Highly luminescent anthracene sulfides. Synthesis, experimental and DFT study of their optical properties and interaction with electron deficient nitroaromatic compounds

Article abstract of DOI:10.1016/j.jphotochem.2017.07.011

We report on the synthesis, thermal and photooxidative behavior and optical properties of a series alkyl and aryl sulfides with anthracene cores. All anthracene sulfides studied here were highly fluorescent. Their optical behavior was compared with that o

Full text of DOI:10.1016/j.jphotochem.2017.07.011

Accepted Manuscript
Title: Highly luminescent anthracene sulfides. Synthesis,
experimental and DFT study of their optical properties and
interaction with electron deficient nitroaromatic compounds
Authors: Claudia M. Hegguilustoy, Rosana S. Montani, Pablo
´
G. Del Rosso, Maria J. Romagnoli, Raul O. Garay
PII:
S1010-6030(17)30733-5
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.jphotochem.2017.07.011
JPC 10737
To appear in:
Journal of Photochemistry and Photobiology A: Chemistry
Received date:
Accepted date:
28-5-2017
7-7-2017
Please cite this article as: Claudia M.Hegguilustoy, Rosana S.Montani, Pablo
´
G.Del Rosso, Maria J.Romagnoli, Raul O.Garay, Highly luminescent anthracene
sulfides.Synthesis, experimental and DFT study of their optical properties and
interaction with electron deficient nitroaromatic compounds, Journal of Photochemistry
and Photobiology A: Chemistryhttp://dx.doi.org/10.1016/j.jphotochem.2017.07.011
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.

Highly luminescent anthracene sulfides. Synthesis, experimental and DFT study of their
optical properties and interaction with electron deficient nitroaromatic compounds.
Claudia M. Hegguilustoy, Rosana S. Montani, Pablo G. Del Rosso, Maria J. Romagnoli, and
Raúl O. Garay*
INQUISUR, Departamento de Química, Universidad Nacional del Sur (UNS)-CONICET,
Av. Alem 1253, 8000 Bahía Blanca, Argentina
*Corresponding author: Prof. Raúl O. Garay. Tel.: +54 291 4595101; fax: +54 291
4595187. E-mail address: rgaray@criba.edu.ar
1

Graphical abstract
Research Highlights
Anthracene and pyrene cores bearing alkyl- and arylsulfide groups were synthetized
Anthracene tetrasulfides as well as the dialkoxydisulfides are highly fluorescent
The spectroscopic behavior of the tetrasubstituted anthracene and pyrene sulfides exhibit
quite a few similarities
The quenching efficiencies with NACs are dominated by steric hindrance rather than by
electronic factors
Trends in absorption maxima are well reproduced by TD-DFT-BHLYP calculations
Abstract
We report on the synthesis, thermal and photooxidative behavior and optical properties of a
series alkyl and aryl sulfides with anthracene cores. All anthracene sulfides studied here were
highly fluorescent. Their optical behavior was compared with that of related pyrene bisulfide
and tetrasulfides. All sulfides are crystalline compounds and show good chemical stability in
condensed phase. In solution, the pyrene derivatives are very stable but the anthracene
derivatives suffer a slow photooxidation process under air and ambient light. While the
bisulfides of anthracene and pyrene have very different spectroscopic behavior, the
tetrasubstituted anthracene and pyrene sulfides exhibit quite a few similarities. Thus, the
chromophoric core of the compounds consists of the aromatic core and the sulfur or oxygen
atoms, so the variation from alkyl to aryl substituents does not alter the profiles of the
absorption and emission spectra. All they show intense fluorescence, but sulfides bearing
SCH2 or OCH2 groups display higher fluorescence efficiencies than the ones with aryl
substituents. In addition, their quenching efficiencies with nitroaromatic compounds of both
2

series is dominated by steric hindrance rather than by electronic factors. Bathochromic shifts
in the absorption spectra are well reproduced by TDDFT. Moreover, calculations indicate that
the lower energy excitations are of non-ionic type in agreement with experimental results
which showed that both absorption and emission spectra are devoid of noticeable
solvatochromic effects.
Keywords
Polysulfurated;Anthracene;Pyrene;Optical properties;TDDFT;Photooxidation;Chemosensor
1. Introduction
Aryl sulfides are notable building blocks for organic materials whose valuable properties
were initially expressed in a polymeric structure, poly-p-phenylene sulfide, PPS, which
became one of the most important high temperature thermoplastic polymers. Upon oxidation,
PPS was also one of the first conductive polymers to be studied [1]. Later, aryl sulfides were
included in a variety of molecular architectures such as oligomers [2,3], discotic liquid
crystals [4,5], asterisks (a kind of star oligomers) [6], and dendrimers [7]. The interest in
polysulfurated arenes springs from the interplay between the extended pi system and sulfur
atoms with moderate electronegativity and d orbitals that originate electrophores with dual
character, that is, they are able to stabilize either positive or negative charges. Thus,
hexakis(arylthio)benzenes have been studied as ligands in analogues of biological electron
transfer clusters [8], the electrical conductivities of tetrakis(alkylthio) bisphthalocyanines [9]
and radical cation salts of tetrakis(alkylthio)pyrenes [10,11] have been assessed, the redox
and electrochromic properties of coronene asterisks [12] and polysulfurated pyrene-cored
dendrimers [7] were analyzed and the organic field-effect transistors properties of
bis(methylthio)tetracenes [13] have also been tested.
By contrast, the photophysical properties of sulfurated arenes have received less attention
than electronic ones. Recently, it was reported that persulfurated benzenes show very intense
phosphorescence in conformationally restricted environments [14] that was used to develop a
sensing system responsive to magnesium and fluoride [15]. But on the whole, the
photophysical properties of polysulfurated arenes have remained rather unexplored since it
was initially observed that persulfurated arenes with benzene [14] or corannulene cores [16]
display no appreciable emission in solution. However, a few studies reported that
polysulfurated pyrenes exhibit fluorescence in solution [7,17] while the performance of
bis(alkylthio)- and bis(arylthio)anthracenes in electroluminescent devices [18] was measured.
In the latter case, it was reported that quantum yields for the emission of the anthracene 9,10-
bissulfides were quite low. Indeed, the low emission efficiencies displayed by some
disubstituted anthracene derivatives have been often attributed to intersystem crossing
3

