and ex-
hibits an opaque texture. The emission of the resulting
amorphous state is 480 nm and is redshifted by 32 nm rela-
tive to the crystalline state (448 nm). In contrast, the PL
emission of DPP remains unchanged in the solid state rela-
tive to when it is dispersed in a PMMA matrix, which sug-
gests that there is negligible change in structural conforma-
tions of DPP in both the amorphous and crystalline states.
Chem. Eur. J. 2012, 18, 16037 – 16045
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16039

M. P. Aldred, C. Li, M.-Q. Zhu
termolecular effects. Excimers in DPP-type materials have
been demonstrated before.[16]
Photo-oxidation behaviour (solution and nanoparticle state):
During PL investigations of TPE we accidently observed
that well-defined vibronic peaks in the range 350–400 nm in-
creased upon repeated measurement. These peaks are remi-
niscent of phenanthrene and led us to the conclusion that
photocylisation and photo-oxidation were taking place
under our experimental conditions. This sensitivity to UV
light encouraged us to further investigate the photo-oxida-
tion of TPE-based fluorophores upon UV-light exposure.
Oxidation of stilbene-type molecules is well known and has
been used extensively with respect to photochemistry and as
a synthetic tool. The latter was made possible by using
iodine as the catalyst, in which large-scale oxidations can be
carried out, and is most commonly known as the Mallory re-
action (hv and I2)[17] or the Katz-modified Mallory reaction
(using methyloxirane as the scavenger).[18]
Figure 3. PL spectra of TPE in the pure form and in a PMMA matrix
(80 wt%) before and after thermal treatment. Also DPP in pure form
and in a PMMA matrix (80 wt%).
By using our fluorescence spectrofluorometer, photo-oxi-
dation kinetic measurements of TPE, TPE-OCT and TPE-
PEG in different solvents were performed (Figure 5A),
whereby the emission at a chosen wavelength was followed
over a period of time at fixed excitation and emission wave-
lengths. Typically, the excitation wavelength was the maxi-
mum absorption wavelength (labs), the emission wavelength
was the maximum emission wavelength (lem) and the experi-
ments were performed under ambient conditions using
dry/purified solvents. Although the photo-oxidation tenden-
cy can be curbed by using lower-energy excitation wave-
lengths (i.e., away from its maximum absorbance), this is of
course at the expense of its maximum potential brightness,
therefore, the excitation wavelength at labs was used. The
power of the light source was approximately 1 mWcmÀ2.
After 2 h of irradiation the emission spectra of TPE and
TPE-OCT in hexane and THF exhibit defined emission
bands, which is reminiscent of phenanthrene (Figure 5B–E).
The emission spectrum of TPE in hexane is noticeably dif-
ferent from that in THF, with a broad emission peak at
540 nm, which is more redshifted than that in the solid state.
Similar emission characteristics of oligo-fluorenes end-
capped with TPE in hexane have also been demonstrated
previously.[4] This unusual emission behaviour could be due
to excimer formation in solution, however, this seems un-
likely. Firstly, because the oligo-fluorene end-capped TPE
compounds exhibit structural features that would essentially
prevent excimers owing to the long alkyl chains at the fluo-
rene 9-position. Secondly, TPE is non-planar, which should
prevent the close intermolecular packing necessary for exci-
mer formation. Significant spectral shifts in emission at dif-
ferent temperatures in 3-methylpentane have been attribut-
ed to different structural geometries of TPE.[9c] Therefore,
the most likely explanation is due to different geometrical
configurations of the TPE core in hexane, which seems to
be a more planar configuration. The PL spectra of TPE-
OCT and TPE-PEG in 99% water/THF (Figure 5F,G) are
different because the molecules in the initial state are “ag-
An interesting comparison is the emission properties of
DPP. Here, two of the phenyl rings are “locked”, which re-
duces the internal bond motions and as a result should in-
crease the PLQE in solution. Indeed, the PLQE is enhanced
more than tenfold from 0.5 (TPE) to 5.3% (DPP) and upon
addition of water the PL intensity is reduced due to ACQ.
