Monitoring: Cis -to- trans isomerization of azobenzene using terahertz time-domain spectroscopy

Article abstract of DOI:10.1039/c8cp04570d

We present terahertz time-domain spectroscopy (THz-TDS) to explore the conformational dynamics of thermally induced and photoinduced isomerization of azobenzene. The essence of the method is that isomerization of azobenzene proceeds via large structural c

Full text of DOI:10.1039/c8cp04570d

PCCP
View Article Online
View Journal | View Issue
PAPER
Monitoring cis-to-trans isomerization of
azobenzene using terahertz time-domain
spectroscopy
Cite this: Phys.Chem.Chem.Phys.,
2
018, 20, 27205
a
a
a
b
b
Lu Zhou,† Ligang Chen,† Guanhua Ren,† Zhongjie Zhu, Hongwei Zhao,
c
ad
a
Huabin Wang, Weili Zhang
*
and Jiaguang Han
*
We present terahertz time-domain spectroscopy (THz-TDS) to explore the conformational dynamics of
thermally induced and photoinduced isomerization of azobenzene. The essence of the method is that
isomerization of azobenzene proceeds via large structural changes in the molecule, while the THz response
is sensitive to these changes. We experimentally demonstrate that the THz spectra of azobenzene show
remarkable variations upon heating and irradiation, and as such quantitatively recorded and identified THz
spectroscopy can be used to monitor the isomerization process. Specifically, the measured THz spectra
clearly reveal that the rate of thermal-isomerization from cis-to-trans in non-polar solvents is faster than that
in polar solvents, and an about 6-fold acceleration of the rate could be achieved when Au NPs were
introduced as a catalyst into azobenzenes. Moreover, we provide evidence that the temperature and Au NP
catalyst do not have an obvious influence on the photoinduced isomerization of azobenzene. The presented
example illustrates the power of the THz-TDS method to open up a novel avenue for exploring molecular
dynamics.
Received 19th July 2018,
Accepted 9th October 2018
DOI: 10.1039/c8cp04570d
rsc.li/pccp
methods including UV-vis absorption, Raman, nuclear magnetic
resonance (NMR), and so on. For example, Quick et al. studied
Introduction
13
Azobenzene (AB), a diazene (HNQNH) derivative, which con- the photo-isomerization dynamics and pathways of cis- and trans-
sists of two benzene rings joined by a NQN double bond, can AB in solution by using UV-vis absorption. The excited-state
exist in either the cis or trans conformation. When irradiated structure and dynamics of cis- and trans-AB were investigated by
14
with ultraviolet (UV) or visible (vis) light, the isomerization Stuart et al., where the method of resonance Raman was
15
from the stable trans to the metastable cis is induced, and the adopted. Curtis et al. obtained the nitrogen-15 NMR spectra of
1
,2
reverse process can occur either by irradiating or heating.
AB. The previous reported methods can monitor the thermal- or
This reversible geometric effect was first noted in 1937 by photo-isomerization of AB well, and thus play important roles in
3
Hartley, when he observed the cis isomer of AB upon irradia- the studies of AB. However, of particular note, it is well known
tion. Since then, the isomerization of AB has attracted intense that AB is sensitive to light, and the above methods unavoidably
interest because of its potential applications in optical cause additional irradiation of AB. Therefore, developing new
4
–6
7,8
9,10
switches,
memory storage, molecular machines,
and approaches with minimum interference to quantitatively monitor
1
1,12
light-sensitive biomolecules.
The isomerization processes the conformational dynamics of isomerization of AB holds great
of AB have been monitored and characterized by a variety of promise in determining the course of chemical reactions and thus
enabling novel applications.
Terahertz time-domain spectroscopy (THz-TDS), containing
a
Center for Terahertz Waves and College of Precision Instrument and
abundant physical and chemical information on materials, has
Optoelectronics Engineering, Tianjin University, Tianjin 300072, China.
1
6
shown many promising applications in the fields of physics,
E-mail: weili.zhang@okstate.edu, jiaghan@tju.edu.cn
1
7
18
19
b
c
chemistry, materials science and biology. As a non-
destructive data acquisition and coherence detection technology,
THz-TDS provides a reliable and unique analytical method to study
the characteristics of molecules. In particular, the low-frequency
torsional vibration or rotation energy levels of most molecules lie
in the THz region, and low-frequency vibrational or rotational THz
spectroscopy is highly sensitive to the conformation and structure
Shanghai Institute of Applied Physics, Chinese Academy of Sciences,
Shanghai 201800, P. R. China
Chongqing Engineering Research Center of High-Resolution and Three-Dimensional
Dynamic Imaging Technology, Chongqing Institute of Green and Intelligent
Technology, Chinese Academy of Science, Chongqing 400714, China
School of Electrical and Computer Engineering, Oklahoma State University,
Stillwater, Oklahoma 74078, USA
d
†
These authors contributed equally to this work.