[19,20]. It is difficult however to establish an emission pattern for anthracene derivatives.
Thus, the quantum yields of 9,10-dimethoxyanthracene approach unity, 9-chloroanthracene
shows no emission but the 9,10-dichloride does while both 9,10-dibromo- [20] and 9,10-
bis(dimethylamino)anthracene [21] display very low quantum yields. Moreover, 1,5-, 2,6-
and 9,10 distyrylanthracenes are luminescent organic semiconductors [22] while a 2,6,9,10-
tetra(p-substituted-styryl)anthracene derivative shows very weak emission in THF solution
[23].
Currently, we are interested in the use of aromatic fluorophores as the active elements of
sensing materials either in polymeric assemblies or discrete molecules. In particular, we
found that a fluorescent diphenylanthrylene-based polymer showed significant efficiency in
chemosensing nitroaromatics [24]. We also noted during a reactivity study involving some
anthracene tetrasulfides as products [25] that their chloroform solutions appeared highly
fluorescent. So, they could be foreseen as very interesting organic semiconductors if their
luminescent properties indeed parallel their excellent electronic properties. Therefore we
undertook an investigation of their spectroscopic properties. To this end, the effect of
additional phenylthio or alkoxy groups at the 2,6 positions of the structure of the weakly
emitting 9,10-bis(phenylthio)anthracene 1 as well as the effect of replacing the phenylthio
groups by alkylthio groups on the photophysical properties was studied (Scheme1).
Additionally, their optical behavior was compared with that of related bis(phenylthio)-,
tetrakis(phenylthio)- and tetrakis(alkylthio)pyrene. The thermal properties, photooxidative
behavior and fluorescence quenching by nitrobenzene and 3,4,6-trinitrotoluene of this series
of highly fluorescent anthracene and pyrene sulfides are also reported. Conclusions from
experimental findings were discussed with the assistance of quantum chemistry calculations
of the electronic structure (DFT) and absorption transitions (TD-DFT) of the polysulfides.
2. Experimental
2.1. Materials, general methods and analytical equipment
Reagents and solvents were purchased from Aldrich and used without further purification
unless otherwise specified. Melting points are not corrected. 1H NMR (300 MHz) and 13C
NMR (75 MHz) spectra were recorded on a Bruker AVANCE III spectrometer at 25 °C.
Mass spectra were obtained with a GC/MS instrument (HP5-MS capillary column, 30 m ×
0.25 mm × 0.25 μm) equipped with a 5972 mass selective detector operating at 70 eV (EI).
Elemental analyses (C, H) were performed on an EXETER CE-440 instrument. Thermal
analysis was carried out on a TA Q20 instrument under nitrogen flow at a scan rate of 10
ºC/min. The thermal behavior was also analyzed by polarizing optical microscopy (POM)
using an optical polarizing microscope (Leitz, Model Ortolux) equipped with a hot stage
(Mettler). UV/vis spectra were obtained from a UV-Visible GBC Cindra 20 spectrometer.
The absorption measurements were done on dilute samples (less than 0.01 g/ml). The molar
4

extinction coefficients (ε) were obtained from the slope of the plot of the absorption with six
solutions of different concentrations vs. the concentration (correlation values ≥ 0.99).
Fluorescence studies were conducted using a SML model 4800 spectrofluorometer at 25 ºC.
The emission measurements were carried out on dilute samples (less than 0.02mg/ml) using a
quartz cuvette with a path length of 1 cm and keeping the optical densities below 0.1.
Fluorescence quantum yields of compound in THF solution were determined relative to
equiabsorbing solutions of quinine sulphate (ΦF = 0.546 in 0.5 M sulphuric acid) either in air
equilibrated solutions or in nitrogen deaerated solutions.
2.2. Computational Details
All calculations were performed with the ORCA program package [26]. Pre- and post-
processing operations were performed by using the graphical interface Gabedit [27]. To
reduce the computational expenses, the flexible side chains in derivatives of anthracene and
pyrene were replaced by methyl groups. It has been shown that the effect of side chains on
steady state optical properties is negligible [28,29]. Ground-state geometry optimizations
used the B3LYP functional, the def2-TZVP(-f) basis set [30], a conductor-like screening
model (COSMO) correction [31,32] for solvation effects and Grimme’s dispersion energy
term (D3) [33]. The COSMO term used the predefined parameters appropriate to THF. The
vibrational spectrum was checked to ensure that no imaginary frequency was present. Thus
the analytical determination of the Hessian matrix was made consistently using the same
methodology as in the previous step. The excited state calculations were done at the
optimized ground-state geometries, hence they correspond to vertical transitions. The
absorption transitions were calculated on the basis of the time-dependent density functional
theory (TD-DFT) within the TDA approximation [34] employing the B3LYP and the BHLYP
hybrid functionals and the def2-TZVP(-f) basis set. All two-electron integral transformations
were carried out applying the RI approximation [35] with auxiliary basis set [36].
3. Results and Discussion
3.1. Synthesis
The bromides used as synthetic precursors for the corresponding sulfides were obtained by
brominating the aromatic hydrocarbons (Scheme 1). Mild synthetic conditions were used to
prepare the bis- and tetrakis(arylthio) derivatives by nucleophilic polysubstitution of the
corresponding bromides using an excess sodium phenylthiolate in DMI at 40 °C for extended
periods of time. This procedure is chemoselective and useful in the case of 3 since at that
mild temperature the sodium 2-chlorobenzenethiolate self-attack does not occur [24]. The
tetrakis(alkylthio) derivatives were synthesized using sodium alkylthiolate in DMI or DMF.
Synthetic details are included in the Supporting Information Section. All compounds are
soluble in common organic solvents, e.g., toluene, benzene, CHCl3 and THF, though
solubility usually increases with the number of substituents and the replacement of aryl by
5

alkyl or ethyleneoxy substituents. Full structural characterization of two of the anthracene
sulphides studied here had been reported before, namely 1, that was prepared by a different
synthetic route than the one implemented for this work [25] and 3 which was previously
reported by us [24]. We had also reported preliminary characterization data and conditions to
obtain 2 and 5 but they were not adequately purified and characterized. Curiously, the
structurally simple 1P had not been yet synthetized and for the pyrene 4P only the melting
point had been reported in the literature [9].
3.2. Thermal and photooxidative behavior
All sulfides are crystalline compounds according to evidence from DSC and POM
experiments. The DSC results indicate that they can be repeatedly cycled to temperatures
above their melting points in nitrogen atmosphere. Their DSC traces registered from -20°C
up to 260 °C showed only distinct melting and crystallization transitions while no glass
transitions were observed; see Figure S1 and Table 1. All of them showed only one melting
transition in the second heating cycle, strong supercooling in the cooling cycle and no sign of
the presence of a liquid crystalline phase either in the heating or the cooling cycles. Their
non-mesomorphic nature was confirmed by POM visual observations in thin films placed
between cross polarizers; only generic colorful crystal textures could be observed after
cooling from the isotropic liquid state. As expected, the compounds with long flexible chains
showed low melting transitions, i.e., below 100 °C, while the derivatives with aryl sulfide
groups have higher melting transitions, ca. 200 °C, stronger supercooling and, on occasions,
several very close melting and crystallization transitions which could be related to the
presence of multiple polymorphs originated by the conformational freedom of the
arylsulfides groups. For instance, 3 showed in the first heating cycle a melting transition at
229 °C coupled with a cold crystallization at 233 °C to finally melt at 245 °C, while in the
second heating cycle only a transition at 234 °C to a metastable polymorph was detected (Fig.
S2). Simple space-filling considerations indicate that the aryl groups linked by a sulfur atom
with a C-S-C angle of ~109° cannot lie in the same plane as the central aromatic core. These
substituent groups must instead adopt conformations above or below such plane. Indeed, the
DFT modeling of the geometry of the sulfides presented in the Section 3.4 supports this
conception, see Fig. 4.
These sulfides show good chemical stability in condensed phase. Thus, their microcrystalline
powers can stay months in open atmosphere exposed to light and air. In solution, the pyrene
derivatives are also very stable but the anthracene derivatives degrade slowly under ambient
conditions. Thus, a perceptible darkening occurred after a day or two when yellow-greenish
chloroform solutions (~1x10-3 M) of the sulfides compounds were kept under both air and
ambient light.
Indeed, the solution 1H NMR spectra of 1 in CDCl3 showed new aromatic signals at 8.33 ppm
and 7.81 ppm whose intensity grew with time, which can be assigned to anthraquinone, while
6

