Julolidine fluorescent molecular rotors as vapour sensing probes in polystyrene films

Article abstract of DOI:10.1016/j.dyepig.2014.07.025

We introduce a new sensing polymer system for detection of volatile organic compounds (VOCs) based on the optical response of polystyrene (PS) films doped with julolidine fluorescent molecular rotors (FMRs). The julolidine FMRs exhibited viscosity-dependent changes in the fluorescence intensity, that was enhanced when glycerol was added to ethanol solutions and when they were dispersed in PS films. Thus, reduction in medium mobility slowed down internal motions and allowed for a major radiative decay pathway. The FMR/PS films were exposed to several VOCs, and showed a significant decrease in fluorescence emission when exposed to chloroform, whereas a negligible variation in their emission occurred when methanol was utilized. This vapour sensing behaviour was much more evident when a perfluorodecyl chain was linked to the julolidine core being the molecule segregated at the film surface. This responsive behaviour was affected by solvent composition and its reproducible response was easily determined by luminescence experiments.

Full text of DOI:10.1016/j.dyepig.2014.07.025

Dyes and Pigments 113 (2015) 47e54
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Julolidine fluorescent molecular rotors as vapour sensing probes
in polystyrene films
Giulio Martini , Elisa Martinelli b, Giacomo Ruggeri a b, Giancarlo Galli a b,
Andrea Pucci
a
a
a
,
,
,
a, b, *
Dipartimento di Chimica e Chimica Industriale, Universit aꢀ di Pisa, Pisa, Italy
b
INSTM, UdR Pisa, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 10 June 2014
Received in revised form
We introduce a new sensing polymer system for detection of volatile organic compounds (VOCs) based
on the optical response of polystyrene (PS) films doped with julolidine fluorescent molecular rotors
(
FMRs). The julolidine FMRs exhibited viscosity-dependent changes in the fluorescence intensity, that
17 July 2014
was enhanced when glycerol was added to ethanol solutions and when they were dispersed in PS films.
Thus, reduction in medium mobility slowed down internal motions and allowed for a major radiative
decay pathway. The FMR/PS films were exposed to several VOCs, and showed a significant decrease in
fluorescence emission when exposed to chloroform, whereas a negligible variation in their emission
occurred when methanol was utilized. This vapour sensing behaviour was much more evident when a
perfluorodecyl chain was linked to the julolidine core being the molecule segregated at the film surface.
This responsive behaviour was affected by solvent composition and its reproducible response was easily
determined by luminescence experiments.
Accepted 18 July 2014
Available online 8 August 2014
Keywords:
Fluorescent molecular rotors
Julolidine derivatives
Polystyrene films
Volatile organic compounds
Vapour sensing behaviour
Reversible behaviour
© 2014 Elsevier Ltd. All rights reserved.
1. Introduction
The optical response to vapours is often associated with changes
in the weak metalemetal interactions of coordination complexes as
Luminescent materials have been actively pursued recently,
owing to their various promising applications in diverse fields,
ranging from solar energy conversion [1], to optoelectronic devices
a result of analyte vapour sorption [9]. As an alternative, sol-
vatochromic fluorescent organic species are used for the detection
of organic vapours, since their wavelength emission depends on the
polarity of VOCs [12,13]. In the last years, organic luminophors such
as tetraphenylethene derivatives [14e17], have been proposed as
compounds with effective vapour sensitivity. They display aggre-
gation induced luminescent properties [10,18] and the vapour up-
take caused changes in their intermolecular interactions thus
affecting molecular packing and emission [19].
[2] and chromogenic materials [3e6]. Chromogenic systems are
capable to respond to various stimuli (e.g. light, heat, mechanical
stress and chemical stimuli) through a macroscopic optical output
[4,7,8]. The energy of the stimulus is properly transduced into op-
tical variations (i.e., absorption, emission, refractive index) as a
function of external interference.
Recently, luminescent materials which display reversible colour
changes upon exposure to vapours of volatile organic compounds
For practical applications, those molecules are fabricated into
thin solid films or incorporated in polymer matrices. The success of
vapour sensing polymer films is largely due to the ability of volatile
compounds to spread rapidly inside the polymer matrix and
interact with the sensor molecule giving a fast and reliable
response [20,21]. The use of visible-light transparent polymer
matrices with good film-forming capacity allows also the prepa-
ration of large area devices at ambient conditions by low-cost
fabrication techniques.
(
VOCs), have also rapidly evolved due to their potential application
for chemical vapour detection [9e11]. The detection of VOCs can
often occur by the naked eye, thus suggesting such systems as
promising tools for environmental monitoring and safety systems
at workplaces.
Fluorescent molecular rotors (FMRs) are fluorescent molecules
composed by an electron donor unit in conjugation with an elec-
tron acceptor moiety and are reported to undergo non-radiative
relaxation from the fluorescent excited state [22e24]. More
*
Corresponding author. Dipartimento di Chimica e Chimica Industriale, Uni-
versit aꢀ di Pisa, Via Risorgimento 35, 56126 Pisa, Italy. Tel.: þ39 050 2219270;
fax: þ39 050 2219260.
E-mail address: andrea.pucci@unipi.it (A. Pucci).
http://dx.doi.org/10.1016/j.dyepig.2014.07.025
0143-7208/© 2014 Elsevier Ltd. All rights reserved.