A comparison of the “monomeric” and “aggregate” PL
spectra of TPE and DPP (Figure 1) clearly illustrates both
AIE and ACQ phenomena in analogous materials differing
by reducing its internal motions. The low-energy emission
band at 474 nm was further investigated to understand its
origin. Figure 4 shows a comparison of the PL spectra of
DPP in solution (THF), solid state and in a PMMA matrix
at 9 wt%. In contrast to TPE, it is clear that the low-energy
emission band of DPP in the solid state is largely due to in-
Figure 4. A comparison of the PL spectra of DPP in solution (THF), a
PMMA matrix at 9 wt% and the solid state.
16040
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16037 – 16045

Tetraphenylethene-Based Fluorophores
FULL PAPER
Figure 5. A) The kinetic scans of TPE, TPE-OCT and TPE-PEG in different solvents at a concentration of 10À5 m, during 2 h of UV-light irradiation. PL
spectra of TPE in B) hexane, C) THF; of TPE-OCT in D) hexane, E) THF; of TPE-OCT in F) 99% water/THF; and of TPE-PEG in G) 99%
water/THF under repeated UV-light exposure (UV max, ꢀ1 mWcmÀ2). Concentration=10À5 m.
gregated”, compared to the “monomeric” emission in THF
and hexane. Therefore, the broad emission peak around
470 nm at 0 s decreases as the irradiation time elapses owing
to the depletion of TPE-OCT and TPE-PEG and perhaps
the resulting lower PL efficiencies of the oxidised TPE-OCT
and TPE-PEG aggregate. Unlike DPP, the oxidation forms
of TPE-OCT/TPE-PEG seem to form fewer excimeric spe-
cies than DPP in the aggregate state (see below). The slight
increase of the defined emission peaks around 368 and
387 nm is probably due to the emission from the oxidation
product (phenanthrene). The kinetic scans demonstrate
three features about the rate of oxidation: 1) TPE-OCT>
TPE, 2) THF>hexane and 3) solution>“aggregate”/aque-
ous solution. In all cases the photo-oxidation during the 2 h
irradiation was incomplete owing to the low power
(ꢀ1 mWcmÀ2) of the light source. Due to the incomplete
photo-oxidation, especially in the aggregate state, we further
carried out some photo-oxidation tests (Figure 6) for TPE,
TPE-OCT and TPE-PEG at 99% water/THF with higher
power UV-light irradiation (ꢀ5 mWcmÀ2).
During the photo-oxidation of TPE, the aggregate emis-
sion at 468 nm decreases and after 20 min is undetectable.
This is due to the depletion of TPE and the subsequent
lower fluorescence efficiency of DPP in the aggregate state,
which is most likely an excimer. In comparison, the broad
emission peak at 484 nm for TPE-OCT during photo-oxida-
tion also rapidly decreases, but with a slight increase of the
defined peaks at 368 and 387 nm. These defined peaks are
most likely due to the phenanthrene core from TPE-OCT
oxidation. The absence of these peaks during TPE oxidation
(i.e., for DPP) suggests that excimer formation in oxidised
TPE-OCT is much less than in TPE, because of the long
octyl side chain that prevents sufficient aggregation that is
needed for excimer formation. TPE-PEG behaves similar in
that the broad emission peak around 485 nm also decreases
upon UV-light irradiation, however, the emission peaks at
Chem. Eur. J. 2012, 18, 16037 – 16045
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16041

M. P. Aldred, C. Li, M.-Q. Zhu
Figure 7. Gas chromatograms of TPE in THF and hexane after 4 h UV-
light exposure at 302 nm (ꢀ5 mWcmÀ2). For reference, GC traces for
pure TPE and DPP are also shown.
Figure 6. PL spectra of A) TPE, B) TPE-OCT and C) TPE-PEG in 99%
water/THF (10À5 m) with UV irradiation=302 nm (ꢀ5 mWcmÀ2), excita-
tion=310 nm; and the corresponding photos excited over 302 illumina-
tion.
