Photophysics, photochemistry and thermal stability of diarylethene-containing benzothiazolium species

Article abstract of DOI:10.1016/j.jphotochem.2014.12.015

The photophysics, photochemistry and thermal stability of four 1-aryl-2-(N-methyl-2-benzothiazolium) ethene iodides (aryl: phenyl, 1-naphthyl, 9-phenanthryl and 9-anthryl) were studied. Although the absorption spectra are found to be hypsochromic-shifted, fluorescence spectra are bathochromic-shifted. The dipole moment in the relaxed excited state was found to be larger than that in the ground state. To investigate effects of N-methylation and the aryl-ring size, a detailed comparison was made between those in the present work (of charged compounds) and previous studies of their neutral analogues, with computed electron affinity and ionisation potentials serving to rationalise the experimentally observed bathochromic shifts in absorption and emission spectra. The kinetics of thermal isomerisation depend strongly on the nature of the aryl moiety and solvent; the larger the aryl ring, the slower the rate of isomerisation. The fastest isomerisation process was found to take place in MeOH. The anthryl derivative did not isomerize either by light- or heat-exposure, due to high energy barriers of rotation around the ethenic bond. Based on the significant blue-shift of the Z-isomer absorption maximum relative to that of the E-isomer, and the high percentage of Z-isomers in the photostationary state, these compounds may serve as potential promising candidates for optical data-storage applications.

Full text of DOI:10.1016/j.jphotochem.2014.12.015

Accepted Manuscript
Title: Photophysics, Photochemistry and Thermal Stability of
Diarylethene-Containing Benzothiazolium Species
Author: Morad M. El-Hendawy Tarek A. Fayed Mohamed K.
Awad Niall J. English Safaa Eldin H. Etaiw Ahmed B. Zaki
PII:
S1010-6030(15)00004-0
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.jphotochem.2014.12.015
JPC 9818
To appear in:
Journal of Photochemistry and Photobiology A: Chemistry
Received date:
Revised date:
Accepted date:
14-8-2014
3-11-2014
31-12-2014
Please cite this article as: Morad M.El-Hendawy, Tarek A.Fayed, Mohamed
K.Awad, Niall J.English, Safaa Eldin H.Etaiw, Ahmed B.Zaki, Photophysics,
Photochemistry and Thermal Stability of Diarylethene-Containing Benzothiazolium
Species, Journal of Photochemistry and Photobiology A: Chemistry
http://dx.doi.org/10.1016/j.jphotochem.2014.12.015
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Photophysics, Photochemistry and Thermal Stability of Diarylethene-
Containing BenzothiazoliumSpecies
Morad M. El-Hendawy1,2,3,Tarek A. Fayed 1, Mohamed K. Awad*1, Niall J. English*2, SafaaEldin H.
Etaiw,1 Ahmed B. Zaki1
1 Chemistry Department, Faculty of Science, Tanta University, Tanta, Egypt
2The SFI Strategic Research Cluster in Solar Energy Conversion and the Centre for Synthesis and
Chemical Biology, School of Chemical and Bioprocess Engineering, University College Dublin,
Belfield, Dublin 4, Ireland
3Chemistry Department, Faculty of Science, Kafrelsheikh University, Kafrelsheikh, Egypt
Corresponding authors:
Prof. Mohamed K. Awad, E-mail: mohamed.awad1@science.tanta.edu.eg
Dr. Niall J. English, Email: niall.english@ucd.ie
Highlights
 Ring-size and solvent effects on photophysics, photo- and thermal-isomerisation of
diaylethene-containing benzothiazolium moieties
 Effect of N-methylation on photo-properties neutral dyes.
 Investigation of mechanism of cis-trans thermal isomerisation reaction.
Abstract
The photophysics, photochemistry and thermal stability of four 1-aryl-2-(N-methyl-2-
benzothiazolium)ethene iodides (aryl:phenyl, 1-naphthyl, 9-phenanthryl and9-anthryl) were
studied.Although the absorption spectra are found to behypsochromic-shifted, fluorescence spectra are
bathochromic-shifted. The dipole moment in the relaxed excited state was found to be larger than that
in the ground state.To investigate effects of N-methylation and the aryl-ring size, a detailed
comparison was made between those in the present work (of charged compounds) and previous
studies of their neutral analogues, with computed electron affinity and ionisation potentials serving to
rationalise the experimentally observed bathchromic shifts in absorption and emission spectra. The
kinetics of thermal isomerisation depend strongly on the nature of the aryl moiety and solvent; the
larger the aryl ring, the slower the rate of isomerisation. The fastest isomerisation process was found
1

to take place in MeOH. The anthryl derivative did not isomerize either by light- or heat-exposure, due
to high energy barriers of rotation around the ethenic bond. Based on the significant blue-shift of the
Z-isomer absorption maximum relativeto that of the E-isomer, and the high percentage of Z-isomers in
the photostationary state, these compounds may serve as potential promising candidates for optical
data-storage applications.
Keywords: Benzothiazolium; Isomerisation
1- Introduction
Photo-induced Z-E isomerisation about double bonds has long been a subject of intense research,
motivated by either pure-chemistry or more applied, industrial, perspectives. For instance,
photoisomerisation about a carbon-carbon double bond in the rhodopsin chromophore is the initial
trigger for light recognition in human eyes [1]. The photoisomerisation of urocanic acid acts as a
‘natural sunscreen’, protecting DNA from photo-damage [2]. Owing to photoisomerisation,
diarylethenesare a popular example of photo-isomerizable molecules; they have found widespread
application as sensitisers and other additives in the photographic industry [3-7],ion recognition [8],as
optical-recording media in laser disks [9], as flexible dyes [10], laser dyes [11], and, indeed, as optical
sensitisers in various other fields [12–15]. It has been reported that this class of dyes act as imaging
agents for Alzheimer disease [16], as effective biological markers [17], and as promising candidate
dye sensitisers in solar cells [18].Azastilbenes and their quaternary salt, stilbazolium, constitute good
representative examplesofthe diarylethene family (cf. Fig. 1). Motivated by the desire to make further
progress into potential modern technological applications, the synthesis of new photo-isomerisable
molecules and characterisation of their photophysics, photochemistry and thermal stability in different
environments are sine qua non.
The Z↔E photoisomerisation of neutral azastilbene derivatives has been the subject of several
studies [19-21]. The introduction of a positive charge (upon quaternisation) into them give rise to a
new phenomenon, driven and controlled by heat, namely thermal Z→E isomerisation [20, 22-29]; this
is the ‘dark’ counterpart of the E→Z photoisomerisation reaction (cf. Fig. 1). Quaternary salts of 1-
alkyl-4-(-4’-R-styryl) quinolinium iodide (e.g., R = NO2, H or CH3) exhibit thermal Z→E
isomerisation at room temperature [24-26], whilst neutral analogues do not [19]. For stilbazolium
salts, the rate constant has been found to depend significantly on the polarity of the solvent and to a
smaller extent on the substituent in the 4-position of the styrene ring, as well as on the nature of the
anion [26].
2