48
G. Martini et al. / Dyes and Pigments 113 (2015) 47e54
specifically, in the ground state the FMR is almost planar and highly
conjugated as well as its locally excited (LE) state. Nevertheless,
solvent relaxation and rapid internal torsional motion occur, thus
resulting in a twisted intramolecular charge transfer (TICT) excited
state, which rapidly decays in a non-radiative way through internal
rotation [22,25e27]. Moreover, non-radiative deactivation of the
first excited state is controlled by rapid internal torsional motion,
which is substantially restricted in viscous media [28,29] When this
internal rotation is hindered, e.g. due to an increase in viscosity or
sterical constraints, the radiative decay of LE state is favoured, and
an increase in quantum yield is obtained [27].
aqueous solution of sodium hydroxide (2 M) and the mixture was
stirred at 0 C for 4 h. The organic layer was extracted with diethyl
ether, dried over Na SO and evaporated to dryness under reduced
2 4
pressure. The crude product was purified by column chromatog-
raphy on silica gel (230ꢀ400 mesh) using diethyl ether/n-hexane
ꢁ
(3/7 v/v) as eluent mixture (R
f
¼ 0.37) (60% yield).
ꢀ1
FT-IR (KBr, cm ): 2950, 2895, 1662, 1600, 1320, 900, 720.
1
H NMR (CDCl
(t, 4H NCH ), 2.7 (t, 4H NCH
C NMR (CDCl
3
) (ppm): 9.6 (s, 1H, CHO), 7.3 (s, 2H, aromatic), 3.3
2
2
CH CH ), 1.9 (m, 4H NCH CH ).
2
2
2
2
13
3
) (ppm): 191.3 (CHO), 149.1 (¼CeN aromatic),
128.5 to 122.0 (aromatic), 49.3 (NCH
2
), 28.1 to 20.4 (NCH
2
CH
2
CH
2
).
þ
The apparent sensitivity to fluid motion of FMRs like julolidine
derivatives is also found to be an indirect effect of a photo-
isomerization reaction [30].
EI-MS m/z (%): 201 (100, M ).
The spectral properties of this compound are in agreement with
those previously reported [33].
FMRs have received popularity in the last 5ꢀ10 years thanks to
their easy applicability as non-mechanical viscosity sensors, tools
for protein characterization and local microviscosity imaging
2.3. Synthesis of 1H,1H,2H,2H-perfluorodecyl cyanoacetate (2)
[
31e34]. Moreover, the high sensitivity towards viscosity changes
A solution of 1.03 g (5 mmol) of DCC in 5 mL of anhydrous
dichloromethane was added dropwise to a solution of cyanoacetic
acid (0.43 g, 5 mmol) and 1H,1H,2H,2H-perfluorodecyl alcohol
(2.32 g, 5 mmol) in anhydrous dichloromethane (10 mL). The
has reached a precision comparable to commercial mechanical
rheometers with shorter measurement time [35].
While the application of julolidine FMRs as viscosity sensors is
widespread, their behaviour within polymer matrices is still under
debate. A few examples have been reported for the determination
of the molecular weight dependence of viscosity in polymer melts
ꢁ
mixture was kept under stirring for 24 h at 25 C. Then, it was
diluted with dichloromethane (10 mL) and the precipitate formed
during the reaction was filtered off. The filtrate was dried under
vacuum and the residue was purified by column chromatography
on silica gel (230e400 mesh) using dichloromethane as eluent
[
36] and for sensing free volume and plasticity in thermoplastic
polymers [37,38].
Herein, we report on the emission properties of different julo-
lidine FMRs dispersed (~0.05 wt.%) within a transparent and
amorphous PS matrix as a function of the exposure to different
VOCs and the results are discussed in terms of sensitivity and
reproducibility of the fluorescence response of the systems.
Different julolidine FMRs were utilized, i.e. DCVJ, 9-(2-carboxy-2-
(R
f
¼ 0.91) (41% yield).
ꢀ1
FT-IR (KBr, cm ): 2980, 2960, 2188, 1756, 1355ꢀ1055, 660.
1
H NMR (CDCl
3
) (ppm): 4.6 (t, 2H, CH
COO), 2.6 (m, 2H CH CF ).
) (ppm): 164.2 (COO), 114.6 (CN), 123 to 107 (CF),
CH ), 26.0 (CH CN).
: C, 29.40; H, 1.14. Found: C, 30.0; H,
2 2 2
CH CF ), 3.5 (s, 2H,
CNCH
2
2
2
13
C NMR (CDCl
3
56.3 (OCH
Anal. Calcd for C13
1.0.
2
), 31.8 (OCH
2
2
2
cyanovinyl)julolidine
(CCVJ)
and
9-(2-(1H,1H,2H,2H-per-
6 2
H F17NO
fluorodecyloxycarbonyl)-2-cyanovinyl)julolidine (F8CVJ), in order
to identify best suited molecular rotors for sensor performance. The
perfluorodecyl chain was selected to favour fluorophore segrega-
tion near the filmeair interface as to be more responsive to VOC
exposure.
2.4. Synthesis of 9-(2-(1H,1H,2H,2H-perfluorodecyloxycarbonyl)-2-
cyanovinyl)julolidine (F8CVJ)
Triethylamine (0.3 mL, 2.14 mmol) was added to a solution of 2
2
. Experimental part
(0.83 g, 1.56 mmol) and 1 (0.21 g, 1.06 mmol) in tetrahydrofuran
ꢁ
(8 mL) and the mixture was stirred at 50 C for 10 h. The solvent
2
.1. Materials and methods
was then evaporated and the residue was purified in a first step by
column chromatography on silica gel using dichloromethane/n-
hexane (4/6 v/v) as mobile phase. The obtained product was further
purified by elution on preparative TLC plates using ethyl acetate/n-
0
Julolidine, N,N -dicyclohexylcarbodiimide (DCC), phosphorous
oxychloride,
9-(2,2-dicyanovinyl)julolidine,
9-(2-carboxy-2-
cyanovinyl)julolidine were purchased from Aldrich and used as
received. Cyanoacetic acid (Aldrich) was recrystallized from a
mixture of toluene/acetone 2:3 v/v. N,N-dimethylformamide and
hexane (4/6 v/v) as eluent mixture (R
f
¼ 0.59) (20% yield).
ꢀ
1
FT-IR (KBr, cm ): 2925, 2855, 2215, 1720, 1615ꢀ1525, 1450,
1322ꢀ1130, 660.
1
dichloromethane (Aldrich) were refluxed over CaH
2
for 2 h and
H NMR (CDCl
3
) (ppm): 7.9 (s, 1H, CNCCH), 7.5 (s, 2H, aromatic),
), 3.3 (t, 4H, NCH ), 2.7 (t, 4H, NCH CH CH ), 2.6
), 1.9 (m, 4H, NCH CH ).
C NMR (CDCl
) (ppm): 164.4 (COO), 154.8 (PhCH ¼ ), 147.8
(¼CeN aromatic), 128.2 to 121.0 (aromatic), 117.6 (CN), 123ꢀ107
(CF), 57.3 (OCH ), 50.2 (NCH ), 27.2e21.0 (NCH CH CH ).
NMR (CDCl /CF ), ꢀ38 (2F,
COOH) (ppm): ꢀ6 (CF
), ꢀ49 to ꢀ47 (10F, CF ), ꢀ52 (2F, CF CF ).
EI-MS m/z (%): 95 (48), 186 (15), 251 (15), 267 (16), 463 (8).
distilled under nitrogen. Tetrahydrofuran (Aldrich) was refluxed
over Na/K alloy for 3 h and distilled under nitrogen. Triethylamine
4.5 (t, 2H, COOCH
(m, 4H, CH CF
2
2
2
2
2
2
2
2
2
13
(
Aldrich) was refluxed over KOH for 3 h and distilled under nitro-
gen. 1H,1H,2H,2H-perfluorodecyl alcohol (Fluorochem) was used as
received. Atactic polystyrene (PS, Repsol, M
3
W
¼ 86,000 g/mol) was
2
2
2
2
2
19
used as received. Spectroscopy grade solvents (Carlo Erba or
F
3
3
3
Aldrich) were utilized without further purification.
CH
2
CF
2
2
2
3
2.2. Synthesis of 9-formyljulolidine (1)
2.5. Preparation of polymeric films
The synthesis of 9-formyljulolidine (1) was carried out modi-
fying a reported procedure [33]. In brief, phosphorous oxychloride
Films of julolidine derivative/PS mixtures were prepared by
(
(
0.29 mL, 3.17 mmol) was added dropwise to a solution of julolidine
0.5 g, 2.88 mmol) and N,N-dimethylformamide (0.27 mL,
dissolving 1 g of PS and the desired amount of dye (0.005e0.1 wt.%)
in 150 mL of CHCl . After solvent evaporation, the polymer mixture
3
3
.45 mmol) in anhydrous dichloromethane (5 mL) and the mixture
was melt-pressed between two Teflon foils in a Carver 3851-0 press
at 150 C and 4 tons of pressure for 5 min. After removal from the
ꢁ
ꢁ
was stirred for 8 h at 25 C. The reaction was treated with an

G. Martini et al. / Dyes and Pigments 113 (2015) 47e54
49
press, the films (180ꢀ220
m
m thick) were allowed to reach slowly
Solvent uptake by polymer films was determined by placing the
aluminium foil/polymer film system in a Gibertini E 50 S/2
analytical balance with five significant figures.
room temperature.
2
.6. Characterization
Thermogravimetric scans were carried out by a Mettler Toledo
e
Starc System (TGA/SDTA851 ). Samples were heated from 25 to
1H NMR and 19F NMR spectra were recorded with a Varian
Gemini VRX300 spectrometer on CDCl or CDCl /CF COOH solu-
tions, respectively, whereas C NMR spectra were accomplished
with a Varian Gemini 600 spectrometer on CDCl solutions. EI-MS
spectra were measured at 70 eV by GLC/MS. Infrared spectra
were recorded by a Fourier transform infrared spectrometer
ꢁ
ꢁ
800 C at 10 C/min under a nitrogen flow.
®
Film thickness was measured with a Starrett V230MXFL
outside micrometer having an accuracy of ±0.002 mm.
3
3
3
13
Angle-resolved X-ray photoelectron spectroscopy (XPS) spectra
were recorded by using a PerkineElmer PHI 560 spectrometer with
3
a standard Al-K
a source (1486.6 eV) operating at 350 W. The
ꢀ8
(
Spectrum One, PerkinElmer) on KBr windows. UVeVis spectra of
working pressure was less than 10 Pa. The spectrometer was
calibrated by assuming the binding energy (BE) of the Au 4f7/2 line
to be 84.0 eV with respect to the Fermi level. Extended (survey)
spectra were collected in the range of 0e1350 eV (187.85 eV pass
energy, 0.4 eV step, 0.05 s/step). The spectra were recorded at the
THF solutions were recorded at room temperature in isotropic
conditions with a Perkin Elmer Lambda 650.
Fluorescence spectra of solutions were recorded (
l
exc. ¼ 450 nm)
®
at room temperature with a Horiba Jobin-Yvon Fluorolog -3 spec-
trofluorometer. The fluorescence quantum yield (
F
f
) in ethanol was
two photoemission angles q (between the path taken by the pho-
toelectrons to the detector and the surface normal) of 70 and 20 .
The standard deviation (SD) in the BE values of the XPS line was
s
ꢁ
ꢁ
determined at room temperature relative to fluorescein (f ¼ 0.79
f
in 0.1 N NaOH) using Equation (1):
0
.10 eV. The atomic percentage, after a Shirley type background
!
ꢀ
ꢁ
2
Grad
GradST
x
n
subtraction [40], was evaluated using the PHI sensitivity factors. To
take into account charging problems, the C(1s) peak was consid-
ered at 285.0 eV and the peak BE differences were evaluated. The
x
F
x
¼ FST
(1)
2
n
ST
effective information depth varies according to d ¼ d
is the maximum information depth (d ~ 10 nm for the C(1s) line
with an Al-K source).
0
cosq, where
where the subscripts ST and X are standard and dye respectively,
Grad the gradient from the plot of integrated fluorescence intensity
vs absorbance for different solutions of standard and dyes. In order
to minimize re-absorption effects, absorbances never exceeded 0.1
in correspondence and above the excitation wavelength. n is the
refractive index of the solvent, i.e. 1.3611 for ethanol, 1.4746 for
glycerol and 1.333 for water.
d
0
0
a
3. Results and discussion
DCVJ and CCVJ are commercially available, whereas novel F8CVJ
was synthesized from julolidine (Fig. 2).
The viscosity of EtOH/glycerol mixtures was predicted with
good approximation by a simple additive rule: [39]
In detail, commercially available julolidine was formylated with
phosphorus oxychloride and dimethylformamide to afford 9-
formyljulolidine. The fluorinated cyanoacetic ester was obtained
via dicyclohexyl carbodiimide-induced esterification of cyanoacetic
acid with 1H,1H,2H,2H-perfluorodecyl alcohol. Condensation of the
ester with 9-formyljulolidine in the presence of triethylamine
produced the desired F8CVJ.
2
X
log h ¼
fi log hi
(2)
i¼1
where
h
is the viscosity of solvent mixture,
f
i
and
h
i
are the volume
fraction and viscosity of the component i, respectively.
Fluorescence spectra of polymer films were recorded
(
l
exc. ¼ 450 nm) in the dark by using the F-3000 Fiber Optic Mount
3
.1. Spectroscopic characterization of DCVJ, CCVJ and F8CVJ
apparatus coupled with optical fibre bundles. Light is focused from
the excitation spectrometer into the optical fibre bundles, and then
directed to the sample. Fluorescence emission from the sample is
directed back through the bundle and into the front-face collection
port in the sample compartment.
solutions
Absorption and emission maxima of DCVJ, CCVJ and F8CVJ in
ethanol solutions were all in the blue-green range, with emission
maxima between 470 and 500 nm. Quantum yields were negligible
The fluorescence variation of the films was tested by exposing
the film at room temperature to VOCs, such as hexane, heptane,
toluene, chloroform, tetrahydrofuran, acetone, dioxane and meth-
anol. The film was attached to an aluminium foil, which covers a
closed container as shown in Fig. 1.
(
Table 1), due to the formation of a TICT excited state, which rapidly
decayed in a non-radiative way through internal rotation [23,25].
The higher quantum yield of F8CVJ with respect those of DCVJ
and CCVJ was attributed to a more restricted intramolecular
twisting motion which favours the emission of light from the LE
state of the dye [23,25].
All compounds exhibited a noteworthy viscosity-dependent
fluorescence emission when glycerol was added to ethanol solu-
tions (Fig. 3 and Table 2).
All FMRs experienced a strong increase in quantum yield (about
1
0e20 times higher) when dissolved in viscous environments like
ꢁ
glycerol solutions (viscosity
h
¼ 945 mPa s at 20 C for glycerol, as
compared to 1.2 mPa s for EtOH). According to the literature [27], in
viscous media the molecular internal rotation is hindered, thus
favouring a radiative decay of the LE state and an increase in
quantum yield. Conversely, the emission wavelength appeared less
sensitive to polarity variations (dielectric constant ε ¼ 45.2 for
glycerol, as compared to 25 for EtOH) and only a red-shift by about
10 nm occurred for all compounds. DCVJ, CCVJ and F8CVJ FMRs can
Fig. 1. Schematic apparatus used to study the vapour sensing behavior.

50
G. Martini et al. / Dyes and Pigments 113 (2015) 47e54
Fig. 2. 9-Dicyanovinyljulolidine (DCVJ, left), 9-(2-carboxy-2-cyanovinyl)julolidine (CCVJ, middle) and 9-(2-(1H,1H,2H,2H -perfluorodecyloxycarbonyl)-2-cyanovinyl)julolidine
F8CVJ, right).
(
be therefore utilized as viscosity probes even in environments
where polarity changes are expected.
The slight difference in emission spectral shape can be possibly
attributed to the different dyeepolymer interactions caused by the
diverse nature of the FMR functional groups [43].
All the FMRs followed a F o€ rster-Hoffmann behaviour according
to Equation (3) (Fig. S1, Supplementary information), which relates
the (double logarithmic) correlation of quantum yield with
viscosity:
The nature of the functional julolidine moieties also affected the
intensity of the emitted light of FMR/PS films. The higher emission
intensity of CCVJ/PS with respect to DCVJ/PS reflected the behav-
ꢀ
2
ꢀ2
iour in solution, i.e.
F
CCVJ ¼ 6.0$10 and
F
DCVJ ¼ 4.2$10 (see
logF ¼ C þ x$logh
(3)
f
Table 2), whereas the strong luminescence of F8CVJ/PS film must be
due to some peculiar feature of the perfluorodecyl chain. It has
been recognized that perfluorinated alkyl chains can be used to
impose phase-segregating power onto a given molecule and
where C and x are constants.
h
was calculated by using Equation (2)
(
see Experimental part).
In all cases, the x parameter, i.e. the viscosity sensitivity of the
FMR, was found to be between 0.36 and 0.57, in agreement with the
values reported in literature [41]. Interestingly, F8CVJ, which
showed the highest fluorescence emission in EtOH, appeared as the
FMR with the lowest viscosity sensitivity. This suggests that in
solution less emission intensity at low viscosity regime would help
the FMR be more sensitive to viscosity variation than a brighter dye.
3.2. Spectroscopic characterization of FMR/polystyrene films
DCVJ, CCVJ and F8CVJ FMRs were dispersed in polystyrene (PS)
at different concentration in the range between 0.005 and 0.1 wt.%.
All the films appeared well homogeneous, optically transparent
with absorption features similar to those collected in solution. The
concentration of 0.05 wt.% was then selected for all FMRs in order
to provide PS films with appropriate luminescent responses
without encountering possible quenching phenomena due to
concentration.
PS is an amorphous polymer with a glass transition temperature
ꢁ
of about 90e100 C and negligible absorption and emission fea-
tures in the wavelength range investigated (Fig. S2). Thus, the dyes
were dispersed in a glassy matrix in which the intramolecular ro-
tations of their julolidine fluorophores were in fact completely
arrested. This would favour the emission of light from their LE
states. Recently, Iasilli et al. found that when tetraphenylethylene
(
TPE) FMR was dispersed in a glassy PS matrix, the arrested intra-
molecular rotations of its aryls resulted in a strong emission of light,
whereas fluorescence significantly weakened when viscous but not
glassy polymer matrices were used [42].
Consistent with those findings, the fluorescence spectra of DCVJ/
PS, CCVJ/PS and F9CVJ/PS films (Fig. 4) showed about 10 nm blue-
shifted emission intensities similar to those collected from EtOH/
glycerol 10:90 v/v solutions.
This blue shift is ascribed to the lower dielectric constant of the
polymer matrix (ε ¼ 2.6) as compared to that of solvent mixtures.
Table 1
ꢀ
5
Optical features of 10 M DCVJ, CCVJ and F8CVJ ethanol solutions.
a
Compound
Absorption max (nm)
Emission max (nm)
F
f
ꢀ
ꢀ
ꢀ
3
3
3
DCVJ
CCVJ
F8CVJ
458
444
448
496
474
488
1.2$10
1.5$10
3.8$10
a
Fluorescence quantum yield (
F
f
) determined at room temperature relative to
exc. ¼ 450 nm.
Fig. 3. Fluorescence spectra of (a) DCVJ, (b) CCVJ and (c) F8CVJ dilute solutions
(5$10 M) in EtOH/glycerol mixtures with different glycerol volume contents.
s
ꢀ6
fluorescein (ff ¼ 0.79 in 0.1 N NaOH).
l

G. Martini et al. / Dyes and Pigments 113 (2015) 47e54
51
Table 2
the effect of several VOCs on the fluorescence properties of FMR/PS
films, by selecting different kinds of organic solvents.
ꢀ
6
Optical features of 5$10 M DCVJ, CCVJ and F8CVJ EtOH/glycerol 10:90 v/v solutions.
a
Compound
Absorption max (nm)
Emission max (nm)
F
f
An illustration of the fluorescence emission dependence on
exposure time to chloroform vapours is shown in Fig. 5. Notably, the
0.05 wt.% DCVJ/PS films underwent a significant variation in
emission intensity when exposed to a saturated atmosphere of
chloroform (Fig. 5).
ꢀ
ꢀ
ꢀ
2
2
2
DCVJ
CCVJ
F8CVJ
469
452
457
506
484
497
4.2$10
6.0$10
3.3$10
a
Fluorescence quantum yield (
f
F ) was determined at room temperature relative
to fluorescein.
l
exc ¼ 450 nm.
The emission intensity dropped by about 50% just after 420 s of
exposure. This solvent dependence was ascribed to the FMR
sensitivity of the DCVJ molecule and resides in the reorganization
energy of the excited transition state (from LE to TICT states) with
increasing solvent uptake. PS matrix is in the glassy state with an
associated large fraction of free volume in the form of channels and
holes of molecular dimensions. When chloroform vapours fill these
empty spaces, diffusion and swelling of the polymer start from the
outer surface layers inwards. As a consequence, the swollen layers
contain the polymer material in a viscous state, thus making the
intramolecular rotations of julolidine moieties to occur more freely
fluorinated materials can be used as surface-active components of
polymer mixtures [44,45]. Analogously, the perfluorodecyl chain
could force F8CVJ molecules to segregate from PS, thus favouring
their effective migration to the surface of the polymer films [46,47].
In order to investigate the effects of perfluorodecyl chain on the
surface migration and enrichment of dispersed F8CVJ, F8CVJ/PS
films were analyzed by angle-resolved X-ray photoelectron spec-
troscopy (AR-XPS). Spectra were recorded at different photoemis-
ꢁ
ꢁ
sion angles,
q
¼ 70 and 20 , corresponding to increasing probing
[27]. This phenomenon resulted in a decreased emission intensity
depths in the range 3e10 nm of the outermost surface. The
of DCVJ. The concomitant shift to longer wavelengths, i.e. from 480
to about 500 nm, was possibly due to the higher dielectric constant
of chloroform with respect to PS (εchloroform ¼ 4.81 and εPS ¼ 2.6).
CCVJ/PS and F8CVJ/PS films exhibited a quite different behaviour
upon exposure to chloroform vapours (Fig. 6). The fluorescence
variation appeared more pronounced and reached a plateau after
about 350e400 s for both films, in contrast to DCVJ/PS films, which
displayed a lesser reduction in fluorescence emission without
levelling off to a constant value. The fastest response observed for
F8CVJ/PS films, whose emission amplitude decreased by 50% just
after 100 s of exposure, suggests that a larger variation in lumi-
nescence would result for FMRs that are more sterically constrained
and carry functional groups that adversely affect their phase
dispersion within PS. The carboxylic unit in CCVJ and, more
evidently, the perfluorodecyl chain in F8CVJ enable the julolidine
FMR to interact more readily with chloroform. Such interaction was
facilitated for the latter FMR that was distributed closely to the film
surface.
experimental intensities of C(1s) and F(1s) signals (at binding en-
ergies of ~290 eV and ~690 eV, respectively) were used to evaluate
ꢁ
C and F atomic abundances. The C/F ratio at both
q
q
¼ 70 (~30) and
ꢁ
¼ 20 (~15) was much lower than the theoretical one (~2500),
calculated on the basis of the known stoichiometric composition of
the F8CVJ/PS film. Thus, F8CVJ was preferentially concentrated in a
~10 nm outer layer of the film surface, owing to the selective
segregation of the low surface energy fluorinated fluorophore at
the polymereair interface.
An AR-XPS analysis was also carried out on DCVJ/PS film. The
film surface was found to contain only carbon, whereas neither
oxygen nor nitrogen could be recorded, being below the detection
limit of XPS sensitivity. DCVJ did not migrate to the near-surface
region of the polymer film but was buried in the bulk of the PS
matrix.
3.3. Effect of vapour exposure on the fluorescence emission of FMR/
PS films
The curves actually reflected the weight of chloroform pro-
gressively adsorbed with time by PS films (Fig. S3). After 420 s, the
system reached an equilibrium since the film across its whole
thickness was involved in solvent permeation. All FMR molecules
were embedded in a polymer environment with homogeneous
microviscosity and their emission did not change any longer for
prolonged exposure times.
The fluorescence variation of FMR/PS films appeared similar
when toluene was utilized as a probing VOC (Fig. 7). However, the
amplitude decrease of the emission reduced in extent and was
significantly slower, by about 6 times, than that shown with
One of the peculiarities of FMRs is their fluorescence variation in
response to changes in viscosity or sterical constraints. Those ef-
fects can also be provided by variations in the free volume of the
matrix in which they are dispersed. In fact, the interactions be-
tween a polymer and vapours of a suitable solvent are able to
induce a relaxation of macromolecular chains which is followed by
a greater mobility with an increase in the free volume and a
consequent decrease in the local microviscosity [48]. It is therefore
expected that the fluorescence emission of an FMR embedded in a
polymer matrix could be affected by vapour exposure. We explored
6
6
exposure time (s)
0
5x10
1.0x10
F8CVJ
CCVJ
DCVJ
60
4
0.8
0.6
0.4
0.2
120
00
420
3
3
2
1
0
6
7
1
00
80
200
460
480
500
520
540
560
580
600
460
480
500
520
540
560
580
600
wavelength (nm)
wavelength (nm)
Fig. 4. Emission spectra of 0.05 wt.% DCVJ/PS, CCVJ/PS and F9CVJ/PS films and (inset)
images of the films taken under the illumination at 450 nm.
Fig. 5. Progressive changes in the fluorescence emission of 0.05 wt.% DCVJ/PS film as a
function of the exposure to chloroform vapours.

52
G. Martini et al. / Dyes and Pigments 113 (2015) 47e54
0.0
composition (Fig. 8). Methanol was actually utilized as a sort of
diluent for chloroform.
Comparing the fluorescence variations recorded by exposing the
CCVJ/PS (Fig. 8a) and F8CVJ/PS (Fig. 8b) films to solvent mixtures of
different vol.% of chloroform, the fluorescence variation became
progressively smaller in extent, according to the decreasing amount
of chloroform in the vapour composition. Moreover, the curves
recorded with solutions rich in methanol showed a continuous
descending response without levelling off, thereby suggesting
incompleteness of the phenomenon within the collection time of
450 s.
Notably, the F8CVJ/PS films appeared more sensitive to solvent
vapours than CCVJ/PS films, with the fluorescence emission being
responsive even to 25 vol.% chloroform in the solution. This result
was in agreement with the F8CVJ characteristics, being the per-
fluorodecyl chain properly designed to provide the julolidine FMR
with a highest sensitivity to VOCs, thanks to its favourable distri-
bution close to the PS film surface. The PS/F8CVJ films sensitivity
limit of 25 vol.% chloroform in the solution corresponds to about
45e50 vol.% in the vapour phase. [52]
-
-
-
-
0.2
0.4
0.6
0.8
DCVJ
CCVJ
t=0
t=420
F8CVJ
200
0
100
300
400
time (s)
Fig. 6. Variation of the fluorescence maximum intensity with exposure time for all
0
.05 wt.% FMR/PS films with inset images of the 0.05 wt.% F8CVJ/PS film before (t ¼ 0 s)
and after (t ¼ 420 s) chloroform exposure.
chloroform. This behaviour can be attributed to the lower vapour
pressure of toluene (i.e., 2.9 kPa for toluene against 21.3 kPa for
chloroform, at 25 C) [49], which can delay solvent adsorption by
ꢁ
The experimental setup consists in vapour exposure at ambient
temperature and pressure, and it did not affect sample size or
morphology within the time interval investigated (<20 min). Sol-
vent then gradually desorbed at room temperature and pressure
from the film when removed from the apparatus, allowing com-
plete recovery of the FMR emission after 24 h. This behaviour
suggested to monitor changes in FMR/PS fluorescence during suc-
cessive solvent exposures. Fig. 9 illustrates the response of F8CVJ/PS
films to a sequence of solvent adsorption and subsequent slow
desorption at room temperature and pressure for 24 h.
the film during the initial stages of exposure. However, the fluo-
rescence emission did not experience any variation in amplitude
when methanol was used as a VOC (Fig. 7).
This behaviour can arise from a combination of effects: the
lower vapour pressure of methanol (i.e., 12.7 kPa for methanol
ꢁ
against 21.3 kPa for chloroform, at 25 C) [49], which can delay
solvent adsorption by the film, and the limited affinity between
methanol and PS. This feeble interaction hampers methanol uptake
by the polymer, thus making the fluorescence variation of the films
negligible. This is in agreement with the more unfavourable Flor-
Notably, the fluorescence variation of the samples was well
retained also after repeated cycles of solvent adsorption/desorp-
tion, thereby indicating a very good reproducibility in the optical
response. All results were confirmed when desorption steps were
yeHuggins interaction parameter (c) of 2.44 for methanol [50]
compared to 0.52e0.17 for chloroform [51] and 0.42e0.31 for
ꢁ
toluene [51] at 25 C. Moreover, PS films appeared responsive also
to tetrahydrofuran (
dioxane (
¼ 0.35, Fig. S4), whereas the emission was barely
affected by hexane, cyclohexane and heptane (
c
¼ 0.16e0.70), acetone (
c
¼ 0.81e0.94) and
ꢁ
carried out at 50 C for 1 h (data not shown). Similar trends were
c
also collected by using toluene as VOC, even if slightly lower
reproducibility of the fluorescence variation was observed during
initial solvent uptake. Whilst toluene may have a larger influence
c
¼ 1.49e1.14) va-
pours. These results suggest that the selectivity of FMR/PS films is
determined by the chemical affinity of PS for the solvent vapours.
More specifically, solvents with
interact with the PS matrix, thus providing the vapour sensing
behaviour.
c
values lower than 1ꢀ2 well
The PS films based on CCVJ and F8CVJ, i.e. those FMRs charac-
terized by the largest vapour response, were then exposed to
chloroform/methanol mixtures in order to determine whether the
system could be sensitive and selective to mixed vapours of varied
time (s)
200
0
100
300
400
methanol
0.0
-0.2
-0.4
-0.6
DCVJ
CCVJ
F8CVJ
800
0
200
400
600
1000
1200
time (s)
Fig. 7. Variation of the fluorescence maximum with exposure time to toluene and
methanol (superimposed curves with no variation) vapours for 0.05 wt.% FMR/PS films.
Fig. 8. Variation of the fluorescence maximum with exposure time to vapours of
chloroform/methanol mixtures (v/v) for 0.05 wt.% (a) CCVJ/PS and (b) F8CVJ/PS films.

G. Martini et al. / Dyes and Pigments 113 (2015) 47e54
53
References
[
[
1] Hagfeldt A, Boschloo G, Sun L, Kloo L, Pettersson H. Dye-sensitized solar cells.
Chem Rev 2010;110(11):6595e663.
2] He GS, Tan L-S, Zheng Q, Prasad PN. Multiphoton absorbing materials: mo-
lecular designs, characterizations, and applications. Chem Rev 2008;108(4):
1245e330.
[3] Prampolini G, Bellina F, Biczysko M, Cappelli C, Carta L, Lessi M, et al.
Computational design, synthesis, and mechanochromic properties of new
thiophene-based p-conjugated chromophores. Chem Eur
996e2004.
J 2013;19(6):
1
[
4] Ciardelli F, Ruggeri G, Pucci A. Dye-containing polymers: methods for prep-
aration of mechanochromic materials. Chem Soc Rev 2013;42(3):857e70.
5] Pucci A, Ruggeri G, Bronco S, Bertoldo M, Cappelli C, Ciardelli F. Conferring
dichroic properties and optical responsiveness to polyolefins through organic
chromophores and metal nanoparticles. Progr Org Coat 2007;58(2e3):
[
105e16.
[
[
[
6] Minei P, Battisti A, Barondi S, Lessi M, Bellina F, Ruggeri G, et al. Light-
responsive polystyrene films doped with tailored heteroaromatic-based flu-
orophores. ACS Macro Lett 2013;2(4):317e21.
7] Ciardelli F, Bertoldo M, Bronco S, Pucci A, Ruggeri G, Signori F. The unique
optical behaviour of bio-related materials with organic chromophores. Polym
Int 2013;62(1):22e32.
8] Makowski B, Kunzelman J, Weder C. Stimuli-driven assembly of chromogenic
dye molecules: a versatile approach for the design of responsive polymers. In:
Urban MW, editor. Handbook of stimuli-responsive materials. Weinheim,
Germany: Wiley-VCH Verlag GmbH & Co. KGaA; 2011. p. 117e38.
9] Wenger OS. Vapochromism in organometallic and coordination complexes:
chemical sensors for volatile organic compounds. Chem Rev 2013;113(5):
[
3686e733.
[
10] Hong Y, Lam JWY, Tang BZ. Aggregation-induced emission: phenomenon,
Fig. 9. Variation of the fluorescence maximum of F8CVJ/PS films for successive cycles
mechanism and applications. Chem Commun 2009;29:4332e53.
of solvent exposure/desorption to (a) chloroform and (b) toluene vapours.
[11] Chi Z, Zhang X, Xu B, Zhou X, Ma C, Zhang Y, et al. Recent advances in organic
mechanofluorochromic materials. Chem Soc Rev 2012;41(10):3878e96.
[
12] Yoon J, Chae SK, Kim J-M. Colorimetric sensors for volatile organic compounds
VOCs) based on conjugated polymer-embedded electrospun fibers. J Am
Chem Soc 2007;129(11):3038e9.
[13] Kim S-H, Lee S-Y, Gwon S-Y, Son Y-A, Bae J-S. D-
on the PS film morphology during the first cycle of solvent
adsorption/desorption, F8CVJ appeared to experience no substan-
tial modification of its distribution within PS film across the whole
experiments.
(
p-A solvatochromic charge
transfer dyes containing a 2-cyanomethylene-3-cyano-4,5,5-trimethyl-2,5-
dihydrofuran acceptor. Dyes Pigments 2010;84(2):169e75.
[
14] Zhang X, Chi Z, Li H, Xu B, Li X, Zhou W, et al. Piezofluorochromism of an
aggregation-induced emission compound derived from tetraphenylethylene.
Chem Asian J 2011;6(3):808e11.
4
. Conclusions
[15] Zhang X, Chi Z, Li H, Xu B, Li X, Liu S, et al. Synthesis and properties of novel
aggregation-induced emission compounds with combined tetraphenyl-
We have shown that julolidine FMRs can be utilized as effective
dyes for the preparation of plastic films with vapour sensing
characteristics.
ethylene and dicarbazolyl triphenylethylene moieties.
011;21(6):1788e96.
16] Chen Q, Wang J-X, Yang F, Zhou D, Bian N, Zhang X-J, et al. Tetraphenyl-
J Mater Chem
2
[
[
FMRs exhibited viscosity-dependent emission properties when
dissolved in solvents with different viscosity or dispersed at low
loadings (<0.1 wt.%) in PS plastic films. The exposure of FMR/PS
films to a saturated atmosphere of well-interacting VOCs caused a
significant drop of FMR fluorescence due to their favoured relaxa-
tion from the non-emissive TICT excited state. This optical behav-
iour was allowed by the increased mobility of the PS
macromolecular chains with solvent uptake, and was largely
enhanced for the F8CVJ FMR that was preferentially distributed at
the outer film surface. By contrast, the film emission appeared
unaffected when methanol and hexane were used as solvents
barely interacting with PS. This characteristic response made
plastic films able to discern vapours of different composition,
providing also a sensitive and reproducible fluorescence response
in successive cycles of solvent exposure. FMR/PS films might be
exploited for the realization of reliable vapour sensing plastic ma-
terials as new tools to detect vapours of different classes of VOCs.
ethylene-based fluorescent porous organic polymers: preparation, gas sorp-
tion properties and photoluminescence properties.
011;21(35):13554e60.
J
Mater Chem
2
17] Shustova NB, McCarthy BD, Dinca M. Turn-On fluorescence in
tetraphenylethylene-based metal-organic frameworks: an alternative to
aggregation-induced emission. J Am Chem Soc 2011;133(50):20126e9.
18] Hong Y, Lam JWY, Tang BZ. Aggregation-induced emission. Chem Soc Rev
[
[
2011;40(11):5361e88.
19] Wang J, Mei J, Hu R, Sun JZ, Qin A, Tang BZ. Click synthesis, aggregation-
induced emission, E/Z isomerization, self-organization, and multiple chrom-
isms of pure stereoisomers of a tetraphenylethene-cored luminogen. J Am
Chem Soc 2012;134(24):9956e66.
20] Kumpfer JR, Taylor SD, Connick WB, Rowan SJ. Vapochromic and mechano-
chromic films from square-planar platinum complexes in polymethacrylates.
J Mater Chem 2012;22(28):14196e204.
21] Lim SH, Feng L, Kemling JW, Musto CJ, Suslick KS. An optoelectronic nose for
the detection of toxic gases. Nat Chem 2009;1:562e7.
22] Haidekker MA, Nipper M, Mustafic A, Lichlyter D, Dakanali M, Theodorakis EA.
Dyes with segmental mobility: molecular rotors. Springer Ser Fluoresc 2010;8:
[
[
[
267e308. Advanced Fluorescence Reporters in Chemistry and Biology I.
[
[
[
23] Haidekker MA, Theodorakis EA. Molecular rotors-fluorescent biosensors for
viscosity and flow. Org Biomol Chem 2007;5(11):1669e78.
24] Haidekker MA, Theodorakis EA. Environment-sensitive behaviour of fluores-
cent molecular rotors. J Biol Eng 2010;4(11):1e14.
25] Zhou F, Shao J, Yang Y, Zhao J, Guo H, Li X, et al. Molecular rotors as fluorescent
viscosity sensors: molecular design, polarity sensitivity, dipole moments
changes, screening solvents, and deactivation channel of the excited states.
Eur J Org Chem 2011;2011(25):4773e87. S/1eS/70.
Acknowledgements
This work was supported by MIUR-PRIN (2010XLLNM3). The
authors are grateful to Dr. A. Glisenti (University of Padova) for
assistance with XPS measurements.
[
26] Amdursky N, Erez Y, Huppert D. Molecular rotors: what lies behind the high
sensitivity of the thioflavin-T fluorescent marker. Acc Chem Res 2012;45(9):
1548e57.
[
[
27] Valeur B, Berberan-Santos MN. Molecular fluorescence: principles and ap-
plications. Weinheim (Germany): Wiley-VCH; 2012.
28] Zhu L-L, Qu D-H, Zhang D, Chen Z-F, Wang Q-C, Tian H. Dual-mode tunable
viscosity sensitivity of a rotor-based fluorescent dye. Tetrahedron 2010;66(6):
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.dyepig.2014.07.025.
1254e60.

54
G. Martini et al. / Dyes and Pigments 113 (2015) 47e54
[
29] L aꢀ Zhu, Li X, Ji F-Y, Ma X, Wang Q-C, Tian H. Photolockable ratiometric vis-
cosity sensitivity of cyclodextrin polypseudorotaxane with light-active rotor
graft. Langmuir 2009;25(6):3482e6.
[41] Sutharsan J, Lichlyter D, Wright NE, Dakanali M, