Trans/cis-isomerization of fluorene-bridged azo chromophore with significant two-photon absorbability at near-infrared wavelength

Article abstract of DOI:10.1002/asia.201402896

Azo-containing materials have been proven to possess second-order nonlinear optical (NLO) properties, but their third-order NLO properties, which involves twophoton absorption (2PA), has rarely been reported. In this study, we demonstrate a significant 2P

Full text of DOI:10.1002/asia.201402896

COMMUNICATION
DOI: 10.1002/asia.201402896
trans/cis-Isomerization of Fluorene-Bridged Azo Chromophore with
Significant Two-Photon Absorbability at Near-Infrared Wavelength
Chih-Chien Chu,*[a, b] Ya-Chi Chang,[a] Bo-Kai Tsai,[c] Tzu-Chau Lin,[c] Ja-Hon Lin,[d] and
Vincent K. S. Hsiao[e]
Abstract: Azo-containing materials have been proven to
possess second-order nonlinear optical (NLO) properties,
but their third-order NLO properties, which involves two-
photon absorption (2PA), has rarely been reported. In this
study, we demonstrate a significant 2PA behavior of the
novel azo chromophore incorporated with bilateral dipheny-
laminofluorenes (DPAFs) as a p framework. The electron-
donating DPAF moieties cause a redshifted p–p* absorption
band centered at 470 nm, thus allowing efficient blue-light-
induced trans-to-cis photoisomerization with a rate constant
of 2.04ꢀ10À1 minÀ1 at the photostationary state (PSS). The
open-aperture Z-scan technique that adopted a femtosecond
(fs) pulse laser as excitation source shows an appreciably
higher 2PA cross-section for the fluorene-derived azo chro-
mophore than that for common azobenzene dyes at near-in-
frared wavelength (lex =800 nm). Furthermore, the fs 2PA
response is quite uniform regardless of the molecular geom-
etry. On the basis of the computational modeling, the intra-
molecular charge-transfer (ICT) process from peripheral di-
phenylamines to the central azo group through a fluorene p
bridge is crucial to this remarkable 2PA behavior.
hibit facile geometric isomerization between their trans and
cis forms.[2] Generally, trans-azobenzene shows intense p–p*
absorption in the UV region, and noncoherent UV light can
rapidly induce the trans-to-cis isomerization. The cis isomer
has an enhanced n–p* absorption in the visible region; thus,
the trans isomers are regenerated from the cis state through
visible-light irradiation. Accordingly, trans/cis azobenzenes
can be readily switched by alternating UV and visible-light
exposure. This light-induced interconversion allows the
system that incorporates azo chromophores to be applied in
phototriggered molecular switches and machines.[3] With
electron donor–acceptor (D–A) ring substitution, azo chro-
mophores have been extensively studied for second-order
nonlinear optical (NLO) applications.[4] Irradiation in the
absorption band of poled azo polymers with a pulse laser
beam, for example, can cause significant second-harmonic
generation (SHG) decay correlated with trans-to-cis isomeri-
zation, followed by a rapid recovery to the initial SHG level
when the cis isomer thermally relaxes back to the trans
state.[5] This marked photoswitching of the “on/off” SHG
signal is mainly attributed to the distinctive hyperpolariza-
bility of trans and cis azobenzenes. A more p-conjugated
trans isomer has a higher hyperpolarizability, thus enabling
the NLO response to be more readily switched between the
trans and cis states.