This journal is ©the Owner Societies 2018
Phys. Chem. Chem. Phys., 2018, 20, 27205--27213 | 27205

View Article Online
Paper
PCCP
of molecules. Therefore, the high sensitivity of THz spectroscopy Finally, the Au NPs were obtained and stored at ꢁ4 1C for
provides us with a powerful and unique fingerprint to probe the further use.
conformational dynamics of molecules. A major difference
between THz-TDS and Raman or UV-vis spectroscopy is that
Thermally induced cis-to-trans isomerization of AB
THz-TDS is a coherent measurement process, in which one could Thermally induced cis-to-trans isomerization of AB was per-
directly obtain the optical parameters of a sample such as refractive formed as follows. The 0.14 M solutions (8 mL) of cis-AB in
20,21
index and dielectric constants.
4
Another significant advantage different solvents (ethanol, acetone, n-hexane and CCl ) were
of the use of THz-TDS is that THz waves do not give rise to direct prepared and then the solutions in closed brown bottles were
electronic excitations in the sample and therefore avoid causing kept in a water bath at a temperature of 50 1C for a certain time
damage or other photochemical reactions due to their low photon interval. After the allotted time, the solid powder of AB was
energies and non-ionization.
obtained in a gentle flow of dry air at room temperature for THz
Thermal- or photo-isomerization of AB proceeds via large measurement.
structural changes and THz spectroscopy is thus very useful to
In addition, Au NP catalyzed thermally induced cis-to-trans
monitor these dynamical processes. In light of that, herein, isomerization of AB in ethanol solution was also performed.
we apply the THz-TDS method to quantitatively monitor the Similarly, 300 mL of the Au NP colloid was added into the 0.14 M
cis–trans isomerization of AB. The quantitative thermally (8 mL) cis-AB (in ethanol solution) in a brown bottle and the
induced cis-to-trans isomerization of AB in different polar mixed solution was kept in a water bath at 50 1C for a certain
solvents and under gold nanoparticle (Au NP) catalysis was time interval. Then, the mixture was washed by centrifugation
characterized by THz-TDS. Our measured results demonstrate at 12 000 rpm for 15 min to remove the Au NP colloid. After-
that the rate of thermal cis-to-trans isomerization of AB in non- wards, the solid powder of AB was obtained in a gentle flow of
polar solvent is faster than that in polar solvent. The thermal dry air at room temperature for THz measurement.
cis-to-trans isomerization reaction of AB is highly accelerated
Photoinduced isomerization of AB from cis-to-trans
when Au NPs are introduced as a catalyst. Moreover, the
photoinduced isomerization of AB from cis-to-trans under The photoinduced isomerization of AB from cis-to-trans under
different conditions was also investigated by THz-TDS. It was different conditions was carried out and the steps of the
found that the temperature and catalyst have no remarkable experiment are similar to the above procedure for thermal-
influence on the photo-isomerization of AB.
isomerization. Briefly, 0.14 M cis-AB in different solvents (ethanol,
acetone, n-hexane and CCl ) was irradiated with visible light at
4
445 nm (210 mW) for a certain time interval under vigorous
Experimental
Chemicals
stirring at room temperature and 50 1C. Then the solid powder
of AB was obtained, in a gentle flow of dry air at room temperature
for THz measurement.
In addition, Au NP catalyzed photo-isomerization of AB from
cis-to-trans in ethanol solution was also performed. Similarly,
trans-Azobenzene (trans-AB) was purchased from J&K Scientific
Co., Ltd. The cis-azobenzene (cis-AB) was obtained by illuminat-
ing an ethanol solution of trans-AB with an ultraviolet lamp
300 mL of the Au NP colloid was added into the 0.14 M (8 mL)
(consumed power 250 W) at 365 nm. About 40% of cis-AB was
cis-AB (in ethanol) solution in a closed beaker and the mixed
solution was irradiated at room temperature for a certain time
interval. Then, after the allotted time, the mixture was washed
by centrifugation at 12 000 rpm for 15 min to remove the Au NP
colloid. Afterwards, the solid powder of AB was obtained, in a
gentle flow of dry air at room temperature, for THz
measurement.
produced within two hours then purified by column chromato-
graphy with silica gel (200 mesh) as the stationary phase and
petroleum ether and ethyl acetate mixed solvent (40 : 1, v/v) as
eluents. COC (cyclic olefin copolymer) powder (particle size
5
0–70 mm) was purchased from Terahertz Photonics Co., Ltd.
Acetone, tetrachloromethane (CCl ), ethanol and n-hexane were
obtained from Real and Lead Chemical Co., Ltd. Hydrogen
tetrachloroaurate(III) trihydrate (HAuCl ꢀ3H O) was purchased Sample preparation
4
4
2
from Sigma-Aldrich. Trisodium citrate (Na C H O ꢀ2H O) was
3
6
5
7
2
For the THz-TDS measurements, the powder AB samples were
pressed into disks with a hydraulic press at a pressure of
purchased from Aladdin.
1
0
5 MPa. The thickness of the sample tablets was approximately
.7 mm, and the diameter was 13 mm. As for the low tempera-
Synthesis of Au nanoparticles
Au nanoparticles (Au NPs) with an average radius of 20 nm were ture experiments, the AB powder was blended with COC
prepared according to the procedure developed by Turkevich powder, which was nearly transparent in the THz region, at a
2
2
et al. 2 mL (1%) sodium citrate aqueous solution was added mass ratio of 1 : 3 (50 mg : 150 mg) and then pressed into disks
ꢁ4
ꢁ1
immediately to 100 mL of a 10
g mL
HAuCl4 boiling of 1.4 mm thickness.
solution under vigorous stirring. The mixture was stirred
vigorously and boiled for an additional 20 min. Then the
Instrumentation
obtained Au colloids were purified by 3 cycles of re-dispersion UV-vis absorption spectra were measured with a spectrometer
in ethanol and centrifugation at 12 000 rpm for 15 min. (TU1901, Pgeneral). An SU-70 FESEM instrument was used to
2
7206 | Phys. Chem. Chem. Phys., 2018, 20, 27205--27213
This journal is ©the Owner Societies 2018

View Article Online
PCCP
Paper
record the scanning electron microscopy (SEM) images under
an accelerating voltage of 5 kV. Powder X-ray diffraction (PXRD)
patterns of cis- and trans-AB were recorded using an X’Pert Pro
MPD (Cu source, 40 kV voltage, 40 mA filament emission). The
data were collected over a scan range from 101 to 401 (2y).
Photo-isomerization of AB was achieved using a diode pumped
solid state laser (DPSSL, 210 mW, Shanxi Tongwei Optoelec-
tronic Equipment Co., Ltd.) with a wavelength of 445 nm.
Quantum chemical calculations
Calculations based on DFT using the Cambridge Sequential
Total Energy Package (CASTEP) program as a part of the
Materials Studio package from Accelrys were performed to
Fig. 1 (a) The THz absorption spectra and refractive index of trans- and
analyze the observed THz spectra. After verification by compar- cis-AB and (b) the corresponding complex dielectric functions. The
ing measured PXRD spectra (see insets of Fig. 2(a and d)) of the temperature effect on the THz absorption spectra of (c) trans-AB and
(
d) cis-AB.
samples with those calculated from the previous data pub-
lished, crystal cell parameters of trans- and cis-AB for calcula-
tions were taken from ref. 23 and 24 respectively. The trans-AB
belongs to the monoclinic crystal system, and the space group
ꢁ
1
less than 3 cm below 1.0 THz and then gradually increases
with increasing frequency. The corresponding refractive index
shows a comparatively smooth feature at a value of about 1.7 in
our measured frequency range. Below 2.3 THz, the cis-AB has a
much higher absorption coefficient than trans-AB, especially at
the absorption peaks of 0.60 and 1.27 THz. Fig. 1(b) shows the
is P2 /a (Z = 4), with lattice dimensions of a = 12.144 Å, b = 5.756 Å,
c = 15.396 Å, a = 901, b = 1141 8 , g = 901. The cis-AB is suggested to
1
0
be orthorhombic in structure with the space group Pbcn (Z = 4),
and the corresponding lattice parameters are a = 7.57 Å,
b = 12.71 Å, c = 10.32 Å, a, b, g = 901. The calculations were
performed on the crystalline state within the generalized gra-
dient approximation (GGA) at the Perdew–Burke–Ernzerhof
r i
real part e and imaginary part e of the measured complex
dielectric functions of all samples, where the frequency-
dependent complex dielectric function is obtained through
(PBE) correlation functional. Grimme’s dispersion-corrected
2
6
2
2
the formula: er = n ꢁ (al /4p) and e = an l /2p.
method was used for DFT-D correction and norm-conserving
pseudopotential for CASTEP.
r
0
i
r 0
Additionally, we measured the THz spectra of the two
isomers within a temperature range ranging from ꢁ190 to
22 1C. For trans-AB, as shown in Fig. 1(c), the absorption peaks
THz-TDS measurements
become stronger and sharper and suffer a distinct blue-shift
with the decrease of temperature. The two peaks at 1.39 and
1.74 THz were blue-shifted to 1.65 and 1.98 THz, respectively.
Such a shift is usually ascribed to the increased bond length
The measurements were performed with two THz-TDS systems,
one is a photoconductive switch-based 8-F confocal THz-TDS
system, the other is a TAS7400TS THz-TDS system (Advantest
Corporation, Japan), which is used for the low temperature
experiments in this work. The effective spectrum band is 0.2 to
2
5
2
7
due to thermal expansion. Meanwhile, the shoulder peak at
.28 THz split into two peaks at 2.34 THz and 2.46 THz, and a
2
2.5 THz. The THz beam pathway was purged with dry air to
weak peak at 1.48 THz gradually appeared. The cis-AB shows
similar behaviors to the trans-AB. The absorption peaks became
stronger with decreasing temperature, and then two small
peaks appeared at 1.64 and 2.09 THz at low temperature.
We also noticed that as the temperature decreased, the peak
at 0.6 THz has a significant red-shift rather than a blue-shift
like other peaks, and the peak at 1.27 THz shows a negligible
frequency shift. The intensities and frequency positions of
peaks usually show a temperature dependent change in the
keep a relative humidity of about 2.5%. We extracted the
absorption coefficient, refractive index and complex dielectric
function of the samples from their transmitted THz pluses.
Temperature-dependent THz spectra of AB were measured
using a temperature controller (Specac Ltd UK, accuracy ꢂ0.5 1C)
which is equipped with COC windows. THz spectra were recorded
over the temperature range from ꢁ190 to 22 1C (83–295 K).
2
8–31
THz spectrum.
The stronger and sharper peaks are the
Results and discussion
THz spectra of AB
result of temperature-dependent changes in distribution of the
28
energy vibrational states. The temperature induced frequency
Fig. 1 shows the THz power absorption coefficient a, refractive shifts have been considered as the consequence of multiple
indices n, and complex dielectric functions e, of trans- and mechanisms. The origins of the blue shift for most peaks are
cis-AB in the range of 0.2–2.5 THz at room temperature. trans- the anharmonicity of vibrational potential and the temperature-
28,29,32
AB has two distinct absorption peaks at 1.39 and 1.74 THz, and dependent change of the volume.
The anomalous red shift
a shoulder peak at 2.28 THz. cis-AB has four obvious absorption at 0.60 THz can be interpreted as the interaction of weak inter-
peaks located at 0.6, 0.9, 1.27 and 2.19 THz. It is found that the molecular bonding forces such as hydrogen bonds and van der
32
absorption coefficient of the trans-AB is extremely low, and is Waals forces. The behaviour seen at 1.27 THz, almost unchanged
This journal is ©the Owner Societies 2018
Phys. Chem. Chem. Phys., 2018, 20, 27205--27213 | 27207

View Article Online
Paper
PCCP
with temperature, can be explained by the combination of two cis- and trans-AB mixtures were plotted in Fig. 3(a). On decreasing
effects which lead to opposite frequency shifts.
the content of cis-AB, the absorption peaks of cis-AB at 0.6,
.9, 1.26 and 2.20 THz decrease before vanishing, and mean-
0
Quantum chemical calculation of trans- and cis-AB
while the peaks of trans-AB at 1.39 and 1.74 THz emerge and
increase gradually. Particularly, the peaks at 0.6 THz exhibit a
clean gradient change process upon decreasing the content of
cis-AB. As such, the peak at 0.6 THz was used to evaluate the
content of cis-AB in the two isomers. The dose–response fit
curve of the peak intensity at 0.6 THz for different ratios of
cis-AB shown in Fig. 3(b) obeys the linear regression equation:
Quantum chemical calculations were performed to understand the
experimentally measured THz responses of AB. The simulated THz
spectra of trans- and cis-AB using solid-state DFT calculations are
plotted in Fig. 2(a and d), where the simulated results are overall in
good agreement with the experimental ones. Four typical vibra-
tional modes of trans- and cis-AB are demonstrated in Fig. 2(b–f).
For trans-AB, four molecules are arranged in the unit cell with
2
y = ꢁ0.95 + 0.45x, correlation coefficient R = 0.997. From the
dose–response, the absorption coefficient of the designated
peak was proportional to the cis-AB content in mixtures across
the whole range.
1
space group symmetry P2 /a, with the centers of each molecule
lying on a center of symmetry of the lattice. According to ref. 23,
the molecules centered on (0, 0, 0) and those centered on (0, 0, 1/2)
are non-equivalent, and this non-equivalent property has also been
reflected in the calculated vibrational modes as shown in Fig. 2. Thermally induced cis-to-trans isomerization of AB
The experimental peak at 1.65 THz corresponds to the vibrational
mode calculated at 1.44 THz, primarily arising from the translation
of molecules centered on (0, 0, 0) along with milder translation of
The thermally induced cis-to-trans isomerization process of AB
in different polarity solvents (ethanol, acetone, hexane and
CCl ) at 50 1C was monitored by using the THz-TDS method.
4
molecules on (0, 0, 1/2). The measured peak at 1.98 THz is relative
to the calculated peak at 2.00 THz, arising from the twisting of the
benzene rings with translation of the NQN bond of the molecules
centered on (0, 0, 1/2) while the molecules centered on (0, 0, 0)
show a negligible translation. For cis-AB, the peak at 0.6 THz at
room temperature, which takes a significant red-shift with decreas-
ing temperature, can be attributed to the calculated mode at
Fig. 4(a) illustrates the THz absorption spectra of the thermal
cis-to-trans isomerization process of AB in ethanol solution at
50 1C. Upon heating, the metastable cis-AB is isomerized into
the thermostable trans-AB. The THz absorption peaks of the
cis-AB at 0.6, 0.9, 1.26 and 2.20 THz decrease with increasing
heating time. Meanwhile, the new peaks at 1.39 and 1.74 THz
emerge after 4 h and then become stronger with heating time.
After 38 h, the peaks for cis-AB vanished and the peaks for trans-
AB reached to the maximum. It means that the cis-AB was
thermally-isomerized to trans-AB. Similarly, Fig. 4(b) shows the
THz absorption spectra of the thermal cis-to-trans isomerization
process of AB in polar solvent (acetone). In Fig. 4(b), the THz
absorption peaks of the cis-AB at 0.6, 0.9, 1.26 and 2.20 THz
0.39 THz, due to the rotational vibration of the whole molecule
(Fig. 2(e)). The mode at 1.60 THz originates from the translation
along with the rotation of the molecules. Table 1 gives the detailed
descriptions of the vibrational modes of trans- and cis-AB.
The THz absorption spectra for mixtures of cis- and trans-AB
To quantitatively study the isomerization process of AB from gradually decrease until they disappear after 38 h. The THz absorp-
cis-to-trans, the THz absorption spectra of different ratios of tion peaks of trans-AB at 1.39 and 1.74 THz appeared and reached
Fig. 2 (a) The experimental and calculated THz spectra of trans-AB, the inset is the measured and calculated PXRD result; the simulated vibrational
modes of trans-AB at (b) 1.44 and (c) 2.00 THz. (d–f) The corresponding results of cis-AB.
2
7208 | Phys. Chem. Chem. Phys., 2018, 20, 27205--27213
This journal is ©the Owner Societies 2018

View Article Online
PCCP
Paper
a
Table 1 Calculated vibrational modes and corresponding experimental peaks of AB
Exp. (THz)
Isomer
22 1C
ꢁ190 1C
Cal. (THz)
Vibrational mode
trans-AB
0.98
1.22
1.44
2.00
2.35
Wagging of benzene ring of M1/2, translation of M
Wagging of benzene ring of M1/2, translation of M
0
1
.48
0
1
1
2
.39
.74
.28
1.65
1.98
2.34
0
Translation of M , weaker translation vibration of M1/2
Twisting of benzene ring with translation of NQN bond of M1/2, translation of M
Rotation of benzene ring of M1/2, twisting of benzene ring with translation of NQN bond of M
0
0
cis-AB
0.6
0.25
1.02
1.29
0.39
1.28
1.60
1.88
2.09
2.28
Rotation
0
1
.9
.27
Rotation perpendicular to benzene ring
Rotation along with translation
Rotation
Twisting of benzene ring and translation of NQN
Rotation
1
2
.64
.09
2.19
42.5
a
0
M stands for the molecules centered on (0, 0, 0), M1/2 stands for the molecules centered on (0, 0, 1/2).
decrease with increasing heating time. These changes are
accompanied with two new enhanced absorption peaks at 1.39
and 1.74 THz, which emerge after 1h. After 20 h, the peaks of cis-AB
vanished and the peaks of trans-AB reached their maximum.
Compared to that in polar solvent (Fig. 4(a and b)), the isomeriza-
tion rate of AB in non-polar solvents is much faster.
To illustrate the solvent polarity-dependent thermal-
isomerization rate of AB from cis-to-trans, the quantitative
relationship between the content of cis-AB and isomerization
time (Fig. 4(f)) was given by measuring the THz absorption
spectra of different ratios of cis and trans-AB mixtures as a
standard curve (Fig. 3(b)). From Fig. 4(f), it can be seen that the
Fig. 3 (a) The THz absorption spectra of the mixtures composed of
different ratios of cis and trans-AB. (b) Plot of the absorption coefficient
versus the ratio of cis and trans-AB in the mixture at 0.6 THz.
their maximum at 38 h, indicating that the cis-AB is completely cis-AB content decreased with increasing thermal-isomerization
isomerized into the trans-AB. Additionally, the THz absorption time in polar solvents (ethanol and acetone), and the cis-AB
spectra of the thermal cis-to-trans isomerization process of AB content is almost close to 0 at 38 h. However, a similar
in non-polar solvents (hexane and CCl
4
) were also explored and conversion process in non-polar solvents (hexane and CCl )
4
are depicted in Fig. 4(c and d), respectively. It can be seen that only takes around 20 h. The measured results indicate that the
the absorption peaks of cis-AB at 0.6, 0.9, 1.26 and 2.20 THz thermal isomerization reaction of AB is dependent on the
Fig. 4 THz absorption spectra of AB during thermal cis-to-trans isomerization in different polarity solvents (a) ethanol, (b) acetone, (c) hexane and
(
4
d) CCl ; (e) THz absorption spectra of AB during thermal cis-to-trans isomerization in ethanol under Au NP catalysis; (f) the ratio of cis-AB as a function
of thermal-isomerization time.
This journal is ©the Owner Societies 2018
Phys. Chem. Chem. Phys., 2018, 20, 27205--27213 | 27209

View Article Online
Paper
PCCP
polarity of the solvent, and that the rate in the non-polar radical cation formation and rotation about an activated NQN
3
8,39
solvent is faster than that in the polar solvent. The reasons bond.
for this are currently not quite clear.
It is worth noting that the effect of Au NPs in
3
3,34
It is known that the different solvents on catalysis is not given here because the
thermal isomerization of AB has two widely recognized major as-prepared Au NPs are hardly dispersed and were inclined to
1
mechanisms: rotation and inversion. We consider that the agglomerate together in other solvents.
rotation is the prevalent mechanism in polar solvents, while
3
5,36
Photoinduced isomerization of AB from cis-to-trans at room
temperature
inversion generally occurs in non-polar media.
Theoretical
calculations predict a significantly larger activation energy for
ꢁ1
rotation (36.2 kcal mol ) as compared to inversion (24.9 kcal The photoinduced cis–trans isomerization of AB irradiated with
ꢁ
1 37
mol ). Thus, the rate of thermal isomerization of AB in the visible (445 nm laser) light in polar (ethanol and acetone) and
non-polar solvent is faster than that in the polar solvent. non-polar (hexane and CCl ) solvents at room temperature was
4
Certainly, verifying this interpretation needs further investiga- studied. Fig. 5(a) shows the THz absorption spectra of the
tion, and a full understanding of this effect requires extensive vis-induced cis-to-trans isomerization in ethanol solution. It
theoretical and experimental work.
can be seen that the THz absorption peaks of cis-AB at 0.6,
So far, we have seen that the thermally induced isomeriza- 0.9, 1.26, and 2.22 THz decreased with increasing irradiation
tion could be monitored quantitatively by the THz-TDS method, time and meanwhile the absorption peaks of trans-AB at 1.74
although the heat-responsive conversion time is long. As such, and 1.39 THz emerged at 45 min and then increased gradually
we explored the catalytic effect of Au NPs upon the thermally with increasing irradiation time. Subsequently, the absorption
induced cis-to-trans isomerization of AB. As shown in Fig. 4(e), peak of cis-AB at 0.9 THz vanished, and the absorption peak at
it can be found that the THz absorption peaks of cis-AB at 0.6, 1.26 THz was obscured by the peak of trans-AB at 1.39 THz at
0.9, 1.26 and 2.22 THz gradually decrease and then completely around 90 min. After 90 min, the THz absorption spectra have
disappear after 6.5 h. Meanwhile, the absorption peaks of trans- no obvious changes and the irradiation of the cis-AB in ethanol
AB at 1.74 and 1.39 THz appeared and reached their maximum solution produced the photo-stationary state composed of
at 6.5 h. The quantitative relationship between the content of mixtures of cis- and trans-AB, as shown in Fig. 5(a). Fig. 5(f)
cis-AB and isomerization time was also investigated in Fig. 4(f). depicts the corresponding changes in the content of constituent
The content of cis-AB decreases with increasing thermal- cis-AB at different irradiation times: 94% at 15 min, 62% at
isomerization time under Au NP catalysis and is almost down 30 min, 43% at 45 min, 29% at 60 min, 20% at 90 min and 19%
to 0 after 6.5 h. The measured results show clearly that the rate at 150 min.
of thermal cis-to-trans isomerization of AB is significantly
Furthermore, Fig. 5(b) shows the THz absorption spectra of
accelerated in the presence of Au NPs. Compared to that in the photo-isomerization process of AB in acetone solutions.
the absence of Au NPs, a nearly 6-fold acceleration in the Likewise, the THz absorption peaks of cis-AB decreased and the
thermal cis-to-trans isomerization of AB is achieved. It is absorption peaks of trans-AB appeared and increased with
ascribed to the participation of Au NP-mediated electron trans- irradiation time. After 90 min, the absorption peak at 0.9 THz
fer in the thermal isomerization of AB, in which Au NPs of cis-AB disappeared, and the THz absorption of 0.6, 1.39 and
mediate electron transfer in the isomerization, followed by AB 1.74 THz were invariant and photo-isomerization of AB reached
Fig. 5 THz absorption spectra of photo-isomerization of AB from cis-to-trans in different polarity solvents (a) ethanol, (b) acetone, (c) hexane and
(
d) CCl
4
; (e) THz absorption spectra of photo-isomerization of AB from cis-to-trans in ethanol under Au NP catalysis; (f) the ratio of cis-AB as a function of
the photo-isomerization time.
2
7210 | Phys. Chem. Chem. Phys., 2018, 20, 27205--27213
This journal is ©the Owner Societies 2018

View Article Online
PCCP
Paper
Fig. 6 THz absorption spectra of AB in photo-isomerization from cis-to-trans at 50 1C in different polarity solvents (a) ethanol, (b) acetone, (c) hexane
and (d) CCl ; (e) the ratio of cis-AB as a function of the thermal-isomerization time.
4
the photo-stationary state. In addition, the THz absorption spectra peaks of cis-AB at 0.6, 0.9, 1.26 and 2.20 THz decrease and
of the photo-isomerization process of AB from cis-to-trans in non- meanwhile the peaks of trans-AB at 1.39 and 1.74 THz emerged
polar solvents (hexane and CCl ) were plotted in Fig. 5(c and d), and increased gradually with time. The whole THz absorption
4
respectively. Similarly, the whole THz absorption spectra of the AB spectra of the AB became invariant after 90 min of irradiation
were almost invariant after 90 min of irradiation. Compared to under heating. Fig. 6(b–d) show the isomerization of AB from
that in polar solvent, the photoinduced conversion of AB from cis-to-trans in acetone, hexane and CCl solutions at 50 1C, respec-
4
cis-to-trans shows no significant difference in non-polar solvents. tively. Similarly, the THz absorption peaks of cis-AB at 0.6, 0.9, 1.26
The quantitative relationship between the content of cis-AB and and 2.20 THz decrease and meanwhile the peaks of trans-AB at
irradiation time is given in Fig. 5(f). We observed that the content 1.39 and 1.74 THz emerged and increased gradually with increasing
of the cis-AB decreased and then approached asymptotically the isomerization time. The THz absorption spectra of the AB are
same photo-stationary composition (20 ꢂ 5% of cis-AB) with invariant after 90 min of irradiation. To further confirm the effects
increased irradiation time in the four solvents at room tempera- of temperature on the photoreaction rate and the photo-stationary
ture. It means that the same photo-isomerization equilibrium is state of AB, we prolong the irradiation time of cis-AB in hexane
established between the cis- and trans-isomers in different polarity solution to 26 h at 50 1C, as shown in Fig. 6(c). It can be seen that
solvents, which indicated that the effect of solvents on photo- the THz absorption spectrum of the AB under irradiation for 26 h
40,41
isomerization of AB from cis-to-trans is almost negligible.
is consistent with that under 90 min. This indicates that the photo-
Additionally, studies of Au NP catalysis on the photoinduced isomerization of AB from cis-to-trans can establish a photo-
isomerization of AB from cis-to-trans in ethanol solution at stationary state after 90 min. The quantitative relationship
room temperature were carried out. As shown in Fig. 5(e), with between the content of cis-AB and isomerization time was plotted
Au NPs, the overall THz absorption spectra of the AB display in Fig. 6(e). It can be found that the cis-AB content decreased and
very similar behaviors to those without Au NPs, as well as the then approached asymptotically the same photo-stationary com-
quantitative relationship between cis-AB content and reaction position (20 ꢂ 5% of cis-AB) with increasing irradiation time in the
time (Fig. 5(f)). Therefore, the cis-AB content during the photo- four solvents at 50 1C. Compared to the photo-isomerization at
isomerization process, in the absence or presence of Au NPs, room temperature, we found that the temperature has no signifi-
was coincident and the Au NPs do not play a catalytic role in the cant influence on the photoreaction rate and the photo-
4
2
photo-isomerization process. According to the literature, the isomerization equilibrium of AB, which was due to the tempera-
catalyst is not responsible for the photo-isomerization of AB.
ture coefficient of the photochemical reactions being relatively
43,44
small (close to 1).
Therefore, in our presented cases, the
Photoinduced isomerization of AB from cis-to-trans under
heating
heating effect is quite slight in photoinduced isomerization of AB.
The THz absorption spectra of the photo-isomerization process of
AB from cis-to-trans irradiated with visible light at 50 1C in the
above four solvents were also measured. Fig. 6(a) gives the THz
Conclusions
absorption spectra of the isomerization process of AB from cis-to- We have presented herein how the THz-TDS technique could
trans in ethanol solution. It can be seen that the THz absorption quantitatively monitor the thermally induced and photoinduced
This journal is ©the Owner Societies 2018
Phys. Chem. Chem. Phys., 2018, 20, 27205--27213 | 27211

View Article Online
Paper
PCCP
isomerization processes of AB. By utilizing THz-TDS, we demon- 13 M. Quick, A. L. Dobryakov, M. Gerecke, C. Richter, F. Berndt,
strate that the rate of thermal cis-to-trans isomerization of AB in
I. N. Ioffe, A. A. Granovsky, R. Mahrwald, N. P. Ernsting and
non-polar solvent is faster than that in polar solvent; and Au NPs
S. A. Kovalenko, J. Phys. Chem. B, 2014, 118, 8756–8771.
as a catalyst can effectively promote the thermal cis-to-trans 14 C. M. Stuart, R. R. Frontiera and R. A. Mathies, J. Phys.
isomerization of AB; we also provide evidence that the solvent
Chem. A, 2007, 111, 12072–12080.
polarity, temperature and Au NPs as a catalyst don’t affect the 15 R. D. Curtis, J. W. Hilborn, G. Wu, M. D. Lumsden,
photo-isomerization of AB obviously. THz-TDS enables quantita-
tive distinguishing of the conversion state from one isomer to the
R. E. Wasylishen and J. A. Pincock, J. Phys. Chem., 1993,
97, 1856–1861.
other that occurs in response to light or heat and thus offer new 16 V. Jelic, K. Iwaszczuk, P. H. Nguyen, C. Rathje, G. J. Hornig,
insights into the isomerization processes. Given the extraordinary
sensitivity of this technology, we believe the decisive and non-
H. M. Sharum, J. R. Hoffman, M. R. Freeman and F. A.
Hegmann, Nat. Phys., 2017, 13, 591–598.
destructive THz-TDS provides exciting possibilities for investiga- 17 S. P. Delaney and T. M. Korter, J. Phys. Chem. A, 2015, 119,
tions and better understanding of the conformational dynamics of
molecules.
3269–3276.
18 N. Goubet, A. Jagtap, C. Livache, B. Martinez, H. Portales,
X. Z. Xu, R. Lobo, B. Dubertret and E. Lhuillier, J. Am. Chem.
Soc., 2018, 140, 5033–5036.
1
9 P. H. Siegel, IEEE Trans. Microwave Theory Tech., 2004, 52,
438–2447.
20 M. Naftaly and R. E. Miles, Proc. IEEE, 2007, 95, 1658–1665.
Conflicts of interest
2
There are no conflicts of interest to declare.
2
1 B. Fischer, M. Hoffmann, H. Helm, G. Modjesch and
P. U. Jepsen, Semicond. Sci. Technol., 2005, 20, S246.
2 J. Turkevich, P. C. Stevenson and J. Hillier, Discuss. Faraday
Soc., 1951, 11, 55–75.
2
Acknowledgements
This work was supported by the National Key Research and 23 C. J. Brown, Acta Crystallogr., 1966, 21, 146–152.
Development Program of China (No. 2017YFA0701004), and the 24 G. C. Hampson and J. M. Robertson, J. Chem. Soc., 1941,
National Natural Science Foundation of China (Grant No.
1875150, 61735012, 61420106006, 61427814 and 61605143).
409–413, DOI: 10.1039/JR9410000409.
25 D. Grischkowsky, S. Keiding, M. van Exter and C. Fattinger,
J. Opt. Soc. Am. B, 1990, 7, 2006–2015.
6
26 J. Han, Z. Zhu, S. Ray, A. K. Azad, W. Zhang, M. He, S. Li and
Y. Zhao, Appl. Phys. Lett., 2006, 89, 031107.
Notes and references
27 T. Beyer and S. L. Price, CrystEngComm, 2000, 2, 183–190.
1
H. M. D. Bandara and S. C. Burdette, Chem. Soc. Rev., 2012, 28 Y. C. Shen, P. C. Upadhya, E. H. Linfield and A. G. Davies,
1, 1809–1825.
Appl. Phys. Lett., 2003, 82, 2350–2352.
C. J. Saint-Louis, Control of molecular geometries using new 29 M. Takahashi, N. Okamura, X. Fan, H. Shirakawa and
4
2
photo-electro-switchable azobenzenes, The University of Alabama,
015.
G. S. Hartley, Nature, 1937, 140, 281.
H. Minamide, J. Phys. Chem. A, 2017, 121, 2558–2564.
30 T. Pan, S. Li, T. Zou, Z. Yu, B. Zhang, C. Wang, J. Zhang, M. He
and H. Zhao, Spectrochim. Acta, Part A, 2017, 178, 19–23.
2
3
4
M. A. Kienzler, A. Reiner, E. Trautman, S. Yoo, D. Trauner 31 S. Zong, G. Ren, S. Li, B. Zhang, J. Zhang, W. Qi, J. Han and
and E. Y. Isacoff, J. Am. Chem. Soc., 2013, 135, 17683–17686.
H. Zhao, J. Mol. Struct., 2018, 1157, 486–491.
M. Dong, A. Babalhavaeji, C. V. Collins, K. Jarrah, O. Sadovski, 32 M. Walther, B. M. Fischer and P. Uhd Jepsen, Chem. Phys.,
5
Q. Dai and G. A. Woolley, J. Am. Chem. Soc., 2017, 139,
3483–13486.
K. G. Yager and C. J. Barrett, J. Photochem. Photobiol., A,
006, 182, 250–261.
Z. F. Liu, K. Hashimoto and A. Fujishima, Nature, 1990,
47, 658.
Z. Li, M. Wang, H. Li, J. He, N. Li, Q. Xu and J. Lu, J. Mater.
Chem. C, 2017, 5, 8593–8598.
Y. Norikane and N. Tamaoki, Org. Lett., 2004, 6, 2595–2598.
2003, 288, 261–268.
1
33 N. Norio, S. Toshinobu, Y. Hideyuki, I. Etsuko, Y. Shunzo
and H. Shigeo, Bull. Chem. Soc. Jpn., 1976, 49, 1381–1387.
34 F. Serra and E. M. Terentjev, Macromolecules, 2008, 41,
981–986.
35 G. Angelini, N. Canilho, M. Emo, M. Kingsley and C. Gasbarri,
J. Org. Chem., 2015, 80, 7430–7434.
6
7
8
9
2
3
36 P. De Maria, A. Fontana, C. Gasbarri, G. Siani and
P. Zanirato, ARKIVOC, 2009, 8, 16–29.
1
1
1
0 T. Muraoka, K. Kinbara and T. Aida, Nature, 2006, 440, 37 N. K. Joshi, M. Fuyuki and A. Wada, J. Phys. Chem. B, 2014,
12–515.
118, 1891–1899.
1 A. A. Beharry and G. A. Woolley, Chem. Soc. Rev., 2011, 40, 38 G. L. Hallett-Tapley, C. D’Alfonso, N. L. Pacioni, C. D.
5
4422–4437.
McTiernan, M. Gonzalez-Bejar, O. Lanzalunga, E. I. Alarcon
2 M. R. Molla, P. Rangadurai, L. Antony, S. Swaminathan,
and J. C. Scaiano, Chem. Commun., 2013, 49, 10073–10075.
J. J. de Pablo and S. Thayumanavan, Nat. Chem., 2018, 10, 39 S. Simoncelli and P. F. Aramend ´ı a, Catal. Sci. Technol., 2015,
59–666. 5, 2110–2116.
6
2
7212 | Phys. Chem. Chem. Phys., 2018, 20, 27205--27213
This journal is ©the Owner Societies 2018

View Article Online
PCCP
Paper
4
4
0 G. S. Hammond, J. Saltiel, A. A. Lamola, N. J. Turro, 42 D. Shin, J. H. Kang, K.-A. Min, S. Hong and B. Hee Hong,
J. S. Bradshaw, D. O. Cowan, R. C. Counsell, V. Vogt and
C. Dalton, J. Am. Chem. Soc., 1964, 86, 3197–3217.
APL Mater., 2014, 2, 092501.
43 W. D. Bancroft and R. P. Allen, Proc. Natl. Acad. Sci. U. S. A.,
1929, 15, 445–448.
1 G. Zimmerman, L. Y. Chow and U. J. Paik, J. Am. Chem. Soc.,
1958, 80, 3528–3531.
44 G. B. Kistiakowsky, Proc. Natl. Acad. Sci. U. S. A., 1929, 15, 194–197.
This journal is ©the Owner Societies 2018
Phys. Chem. Chem. Phys., 2018, 20, 27205--27213 | 27213