About two decades ago, we studied, in various solvents, the N-methylation of 1-(2-naphthyl)-2-(2-
benzothiazolyl)ethene and the kinetics of thermal Z→E isomerisation. Recently, we observed N-
methylation of 1-(9-phenanthryl)-2-(2-benzothiazolyl)ethene (9-PhBE) [21, 22], where the N-
methylated form(9-PhBEI, Fig. 2) back-isomerises thermally [26], whilst, intriguingly, the neutral
analogue does not [22]. We also established the effect of N-methylation on the photophysics on 1-(9-
anthryl)-2-(2-benzothiazolyl)ethene (9-ABE) and its quaternary salt (9-ABEI, Fig. 2)[30];dual
emission was observed for both compounds, which was found to originate from the involvement of
locally-excited (LE) states and either twisted intramolecular charge transfer (TICT) or intramolecular
exciplex formation in the excited state, respectively.Previously [22], we studied the effect of aryl ring
size on the photophysics and photochemistry of styrylbenzothiazole (PBE) via systematic replacement
of the phenyl ring of 1-phenyl-2-(2-benzothiazolyl)ethene with naphthyl (1-NBE) and phenanthryl (9-
PhBE) rings. Increasing the size of aromatic rings served to increase the extent of ICT between the
benzothiazolyl and aryl fragments. This was confirmed by red shifts of their various spectra, along
with increases in their dipole moment and Stokes shift. The percentage of the Z-isomers due to
photoisomerisation was found to be size-dependent for the given solvent and irradiation
wavelength;this was found to be in the order PBE > 9-PhBE > 1-NBE [22]. In interesting work by
other investigators, the solvent and substituent effects on the photophysics ofstyryl-
ethylbenzothiazolium derivatives, possessing different electron-withdrawing or electron-donating
groups,have also been studied [31, 32]. The formation of aggregated structures was also observed at
higher concentrations of the benzothiazolium bromides [31].
Bearing in mind this photochemical activity and previous studies of similar molecular systems as
discussed above, we seek to study in the present work the photophysics and photo-/thermo-chemical
aspects of back E↔Z isomerisation, motivated by the tantalising possibility of realising temperature-
controlled photochromic end-applications. In particular, important goals are the investigation of the
effect of replacing the phenyl ring by larger polycyclic aromatic rings, as well as the nature of solvent
on spectral characteristics and rate of photo- and thermal- isomerisation. We also use Density
Functional Theory (DFT) to study the mechanism of thermal isomerisation. Finally, and importantly,
we report a concise comparison between the photophysics, photo-and thermal reactivities of the tested
(charged) compounds and their neutral analogues.
2- Experimental and computational details
The molecular structure and abbreviations of the compounds under scrutiny are presented in Fig. 2.
The 1-aryl-2-(N-methyl-2-benzothiazolium)ethene iodides were obtained simply by heating equimolar
3

quantities of the neutral analogues [22]with methyl iodide in ethanol for one hour under reflux. The
products were filtered and purified by recrystallisation twice from dry ethanol. Following this, the
structure was confirmed by elemental analysis, IR and UV–Vis spectral measurements. The melting
points of PBEI, 1-NBEI,9-PhBEI, and 9-ABEI are 208-210, 222-226, and 181–183, and 87-89 °C,
respectively. All synthesised compounds were found to be E-isomers.
Spectroscopic-grade MeOH, EtOH, CH3CN, CH2Cl2, CHCl3 and Et-Glycol (Aldrich or Merck) were
used, while water was double-distilled. All solvents were non-fluorescent within the range of
fluorescence measurements. Steady-state electronic absorption spectra were recorded on a Shimadzu
UV-3101 PC spectrophotometer using 1.0 cm-matched silica cells, while steady-state fluorescence
spectra were recorded using a Perkin-Elmer LS 50B scanning spectrofluorometer.Typical
concentrations of 2 ×10−5 M were used for the measurement of the investigated compounds.Although
the fluorescence quantum yields (Φf) were determined at room temperature relative to quinine
bisulphate in 0.1 N H2SO4 (Φf = 0.515)[33], with standard-yield approach of Meech and Phillips [34],
because of the difficulty of comparing the fluorescence quantum yield with two compounds (9-PhBEI
and 9-ABEI) exhibiting dual fluorescence [21 30], a straightforward Φf comparison is very
problematic; therefore, Φfvalues are not reported here.The experimentaluncertainty was ±10%. The
photoisomerisation quantum yields as well as the composition of the photostationary states (PSS) were
calculated as described previously [20], with experimental error of ±10%. The light intensity was
determined using ferrioxalateactinometry [35]. For the inferred activation-energy parameters from
experimental rate data, the experimental error was roughly ±10%, determined from a number of
independent measurements for selected dyes.
The thermal ZE isomerisation of these dyes were tracked spectrophotometrically in solvents in
which the Z-forms are thermally unstable. A solution of each dye was prepared freshly in the dark,
and was irradiated subsequently at 365, 405 and 436 nm light (Mercury Lamp with an intensity of 4.3
x 10-7 E s-1 cm-2, 30 W) for a sufficient period until the PSS was reached. At the PSS, a maximum
population of the Z-isomers was generated. After switching off the irradiation, the sample was
moved quickly to a temperature-controlled spectrophotometer cell holder where the Z-isomer
started to revert back to the E-form in the dark at the desired temperature.This was confirmed by
observing that the spectrum of the PSS reverted to the original (recorded for the fresh sample). This
procedure was repeated at four different temperatures,i.e., 15, 20, 25 and 30◦C.
All DFT calculations were performed using the Gaussian 09 suite [36]. Structural optimisations were
performed using B3LYP/6-31+G(d,p) [37-39], andthe activation parameters of the thermal
isomerisation reactions were calculated. To locate the transition states efficiently, the PM6 level of
4

semi-empirical treatment [40], followed by B3LYP,were used. Vibrational-frequency calculations
were carried out in all cases to confirm the stationary points as either minima or first-order saddle
points, and to obtain the activation parameters. The calculations have been performed under bulk
solvent effects of methanol using the universal solvation model, SMD [41]. Further details on locating
the transition states (TSs) are provided in Supporting Information [42], and Cartesian coordinates of
the TSs are provided therein (cf. Table S1).
3- Results and discussion
3.1 Spectroscopic characteristics
The absorption and fluorescence spectra of 1-aryl-2-(N-methyl-2-benzothiazolium)ethene iodides
(ArBEIs) were measured in H2O, MeOH, EtOH, MeCN, CH2Cl2, and CHCl3. The relevant
spectroscopic data of PBEI and 1-NBEI are gathered in Table 1, while those of 9-PhBEI and 9-ABEI
have already been published in refs. 21 and 30.The absorption spectra of ArBEIs in three different
solvents, namely H2O, MeOH and CHCl3, are presented in Fig. 3. The spectra show a broad band,
with maxima between 371 and 528 nm, usually in and around ~420 nm, depending on the solvent and
the size of the aryl moiety. One absorption maximum for PBEI was observed at around 225 nm while
in the case of 1-NBEI, two absorption bands were observed around 280 and 320 nm. More detailed
information about the spectroscopic characteristics of 9-PhBEI and 9-ABEI have been reported
previously [21, 30]. For the present spectra, although it could perhaps be argued (albeit somewhat
implausibly) that this corresponds to a charge-transfer (CT) transition from the aryl to the
benzothiazolium moiety, as could those of other absorption bands at shorter wavelengths for local
transitions in the aryl or benzothiazolium fragment, these transitions are much more likely to originate
from electron transfer from the iodide ion to the benzothiazolium ion in an intimate ion pair. This is
because of the key rôle of the iodide counterion in promoting aggregation.In chloroform, the solutes
are expected to exist exclusively as ion pairs; indeed, in their seminal study of charge transfer in
pyridinium iodides, Kosoweret al selected chloroform as the solvent for this reason, and identified the
two lowest-energy electronic transitions as CT bands for transfer from I- to N-MePy+, withthe exact
correspondence of separation between the CT bands to two energy states of the iodine atom supporting
that assignment [43]. In the case of the present spectra (cf. Fig. 3), their similarity in all solvents
suggests a common origin for the spectroscopic transition – likely to be electron transfer from the
iodide to the pyridinium ion.The present ArBEIs exhibit negative solvatochromic behaviour:
specifically, on going fromCHCl3 to H2O, the maxima of the CT-bands of PBEI, 1-NBEI, 9-PhBEI
5

and 9-ABEI blue-shift by 14, 34, 42 and 55 nm, respectively. Such behaviour reflects a decrease in the
dipole moment upon photo-excitation to the Frank-Condon state. Dyes with similar structures have
been reported to exhibit this behaviour [20, 32, 44-46].On the contrary, the emission maxima of
ArBEIswerered-shifted with increasing the solvent polarity (cf.Fig. 4). The spectral shift ranges from
7 to 33 nm, on going from CHCl3 to water, depending on the size of the aryl moiety. This isdue to the
increase in dipole moment of the relaxed excited singlet state,relative to that of the ground state. From
inspection of Figs. 3 and 4, it is evident that the emission spectra of PBEI, 1-NBEI and 9-ABEI are
‘mirror images’ of their corresponding absorption spectra;this similarity arises from electronic
excitation not altering greatly the nuclear geometry. Hence, the spacing of the vibrational energy
levels of the excited state is similar to that of the ground state. However, the shape of the fluorescence
spectrum of 9-PhBEI deviates from those of the absorption spectrum, indicating a different geometric
arrangement of nuclei in the excited and ground states.Dual emission from 9-PhBEI has been observed
and presented in ref. [21].Also, dual emission was observed for ABEI as shown in Fig. 4. One
fluorescence band around 400 nm (upon excitation ~365 nm) originates from the involvement of
locally-excited (LE) states, whilst the other, separated relative red-shifted band around 500 nm (upon
excitation,~430 nm) originates from the intramolecular exciplex formation in the excited state; for
further details, see ref. [30].The absorption and the locally excited band of the fluorescence spectra of
the ABEI dye are structured with a vibronic progression characteristic of the anthrylchromophore
[30].The localisation of excitation energy at the anthrylchromophorearises as a result of the non-planar
geometry.
The difference in energy between the absorbed and emitted maxima is known as the Stokes shift
(e.g., expressed in nanometres), which serves as a measure to the change in the magnitude of a
chromophore’s dipole moment following excitation; these are listed in Table 1 and Table S2.
Evidently, the Stokes shift increases with solvent polarity. The remarkably large magnitudes(ranging
from 63 up to 132 nm) of the Stokes shifts are attributed to the charge-transfer nature of the long-
wavelength transition and indicates that the relaxed excited state dipole moment is larger than that of
the ground state.The Stokes shift is ring-size dependent,i.e.,in general, the largerthe hydrocarbon
moiety’s size in a dye, the larger the Stokes shift.
The fluorescence excitation spectra of the respective emission bands of the investigated compounds
were measured in various solvents.The maxima of the excitation spectra are not particularly dependent
on the nature of the solvent (cf. Table 1).The excitation spectra of PBEI, 1-NBEI and 9-ABEI [30] do
not display any changes in their shapes, except for changes of intensity at different emission
wavelengths. They resemble each other closely in the lower energy absorption band. This reflects that
6

there is only one species in their ground states. However, 9-PhBEI exhibited two distinct excitation
spectra [21] at f = 520 and 600 nm, due to the formation of ground-state J-aggregate from monomer
molecules.Also, the oscillator strength is not affected greatly by the nature of solvent, but it was
observed to decrease as the size of aryl subunit increases.
3.2 Photochemical activity
Except for 9-ABEI, the photochemical E-Zisomerisations of 1-aryl-2-(N-methyl-2-
benzothiazolium)ethene iodides were carried out at room temperature only in MeOH and H2O due to
the fast backward thermal ZEisomerisation of the irradiated solution in the other solvents such as
EtOH and MeCN. This was confirmed by the finding that the PSS absorption spectra reverted
completely back to the original spectra at rates depending on the solvent and temperature, as reported
previously[20, 21], indicating that the thermal Z-E reaction competes with photoisomerisation.
The photochemical parameters of the investigated compounds are gathered in Table 2. The results
depend on ring size, solvent characteristics and excitation wavelength.As a representative example, the
photochemical parameters of 1-NBEI in MeOH at λirr = 405 nmare described in some detail. Changes
characteristic to EZ isomerism were observed on irradiating a methanol solution of this compound
as illustrated in Fig. 5a. These changes were a reduction in the intensity of the characteristic E-band
accompanied by a small hypsochromic shift (from 417 to 404 nm), and an increase in the intensity of
the absorption in the cissoid region of the spectrum [47]. Moreover, the spectrum of the final isomeric
mixture at PSS indicates a very high degree of conversion of E-isomer, Fig. 5a. Similar behavior was
observed for the other tested dyes (cf. Fig. S1).
The difficulty in monitoring the photoisomerisation of 9-PhBEI in water is attributed to extensive
aggregation even with very dilute concentrations, as reported previously [21]. Three pieces of spectral
evidence suggest the uniformity of the photo-isomeric reaction: (i) firstly,the intensity of the
fluorescence spectrum of 1-NBEI in MeOH diminishes at the PSS, upon irradiation at 436 nm, without
appearance of new bands, as illustrated in Fig. 5b -similar behavior was observed in the other dyes,so
the presence of the non-fluorescent Z-form is most probablyresponsible for the reduction of the
emission intensity; (ii) secondly, the clean interconversion of E-1-NBEI in MeOH during irradiation at
405 nm to the Z-form, without any side reaction, is indicated by the presence of two isosbestic points
at 333 and 275 nm in the UV-Vis spectra, as shown in Fig. 5a; (iii) thirdly, agraphical plot between the
change of absorbance of long-wavelength maximum and that of the maximum in the cissoid region
(A/A diagram) during the period of irradiation exhibitsa linear relationship for 1-NBEI (cf. Fig. 5c). In
7

any event, the linear relationship in Fig. 5c would be expected: the isosbestic points in Fig.5a establish
that this is a two-component closed system.
Fig. 5ddepicts the time-dependence of the absorption of 1-NBEI in water at their maxima during the
irradiation period. For example, upon irradiation at 405 nm, the absorbance at the maximum for NBEI
decreases gradually as the irradiation time proceeds until reaching a constant value confirming the
achievement of PSS.Fig. 6aillustrates the calculated absorption spectra according to Fischer’s method
[48] of the Z-isomer of 1-NBEI in MeOH, as well as those of the corresponding E-isomer. As can be
seen, the absorption maximum of the photoproduct is displaced towards shorter wavelengths than that
of the E-isomer.PBEI and 9-PhBEI showed similar behavior(cf. Fig. S2 [42]).
The percentage of Z-isomer of 1-NBEI at its PSS is increased from 85 % in MeOH to 99.3 % in H2O
upon irradiation at 405 nm. Also, it is found that the larger the ring size of the aryl moiety, the higher
the percentage of Z-isomer. For example, the replacement of phenyl ring by naphthyl or phenanthryl
ring, the % Z increases from 79.1 % to 85 and 89 % in MeOH, respectively, upon irradiation at 405
nm. In addition, the effect of irradiation wavelength on % Z of ArBEIs is clear. In MeOH, the shorter
the irradiation wavelength is the lower the % Z at the PSS. Fig. 6b shows an example for the growth of
% Z with increasing irradiation time until reaching a steady value at the PSS. The high percentage of
composition of Z-isomer at the PSS, coupled with the blue shift of the Z-isomer absorption maximum
vis-à-vis the E-isomer, suggests that these compounds can serve as candidates for data-storage
applications.
Table 2 illustrates that the photoisomerisation quantum yields, EZ and ZE, are high for all dyes
vis-à-vis their neutral analogues [22]. EZ and ZE are comparable in some cases, particularly for
PBEI;at first glance, this may suggest the importance of a singlet mechanism [49], although triplet-
state photoisomerisation remains quite possible [50]. In most cases, the EZ is higher than ZE; this
may be attributed to solvent stabilisation of the Z-isomer, which discourages the formation of E-
isomer from a ‘phantom’ state[51-53]. An exception was found for PBEI in water, where ZE is
higher than EZ, particularly upon irradiation at 405 nm. This behavior can be explained by solvent
stabilisation, where the E-isomer encounters substantial stabilisation through solvation, which reduces
its tendency to isomerize into the Z-form. It is worth mentioning that the highest quantum yield
(0.687) was observed in the case of E→Zisomerisation of 1-NBEI at 436 nm in water. On the other
hand, upon irradiation at 405 nm, the partial rate constants of the forwardisomerisation, kEZ, follow
the trend 1-NBEI > 9-PhBEI > PBEI in methanol,whilst those of the backward kZE reactionsadopt
the 1-NBEI > PBEI > 9-PhBEI. This reflects the fact that 1-NBEI exhibits the fastest rotation of side
moieties around the ethenic bond for E-Z interconversion, and, consequently, is the most kinetically
8

favourable.From a geometric viewpoint, the size of the benzothiazolium moiety is closerin size tothe
naphthyl one than those of phenyl or phenanthryl. Accordingly, this geometric similarity serves to
facilitate mechanical rotation.The photo-stationary state’s composition depends strongly on the
irradiation wavelength, as seen previously in photo-reversible systems [22,49]; in other words, the
shorter the irradiation wavelength, the lower the % Z in the photo-stationary state. Generally, the other
parameters display a different dependence where shorter-wavelength irradiation induces rapid
E↔Zinterconversion with a high quantum yield, with some exceptions in the case of 1-NBEI.
3.3 Thermal Stability
3.3.1 Solvent effects on thermal ZE isomerisation
Except for 9-ABEI, the investigated compounds exhibit thermal ZE isomerisation in MeCN,
MeOH and EtOH. However, in other solvents, such as water, these dyes are thermally stable, even at
higher temperatures over very long time (hour-scales). Accordingly, this study focuses on three
solvents in which thermal ZE isomerisation takes place in a reasonable timeframe. Fig. 7a illustrates
the change in absorbance during thermal ZE isomerisation of PBEI in EtOH at 25 C was followed
at 374 nm. The appearance of an isosbestic point at 322 nm demonstrates the inter-conversion between
Z- and E-isomers. Similar behaviour was found for other compounds. More details about the
behaviour of 9-PhBEI dye have already published in ref. 21; details of the underlying methods for
calculation of parameters are provided in the Supplementary Information [42].
The influence of the solvent polarity on the rate constant and the activation parameters of the thermal
ZE isomerisation for the 1-NBEI dye shall be described in detail as a representative example. As
can be seen from Table 3, the rate constant is almost doubled upon raising the temperature from 15 to
30°C in MeOH (k = 0.88 and 1.54 min-1, respectively). In MeCN, the rate constant of thermal
isomerisation increases by about three orders of magnitude, but in EtOH, it increases by about four. In
general, the isomerisation rate constant for the studied compounds increases in the order
MeOH>MeCN>EtOH, except in the case of PBEI (MeOHEtOHMeCN), (cf. Table S3[42]). As
expected, a higher rate constant is associated with lower activation energy (cf. Table 3). Also, for 1-
NBEI, the activation energy increases from 26.9 to 69.1 kJ/mol,on going from MeOH to EtOH,
respectively. This explains why the rate of thermal isomerisation is faster in MeOH than that in EtOH,
and also shows the effect of the dielectric constant of the solvent in lowering the activation barrier in
this type of reaction. However, the E# value in MeOH is much smaller than in MeCN, demonstrating
the probability of forming hydrogen bond under the effect of temperaturewhich may facilitate thermal
9

isomerisation. The variation of the E# value takes place when the reactant is in a solvent environment
which destabilises or stabilises the reactant ground state vis-à-vis that of the transition state in the
respective solvents. Similar results were observed in the case of the other dyes, as illustrated in Table
S3 [42].
It has been suggested that reaction-generated electric charges exhibit negative entropies of activation
[48]. The observed negative entropies of activation in this study for thermal isomerisation (between -
94.1 and -167 J/mol.K) indicate that the transition states have a greater degree of charge separation
vis-à-vis the ground state. Because of their highly polar character, the transition state requires a greater
degree of ordering of the solvent molecules than the ground state [54]. The negative entropy of
activation in the case of 1-NBEI dye increases upon going from MeOH to MeCN. This suggests that
the reaction is more highly ordered in polar protic solvents in comparison to their polar aprotic
counterparts. On the other hand, the S value increases from MeOH to EtOH, reflecting the role of
the dielectric constant of the solvent in controlling the ordering of the system in both the ground and
transition states. Similar results are observed in the case of the remaining dyes, as illustrated in Table
S3 [42].
3.3.2 Effects of size of aryl moiety on the thermal ZE isomerisation
The effect of the replacement of the phenyl moiety of the styrylbenzothiazolium dye by larger
aromatic polycyclic ones on the rate of thermal ZE isomerisation was elucidated from the calculated
data in MeOH (cf. Table 4). The kinetic data reveals that the increase in the size of aryl moiety retards
the rate of ZE isomerisation. For example, the k values of the isomerisation in MeOH are 1.53,1.26
and 0.90 min-1 for PBEI, 1-NBEI and 9-PhBEI, at 22.5C (the average experimental temperature),
respectively.This can be interpreted on the basis of the rotation-energy barriers in these dyes,
computed from the Arhenius equation (Supplementary Information, eqn. 2). Ongoing from PBEI to 9-
PhBEI, the rotational barrier increases from 25.8 to 47.1 kJ/mol. In conclusion, PBEI is the fastest
interconverting molecule,as seen from the shortest half-life time of the isomerisation reaction,
t1/2(determined as in Supplementary Information, eqn. 4), and the lowest rotation-energy barrier.
The data in Table 4 reveal also that the entropies of activation become less negative on going from
the phenyl to phenanthryl derivative (from -167.0 to -94.1 J/mol. K), revealing that the transition
stateof PBEI compound is more organised than that of 1-NBEI and 9-PhBEI derivatives. This may be
attributed to reduction of the formal charge on the benzothiazolium moiety in the transition state as a
result of increasing the donation ability of the aryl group as its size increases and consequently
decreases the charge separation over the whole molecule. Actually, DFT supports this expectation
10

where the group charge on benzothiazolium moiety of PBEI, 1-NBEI and 9-PhBEI in their transition
states are 0.117, -0.010 and -0.095 e, respectively. Otherwise, the dipole moment is a measure of
charge separation in a molecule. As shown in Table 5, the dipole moment of transition state decreases
as the aryl size increases, also supporting the above expectation.
The largest activation energy for the slowest reaction (i.e., that of 9-PBEI, cf. Table 4) indicates that
the reaction is enthalpy-controlled, within the reaction series. The variation in the rate may be caused
by changes in either the enthalpy or entropy of activation, or both. Enthalpy and entropy of activation
are correlated by
, the isokinetic relationship, where β is the isokinetic
temperature. When the experimental temperature T<β, the reaction rate is controlled mainly by the
enthalpy change. The isokinetic relationship between Hand S of thermal isomerisation of the dyes
in different solvents gives a straight line whose slope  is equal to 119 °C with a good correlation
coefficient (r = 0.963; cf. Fig. 8). This result provides evidence that isomerisation in different solvents
follows the same mechanism, i.e., the same rate-determining step [55]. It is evident that the isokinetic
temperature is higher than the average of the experimental value (22.5 °C); therefore, the
isomerisation process is an enthalpy-controlled reaction [55]. At this temperature, it is expected that
the variation of aryl moiety and/or solvent has no influence on the rate of the thermal ZE
isomerisation of the concerned dyes. From the isokinetic plot, it can be concluded that the
isomerisation reactions for all dyes proceeds via identical common mechanistic pathway.
3.3 Mechanism of Z→E thermal isomerisation
The structures for Z, TS and E forms of the investigated dyes were optimised by DFT in the gas
phase, followed by optimisation in methanol. Fig. 9 shows those of PBEI as a representative example.
The aryl moiety of the optimised Z-isomer is not on the same plane formed by the central double bond
due to steric repulsion between the aryl moiety and the benzothiazole moiety. Lacking steric repulsion,
the optimised E-isomer is almost planar. Selected optimised parameters for the Z-isomer and the
transition state for PBEI, 1-NBEI and 9-PhBEI are shown in Table S4[42]. For any given substrate,
upon going from Z to TS, shows (i) lengthening of C2=C3 bond, (ii) shortening of C1-C2 and C3-C4
bonds, (iii) substantial increase of the <C1C2C3C4 dihedral angle from about 12° to around 93°,
indicating that a rotation takes place around the C2-C3 bond, and (iv) an increase in the dipole
moment. All of these changes in geometrical parameters for the transition state relative to the Z-isomer
indicate a rotation mechanism for Z-to-E isomerisation, in which rotation around the ruptured C2-C3
double bond leads to a dipolar transition state. Such a pathway is analogous to the rotational
mechanism characteristic of ‘push-pull’ (donor/acceptor-substituted) azobenzenes [50,56].
11

Although DFT can reproduce qualitatively experimental trends of the activation parameters of the
isomerisation process, as shown in Table 4, it is also evident that DFT overestimates them somewhat
for all investigated compounds, except for those of PhBEI. B3LYP often tends to lead to systematic
errors in estimating energies for a family of compounds, often with the disparities growing with
molecule size (as suggested by Table 4) [57,58]. This is attributable predominately due to
inadequacies in treating longer-range non-bonded attractiveinteractions (dispersion), and it is to be
hoped that more sophisticated functionals with better treatment of this would lead to performance
improvements in future.The computed values of activation energy and enthalpy for PhBEIreplicate
trends observed experimentally, although are not in quantitative agreement per se. On the contrary, the
E# andH#of PBEI or 1-NBEI isomerisation reaction are overestimated by about 15 and 17.5 kJ/mol,
respectively. Fig. 9 shows the energy profile for the isomerisation pathway clarifying the internal
rotation around the ethenic bond. It is clear that the thermal isomerisation of PBEI is the most
kinetically favoured reaction, whilst that of 9-PhBEI is the most thermodynamically favourable. It is
well known that the activation energy is governed by the energy of rotational transition state with
respect to that of the Z-isomer [50,56]. The activation energy is affected primarily by two factors:
(a) The degree of electronic interaction between the benzothiazolium and aryl moieties across the
central ethenic double bond. This includes the steric hindrance and potential putative ICT from the
aryl ring to the benzothiazoliumcation.
(b) The degree of the interaction between the solvent molecules and the three subsystems (aryl, bridge
and benzothiazolium) of the dye.
The interplay of both factors leads to the following order of activation energy: PBEI < 1-NBEI < 9-
PhBEI. Accordingly, the number of molecules reaching the rotational transition state and then the rate
of thermal ZE isomerisation increases in the order PBEI  1-NBEI  9-PhBEI.
3.4 Electron affinities and ionisation potentials
Table 6 and Fig. 10 demonstrate a comparison between the current N-methylated derivatives and
their neutral analogous [21, 22, 30]. The extent of any potential, putative ICT between the two
moieties (aryl and benzothiazol(e)/ium) is determined by their ionisation potentials and electron
affinity, as determined by HartreeFock/6-31(d) calculations. The electron affinity of benzothiazolium
(-2.69 eV) in comparison to benzothiazole (2.76 eV) suggests that any putative ICT may be stronger
under the effect of N-methylation, and the ionisation potential of the aryl moiety was found to
decrease as its size increases, either in the neutral or charged case.However, this is a marginal
difference in electron affinity, rendering aryl- benzothiazole ICT less likely, which is not inconsistent
12

with the notion that iodide anion-to-stilbazolium CT is more likely. The computed ionisation
potentials of benzene, naphthalene, phenanthacene, anthracene were 8.98, 7.72, 7.59 and 6.90,
respectively. This order may well serve to rationalise the experimentally observed bathochromic shifts
in absorption and emission spectra.
4- Conclusions
The ultimate thrust of the present work has been to demonstrate the effect of replacing the phenyl
ring of the styrylbenzothiazolium dye by a larger polycyclic ring, as well as the solvent effect(if
applicable) on its photophysics, photochemistry and thermal stability, motivated by real-world
applications. Increasing the size of aromatic rings led tored-shifting of absorption and
fluorescencespectra, along with dipole-moment increases and Stokes shifts.The computedelectron
affinity and ionisation potentials help to rationalise the experimentally observed bathchromic shifts in
absorption and emission spectra as the aromatic ring size increases. PBEI is the fastest thermally-
interconverting amongst the investigated compounds, as it has the lowest rotation-energy barrier. 1-
NBEI is the fastest for photochemical interconversion, as the rotation around the bridge is the most
kinetically favourable, in this case.
The mechanism of thermal Z→E transformation was found to proceed via rotation around the ethenic
bridge by forming a ‘perpendicular’ transition state, although that for the anthryl derivative possesses
a high rotational barrier around the ethenic bond, rendering light- or heat-driven activation difficult.
Based on the significant blue shift is the Z-isomer’s absorption maximum relative to that of the E-
isomer (~40-110 nm), and the high percentage of Z-isomers in the PSS, these compounds may serve as
potential promising candidates for data-storage applications. The B3LYP functional reproduced
successfully the experimental trends in activation parameters, albeit with somewhat overestimated
values.
13

References
[1] R. A. Mathies, S. W. Lin, J. B. Ames, W. T. Pollard, Annu. Rev. Biophys. Chem., 20 (1991) 491-
512.
[2] B. Li, K. M. Hanson, J. D. Simon, J. Phys. Chem. A, 101 (1997) 969-972.
[3] J. C. Banerji, A. K. Mandal, B. K. Banerje, Dyes Pigm.,3 (1982) 273-280.
[4] B. N. Jha, J. C. Banerji, Dyes Pigm.,6 (1985) 213-226.
[5] B. N. Jha, R. K. Jha, J. C. Banerji, Dyes Pigm., 7 (1986) 133-152.
[6] K. Heilbron, B. Walter, J. Chem. Soc., 127 (1925) 690-696.
[7] Y. Nakazawa, Y. Nakamura, T. Sueyoshi, A. Sato, DE2147586 (Fuji Photo Film Co. Ltd;
1972)Available at: http://www.google.com/patents/EP0611109A1?cl=en (accessed 23 Jun 2014)
[8] Thomas, K. J.; Thomas, K. G.; Kumar, T. K. M.; Das, S.; George, M. V. Proc. Ind. Acad. Sci.
(Chem. Sci.) 1994, 106, 1375.
[9] L. Qun, B-X Peng, Photograph. Sci. Photochem., 12 (1994) 150-165.
[10]M. S. Antonious, Spectrochim. ActaA Mol. Biomol. Spectr.,53 (1997) 317-324.
[11]J-L Yang, GaodengXuexiaoHuaxueXuebao, 3 (1990) 286.
[12]J. H. H. Keller, US Patent 3888668 (1975).
[13]H. Berneth, F.K. Bruder, W. Haese, R.Hagen, K. Hassenrueck, S. Kostromine, P. Landenberger,
R. Oser, T. Sommermann, J.W. Stawitz, T. Bieringer, WO2002089128 A1 (2002).
Available at: http://www.google.st/patents/WO2002089128A1?cl=en (accessed 23 Jun 2014)
[14]J. R. Manhardt, N. H. Nashua, US Patent 3630733 (1971).
[15]H. Tsuhahara, Japanese Patent 7405467 (1975).
[16]K. Cisek, J.R. Jensen, N.S. Honson, K.N. Schafer, G.L. Cooper, J. Kuret, Biophys. Chem. 170
(2012) 25-33
[17]V. Kovalska, K. Volkova, M. Losytskyy, O. Tolmachev, A. Balanda, S. Yarmoluk, Spectrochim.
Acta A.65 (2006) 271 -277.
[18]Z-S. Wang, F-Y. Li, C-H. Huang, Chem. Commun., (2000) 2063–2064.
[19]U. Mazzucato, Pure Appl. Chem., 54 (1982) 1705-1721 and references therein.
[20]T.A. Fayed, S. El-Din H. Etaiw, Monash. Chem,130 (1999) 1319-1333.
[21]T.A. Fayed, S.E.H. Etaiw, M.K. Awad, M.M. El-Hendawy,J. Photochem. Photobiol. A, 222(1)
(2011) 276-282.
[22]M.K. Awad, M.M. El-Hendawy, T.A. Fayed, S.E.H. Etaiw, N J. English, Photochem. Photobiol.
Sci.12(7) (2013) 1220-1231.
[23]H. Görner, D. Schulte-Frohlinde, Chem. Phys. Lett., 101 (1983)79-85.
14

[24]H. Güsten, D. Schulte-Frohlinde, Tetrahedron Lett.,11 (1970)3567-3570.
[25]D. Schulte-Frohlinde, H. Güsten, Liebigs Ann. Chem.,749 (1971)49-55.
[26]H. Güsten, D. Schulte-Frohlinde, Z. Naturforsch., 34b (1979)1556- 1566.
[27]H. Güsten, D. Schulte-Frohlinde, Chem. Ber.104 (1971) 402-406.
[28]K. Takagi, K. Aoshima, Y. Sawaki, H. Iwamura, J. Am. Chem. Soc.107(1985) 47-52.
[29]S. T. Abdel-Halim, M. H. Abdel-Kader, U.E. Steiner, J. Phys.Chem., 92 (1988) 4324-4328.
[30]S. H. Etaiw, M. K. Awad, T. A. Fayed, M. M. El-Hendawy, J. Mol. Struct., 919 (2009) 12–20.
[31]A. Gaplovsky, J. Donovalova, P. Magdolen, S. Toma, P. Zahradnik, Spectrochim.Acta A,58
(2002) 363–371.
[32]J. Kabatc, B. Jedrzejewska, P. Orlinski, J. Paczkowski, Spectrochim.Acta A,62 (2005) 115–125.
[33]W. H. Melhuish, J. Phys. Chem.65 (1961) 229-235.
[34]S. R. Meech, D. Phillips, J. Photochem.23 (1983) 193-217.
[35]S. L. Murov, I. Carmichael, G. L. Hug, Handbook of Photochemistry, 2nded, Marcel Dekker, New
York, 1993.
[36]Gaussian 09, Revision B.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M.
Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E.
Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi M. Cossi, N. Rega, J. M. Millam, M. Klene, J.
E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O.
Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G.
Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J.
B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2010.
[37]A. D. Becke, J. Chem. Phys.,98 (1993) 5648-5652.
[38]C. Lee, W. Yang, Parr, R. G. Phys. Rev. B, 1988, 37, 785-789.
[39]P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys. Chem.,98 (1994) 11623-
11627.
[40]J. J. P. Stewart, J. Mol. Model., 13 (2007) 1173-213.
[41]A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B, 113 (2009) 6378-6396.
15

[42]See Electronic Supp. Info. Accompanying this article
[43]E.M. Kosower, J.A.Skorcz, W.M. Schwartz, Jr., J.W. Patton, J. Am. Chem. Soc.,82 (1960) 2188-
2191
[44]Ph. Hébert, G. Baldacchino, Th. Gustavsson, J. C. Mialocq, J. Photochem. Photobiol. A:Chem.,
84 (1994)45-55.
[45]A. Mishra, G. B. Behera, M. M. G. Krishna, N. Periasamy, J. Lumin., 92 (2001)175-188.
[46]E. Lippert, W. Lüder, F. Moll, Spectrochim. Acta A.,10 (1959) 858–869
[47]D. Jansone, M. Fleisher, G. Andreeva, L. Leite, J. Popelis, E. Lukevics, Chem. Heterocycl.
Comp.,39 (2003) 1584-1590.
[48]E. Fischer, J. Phys. Chem.,71(1967) 3704–3706.
[49]G. Baratoci, F. Masetti, U. Mazzucato, A. Spalletti, G. Orlandi, G. Poggi, J. Chem. Soc. Faraday
Trans.,84 (1988) 385-399.
[50]H. Görner, H.J. Kuhn, Adv. Photochem.19(1995) 1-117.
[51] J. Saltiel, J. Am. Chem. Soc. 89 (1967) 1036-1037.
[52] J. Saltiel, J. Am. Chem. Soc. 90 (1968) 6394-6400.
[53] J. Saltiel, J.L. Charlton, ‘Cis-Trans Isomerization of Olefins’. In ‘Rearrangements in Ground
and Excited States’; P. de Mayo (Ed.); Academic Press, New York, (1980); Vol III, pp 25-89
[54]A. F. Carey, J. R. Sundberg, "Advanced Organic Chemistry, Part A: Structure andMechanism",
2nd ed. Plenum Press: New York, (1984).
[55]E. Grunwald, S. Winstein, J. Am. Chem. Soc.,70 (1948) 846-854.
[56]J. E. Leffler, J. Org. Chem., 20 (1955) 1202-1231.
[57]M.N. Glukhovtsev, R.D. Bach, A. Pross and L. Radom, Chem. Phys. Lett., 260 (1996) 558-564.
[58]M.D. Wodrich, C. Corminboeuf and P. von RaguéSchleyer, Org Lett., 8 (2006) 3631-3634.
16

Table 1. Spectroscopic data of PBEI and 1-NBEI in various solvents at 25 ºC
amax (nm)

max(M-1cm-1)
f
ex (nm)
f (nm)
PBEI
Stokes Shift
(nm)
CHCl3
DCM
385
385
376
372
373
371
34148
33860
31561
35489
34579
31705
0.804
0.766
0.760
0.871
0.868
0.797
397
398
394
395
392
395
448
467
456
458
457
458
63
82
80
86
84
87
EtOH
MeCN
MeOH
H2O
1-NBEI
CHCl3
DCM
442
439
419
415
417
408
25111
25988
23414
23414
24045
21374
0.667
0.562
0.541
0.536
0.550
0.557
398, 422, 441
397, 422, 441
403, 425, 443
400, 423, 438
404, 424, 443
402, 421, 437
519
529
522
526
527
526
77
90
EtOH
MeCN
MeOH
H2O
103
111
110
118
17

Table 2.Photochemical data for EZphotoisomerisation of PBEI, 1-NBEI, and 9-
PhBEI in methanol and water.
MeOH
H2O
405 a/436 nm
365a/405 nm
405 a/436 nm
365 a/405 nm
KE
%
%
KZE
KEZ
KZE
KEZ
KZE
KEZ
KZE
Z
% Z
% Z
74.1
99.4
___
ZEt
EZ ZE
EZ ZE
EZ ZE
EZ
10-
10-4s-1
10-4s-1
10-4s-1
10-4s-1
10-4s-1
10-4s-1
10-4s-1
Z
Z
4s-1
0.22
0.46
7
0.25
0.71
___
0.52
0.02
___
77
0.35
0.27
2.5
1.94
79
0.29
0.20
1.1
0.73
66.1
0.46
0.54
1.51
1.77
PBEI
0.69
___
0.02
2
85
89
0.34
0.33
0.25
0.12
2.1
1.5
1.53
0.56
90
92
0.27
0.27
0.28
0.11
1.10
1.00
1.20
0.41
99.3
___
0.38
___
0.57
___
0.02
___
1-NBEI
0.008
___
___
9-PhBEI
a refers to the irradiation wavelength that is used only for PBEI.Notes: % Z: the quantity of Z-isomers
at the PPS. ΦE→Z& ΦZ→E: the quantum yields of photo-isomerisation of the E → Z and Z → E
reactions. kE→Z&kZ→E: the partial rate constants for the forward and the backward reactions. The
isomerisation process of 9-PhBEI could not monitored in water due to the heavy aggregation of the dye
molecules.To monitor photo-isomerisation, a mercury lamp with highly intense peaks at 365, 405 and
436 nm was used. The absorption maximum of PBEI lies around 385 nm, thus 365 and 405 nm lamp
excitation is appropriate: both frequencies match high-absorptivity regions around the maximum, but
436 nm will not be of benefit, as it matches the tail of PBEI’s spectrum. For 1-NBEI and 9-PhBEI,405
and 436 nm match the high-absorptivity regions around their 365 nm maxima. See also Fig. 5.
18

Table 3.Rate constant (k) of the ZE thermal isomerisation of 1-NBEI, activation
parameters in different solvents
E#
H#
S#
T
(°C)
15
20
25
30
15
20
25
30
15
20
k
t1/2
(min)
0.78
0.58
0.55
Solvent
MeOH
(min-1)
0.8833
1.1911
1.2607
1.5421
0.0303
0.0378
0.0607
0.0998
0.0010
0.0016
0.0027
(kJ/mol)
(kJ/mol)
(J/mol. K)
26.9
57.5
69.1
24.4
55.0
-159.6
-82.3
0.45
22.87
18.34
11.42
6.95
693.14
433.21
MeCN
EtOH
-69.4
66.7
25
30
256.72
161.20
0.0043
Table 4.Rate constant (k) of the thermal ZE isomerisation of PBEI, 1-NBEI and 9-
PhBEI, activation parameters in MeOH at 22.5 °C. The B3LYP activation parameters
are presented in brackets
E#
H#
S#
k
t1/2
(min)
Compound
PBEI
(min-1)
(kJ/mol)
(kJ/mol)
(J/mol. K)
1.5262
1.2607
0.45
0.55
24.4 (40.4) 22.0 (40.4)
-167.0
1-NBEI
26.9 (40.8) 24.4 (40.9)
-159.6
9-PhBEI
0.8951
0.77
47.4 (49.5) 45.0 (49.4)
-94.1
Table 5. The computed dipole moments (Debye) for Z, E and transition state (TS)
forms of the investigated compounds at B3LYP/6-31+G(d,p) level
Z
TS
E
PBEI
12.05
9.16
9.46
13.58
10.13
10.16
9.04
8.04
7.06
1-NBEI
9-PhBEI
19

Table 6.Comparison between the tested compounds and their neutral analogues
Neutral compounds
Charged Compounds
Solvatochromism
Dipole moment
Positive solvatochromism in both Negative solvatochromism in
absorption and fluorescence spectra absorption spectra and positive
in the fluorescence spectra
FC and relaxed excited states are FC state is less polar than
more polar than ground state
ground state but the relaxed
excited state is larger
Relatively strong
Relatively high
ICT
Relatively moderate
Relatively moderate
low
Stokes shift
Fluorescence quantum yield
low
Reached to 95%
Reached to 99%
% Z
Reached to 0.138
Reached to 0.461
E→Z
Abnormal
behaviour
spectroscopic Dual emission in the case of 9- Dual emission in the case of 9-
ABE due to the emission from LE PhBEI (from monomer and
and TICT [30]
aggregates [21]) & 9-ABEI (due
to the extra emission from
intramolecular exciplex [30])
Aggregation of 9-PhBEI in
water with remarked change in
the colour [21]
Photoreactivity of the anthryl No isomerisation
derivative
No isomerisation
20

Figure captions
Fig. 1.Introduction of positive charge to azastilbene adapts it to convert thermally by heat.
Fig. 2. The molecular structure of 1-aryl-2-(N-methyl-2-benzothiazolium)ethene iodides
(ArBEIs) with their abbreviated names.
Fig. 3.ArBEIs’electronic absorption spectra in different solvents (concentration ~2 ×10−5 M).
Fig. 4. Normalised fluorescence spectra of PBEI, 1-NBEI and 9-PhBEI as well as those of
9-ABEI in different solvents.Concentrations were ~2 ×10−5 M.
Fig. 5. (a)Electronic absorption spectra recorded during the photoisomerisation of 1-
NBEIin MeOH at λirr = 405 nm as a function of irradiation time.
(b)Fluorescence spectra recorded during the irradiation of 1-NBEI in MeOH.
(c)Demonstration of linear correlation between the monitored absorbance changes at
the two maxima of 1-NBEI in MeOH, upon irradiation at 436 nm.
(d)Plot of absorbance at λa
versus irradiation time of 1-NBEI in H2O
max
confirming the achievement of PSS.
Fig. 6. (a) Calculated electronic absorption spectra of E- and Z-isomers of 1-NBEI in
MeOH according to Fischer method [48]
(b)Growth of the percentage of Z-1-NBEI (Z%) in MeOH with the irradiation time.
Fig. 7(a)The thermal-back isomerisation of PBEI wasmonitored by recording its
absorption spectra as function of timein EtOH at 25 °C.
(b) Absorbance-time curves for the thermal ZE isomerisation of PBEI in MeCN
at the indicated temperatures.
Fig. 8. Isokinetic relationship for the thermal Z→E isomerisation of PBEI, 1-NBEI and 9-
PhBEI in EtOH, MeCN and MeOH (referred to as 1, 2 and 3, respectively).
Fig. 9.Computed energy profile of rotational Z-to-E isomerisation pathway. The values
refer to the relative energy change with respect to the reactants, where the value
of energy of all Z-isomers is taken zero as a reference for other ones. Inset:
B3LYP/6-31+G(d,p) optimised structures for Z, E and rotational transition state
(TS) forms of PBEI in methanol.
Fig.10. Effect of N-methylation on absorption (abs.) and fluorescence(flu.) spectra; the
excitation wavelengths(λex) inducing fluoresenceare shown.The neural compounds
are PBE, 1-NBE, 9-PhBE and 9-ABE, while their N-methylated forms are PBEI,
1-NBEI, 9-PhBEI and 9-ABEI. Concentrations were ~2 ×10−5 M.
21

N
Thermally isomerized
N
E-stilbazolium
Z-stilbazolium
N
NOT thermally isomerized
N
E-azastilbene
Z-azastilbene
Fig. 1
S
N
Ar
I
CH3
1-aryl-2-(N-methyl-2-benzothiazolium)ethene iodides (ArBEIs)
1
Ar =
9
9
PBEI
1-NBEI
9-PhBEI
9-ABEI
Fig. 2
22

Fig. 3

Fig. 4
24

Time (min)
0
2
4
MeOH
irr = 405 nm
Fresh solution
MeOH
irr = 436 nm
0.5
0.4
0.3
0.2
0.1
0.0
45
30
15
0

7
12
18
24
30
40
53
PSS solution
333 nm
275 nm
300
400
500
500
550
600
650
Wavelength (nm)
Wavelength (nm)
(a)
(b)
0.3
0.2
0.1
0.5
0.4
0.3
0.2
1-NBEI, irr = 405 nm
irr =436nm; r =1
0
40
80
120
0.18
0.21
0.24
Time (min)
A305
(c)
(d)
Fig. 5.
25

Pure E
MeOH
(a)
irr = 436 nm
12000
9000
6000
3000
0
Pure Z
300
360
420
480
Wavelength (nm)
b
90
60
30
0
(b)
MeOH
irr = 436 nm
0
22
44
66
Time (nm)
Fig. 6.
26

S
(a)
Time (min)
0.9
0.6
0.3
0.0
90
75
60
45
33
23
13
5
N
I CH3
0
322 nm
a
250
300
350
400
450
Wavelength (nm)
30° C
25° C
(b)
S
0.75
0.60
0.45
N
CH3
I
20° C
15° C
a
0
15
30
45
Time (min)
Fig, 7
27

80
60
40
20
 =119 °C; r = 0.963
9-PhBEI3
9-PhBEI2
1-NBEI3
PBEI3
PBEI2
1-NBEI2
9-PhBEI1
1-NBEI1
PBEI1
-150
-100
-50
0
S# (J/mol.K)
Fig. 8
28

Fig. 9
.
29