Recently, third-order NLO behavior that involves a two-
photon absorption (2PA) process has drawn considerable at-
tention because of many promising applications in the
emerging fields of photonics and biophotonics, including op-
tical power limiting, 3D data storage, 3D microfabrication,
bioimaging and tracking, and 2PA-assisted photodynamic
therapy.[6] The 2PA efficiency, namely, the 2PA cross-section
for the organic chromophores, is, in general, closely related
to both the intramolecular charge-transfer (ICT) efficiency
in D–p–D-, A–p–A-, and D–p–A-type structures and the ef-
fective p-conjugation length within a molecule.[7] The p
bridges, similar to a highly rigid fluorene unit, provide de-
localization and conjugation, which are critical for increasing
the 2PA cross-section. Moreover, elongation of the p frame-
works through either alkenyl (C=C) or alkynyl (CꢀC) link-
ages also results in extension of the effective p-conjugation
length and, thus, the enhancement of 2PA efficiency. How-
ever, use of an azo moiety (N=N) as the conjugation linkage
to achieve the 2PA process has been rarely reported.[8] An-
tonov et al. conducted a systematic study on the 2PA cross-
Azobenzene (azo) chromophores have been incorporated
into a wide variety of molecular architectures including
polymers, dendrimers, liquid crystals, self-assembled mono-
layers, and biomaterials.[1] Azo chromophores undergo
a uniquely clean and efficient photoisomerization; they ex-
[a] Prof. Dr. C.-C. Chu, Y.-C. Chang
School of Medical Applied Chemistry,
Chung Shan Medical University, South District, Taichung (Taiwan)
E-mail: jrchu@csmu.edu.tw
[b] Prof. Dr. C.-C. Chu
Department of Medical Education
Chung Shan Medical University Hospital, Taichung 40201 (Taiwan)
[c] B.-K. Tsai, Prof. Dr. T.-C. Lin
Department of Chemistry
National Central University, Jhong-Li 32001 (Taiwan)
[d] Prof. Dr. J.-H. Lin
Department of Electro-Optical Engineering
National Taipei University of Technology, Taipei 10608 (Taiwan)
[e] Prof. Dr. V. K. S. Hsiao
Department of Applied Materials and Optoelectronic Engineering
National Chi Nan University, Puli 54561 (Taiwan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201402896.
Chem. Asian J. 2014, 9, 3390 – 3396
3390
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Chih-Chien Chu et al.
sections in several D–p–A-type azoaromatic compounds and
suggested that the strength of the donor and/or acceptor
groups is significant.[8a] Furthermore, Magennis et al. demon-
strated a 2PA photochromism in a commercial azo chromo-
phore. The trans-to-cis isomerization could be performed
using two-photon excitation by employing 740 nm femtosec-
ond (fs) laser pulses.[8b] Consequently, we envisioned that
the third-order 2PA behavior of azo-containing materials
proven to possess a second-order NLO property should be
explored further.
As a proof-of-concept for this study, we synthesized a D–
p–D compound 1, in which two electron-donating dipheny-
laminofluorenes (DPAFs) were directly linked by an azo
group, to examine the correlation between trans/cis photoi-
somerization and 2PA properties. As shown in Scheme 1,
symmetric azo compound 1, indicated in deep orange, was
successfully synthesized by homocoupling two 2-amino-sub-
stituted DPAF through a copper-catalyzed aerobic oxidative
dehydrogenative process.[9] This symmetric azo-coupling re-
action was executed under mild conditions, using copper
bromide as the catalyst and air as the oxidant. The excellent
yield in this reaction was attributed to the electron-donating
character of the diphenylamine substituents. Thin-layer
chromatography (TLC) analysis, using an ethyl acetate and
hexane mixture (1:19) as the eluent, showed a single compo-
sition for compound 1 at Rf =0.83, thus indicating clearly
a geometric state of either pure trans or pure cis form. The
proton and carbon peaks shown in the 1H and 13C NMR
spectra also confirmed the existence of compound 1 as
Figure 1. a) Micro-Raman spectroscopic analysis for azo compound 1.
The marked peaks denote the characteristic vibration modes of the trans
isomer. b) UV-visible spectroscopic analysis for trans 1 upon 466 nm
LED irradiation. The inset shows the maximum decrement in p–p* ab-
sorbance within 5 s exposure time.
1182 (CN stretch), 1464 (NN
stretch), 1487 (NN stretch), and
1600 cmÀ1 (CC stretch) were
characteristic vibrations of trans
azobenzene, thus indicating that
compound 1 was a thermody-
namically stable trans product.
Moreover, the absence of the
in-plane ring bending mode at
approximately 1440 cmÀ1 might
indicate the rigidity of azo com-
pound 1 based on the fluorene
framework.
Compared with trans aromat-
ic azo compounds that have an
Scheme 1. Synthetic route for azo compound 1 bearing bilateral electron-donating diphenylaminofluorenes
intense p–p* absorption in the
UV region, trans compound
(DPAFs).
1
(Figure 1b), with bilateral
a single component (see the Supporting Information). How-
ever, neither method could determine the trans and cis geo-
metries. Research has shown that trans and cis azobenzenes
can be distinguished according to their characteristic Raman
shifts.[10] Therefore, we adopted micro-Raman spectroscopy
to analyze directly the vibrational spectrum of a solid
sample. Figure 1a shows the Raman spectra for compound
1 based on using a 633 nm He:Ne laser as the excitation
source. The Raman shifts centered at 1142 (CN stretch),
electron-donating DPAF moieties, showed a broad and in-
tense absorption band in the visible region centered at ap-
proximately 470 nm (with a molar extinction coefficient of
e at approximately 4.5ꢀ104 mÀ1 cmÀ1). This redshifted band,
with a coplanar trans configuration, suggested that com-
pound 1 had an elongating p-conjugation length and a lower
bandgap energy of p–p* transition. Therefore, trans-to-cis
photoisomerization for azo compound 1 was expected to
transpire under the excitation of visible light instead of UV
Chem. Asian J. 2014, 9, 3390 – 3396
3391
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Chih-Chien Chu et al.
light. A first attempt that involved using a 467 nm argon ion
laser (approximately 30 mW) as the excitation source could
not produce any change in p–p* absorbance, thus indicating
that compound 1 remained in the all-trans state. For per-
forming the trans-to-cis photoisomerization, a higher radiant
flux and a larger incident beam exposure area were intro-
duced by employing a noncoherent light-emitting diode
(LED) with a higher output energy (approximately 10 W)
and a maximum blue emission of 466 nm. As shown in Fig-
ure 1b, on LED irradiation, absorbance at 470 nm rapidly
reduced within several seconds and remained at a constant
value for a longer exposure time. This partial decrease in ab-
sorbance was apparently due to the formation of a cis prod-
uct with a relatively poor p-conjugation system and blue-
shifted p–p* absorption. The light-induced isomerization for
a dilute solution of aromatic azo compounds generally re-
sults in either an all-trans or an all-cis photostationary state
(PSS). However, the PSS for compound 1, established
through blue LED irradiation and readily visualized using
normal-phase TLC analysis, was composed of both trans and
cis products. The Rf value of the cis isomer (0.55) was lower
than that of the trans isomer (0.83), consistent with the fact
that cis isomers possess a significant dipole moment and
higher polarity. On the assumption that cis 1 would be com-
pletely transparent at 470 nm, cis content was calculated on
the basis of a drop in blue-light absorbance of approximate-
ly 45% during LED irradiation; the trans isomer was the
component in slight excess in this isomeric mixture.
Figure 2. a) Reversible photoswitch for azo compound 1 upon alternating
blue- and UV-light irradiation. 1H NMR spectroscopic analysis for b) all-
trans 1, and c) trans/cis-isomeric mixture by blue-light irradiation for
10 min. H1 and H2 denote the characteristic ortho protons adjacent to the
In addition to a decrease in p–p* absorbance, we ob-
served an increasing absorption band in the UV region of
350–400 nm. We speculated that this increasing band corre-
sponded to the p–p* absorption of the cis isomer. Because
the 2-amino-substituted DPAF precursor possessed intense
UV absorption centered at 330 nm (with a molar extinction
coefficient of e at approximately 2.2ꢀ104 mÀ1 cmÀ1) and be-
cause the cis isomer possessed a poor p conjugation on the
central N=N bond, the cis isomer was expected to have
a blue-shifted p–p* absorption band in the UV region. As
indicated by this UV absorption band, the cis isomer could
be switched back to the trans form with a 365 nm LED
light. As shown in Figure 2a, the absorbance returned to its
initial UV irradiation values within several seconds, and the
trans/cis isomerization in solutions in both toluene and THF
could be reversibly switched by alternating UV and blue-
light irradiation. The photoisomerization of compound
1 manifested a notable difference from conventional azo-
benzene systems in that it possessed a reverse direction of
the photoswitching wavelengths: a trans-to-cis interconver-
sion with visible (blue) light but a cis-to-trans back-isomeri-
zation with UV light.
3
7
À
N=N bond of trans isomer; H
isomer.
H denote the aromatic protons of cis
light irradiation, which ranged from d = 6.9 to 7.3 ppm, indi-
cated the existence of both trans and cis products (Fig-
ure 2c). The distinguishable peaks of H3 to H7 denoted the
characteristic protons on the cis structure. The integral area
of the spectrum from H7 to either H1 or H2 gradually in-
creased as the light exposure time was extended, thereby
suggesting that the cis content in the isomeric mixture could
be increased with a longer irradiation time. On the basis of
the integral ratios between the typical protons of these two
isomers, the calculated maximum cis content was approxi-
mately 30–40% under NMR spectroscopic experimental
conditions. Moreover, after UV-light irradiation, all of the
cis isomers were switched back to the trans form; thus, the
obtained NMR spectrum was identical to the original spec-
trum of all-trans 1. NMR spectroscopic and UV/Vis absorp-
tion analyses confirmed an efficient and reversible trans/cis
photoswitching of this fluorene-based azo compound in con-
centrated and dilute solutions, respectively.
1
One- and two-dimensional H NMR spectra of compound
1 (approximately 10À2 m in CDCl3) from alternating visible
and UV-light irradiation were compared (see the Supporting
Information). Figure 2b shows the partial spectrum for sym-
metric all-trans 1 and the marked aromatic resonances, with
H1 and H2 denoting the ortho protons adjacent to the N=N
bond. The complexity of the aromatic protons after blue-
The PSS for compound 1, established through blue-light
irradiation, contained both trans and cis isomers at a fixed
ratio; further irradiation did not increase the cis content.
The approach of the PSS is attributed to equal reaction
rates of forward (trans-to-cis) and backward (cis-to-trans) in-
Chem. Asian J. 2014, 9, 3390 – 3396
3392
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Chih-Chien Chu et al.
terconversions under 466 nm LED excitation. We assumed
that the incomplete transformation was due to an increase
in the backward reaction rate during light exposure and was
dominated by two main factors. The first factor was the
rapid thermal relaxation of the cis isomer to the thermody-
namic trans product. Following the first-order kinetic shown
in the Supporting Information, the thermal isomerization
rate (kt) calculated according to the recovery of p–p* ab-
sorbance at room temperature was determined to be 1.13ꢀ
10À2 minÀ1, approximately 20-fold faster than common azo-
benzenes, such as 4-butyl-4’-methoxyazobenzene (5.80ꢀ
10À4 minÀ1).[11] The more likely cause for the increasing
backward thermal isomerization of cis 1 was the destabiliza-
tion of the N=N bond by the bilateral electron-donating
DPAF moieties.[12] Secondly, the cis isomer might have un-
dergone a symmetry-allowed n–p* transition in the visible
region to cause the p–p* and n–p* transitions of trans and
cis isomers, respectively, to possess similar energy gaps.
Once the isomeric system contained certain amounts of cis
isomers, blue-light irradiation induced the photoisomeriza-
tion in both directions simultaneously. Thus, the equilibrium
state was soon reached because of an accelerating backward
reaction. We then employed high-performance liquid chro-
matography (HPLC) to separate the trans and cis isomers
quickly and verify their UV/Vis absorption spectra through
on-line photodiode array (PDA) detection. The HPLC chro-
matograms revealed two well-resolved peaks, the trans prod-
uct at 2.9 min and the cis product at 3.1 min (Figure 3a).
The increase in the integral ratio of cis/trans peaks under
different light exposure times also confirmed that the cis
content gradually increased until the PSS was reached. The
absorption profile of trans 1 obtained from PDA detection
in Figure 3b agreed with the UV/Vis spectrum shown in Fig-
ure 1b. Two distinct absorption bands for cis 1 were noticed
in the UV and visible regions, which presumably corre-
sponded to the p–p* and n–p* electronic transitions, respec-
tively. Because of the overlap of the p–p* and n–p* absorp-
tion for trans and cis isomers, respectively, backward cis-to-
trans photoisomerization was also promoted under blue-
light excitation. On the basis of the same molar extinction
coefficients of trans and cis 1 at 300 nm (Figure 3b), the cal-
culation of the integral areas for the two isomers from the
chromatograms at this detection wavelength suggested
a maximum cis content of 35%.
Figure 3. a) HPLC chromatograms for azo compound 1. The cis content
gradually increases with blue-light irradiation. b) UV-visible absorption
profiles for pure trans and cis 1 analyzed by HPLC equipped with an on-
line photodiode array (PDA) detector. c) Time course of trans-to-cis pho-
1
toisomerization. The cis content is determined by H NMR spectroscopy,
The irradiation time course of the cis contents was also es-
tablished by using NMR spectroscopy to obtain the rate
constant of kp at PSS with a blue-light exposure. The experi-
mental data fit the rate law well and yield the values of cis
content (at PSS)=36.3% and kp =2.04ꢀ10À1 minÀ1 (Fig-
ure 3c).[8b] Considering that kp is the sum of the forward (kf)
and backward (kb) reaction rate constants, the two rate con-
stants of kf =0.74ꢀ10À1 minÀ1 and kb =1.3ꢀ10À1 minÀ1 were
obtained. Because kb is composed of both thermal (kt) and
photoisomerization rate constants, the accelerating back-
ward cis-to-trans reaction was mainly found to be dominated
by the n–p* absorption of the cis isomer in the blue-light
region. The calculated cis content at the PSS (approximately
and the solid line is the nonlinear curve fitting integrated rate law. The
calculated cis content at photostationary state (approximately 36%) is
consistent with the experimental value (approximately 35%) obtained
using HPLC analysis.
36%), determined from its fit with the nonlinear curve of
the rate law, was consistent with the experimental values
(approximately 35%) obtained using HPLC analysis.
Because of the efficient nonradiative relaxation process
that involves trans/cis isomerization, the azo compound
1 was nonfluorescent. Therefore, we introduced an open-
aperture fs Z-scan technique to examine its 2PA behavior.[13]
Femtosecond laser pulses were generated by a mode-locked
Chem. Asian J. 2014, 9, 3390 – 3396
3393
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Chih-Chien Chu et al.
Ti:sapphire laser with a near-infrared (NIR) wavelength of
800 nm, a pulse width of 150 fs, and a repetition rate of
82 MHz. Both the trans- and cis-1 were completely transpar-
ent at 700–900 nm (Figure 3b), thus suggesting an absence
of linear optical absorption in the NIR region. During data
collection, laser pulses with a beam waist of approximately
20 mm were focused onto the azo solution in a quartz cuv-
ette with a path length of 1 cm. The incident and transmit-
ted laser powers were monitored as the cuvette moved
along the Z direction, toward and away from the focus posi-
tion. Figure 4a shows the experimental Z-scan curves of
trans 1, which were created by changing the input intensity
per laser pulse (I0), with significant 2PA signatures based on
the decrease in normalized transmittance (T) at Z=0. The
linear dependence between the amplitude of transmittance
change (1/TÀ1) and I0 also confirms a nonlinear 2PA pro-
cess (Figure 4b).[14] Moreover, Figure 4c shows the contrast
of Z-scan traces between trans 1 and Disperse Orange III
(DO3), a commercially available NLO azobenzene chromo-
phore (D–p–A type).[8a,c] Based on the minor change in the
transmittance for DO3 under identical I0, the 2PA response
for trans 1 is considerably higher than that for common azo
dyes.
Considering the equivocal statement that 2PA intensity is
usually lower for azobenzene (Ph-N=N-Ph) than for stilbene
(Ph-C=C-Ph) derivatives, combining azo and rigid fluorene
with bilateral electron-donating groups did enhance the 2PA
cross-section in the NIR wavelength.[8a] Although there was
no clear understanding at the time of this study about the
connection between the remarkable 2PA response and azo
linkage, our results showed that such a structural combina-
tion leads to highly active 2PA chromophores. Furthermore,
as shown in Figure 5a, the computational modeling of the
Figure 5. Optimized isosurface plots of the a) HOMO and b) LUMO for
trans-1 analogue with 9,9-diethyl substitution, energetically minimized at
the density functional B3LYP/6-31G** level under vacuum.
trans-1 analogue with 9,9-diethyl substitution by using the
B3LYP hybrid functional and the 6-31G** basis set shows
that the highest-occupied molecular orbital (HOMO) is
mainly delocalized over the DPAFs. Meanwhile, the elec-
tronic cloud of the lowest-unoccupied molecular orbital
(LUMO) is condensed mostly on the central moiety (Fig-
ure 5b). This preliminary calculation implies an ICT process
from the peripheral diphenylamines (donor) to the central
CNNC group (acceptor) through a fluorene p bridge. More-
over, we conjectured that the ICT for azo compound 1 is
more favored because of the relatively higher electronega-
tivity of the central nitrogen atoms.