368 and 386 nm show a large increase upon further light ex-
posure. This suggests that its oxidised form exhibits higher
fluorescence efficiency than the oxidised forms of TPE and
TPE-OCT owing to the even longer PEG side chain, which
prevents significant aggregation and perhaps because of its
good solubility in aqueous media. One important feature of
the TPE and TPE-OCT photo-oxidation is the transforma-
tion of materials that are AIE active into materials that ex-
hibit ACQ. In all cases, over-oxidation occurs with a notice-
able change in the emission spectra and the manifestation of
an undefined peak at 424, 417 and 420 nm, for TPE, TPE-
OCT and TPE-PEG, respectively. This over-oxidation does
not occur for TPE in THF or hexane (see below), which sug-
gests that it is accelerated in aqueous media. The low con-
centration of this species made it difficult to investigate its
identity, however, it is most likely diphenyldibenzofulvene
and/or dibenzoACHTUNGTRENNUNG[g,p]chrysenes derivatives.
Figure 8. Typical ESI mass spectra from GC-MS analysis of A) TPE in
hexane or THF after 4 h UV-light exposure ([M]+ =332) and B) DPP in
hexane or THF after 4 h UV-light exposure ([M]+ =330).
We used GC-MS analysis to monitor the conversion of
TPE to DPP in hexane and THF at concentrations of 10À3
and 10À4
m after 4 h of UV-light irradiation (302 nm,
ꢀ5 mWcmÀ2). Unfortunately, at 10À5 m the detection of the
species under investigation was limited. The GC traces for
the synthesised materials TPE and DPP were also obtained
for reference purposes (Figure 7). From the corresponding
ESI mass spectra (Figure 8), the peak at approximately
8.07 min in the GC-MS scan was found to represent TPE
and that at approximately 10.25 min was found to represent
DPP. GC-MS analysis of TPE in hexane (10À4 m) after 4 h of
UV-light irradiation and subsequent photo-oxidation shows
that the TPE/DPP ratio is 67:33. The rate of photo-oxida-
tion is faster in THF than in hexane, and consequently when
TPE is dissolved in THF (10À4 m) only DPP can be detected,
which indicates near-complete photo-oxidation. For a con-
centration of 10À3 m, conversion to DPP is less than at a
lower concentration, however, photo-oxidation conversion is
still higher in THF (TPE: 56%, DPP: 44%) than in hexane
(TPE: 78%, DPP: 22%). HPLC was also used to investi-
gate the photo-oxidation of TPE-OCT in hexane after UV-
light irradiation (4 h) and only one major peak was ob-
served. It corresponded to 96% 3-(octyloxy)-9,10-diphenyl-
phenanthrene (see below), which indicates almost complete
photo-oxidation.
Owing to the observed solvent dependence we further in-
vestigated the changes in UV-visible absorbance and PL
16042
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16037 – 16045

Tetraphenylethene-Based Fluorophores
FULL PAPER
Figure 9. PL spectra of TPE in THF at A) 10À4 and B) 10À5 m, and C) the corresponding UV-visible spectra for 10À5 m; TPE in hexane at D) 10À4 and
E) 10À5 m, and F) the corresponding UV-visible spectra for 10À5 m. UV-light exposure=1 h, irradiation at 302 nm, approximately 5 mWcmÀ2, lex =310 nm.
emission of TPE in hexane and THF at both 10À4 and 10À5
m
tensity keeps rising (not shown in Figure 9D). Therefore,
from the PL spectra we can safely say that the photo-oxida-
tion is concentration and solvent dependent: a low concen-
tration and using THF as the solvent favours a fast rate of
photo-oxidation. The APCI mass spectra before and after
UV-light irradiation of TPE-OCT and TPE-PEG in hexane
and THF, respectively, are shown in Figure S1 of the Sup-
porting Information. This confirms that the photo-oxidation
products are phenanthrene-based. The mass spectra of TPE-
PEG shows a Gaussian distribution of molecular ions be-
cause of the polydispersity of TPE-PEG owing to the differ-
ent number of ethylene glycol units (CH2CH2O)n in the side
chain, with an average number of n=12. From cyclic-vol-
tammetry measurements the oxidation potentials of TPE-
OCT and TPE were calculated as 0.78 and 0.89 V, respec-
tively. The lower oxidation potential of TPE-OCT than that
of TPE is due to the positive inductive effect (+I) of the
ether group. A similar observation was made by Rathoreꢀs
group regarding the oxidation of tetraphenylethene deriva-
tives using DDQ/MeSO3H, in which they reported longer
reaction times for the derivatives with higher oxidation po-
tentials.[19] Most water-soluble TPE-based materials used in
biosensor applications are functionalised with +I functional
groups.
concentrations (5 mWcmÀ2). The subsequent UV-visible and
PL spectra during UV-light irradiation are shown in
Figure 9. Unfortunately, we were unable to obtain the UV-
visible time profiles at a concentration of 10À4 m owing to
the high concentration. With regards to TPE in THF at
10À5 m (Figure 9B), the photo-oxidation seems to be near
complete at only 5 min irradiation, with 5 min showing the
sharpest increase in PL with only very slight changes after
this time. At 5 min, the UV-visible spectra (Figure 9C) blue-
shifts and is identical to the pure synthesised DPP, which un-
doubtedly proves that DPP is the photo-oxidation product.
After 1 h there is negligible change in the UV-visible and
PL spectra. In contrast, at higher concentration (10À4 m) the
PL increases gradually up to 1.5 h, again with negligible
changes in PL after this time. Therefore, similar to our GC-
MS experiments photo-oxidation seems to be more efficient
at lower concentration (10À5 m) than at high concentration
(10À4 m).
The rates of oxidation in hexane (Figure 9D–F) are com-
paratively slower than those in THF at both concentrations.
The UV-visible spectra at 10À5 m (Figure 9F) show only a
slow gradual change during UV irradiation, which seems to
be near complete after 30 min because after this time there
is very little enhancement in the PL intensity. Again, there
is negligible change in PL after 1 h. The corresponding UV-
visible spectra show a gradual blueshift with UV-light irradi-
ation and the UV-visible spectrum at 30 min is nearly identi-
cal to that of DPP. At higher concentration (10À4 m), even
after 1 h the photo-oxidation is not complete and the PL in-
The electron-donating oxygen atom in TPE-OCT lowers
the oxidation potential of TPE-OCT relative to that of TPE,
which therefore results in faster rates of photo-oxidation of
TPE-OCT than of TPE in both hexane and THF (Figure 5).
In a 99% water/THF solvent mixture (Figure 6) we ob-
served more “over-oxidation” in TPE-OCT and TPE-PEG
Chem. Eur. J. 2012, 18, 16037 – 16045
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16043

M. P. Aldred, C. Li, M.-Q. Zhu
compared with only a slight “over-oxidation” in TPE, al-
though the rates of photo-oxidation look similar. With re-
gards to TPE and TPE-OCT, the fluorescence decreases as
shown in the photos (Figure 6) and the decrease of the emis-
sion peak around 450-500 nm shown in the PL spectra.
TPE-PEG was synthesised to compare its photo-oxidation
with TPE-OCT in an aqueous environment. In a 99%
water/THF solvent mixture, TPE-PEG is dissolved (al-
though aggregation still takes place) and TPE-OCT is a
“nano-suspension”. From the PL spectra, we can see that
during UV-light irradiation the broad emission around
475 nm decreases and a more defined (0,0 0,1 and 0,2) blue-
shifted emission around 375 nm increases, which is related
to diphenylphenanthrene. These blueshifted phenanthrene
peaks are strong in TPE-PEG and very weak in TPE-OCT.
Why? The conversion of TPE-OCT to its diphenylphenan-
threne analogue induces ACQ, therefore, in the nanoparticle
the fluorescence is weak. However, although TPE-PEG is
“aggregated” in the water, it is also solvated and, therefore,
the conversion of TPE-PEG to its phenanthrene analogue
does not induce ACQ.
Figure 11. PL spectra of TPE-PMMA at different wt% TPE: A) 1, B) 5
and C) 9 wt% during different stages of UV-light irradiation. D) UV-visi-
ble spectra of 9 wt% before and after approximately 5 mWcmÀ2 irradia-
tion. Further irradiation was conducted for 30 min at 302 nm. All thin
films were prepared on quartz substrates, spin-coated from dichlorome-
thane.
PMMA matrix: The concentration dependence was further
investigated by dispersing TPE at different weight percen-
tages (1, 5 and 9 wt%) in a PMMA solid matrix and moni-
toring the change in PL intensity at 372 nm. The resulting
kinetic scans are shown in Figure 10 and clearly illustrate
From the PL spectrum shown in Figure 4, excimer formation
at 9 wt% DPP in PMMA is negligible.
The photochemistry of tetraphenylethene and stilbene are
very similar and tetraphenylethene photo-oxidation pro-
ceeds in a similar manner to that of stilbene-based materials.
In our case no oxidant was used, only solvents and UV
light, so the oxygen from the air and subsequently in the sol-
vent is enough to allow the oxidation to proceed. The small
concentration of the fluorophore means that only a trace of
oxidant is necessary and the oxygen in the solvent is enough
to trap the trans-4a,4b-dihydrodiphenylphenanthrene and
oxidise it to 9,10-diphenylphenanthrene (Figure 12).
Figure 10. The kinetic scans of TPE-PMMA thin films on quartz sub-
strates at different TPE concentration, during 2 h UV-light irradiation at
excitation at 310 nm.
Figure 12. Photocyclisation and oxidation of tetraphenylethene.
Conclusion
that as the weight percentage of TPE is increased the rate
of oxidation decreases. The corresponding PL spectra and
UV-visible spectra (Figure 11) clearly show the transforma-
tion of TPE into DPP at all three weight percentage concen-
trations.
Owing to the probable excimer formation at high concen-
trations of DPP we did not exceed 9 wt% TPE, which could
compromise the validity of our results and interpretation.
We have investigated the optical properties and photo-oxi-
dation behaviour of TPE and its derivatives in solution and
the “aggregate” state. AIE has been observed for all the
TPE derivatives and the opposite behaviour, namely, ACQ,
has been observed for the oxidised derivative of TPE, that
is, 9,10-diphenylphenanthrene. From mass spectrometry,
UV-visible absorption and photoluminescence we have dem-
16044
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16037 – 16045

Tetraphenylethene-Based Fluorophores
FULL PAPER
X. J. Chen, X. Y. Shen, L. Tong, R. R. Hu, J. Z. Sun, B. Z. Tang,
Chem. Eur. J. 2011, 17, 14736–14740; c) M. Wang, G. X. Zhang,
D. Q. Zhang, D. B. Zhu, B. Z. Tang, J. Mater. Chem. 2010, 20, 1858–
1867; d) Y. N. Hong, H. Xiong, J. W. Y. Lam, M. Haussler, J. Z. Liu,
Y. Yu, Y. C. Zhong, H. H. Y. Sung, I. D. Williams, K. S. Wong, B. Z.
Tang, Chem. Eur. J. 2010, 16, 1232–1245; e) X. G. Gu, G. X. Zhang,
D. Q. Zhang, Analyst 2012, 137, 365–369; f) M. Nakamura, T. Sanji,
M. Tanaka, Chem. Eur. J. 2011, 17, 5344–5349; g) T. Sanji, K. Shir-
aishi, M. Nakamura, M. Tanaka, Chem. Asian J. 2010, 5, 817–824.
[8] a) L. Liu, G. X. Zhang, J. F. Xiang, D. Q. Zhang, D. B. Zhu, Org.
Lett. 2008, 10, 4581–4584; b) X. R. Wang, J. M. Hu, T. Liu, G. Y.
Zhang, S. Y. Liu, J. Mater. Chem. 2012, 22, 8622–8628; c) F. Sun,
G. X. Zhang, D. Q. Zhang, L. Xue, H. Jiang, Org. Lett. 2011, 13,
6378–6381; d) Y. N. Hong, S. J. Chen, C. W. T. Leung, J. W. Y. Lam,
J. Z. Liu, N. W. Tseng, R. T. K. Kwok, Y. Yu, Z. K. Wang, B. Z.
Tang, ACS Appl. Mater. Interfaces 2011, 3, 3411–3418.
[9] a) S. Sharafy, K. A. Muszkat, J. Am. Chem. Soc. 1971, 93, 4119–
4125; b) G. Fischer, G. Seger, K. A. Muszkat, E. Fischer, J. Chem.
Soc. Perkin Trans. 2 1975, 1569–1576; c) P. F. Barbara, S. D. Rand,
P. M. Rentzepis, J. Am. Chem. Soc. 1981, 103, 2156–2162; d) H.
Goerner, J. Phys. Chem. 1982, 86, 2028–2035; e) C. L. Schilling,
E. F. Hilinski, J. Am. Chem. Soc. 1988, 110, 2296–2298; f) D. A.
Shultz, M. A. Fox, J. Am. Chem. Soc. 1989, 111, 6311–6320; g) J.
Morais, J. Ma, M. B. Zimmt, J. Phys. Chem. 1991, 95, 3885–3888;
h) Y. P. Sun, M. A. Fox, J. Am. Chem. Soc. 1993, 115, 747–750.
[10] a) G. X. Huang, B. D. Ma, J. M. Chen, Q. Peng, G. X. Zhang, Q. H.
Fan, D. Q. Zhang, Chem. Eur. J. 2012, 18, 3886–3892; b) C. E.
Bunker, N. B. Hamilton, Y. P. Sun, Anal. Chem. 1993, 65, 3460–
3465.
onstrated that upon UV-light exposure the TPE compounds
oxidise to their diphenylphenanthrene derivatives. We hope
that this study will encourage the design and synthesis of
more photostable TPE derivatives, possibly with electron-
withdrawing groups, thus decreasing their susceptibility to
photo-oxidation.
Acknowledgements
This work was financially supported by the NSFC (20874025, 21174045).
M.P.A is the awardee of a NSFC Research Fellowship for International
Young Scientists (21150110141) and the Key Fellowship of China Post-
doctoral Science Foundation (20100480065). The Open Program for Bei-
jing National Laboratory for Molecular Sciences (BNLMS) and the Fun-
damental
Research
Funds
for
the
Central
Universities
(HUST2010MS101) are also thanked. Finally we thank the Analysis and
Test Center of Huazhong University of Science and Technology for sup-
port.
[1] a) J. D. Luo, Z. L. Xie, J. W. Y. Lam, L. Cheng, H. Y. Chen, C. F.
Qiu, H. S. Kwok, X. W. Zhan, Y. Q. Liu, D. Zhu, B. Z. Tang, Chem.
Commun. 2001, 1740–1741.
[2] a) J. W. Chen, B. Xu, X. Y. Ouyang, B. Z. Tang, Y. Cao, J. Phys.
Chem. A 2004, 108, 7522–7526; b) K. Itami, Y. Ohashi, J.-I. Yoshida,
J. Org. Chem. 2005, 70, 2778–2792; c) K. Itami, J.-I. Yoshida, Chem.
Eur. J. 2006, 12, 3966–3974; d) Q. Zeng, Z. Li, Y. Q. Dong, C. A.
Di, A. J. Qin, Y. N. Hong, L. Ji, Z. C. Zhu, C. K. W. Jim, G. Yu,
Q. Q. Li, Z. A. Li, Y. Q. Liu, J. G. Qin, B. Z. Tang, Chem. Commun.
2007, 70–72.
[3] a) Y. N. Hong, J. W. Y. Lam, B. Z. Tang, Chem. Commun. 2009,
4332; b) Y. N. Hong, J. W. Y. Lam, B. Z. Tang, Chem. Soc. Rev. 2011,
40, 5361–5388.
[4] a) M. P. Aldred, C. Li, G. F. Zhang, W. L. Gong, A. D. Q. Li, Y. Dai,
D. Ma, M. Q. Zhu, J. Mater. Chem. 2012, 22, 7515–7528; G. F.
Zhang, M. P. Aldred, W. L. Gong, C. Li, M. Q. Zhu, Chem.
Commun. 2012, 48, 7711–7713.
[11] a) N. Tamai, H. Miyasaka, Chem. Rev. 2000, 100, 1875–1890; b) J.
Quenneville, T. J. Martinez, J. Phys. Chem. A 2003, 107, 829–837;
c) A. Momotake, T. Arai, J. Photochem. Photobiol. C 2004, 5, 1–25;
d) R. J. Sension, S. T. Repinec, A. Z. Szarka, R. M. Hochstrasser, J.
Chem. Phys. 1993, 98, 6291–6315.
[12] K. B. Jørgensen, Molecules 2010, 15, 4334–4358.
[13] A. J. Qin, L. Tang, J. W. Y. Lam, C. K. W. Jim, Y. Yu, H. Zhao, J. Z.
Sun, B. Z. Tang, Adv. Funct. Mater. 2009, 19, 1891–1900.
[14] H. Tong, Y. N. Hong, Y. Q. Dong, M. Haussler, J. W. Y. Lam, Z. Li,
Z. F. Guo, Z. H. Guo, B. Z. Tang, Chem. Commun. 2006, 3705–
3707.
[15] a) X. Q. Zhang, Z. G. Chi, H. Y. Li, B. J. Xu, X. F. Li, S. W. Liu, Y.
Zhang, J. R. Xu, J. Mater. Chem. 2011, 21, 1788–1796; b) H. Y. Li,
Z. G. Chi, B. J. Xu, X. Q. Zhang, X. F. Li, S. W. Liu, Y. Zhang, J. R.
Xu, J. Mater. Chem. 2011, 21, 3760–3767.
[16] Y. Nakamura, T. Tsuihiji, T. Mita, T. Minowa, S. Tobita, H. Shizuka,
J. Nishimura, J. Am. Chem. Soc. 1996, 118, 1006–1012.
[17] F. B. Mallory, C. S. Wood, J. T. Gordon, J. Am. Chem. Soc. 1964, 86,
3094–3102.
[18] L. Liu, B. Yang, T. J. Katz, M. K. Poindexter, J. Org. Chem. 1991, 56,
3769–3775.
[5] a) W. Z. Yuan, P. Lu, S. M. Chen, J. W. Y. Lam, Z. M. Wang, Y. Liu,
H. S. Kwok, Y. G. Ma, B. Z. Tang, Adv. Mater. 2010, 22, 2159–2163;
b) Y. Liu, S. M. Chen, J. W. Y. Lam, F. Mahtab, H. S. Kwok, B. Z.
Tang, J. Mater. Chem. 2012, 22, 5184–5189; c) C. Y. K. Chan, Z. J.
Zhao, J. W. Y. Lam, J. Z. Liu, S. M. Chen, P. Lu, F. Mahtab, X. J.
Chen, H. H. Y. Sung, H. S. Kwok, Y. G. Ma, I. D. Williams, K. S.
Wong, B. Z. Tang, Adv. Funct. Mater. 2012, 22, 378–389; d) J.
Huang, N. Sun, J. Yang, R. L. Tang, Q. Q. Li, D. G. Ma, J. G. Qin, Z.
Li, J. Mater. Chem. 2012, 22, 12001–12007; e) J. Huang, X. Yang,
J. Y. Wang, C. Zhong, L. Wang, J. G. Qin, Z. Li, J. Mater. Chem.
2012, 22, 2478–2484.
[19] T. S. Navale, K. Thakur, R. Rathore, Org. Lett. 2011, 13, 1634–1637.
[6] J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks, K.
Mackay, R. H. Friend, P. L. Burns, A. B. Holmes, Nature 1990, 347,
539–541.
[7] a) Y. Liu, C. M. Deng, L. Tang, A. J. Qin, R. R. Hu, J. Z. Sun, B. Z.
Tang, J. Am. Chem. Soc. 2011, 133, 660–663; b) Y. Liu, A. J. Qin,
Received: July 29, 2012
Revised: September 4, 2012
Published online: October 18, 2012
Chem. Eur. J. 2012, 18, 16037 – 16045
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16045