no additional signals above 9 ppm could be observed (Figure 1a). Although no attempts to
isolate the oxidation products were made, the GC-MS analysis of a solution of 1 in CDCl3
aged for 3 days showed the disulfide 1, anthraquinone and PhSSPh as the predominant peaks
in the GC-MS trace. We also found new signals at 8.09 ppm and 7.98 ppm in the spectrum of
2 after its recrystallization from toluene at 110 °C under air and ambient light (Fig. 1b), these
signals could be tentatively assigned to aromatic protons of the 2,6-
bis(benzylthio)anthraquinone. This rather clean photo-oxidative reactivity pattern matches
the one observed in higher acenes and some of their disulfide derivatives [38]. The extension
of the proposed mechanism to our systems suggests that the path towards decomposition for
the 9,10-thioanthracene sulfides here observed consists first in the formation of 1O2 adducts
in the 9,10 position and the subsequent concerted extrusion of the disulfide from the
endoperoxide intermediate to render the corresponding anthraquinone and the dialkyl or
diaryl disulfide.
The photooxidation reaction was also monitored for the other anthracene sulfides. All of
them show signals whose intensity increased with time in their 1H NMR spectra which can be
attributed to aromatic oxidation products. The alkylthio derivatives were found to be more
stable than the phenylthio ones while the 2,6-dialkoxy derivatives were the more stable ones
though they showed a more complex reaction pattern than 1 or 2. Thus, the steady grow of at
least three predominant aromatic oxidation products of 5 was monitored by 1H NMR spectra
for a period of sixteen days (Fig. S3). An in-depth analysis of these reactions was not pursued
because our main purpose here was to find handling and working conditions for the sulfides.
As expected, the time dependent 1H NMR analysis of anthracene sulfides showed that the
oxidation reaction does not proceed if the air equilibrated solutions are kept in the dark.
Therefore, the synthesis and the recrystallization steps of the anthracene sulfides were done in
darkened glassware while purification by column chromatography was avoided. On the other
hand, pyrene-based polysulfides dissolved in CHCl3 or THF were fully stable in ambient
conditions for weeks and no additional signals were found in their 1H NMR spectra. It must
be stressed, however, that all the polysulfides studied here, either anthracene- or pyrene-
based, are stable for weeks if solutions are kept in the dark or for prolonged periods of time in
solid state.
3.3. Photophysical properties
The photophysical data of the sulfides in THF solution, along with data of anthracene, A, and
pyrene, P, included for comparison purposes, is reported in Table 1 while the absorption and
emission spectra are shown in Fig. 2 and 3. The luminescence of sulfides was studied in both
aerated and deaerated THF solutions. The profiles of the emission spectra obtained in the
presence and absence of dioxygen were similar. It is relevant to note that the anthracene core
presents all optical signatures of rigid chromophores having S0→S1 transitions with slight
geometry differences between planar S0 and S1 states (Fig. S4), that is, a very small Stokes
7

shift, SS, of 209 cm-1, strong vibronic coupling and mirror image symmetry [39]. In the
anthracene sulfides 1, 2, 3 and 4 (Fig. 2a and 2b), the absorption spectra partially retain the
vibronic splitting of A indicating either a small distribution of ground-state geometries or that
the rotamers generated by the substituent groups do not bear upon the electronic properties of
the central rigid core. However, the mirror image symmetry is lost. Increasing bathochromic
shifts of the absorption λmax are observed with increasing number of phenylthio substituents,
e.g., A (377 nm), bis(phenylthio) 1 (416 nm) and tetrakis(phenylthio) 2 (448 nm). Thus, the
9,10-disubstitution produces a red shift of 0.31 eV and then the additional 2,6-disubstitution
further increases the red shift by 0.2 eV. Clearly, the sulfur atoms became integrated into the
chromophoric core and a larger effect of the 9,10- than the 2,6-substitution is expected
considering that the dipole moment of the S0→S1 transition corresponds to a transition dipole
moment along the short anthracene axis (Fig. S4). The substitution of an ortho-hydrogen
atom for a chlorine atom in the phenyl rings of 2 increases the absorption cross section of 3
but leaves its absorption λmax almost unaltered, thus indicating that little interaction occurs
between the polysulfurated central aromatic core and the aryl substituents. Likewise, only
small changes in the absorption spectra are observed when the aryl groups are replaced by
their alkyl counterparts, namely, between 2 and 4 (Fig. 2b) or between 5 and 6 (Fig. 2c). The
molar extinction coefficients, ε, showed small variations and lay in a range of ± 30% of the
value of A, with the exception of the above mentioned case of 3.
The emission bands of the sulfides are devoid of vibronic structure, show moderate Stokes
shifts and are markedly red-shifted compared to that of anthracene. Thus, the increase of
thioether groups in the anthracene core produces a red shift from 380 nm (A) to 458 nm in the
disulfide (1) and then to ~480 nm in all the anthracene tetrasulfides. The bis(phenylthio) 1
shows a large SS (2204 cm-1) and very low quantum yield, ΦF = 0.02 either in air equilibrated
or deaerated THF solutions, that is, a value close to that reported for 1 in air-equilibrated
CHCl3 solution (ΦF = 0.038) [18]. In order to be certain that the low yields were due to an
intrinsic nonradiative path, measurements were done in deaerated THF solutions and
immediately after the solutions were prepared and kept in darkened vials to avoid the
presence of even minute quantities of the photooxidation products. On the contrary, smaller
SS and augmented QY were observed with the additional 2,6-substitution. Thus, the
tetrasulfides 2, 3 and 4 have Stokes shifts of ~ 1550 cm-1 and ΦF in the range of 0.24-0.49 in
air equilibrated solutions and 0.26-0.59 in aerated solutions, with the highest efficiencies
corresponding to the tetraalkyl sulfide 4.
The substitution of sulfur by oxygen at the 2,6-positions of anthracene produces a small
hypsochromic shift in the absorption spectra, that is, the λmax is displaced from 448 nm in 2 to
440 nm in 5. Similarly, a blue shift of nearly 10 nm is observed when going from 4 to 6 (Fig.
2c). Therefore, chalcogen variation from S to O at the 2,6-positions have a minor effect on
the λmax of the lowest energy absorption transitions and parallels equal chalcogen substitution
8

at the 9,10-positions of the anthracene ring, e.g., the 409 nm absorption maxima of 9,10-
bis(methylthio) anthracene shifts slightly to 404 nm in 9,10-dimethoxyanthracene [40].
Interestingly, despite the presence of the 9,10-disulfide substituents, both 5 and 6 closely
resemble the characteristic absorption spectra of 2,6-dialkoxyanthracenes which present two
well defined absorption bands at 403 nm and 339 nm [41]. Thus, two vibronically structured
electronic transitions with origin at 448 nm and 360 nm can be observed in the absorption
spectrum of 6 which are somewhat less defined in 5. The fluorescence of mixed 2,6-dialkoxy-
9,10-disulfides 5 and 6 have moderately blue shifted λmax and slightly larger Stokes shift (~
1900 cm-1) and higher quantum efficiencies compared to the tetrasulfides. Again, the QY
near unity in the absence of oxygen of 6 resembles the one presented by 2,6-
dialkoxyanthracene.
The pyrene tetrasulfides are yellow powders except the disulfide 1P which has a whitish
appearance. As in anthracene, the ground state and the first relaxed state of pyrene are planar
[41] but its apparently large SS of 2951 cm-1 arises because the strong absorption at 336 nm
corresponds to the high energy bright S0 → S2 transition while emission is originated by the
S0 → S1 transition. The lowest energy dark S0 → S1 transition, which has been observed in
gas phase at ca. 367 nm, [42] is very weak and not usually observed in standard absorption
measurements in solution (Fig. S4). So, even though the anthracene and pyrene cores absorb
in rather different wavelengths they both emit in the same region, yet P exhibits larger εmax
and ΦF than A. In contrast to P, as a result of the influence of the substituents in 1, 3, 6 and 8
positions in both S0 → S1 and S0 → S2 excitations, the low energy absorption transitions of
the pyrene sulfides extend into the visible, are strongly allowed, showed a significant overlap
with their fluorescence bands and bear some mirror-image relationship to the emission
spectra.
A comparison with the reported λmax of 367 nm of the dark S0 → S1 transition of P indicated
that, as the number of phenylthio substituents in the pyrene ring increased, the bathochromic
shifts of the λmax increased (Fig. 3a). Likewise, the emission maximum shifts from 373 nm
(P) to 412 nm (1P) and to ~450 nm (2P, 3P and 4P). Replacement of the pendant aryl moiety
of 2P by R-CH2- groups produces a very small blue shift (~5 nm) in both the absorption and
the emission spectra of benzylsulfide 3P as well as in dodecylsulfide 4P. Both sulfides 3P
and 4P have nearly equivalent bands (Fig. 3b) thus indicating that insertion of a single sp2
carbon suffices to break whatever small conjugation is occurring through the sulfur atom
between the pyrene and phenyl groups. Therefore, the similar influence of the substituents on
the absorption and emission bands, the small Stokes shift and high extinction coefficients
indicate that the lowest energy absorption bands could correspond to the overlap of a now
bright lower energy S0 → S1 transition and a higher energy S0 → S2 transition which are both
bathochromically shifted as has been observed in 1-substituted pyrenes [43]. Analogous to
the anthracene sulfides, the pyrene sulfides here studied showed smaller Stokes shifts with
9

increased substitution. However, quantum yields are influenced by the nature of the pendant
groups instead of its number. Thus, arylsulfides 1P and 2P showed Φem values ca. 0.3 while
alkylsulfides 3P and 4P exhibit larger Φem values in the range of 0.7-0.9 in deaerated
solutions.
The absorption and emission spectra of 1, 2, 5, 1P and 2P were measured in other solvents of
variable polarity (EtOH, CH3CN and CCl4) to observe any potential solvatochromic effects
(Table 2). The shifts of the maxima with respect to the one observed in CCl4 are below 5 nm;
with the exception of the one observed for the emission of 1 in acetonitrile. So, no
considerable variations in the λmax or in the λem were observed as the solvent polarity
increases suggesting that these transitions do not imply significant reorganization of the
molecular electronic density upon excitation. All sulfides were soluble in non-polar solvents
(CCl4 and THF) but solubilities decreased markedly in acetonitrile and even more in ethanol.
Excimer formation is often an undesired property when the potential applications of electron
rich aromatic chromophores are considered. However, for these series of sulfides, solutions in
CCl4 or THF did not show additional emission bands which would indicate excimer
formation even at concentrations as high as 5x10-3 M.
3.4. Theoretical calculations
The geometries of the conformers up-down, ud, and up-up, uu, of 1 displayed in Figure 3a
and 3b were investigated at the B3LYP-D3/def2-TZVP(-f)/COSMO(THF) level of theory.
The relative energies were found to be quite similar (Eud – Euu = 0.02 Kcal mol-1). This
suggests that both conformers contribute almost equally to the absorption spectra. Moreover,
the values of the vertical absorptions of both conformers calculated at the TDDFT/BHLYP
level are also alike, λvert-a(ud) = 358 nm and λvert-a(uu) = 359 nm. Similar results have been
obtained for multiarmed aryl ethers carrying from two to eight arylmethylene side-chains in a
variety of aromatic cores, that is, the computed UV-vis spectra of several conformers were
essentially the same [44].
Therefore, only a selected structure with opposite groups in their up and down conformations
was considered for each arylsulfide in the following simulations, e.g. the udud conformation
of 2, see Figure 3c. The flexible side chains were replaced by methyl groups in order to scale
down the computational expenses. The computed geometry of the sulfides have the methyl
groups of the 2,6-dialkoxy and 2,6-disulfide groups situated in the plane of the central core
while the methyl groups in the 9,10-anthacenyl sulfide groups adopt up and down
conformations.
The one-electron picture of an excited state is a simple model where the excitation of an
electron is visualized as occurring from an occupied to a virtual molecular orbital. So, in
order to discuss structure-properties relationships, it is customary to correlate the
experimental optical band-gaps measured from the absorption spectra, Egopt, with the
difference in energy between frontier molecular orbitals (FMO), ELUMO-EHOMO, obtained by
10

the geometry optimization process. The Egopt of the polysulfides determined from the onset of
the energetically lowest absorption band along with HOMO-LUMO gaps computed with the
DFT/B3LYP-D3/COSMO(THF) approach are listed in Table 3. The DFT band-gaps thus
calculated are systematically 0.3-0.5 eV higher than experimental optical gaps. Interestingly,
the calculations indicate that sulfur substitution affects both HOMO and LUMO levels in a
similar pattern irrespective of the nature of the central aromatic core. Thus, both bis(phenyl
sulfide)s 1 and 1P present a decrease in the energy of their HOMO and LUMO levels, albeit
the effect is stronger on the LUMO level, so these disulfides present a bathochromic shift
with respect to their parent aromatic hydrocarbons.(Fig. 4). The tetrakis(phenylsulfide)s 2 and
2P show a red shift compared to their related bis(phenyl sulfide)s 1 and 1P due to an increase
in the HOMO energies and a decrease in the LUMO energies. The replacement of the
aromatic moieties by alkyl chains produces an increase in the energy of both FMOs of 4 and
4P, so the band gaps change only slightly. Therefore, the modeling matches the absorption
behavior of the tetraalkyl and the tetraaryl derivatives, either anthracene-based or pyrene-
based whose experimental Egopt showed minor variations. Apparently, this insensitiveness
would be circumscribed to the band gap because computation results suggest that alkyl
derivatives should be better donor and worse electron acceptors than the aryl ones. In
addition, the calculations reproduced nicely the insensitiveness of the chlorine substitution in
3 and blue shifts observed in 5 and 6. Thus, while the presence of a chlorine atom on the
phenyl substituents in 3 practically does not bear on the frontier energy levels, the
replacement of the sulfur by more electronegative oxygen atoms in the 2,6 position of the
anthracene derivative 3 widens the FMO gap.
A more elaborate picture of electronic excitations describes them in terms of multielectronic
states where a given optical transition is represented by a linear combination of several
occupied-to-virtual MO excitations. Time-dependent density functional theory (TD-DFT)
permits nowadays the calculation of excited state properties of fairly large systems according
to this approach [45]. Initially, TD-DFT calculations of the vertical absorptions, λvert-a, of the
disulfides in THF were performed with the B3LYP functional but although absolute values of
λvert-a are in the range of the experimental λmax, they showed large dispersions. It was
satisfactory to observe that results obtained with the hybrid BHLYP functional consistently
followed the experimental trend, i.e., the linear relationship between λmax and λvert-a showed a
correlation coefficient of R = 0.987, although the experimental excitation energies were
systematically underestimated by 0.5 eV ±0.04 eV, see Table 3. So, we concluded that
bathochromic shifts are well reproduced by theory and the BHLYP method would be useful
in the design of novel polysulfurated arenes. Moreover, in the computationally difficult case
of pyrene, the bright S0 → S2 transition is correctly described with a dominant contribution of
the HOMO → LUMO transition possessing a sizable oscillator strength, f = 0.353, and higher
energy than the dark S0 → S1 transition which has comparable contributions of the HOMO-1
11

→ LUMO and HOMO → LUMO+1 transitions and a very small f of 0.001 (Table 3).
However, the modeling of hyperchromic/hypochromic shifts is less satisfactory and only
predicts correctly that pyrene derivatives show higher absorptivity values than the anthracene
sulfides(fP-sulfides > fA-sulfides), but the differences between each family are on the whole not
well reproduced.
The analysis of the lowest energy multielectronic states of the anthracene sulfides showed
that the lower energy transitions can be mainly attributed (> 90% contribution) to the HOMO
→ LUMO promotion which correspond to the S0 → S1 transition along the short axis of
anthracene which has a covalent character. So, a π(core)→π*(core) character dominates these
transitions that will show only marginal charge-transfer characteristics as it was evidenced by
the quite small solvatochromic effect observed for the anthracene sulfides. Indeed, neither of
states obtained by TD-DFT described the lower energy excitations as having two comparable
contributions (HOMO→LUMO+1 and HOMO-1→LUMO) which would then present a
mixed n→π*(core)/π(phenyl)→π*(core) nature with significant charge transfer
characteristics. These mixed transitions occurred as high energy states in anthracene and the
anthracene sulfides 4, 5 and 6. Similarly, the TD-DFT calculations of pyrene sulfides
described the lowest energy absorption as a strongly allowed HOMO → LUMO transition.
So, our DT-DFT calculations indicate that the lower energy excitations in both anthracene
and pyrene sulfides are of non-ionic type in agreement with experimental results which
showed both absorption and emission spectra devoid of noticeable solvatochromic effects.
The analysis of the topologies of the frontier orbitals (Fig. 5) confirms the above
observations. It can be observed that the related tetrasulfides 2 and 2P have very similar
electron distribution patterns. That is, their LUMOs are delocalized mostly over the central
aromatic core while the HOMOs are also extended to the sulfur atoms, though only two of
them contribute significantly to the orbital. Thus, the sulfur atoms attached to the 9,10
positions of 2 are weakly integrated to the HOMO while in 2P, due to the symmetric nature
of the chromophore, the HOMO have a notable contribution from two sulfur atoms situated
in opposite positions, e.g.: 1 and 6, while the other two, e.g.: 3 and 8, have reduced
contributions. On the contrary, in sulfide 5 the contribution of the 9,10-sulfur atoms to the
HOMO is higher than that of the 2,6-oxygen atoms. In all cases the HOMO-LUMO
transitions encircling only the aromatic cores and some of the heteroatoms indicate that no
significant electronic reorganizations are expected upon excitation. Since the alkyl and aryl
substituents do not contribute to the orbitals involved in the transition, their effect is limited
to the processing and thermal properties of the sulfides as well as on their fluorescence
efficiencies
12

3.5 Interaction with electron deficient nitroaromatic compounds
We then studied the interaction of the electron rich anthracene sulfides with the electron
deficient nitroaromatic compounds, NACs, whose detection is of interest for environmental
and security reasons [25,46]. Nitrobenzene, NB, and 2,4,6-trinitrotoluene, TNT, were chosen
among the nitroaromatic compounds as quenchers because they show minimal quenching by
resonance energy transfer and inner filter effects even at concentrations as high as 10-3 M
because their absorption maxima lie in regions (> 300 nm) far from the λexc ~ 400 nm used in
the quenching experiments or the emission region of the sulfides (400-500 nm). The
remaining quenching process, i.e., electron transfer from the excited state of the sulfide to the
ground state of TNT [46], requires close encounter of the sulfide with the analyte, thus its
quenching efficiency depends on both electronic and steric factors. The analysis of the DFT
calculations indicates that electronic factors are favorable for both quenchers. Thus,
examination of the HOMO-LUMO energy levels shows that the LUMOs of the NB and TNT
rest below the LUMOs of the anthracene and pyrene sulfides thus providing the necessary
driving force for the oxidative electron transfer (Figure 4). Steric effects are more difficult to
evaluate a priori though for application purposes it is expected that the groups attached to the
periphery of the aromatic central core present steric hindrances large enough to impede the
aggregation-caused self-quenching but not the interaction with analytes [47].
Thus, we carried out fluorescence quenching screening experiments by measuring the
fluorescence intensity responses, (I0-I/I0)x100, of the sulfides upon the addition of the
quencher to explore the ability of the anthracene and pyrene tetrasulfides to interact with
nitroaromatic compounds. Control experiments in the absence of the quencher indicated that
the anthracene sulfides did not show a reduction of fluorescence intensities due to
photobleaching during the course of the quenching observations. The fluorescence intensity
of phenyl substituted sulfides 2, 3 and 2P did not diminish upon the addition of TNT, Figure
6a. However, the partial or total substitution of the bulkier phenyl groups linked to the sulfur
by smaller methylene groups resulted in a moderate decrease in the fluorescence intensity of
4, 5, 6, 3P and 4P. Subsequent experiments showed that all sulfides were more sensitive to
NB than to TNT. In fluorescence quenching titration experiments with 6, the gradual addition
of either TNT or NB caused a steady decrease of the fluorescence intensities (Figure 6b). The
Stern-Volmer analysis of fluorescence quenching titration experiments of 6 afforded linear
relationships with R2 values of > 0.99 with Stern-Volmer constants (KSV) of 118 M-1 for NB
and 56 M-1 for TNT. These KSV values are smaller than those presented by pyrene with NB
(KSV = 315 M-1) [48], thus indicating that steric hindrance is dominant in defining the
quenching ability of these anthracene and pyrene sulfides in processes that require a close
molecular encounter.
13

4. Conclusions
A series of disulfides and tetrasulfides with anthracene or pyrene central cores with aryl
and alkyl pendant groups were obtained by nucleophilic aromatic substitution. All sulfides
are crystalline compounds and show good chemical stability in condensed phase. In solution,
the pyrene derivatives are very stable but the anthracene derivatives suffer a slow
photooxidation process under air and ambient light.
Increasing bathochromic shifts of the absorption and emission bands were observed with
increasing number of phenylthio or alkoxy substituents in the anthracene indicating that the
sulfur and oxygen atoms became integrated into the chromophoric core. The structural
modification of the pendant groups leave absorption and emission bands almost unaltered,
indicating that little interaction occurs between the central aromatic core and the alkyl or aryl
substituents. However, quantum yields are influenced by the nature of the pendant groups.
All tetrasubstituted anthracene sulfides are highly fluorescent with good quantum yields. In
particular, the mixed 2,6-dialkoxy-9,10-disulfides present large quantum efficiencies close to
unity.
While the bisulfides of anthracene and pyrene have very different spectroscopic behavior,
the tetrasubstituted anthracene and pyrene sulfides exhibit quite a few similarities. Thus, the
chromophoric core of the compounds consists of the aromatic core and the sulfur or oxygen
atoms, so the variation from alkyl to aryl substituents does not alter the profiles of the
absorption and emission spectra. All show intense fluorescence, but sulfides bearing SCH2 or
OCH2 groups display higher fluorescence efficiencies than the ones with aryl substituents. In
addition, due to the presence of the substituent groups they do not form excimers and their
quenching efficiencies of both series is dominated by steric hindrance rather than by
electronic factors.
The calculated HO-LU gaps (DFT-B3LYP) broadly described the optical gap trends while
the TD-DFT-BHLYP calculations of the vertical absorptions consistently followed the
experimental trend suggesting that the BHLYP method would be useful in the design of novel
polysulfurated arenes. The analysis of the multielectronic states indicated that the lower
energy transitions can be attributed to the HOMO → LUMO promotion while the topologies
of the frontier orbitals indicated that a non-ionic π(core)→π*(core) character dominate these
excitations in agreement with experimental results which showed both absorption and
emission spectra devoid of noticeable solvatochromic effects.
Acknowledgments
Financial support from SGCyT-UNS and CIC-PBA is acknowledged. CMH and MJR thank
CONICET for a fellowship. PGDR and ROG are members of the research staff of CONICET.
14

15

References
[1] Kricheldorf HR, Nuyken O, Swift G. Handbook of Polymer Synthesis. 2nd ed. New
York: Marcel Dekker; 2005.
[2] Vicente J, Abad JA, Lopez-Nicolas RM. Synthesis of molecular chains: phenylene
thioether and sulfoxide oligomers. Tetrahedron 2008;64:6281–88.
[3] Goyot O, Gingras M. p-Phenylene sulfide oligomers and their properties. A-S couplings
mediated by copper or by fluorine substitutions. Tetrahedron Lett. 2009;50:1977–81.
[4] Kumar S. Self-organization of disk-like molecules: chemical aspects. Chem. Soc. Rev.
2006;35:83–109.
[5] Cammidge AN, Gopee H. Structural factors controlling the transition between columnar
hexagonal and helical mesophase in triphenylene liquid crystals. J. Mater. Chem.
2001;11:773–83
[6] Gingras M, Raimundo JM, Chabre YM. Persulfurated aromatic compounds. Angew.
Chem. Int. Ed. 2006;45:1686–712.
[7] Gingras M, Placide V, Raimundo JM, Bergamini G, Ceroni P, Balzani V. Polysulfurated
pyrene-cored dendrimers: luminescent and electrochromic properties. Chem. Eur. J.
2008;14:10357–63.
[8] Stack TDP, Holm RH. Subsite-differentiated analogs of biological [4Fe-4S] 2+ clusters:
synthesis, solution and solid-state structures, and subsite-specific reactions. J. Am. Chem.
Soc. 1988;110:2484–2494.
[9] Atilla D, Kilinç N, Yuksel F, Gürek AG, Öztürk ZZ, Ahsen V. Synthesis,
characterization, mesomorphic and electrical properties of tetrakis(alkylthio)-substituted
lutetium(III) bisphthalocyanines. Synth. Met. 2009;159:13–21.
[10] G. Heywang, Born L, Roth S. New radical cation salts of substituted pyrenes. Synth.
Met. 1991;41:1073–8.
[11] Heywang G, Born L, Roth S. Radical Cation Salts of 1, 3, 6, 8‐ Tetrakis
(methylthio)‐ pyrene-New Easily Accessible Compounds with High Electrical Conductivity
and Excellent Stability. Angew. Chem. Int. Ed. 1991;30:176–177.
[12] Gingras, M., Pinchart, A., Dallaire, C., Mallah, T., & Levillain, E.. Star‐ Shaped
Nanomolecules Based on p‐ Phenylene Sulfide Asterisks with a Persulfurated Coronene
Core. Chem. Eur. J. 2004;10:2895–2904.
16

[13] Kimoto A, Tanaka K, Kawahata M, Yamaguchi K, Otsubo S, Sakai Y, Ono Y, Ohno A,
Kobayashi K. Bis(methylthio)tetracenes: synthesis, crystal-packing structures, and OFET
properties. J. Org. Chem. 2011;76:5018–25.
[14] b) Fermi A, Bergamini G, Peresutti R, Marchi E., Roy M, Ceroni P, Gingras M.
Molecular asterisks with a persulfurated benzene core are among the strongest organic
phosphorescent emitters in the solid state. Dyes Pigm. 2014;110:113–22.
[15] Fermi A, Bergamini G, Roy M, Gingras M, Ceroni P. Turn-on phosphorescence by
metal coordination to a multivalent terpyridine ligand: a new paradigm for luminescent
sensors. JACS. 2014;136:6395–6400.
[16] Baldridge KK, Hardcastle KI, Seiders TJ, Siegel JS. Synthesis, structure and properties
of decakis(phenylthio)corannulene. Org. Biomol. Chem. 2010;8:53–5.
[17] Fermi A, Ceroni P, Roy M, Gingras M, Bergamini G. Synthesis, Characterization, and
Metal Ion Coordination of a Multichromophoric Highly Luminescent Polysulfurated Pyrene.
Chem. Eur. J. 2014;20:10840–10840.
[18] Nerungsi C, Wanitchang P, Sahasithiwat S, Sadorn K, Kerdcharoen T, Thongpanchang
T. Organic electroluminescence devices based on anthracene sulfide derivatives. Tetrahedron
Lett. 2010;51:6392–95.
[19] Brotin T, Desvergne J‐ P, Fages F, Utermöhlen R, Bonneau R, Bouas‐ Laurent H.
Photostationary fluorescence emission and time resolved spectroscopy of symmetrically
disubstituted anthracenes on the meso and side rings: the unusual behavior of the 1, 4
derivative. Photochem. Photobiol. 1992;55:349–358.
[20] Sotoyama W, Sato H, Matsuura A, Sawatari N. Ab initio configuration interaction
singles (CIS) study on polycyclic aromatic molecules (II): Predicting fluorescence quantum
yields by calculating the excitation energies. J Molec Struct: THEOCHEM. 20059;756:35–
38.
[21] Hasegawa E, Takizawa S, Seida T, Yamaguchi A, Yamaguchi N, Chiba N, Takahashi T,
Ikeda H, Akiyama K. Photoinduced electron-transfer systems consisting of electron-donating
pyrenes or anthracenes and benzimidazolines for reductive transformation of carbonyl
compounds. Tetrahedron 2006;62:6581–6588.
[22] Dadvand, A, Sun, WH, Moiseev, AG, Bélanger-Gariépy, F., Rosei, F., Meng, H., &
Perepichka, D. F. (). 1, 5-, 2, 6-and 9, 10-distyrylanthracenes as luminescent organic
semiconductors. J. Mater. Chem. C, 2013;1:2817–2825.
[23] Sun, M, Zhang, D, Li, Y, Wang, J, Gao, Y, Yang, W. Aggregation-enhanced emission
and piezochromic luminescence of 9, 10-bis (p-dibutylaminostyryl)-2, 6-bis (p-t-butylstyryl)
anthracene. J. Lumin. 2014;148, 55–59.
[24] Del Rosso PG, Romagnolli, MJ, Almassio MF, Barbero CA, Garay RO.
Diphenylanthrylene and diphenylfluorene-based segmented conjugated polymer films as
fluorescent chemosensors for nitroaromatics in aqueous solution. Sens. Actuators B: Chem.
2014;203:612–620.
17

[25] Del Rosso PG, Almassio MF, Bruno M, Garay RO. Mild persubstitution of di- and
tetrabrominated arenes with arylthiolate nucleophiles. Tetrahedron Lett 2010;51:6730–33.
[26] Neese F. The ORCA Program System. Wiley Interdiscip Rev: Comput. Mol. Sci.
2012;2:73−8.
[27] Allouche AR. Gabedit - A graphical user interface for computational chemistry
softwares. J. Comput. Chem. 2011;32:174-82.
[28] Fratiloiu S, Grozema FC, Koizumi Y, Seki S, Saeki A, Tagawa S, Dudek SP, Siebbeles
LDA. Electronic structure and optical properties of charged oligofluorenes studied by vis/nir
spectroscopy and time-dependent density functional theory. J. Phys. Chem. B
2006;110:5984–93.A
[29] Degheili JA, Moustafa RM, Patra D, Kaafarani BR. Effect of chain length on the
photophysical properties of pyrene-based molecules substituted with extended chains. J.
Phys. Chem. A 2009;113:1244–49.
[30] Weigend F, Ahlrichs R. Balanced basis sets of split valence, triple zeta valence and
quadruple zeta valence quality for H to Rn: design and assessment of accuracy. Phys. Chem.
Chem. Phys. 2005;7:3297−305.
[31] Klamt A. Conductor-like screening model for real solvents: a new approach to the
quantitative calculation of solvation phenomena. J. Phys. Chem. 1995;99:2224–35.
[32] Klamt A, Jones V. Treatment of the outlying charge in continuum solvation models. J.
Chem. Phys. 1996;105:9972–81.
[33] Grimme S. Semiempirical GGA-type density functional constructed with a long-range
dispersion correction. J Comput Chem 2006;27:1787–99.
[34] Hirata S, Head-Gordon M. time-dependent density functional theory within the
Tamm−Dancoff approximation. Chem. Phys. Lett. 1999;314:291−9.
[35] Neese F. An improvement of the resolution of the identity approximation for the
calculation of the coulomb matrix. J. Comp. Chem. 2003;24:1740-47.
[36] Weigend F. Accurate coulomb-fitting basis sets for H to Rn: design and assessment of
accuracy. Phys. Chem. Chem. Phys. 2006;8:1057−65.
[37] Kaur I, Jia W, Kopreski RP, Selvarasah S, Dokmeci MR, Pramanik C, McGruer NE,
Miller GP. Substituent Effects in Pentacenes: Gaining Control over HOMO−LUMO Gaps
and Photooxidative Resistances. J. Am. Chem. Soc. 2008;130:16274–86.
[38] Baba M, Saitoh M, Taguma K, Shinohara K, Yoshida K, Semba Y, Kasahara S,
Nakayama N, Goto H, Ishimoto T, Nagashima U. Structure and excited-state dynamics of
anthracene: ultrahigh-resolution spectroscopy and theoretical calculation. J. Chem. Phys.
2009;130:134315.
[39] Kobayashi K, Masu H, Shuto A, Yamaguchi K. Control of face-to-face π-π stacked
packing arrangement of anthracene rings via chalcogen-chalcogen interaction: 9,10-
bis(methylchalcogeno)anthracenes. Chem. Mater. 2005;17:6666–6673.
18

[40] Zhou X, Piland GB, Kurunthu D, Dillon RJ, Burdett JJ, Bardeen CJ. The photophysics
of the 2,6-dialkoxy anthracenes: Evidence for excited state side-chain conformational
relaxation. J. Lumin. 2012;132:2997–3003.
[41] Baba M, Saitoh M, Kowaka Y, Taguma K, Yoshida K, Semba Y, Kasahara S,
Yamanaka T, Ohshima Y, Hsu YC, Lin SH. Vibrational and rotational structure and excited-
state dynamics of pyrene. J. Chem. Phys. 2009;131:224318.
[42] Mangle, EA, Topp MR. Excited-state dynamics of jet-cooled pyrene and some
molecular complexes. J. Chem. Phys. 1986;90:802–807.
[43] Trojanowski P, Plötner J, Grünewald C, Graupner FF, Slavov C, Reuss AJ, Braun M,
Engels JW, Wachtveitl J. Photo-physical properties of 2-(1-ethynylpyrene)-adenosine:
influence of hydrogen bonding on excited state properties. Phys. Chem. Chem. Phys.
2014;16:13875–13888.
[44] Mori T, Grimme S, Inoue Y. A Combined Experimental and Theoretical Study on the
Conformation of Multiarmed Chiral Aryl Ethers. J. Org. Chem. 2007;72:6998–7010.
[45] Laurent AD, Jacquemin D. TD-DFT benchmarks: a review. Int. J. Quantum. Chem.
2013;113:2019–39.
[46] Sun X, Wang Y, Lei Y. Fluorescence based explosive detection: from mechanisms to
sensory materials. Chem. Soc. Rev. 2015;44:8019–61.
[47] Shanmugaraju S, Joshi SA, Mukherjee PS. Fluorescence and visual sensing of
nitroaromatic explosives using electron rich discrete fluorophores. J. Mater. Chem.
2011;21:9130–8.
[48] Meaney MS, McGuffin VL. Investigation of common fluorophores for the detection of
nitrated explosives by fluorescence quenching. Anal. Chim. Acta 2008;610:57–67.
19

a)
b)
Figure 1. Aromatic region of the 1H NMR spectra. a) Solution of 1 in CDCl3 kept for 3 days at ambient
conditions (24 °C, fluorescent lighting). b) Mixture obtained after recrystallization of 2a in toluene, kept for ca.
10 min. at 110 °C, under air and ambient light.
5
6
3
4
1
2
350
400
450
500
550
600
300
350
400
450
500
550
350
400
450
500
550
600
Wavelength (nm)
Wavelength (nm)
Wavelength (nm)
Figure 2. Normalized absorption and emission spectra of THF solutions of the anthracene sulfides.
3P
4P
1P
2P
300
350
400
450
500
550
300
350
400
450
500
550
Wavelength (nm)
Wavelength (nm)
Figure 3. Normalized absorption and emission spectra of THF solutions of the pyrene sulfides 1P-4P.
20

a)
b)
c)
Figure 3. DFT-B3LYP-D3 optimized structures in THF of 1(ud), 1(uu) and 2 (udud).
LUMO
-2
-4
HOMO
-6
A 1
2
3 4
5
6
P 1P 2P 4P
NB
-8
PA
Figure 4. Calculated FMO energies at the B3LYP-D3/def2-TZVP(-f)/COSMO(THF) level of anthracene and
pyrene sulfide and the electro deficient compounds nitrobenzene (NB) and 2,4,6-trinitrotoluene (TNT).
21

LUMO+1
LUMO
HOMO
HOMO-1
2
2P
5
Figure 5. Frontier orbitals of sulfides. Figure uses a isocontour of value 0.025 a.u.; for color figure the reader is
referred to the web version of this article.
22

a
0.3
0.2
0.1
0.0
NB
TNT
20
10
0
0
2
4
6
[Q] / mM
2P
3P 4P
2
3
4
5
6
450
500
550
600
650
Wavelength (nm)
Figure 6. a) Fluorescence intensity responses of sulfides in THF solution upon the addition of NB or TNT in
THF; [Sulfide] = 3 µM, [Q] = 2.4 mM. b). Steady-state fluorescence quenching of 6 in THF solution (3.0 µM)
with TNT (0.15 m M to 1.1 m M; top to bottom); excited at 405 nm in the right angle geometry. Inset refer to
Stern-Volmer plot.
23

X
S
S
X
S
S
X
S
S
X
2, X = H
3, X = Cl
1
b
c
a
S(CH2)7CH3
S(CH2)7CH3
Br
Br
Br
Br
CH3(CH2)7S
Br
CH3(CH2)7S
Br
4
Br
S
O(CH2CH2O)2Et
O(CH2CH2O)2Et
a
c
Et(OCH2CH2)2O
Et(OCH2CH2)2O
Br
S
5
S(CH2)7CH3
O(CH2CH2O)2Et
Et(OCH2CH2)2O
CH3(CH2)7S
6
S
S
S
S
S
1P
Br
S
a
a
2P
Br
RCH2S
SCH2R
Br
d
Br
SCH2R
Br
RCH2S
Br
3P, R = Ph
4P, R = (CH2)10CH3
Scheme 1 Synthesis of the sulfide derivatives. Reactants and conditions: a) PhSH/Na/EtOH, vacuum, DMI/40
°C; b) o-XPhSH/Na/EtOH, vacuum, DMI/40 °C. c) C8H17SH/NaH/DMI/80 °C; d) RSH/NaH/DMF.
24

Table 1. Thermal and photophysical properties of disulfides and tetrasulfides
Cpd(a
Tmb
Absorption
Emision
c
d
f
g
HC
-
CC
-
SC
-
λmax (εmax)
λmax
SSe
ΦF
ΦF
A
1
324, 341, 358, 377
375(sh), 396, 416
396, 434, 448
385, 426, 448
395, 433, 447
397(sh), 416, 440
389(sh), 408, 433
295, 306, 320, 336
367, 384
(0.83)
(1.04)
(0.56)
(1.83)
(0.88)
(1.00)
(0.92)
(5.10)
(3.40)
(3.70)
(5.00)
(2.10)
380, 402, 496 209
-
-
217 149
208 133
234 160
68
75
74
19
30
12
-
458
483
482
482
480
473
2204
0.02
0.24
0.30
0.49
0.52
0.70
-
0.02
0.26
0.63
0.59
0.89
0.98
-
2
1568
1538
1538
1893
1953
3
4
86
72
75
-
67
42
63
-
5
6
P
373, 379, 384 2952
412, 431 1769
1P
2P
3P
4P
216 164
227 197
178 116
108 98
52
30
61
10
0.16
0.28
0.68
0.48
0.31
0.32
0.92
0.68
414, 435
453, 474 (sh) 864
449, 471 (sh) 1066
413(sh), 429
409(sh), 427
446, 472
998
a)A = anthracene
b)Melting transition temperature (°C) determined by DSC at scan rates of 5 °C/min. HC = heating cycle, CC =
cooling cycle, SC = super cooling.
c)Absorption maxima measured in dilute THF solutions (Extinction coefficients of mayor peak of the low energy
bands, in 104 mol-1 cm-1 L). Bold data indicate the major peaks, in nm.
d)Emission maxima measured in dilute THF solutions, in nm.
e)Stokes shifts in dilute THF solutions and on films (λmax,em- λmax,abs, in cm-1).
f)Fluorescence quantum yields measured in air equilibrated THF solutions.
g)Fluorescence quantum yields measured in deaerated THF solutions.
Table 2. Absorption and emission maxima in various solvents
Solvent
1
2
5
1P
2P
ET(30)a) λmax
λem
λmax
451
448
449
451
λem
λmax
440
440
438
437
λem
λmax
385
384
383
384
λem
λmax
429
435
432
432
λem
CCl4
THF
32.4
37.4
418
416
414
413
456
458
467
456
482
483
483
480
477
480
480
477
412
412
414
409
451
453
452
447
CH3CN 45.6
EtOH 51.9
a) Dimroth-Reichardt polarity parameter.
25

Table 3. Comparison of experimental and calculated excitation energies (in eV)
Cpds Exptl a) DFT band-gap b)
Exptl c) TD-DFT d)
Egopt
EHO
ELU
ELU-EHO
λmax
λvert-a State f e)
Configuration f)
A
3.24
-5.49 -1.92 3.57
3.29
3.77
5.94
1
5
0.109 H→L 92
0.523 H-1→L 48
H→L+1 47
1
2
2.82
2.59
-5.76 -2.45 3.30
-5.64 -2.55 3.09
2.97
2.75
3.48
1
0.285 H→L 92
3.34
3.59
1
2
0.334 H→L 91
0.115 H-1→L 91
3
4g)
2.58
2.63
-5.56 -2.54 3.04
-5.25 -2.28 2.97
2.77
2.77
3.22
1
0.170 H→L 93
3.22
3.88
1
2
0.209 H→L 93
0.036 H-1→L 57
H→L+1 36
5g)
6g)
P
2.64
2.72
3.65
-5.34 -2.16 3.18
-5.20 -2.08 3.12
-5.59 -1.75 3.84
2.82
2.86
3.69
3.35
3.99
1
2
0.248 H→L 93
0.073 H-1→L 50
H→L+1 37
3.47
4.07
1
2
0.916 H→L 91
0.056 H-1→L 63
H→L+1 32
4.10
4.18
1
2
0.001 H→L+1 55
H-1→L 45
0.353 H→L 80
H-1→L+1 18
1P
2P
3.02
2.74
-5.69 -2.22 3.47
3.23
3.76
4.18
1
4
0.554 H→L 85
0.112 H-2→L 69
-5.38 -2.36 3.02
-4.86 -1.82 3.04
2.85
2.90
3.33
3.41
1
1
0.997 H→L 90
0.914 H→L 92
4Pg) 2.77
a) Optical HOMO/LUMO gaps were defined by the onset wavelength of the longest wavelength absorption in
THF solution, Eg opt (eV) = 1240/λmaxonset (nm).
b) From optimized geometries at the DFT-B3LYP-D/TZVP(-f)/COSMO (THF) level.
c) Experimental excitation energy for the maximal absorption of the lowest energy transition in THF solution.
d) Vertical transitions energies calculated at the TD-DFT-BHLYP/TZVP(-f)//B3LYP-D/TZVP(-f)/COSMO
(THF) level.
e) Oscillator strengths of the first five states, only transitions with f > 0.035 are listed, except for P.
f) Configuration (with contribution in %, only contributions of > 10% are listed).
g) The calculated values correspond to the methyl analogue of the polysulfide.
26