Figure 4. a) Open-aperture Z-scan traces of azo compound 1 (5ꢀ10À3 m in
toluene) by changing the input intensity of the laser beam (40, 60, 80,
100, 120 mW) at lex =800 nm. b) The correlation of (1/TÀ1) versus I0
[GWcmÀ2]; T and I0 denote the normalized transmittance value at Z=
0 cm and incident energy per laser pulse, respectively. c) The contrast of
We also analyzed the 2PA intensity of the isomeric mix-
ture at PSS, which was established by blue-light irradiation
(approximately 65% of trans and 35% of cis products).
the Z-scan curves between 1 (1ꢀ10À2 m in toluene) and DO3 (1ꢀ10À2
m
in THF) under identical input laser intensity (120 mW).
Chem. Asian J. 2014, 9, 3390 – 3396
3394
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Chih-Chien Chu et al.
101 mg, 0.196 mmol) in toluene (2 mL) was purged with air for 5 min,
and then stirred vigorously at 608C under air balloon for 20 h. After
cooling to room temperature and being concentrated under vacuum, the
residue was purified by silica flash column chromatography using a n-
hexane and ethyl acetate mixture (19:1) as eluent to afford an orange
solid of 1 (83.8 mg, 83%). 1H NMR (CDCl3, 400 MHz): d=7.95 (dd, J=
8.1, 1.7 Hz, 2H), 7.90 (d, J=1.7 Hz, 2H), 7.73 (d, J=8.1 Hz, 2H), 7.61 (d,
J=8.1 Hz, 2H), 7.29–7.25 (m, 10H), 7.14 (d, J=8.7 Hz, 8H), 7.03 (t, J=
7.3 Hz, 6H), 2.05–1.96 (m, 4H), 1.92–1.85 (m, 4H), 1.18–1.00 (m, 24H),
0.79 (t, J=7.1 Hz, 12H), 0.69 ppm (m, 8H); 13C NMR (CDCl3,
100 MHz): d=153.5, 152.1, 151.9, 148.1, 144.0, 135.5, 129.5, 124.3, 123.5,
123.1, 123.0, 121.3, 119.6, 119.1, 117.1, 55.5, 40.6, 31.8, 30.0, 24.1, 22.8,
14.3 ppm; MALDI-TOF-MS: m/z calcd for C74H85N4 [M+H]+: 1029.68;
found: 1029.81.
Prior to the Z-scan experiment, the azo solution was ex-
posed to blue light until the PSS was reached, and was then
excited in situ by using laser pulses to trace the transmit-
tance change at the focal point. Nevertheless, the normal-
ized change in transmittance at Z=0 for the isomeric mix-
ture was nearly identical to that of the all-trans-1 solution
(see the Supporting Information). This result implied that,
for both solutions, the chemical composition at the laser fo-
cusing spot was in the all-trans state. This is presumably due
to the extremely unstable cis structure during the Z-scan
measurement. The rate constants of kt for thermal cis-to-
trans back-isomerization substantially increased as the sur-
rounding temperature increased (see the Supporting Infor-
mation), which suggests that the cis isomer became quite un-
stable with elevating temperatures. Because significant topi-
cal heating by laser pulses in high repetition frequency
could have induced rapid and complete cis-to-trans thermal
isomerization, the influence on 2PA properties in the pres-
ence of the cis isomer could not be determined under exper-
imental conditions. Consequently, the preliminary results
suggested that the fs 2PA behavior for azo chromophore 1 is
quite uniform, regardless of the molecular geometry. To
minimize the laser-induced thermal effect, a 2PA experi-
ment that adopted an fs pulse laser with low repetition rate
of 1 kHz as an excitation source is now under investigation.
In summary, we successfully demonstrated efficient trans/
cis photoisomerization and a significant third-order NLO re-
sponse for the azo chromophore, simply incorporated with
bilateral DPAF as a p framework. Although strong electron-
donating DPAF moieties could destabilize the central N=N
bond, reversibly switching the molecular geometry was still
possible through non-coherent LED-light excitation at two
wavelengths (466 and 365 nm). The PSS established by using
blue light at room temperature was composed of only 36%
cis isomer. This percentage was mainly due to an increasing
cis-to-trans back-isomerization rate through both photo- (n–
p* absorption of cis isomer) and thermal isomerization
routes. Compared with common azobenzene dyes, the fs Z-
scan measurement determined a significant enhancement of
the 2PA intensity for this azo chromophore in its all-trans
state. Because of extremely fast cis-to-trans thermal isomeri-
zation under accumulated and substantial heat generated by
fs laser pulses, the influence on 2PA properties in the pres-
ence of the cis isomer was negligible at this stage. Based on
different p-conjugation domains, we suggest that the trans
and cis isomers exhibit distinct 2PA behaviors. A thermally
stable cis-azo chromophore is necessary to achieve a tunable
NLO response with respect to a change in molecular geome-
try.[11] A study along this line is currently underway, and the
findings will be reported in due course.
Acknowledgements
The authors would like to thank the Ministry of Science and Technology
of Taiwan (MOST102-2113M-040-004) for financially supporting this re-
search. The authors are also grateful to Prof. Toyoko Imae for support
with micro-Raman measurements and to Prof. Ming-Yu Kuo for support
with the computational modeling.
Keywords: absorption · azo compounds · charge transfer ·
chromophores · isomerization
[1] a) Y. Zhao, T. Ikeda in Smart Light-Responsive Materials: Azoben-
zene-Containing Polymers and Liquid Crystals, Wiley, New Jersey,
2009; b) R. Deloncle, A.-M. Caminade, J. Photochem. Photobiol. C
2010, 11, 25–45; c) M. Han, T. Honda, D. Ishokawa, E. Ito, M.
Hara, Y. Norikane, J. Mater. Chem. 2011, 21, 4696–4702; d) S. Sa-
manta, A. A. Beharry, O. Sadovski, T. M. McCormick, A. Babalha-
vaeji, V. Tropepe, G. A. Woolley, J. Am. Chem. Soc. 2013, 135,
9777–9784; e) S.-L. Chen, C.-C. Chu, V. K. S. Hsiao, J. Mater. Chem.
C 2013, 1, 3529–3531.
[2] N. Tamai, H. Miyasaka, Chem. Rev. 2000, 100, 1875–1890.
[3] a) Y. Norikane, N. Tamaoki, Org. Lett. 2004, 6, 2595–2598; b) K. G.
Yager, C. J. Barrett, J. Photochem. Photobiol. A 2006, 182, 250–261;
c) D. P. Ferris, Y.-L. Zhao, N. M. Khashab, H. A. Khatib, J. F. Stod-
dart, J. I. Zink, J. Am. Chem. Soc. 2009, 131, 1686–1688; d) S. K. M.
Nalluri, J. Voskuhl, J. B. Bultema, E. J. Boekema, B. J. Ravoo,
Angew. Chem. Int. Ed. 2011, 50, 9747–9751; Angew. Chem. 2011,
123, 9921–9925; e) I. Tochitsky, A. Polosukhina, V. E. Degtyar, N.
Gallerani, C. M. Smith, A. Friedman, R. N. Van Gelder, D. Trauner,
D. Kaufer, R. H. Kramer, Neuron 2014, 81, 800–813.
[4] J. A. Delaire, K. Nakatani, Chem. Rev. 2000, 100, 1817–1845.
[5] Z. Sekkat, P. PrÞtre, A. Knoesen, W. Volksen, V. Y. Lee, R. D.
Miller, J. Wood, W. Knoll, J. Opt. Soc. Am. B 1998, 15, 401–413.
[6] a) G. S. He, L.-S. Tang, Q. Zheng, P. N. Prasad, Chem. Rev. 2008,
108, 1245–1330; b) R. Rumi, S. Barlow, J. Wang, J. W. Perry, S. R.
Marder, Adv. Polym. Sci. 2008, 213, 1–95; c) H. Xia, W.-Y. Zhang,
F.-F. Wang, D. Wu, X.-W. Liu, L. Chen, Q.-D. Chen, Y.-G. Ma, H.-B.
Sun, Appl. Phys. Lett. 2009, 95, 083118; d) S. Yao, K. D. Belfield,
Eur. J. Org. Chem. 2012, 3199–3217; e) H. Wang, F. Jin, S. Chen, X.-
Z. Dong, Y.-L. Zhang, W.-Q. Chen, Z.-S. Zhao, X.-M. Duan, J.
Appl. Polym. Sci. 2013, 130, 2947–2956.
[7] a) M. Albota, D. Beljonne, J. L. Brꢂdas, J. E. Ehrlich, J.-Y. Fu, A. A.
Heikal, S. E. Hess, T. Kogej, M. D. Levin, S. R. Marder, D. McCord-
Maughon, J. W. Perry, H. Rçckel, M. Rumi, G. Subramaniam, W. W.
Webb, X.-L. Wu, C. Xu, Science 1998, 281, 1653–1656; b) K. D. Bel-
field, Y. Liu, R. A. Negres, M. Fan, G. Pan, D. J. Hagan, F. E. Her-
nandez, Chem. Mater. 2002, 14, 3663–3667; c) S. Yao, H.-Y. Ahn, X.
Wang, J. Fu, E. W. V. Stryland, D. J. Hagan, K. D. Belfield, J. Org.
Chem. 2010, 75, 3965–3974; d) T.-C. Lin, M.-H. Li, C.-Y. Liu, J.-H.
Experimental Section
Synthesis of Azo Compound 1
A mixture that contained CuBr (4.6 mg, 32.1 mmol), pyridine (10 mL,
0.124 mmol), and 2-amino-substituted diphenylaminofluorene (DPAF,
Chem. Asian J. 2014, 9, 3390 – 3396
3395
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemasianj.org
Chih-Chien Chu et al.
Lin, Y.-K. Shen, Y.-H. Lee, J. Mater. Chem. C 2013, 1, 2764–2772;
e) T.-C. Lin, C.-Y. Liu, B.-R. Huang, J.-H. Lin, Y.-K. Shen, C.-Y.
Wu, Eur. J. Org. Chem. 2013, 498–508.
[11] B.-K. Tsai, C.-H. Chen, C.-H. Hung, V. K. S. Hsiao, C.-C. Chu, J.
Mater. Chem. 2012, 22, 20874–20877.
[12] L. Li, F. He, X. Wang, N. Ma, L. Li, ACS Appl. Mater. Interfaces
2012, 4, 4927–4933.
[8] a) L. Antonov, K. Kamada, K. Ohta, F. S. Kamounah, Phys. Chem.
Chem. Phys. 2003, 5, 1193–1197; b) S. W. Magennis, F. S. Mackay,
A. C. Jones, K. M. Tait, P. J. Sadler, Chem. Mater. 2005, 17, 2059–
2062; c) L. De Boni, L. Misoguti, S. C. Zilio, C. R. Mendonca,
ChemPhysChem 2005, 6, 1121–1125; d) T. V. Truong, C.-Y. Chen,
N. V. Tabiryan, Y. R. Shen, J. Opt. Soc. Am. B 2007, 24, 2623–2626;
e) L. D. Boni, A. A. Andrade, S. B. Yamaki, L. Misoguti, S. C. Zilio,
T. D. Z. Atvars, C. R. Mendonca, Chem. Phys. Lett. 2008, 463, 360–
363.
[9] C. Zhang, N. Jiao, Angew. Chem. Int. Ed. 2010, 49, 6174–6177;
Angew. Chem. 2010, 122, 6310–6313.
[10] a) C.-M. Stuart, R. R. Frontiera, R. A. Mathies, J. Phys. Chem. A
2007, 111, 12072–12080; b) Y. B. Zheng, J. L. Payton, C.-H. Chung,
R. Liu, S. Cheunkar, B. K. Pathem, Y. Yang, L. Jensen, P. S. Weiss,
Nano Lett. 2011, 11, 3447–3452.
[13] a) J.-H. Lin, Y.-J. Chen, H.-Y. Lin, W.-F. Hsieh, J. Appl. Phys. 2005,
97, 033526; b) Y.-P. Chan, J.-H. Lin, C.-C. Hsu, W.-F. Hsieh, Opt.
Express 2008, 16, 19900–19908; c) M. Wang, V. Nalla, S. Jeon, V.
Mamidala, W. Ji, L.-S. Tan, T. Cooper, L. Y. Chiang, J. Phys. Chem.
C 2011, 115, 18552–18559.
[14] a) A. A. Said, M. Sheik-Bahae, D. J. Hagan, T. H. Wei, J. Wang, J.
Young, E. W. V. Stryland, J. Opt. Soc. Am. B 1992, 9, 405–414;
b) W. J. Yang, D. Y. Kim, C. H. Kim, M.-Y. Jeong, S. K. Lee, S.-J.
Jeon, B. R. Cho, Org. Lett. 2004, 6, 1389–1392; c) K. D. Belfield,
M. V. Bondar, F. E. Hernandez, O. V. Przhonska, S. Yao, J. Phys.
Chem. B 2007, 111, 12723–12729.
Received: July 29, 2014
Published online: October 7, 2014
Chem. Asian J. 2014, 9, 3390 – 3396
3396
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim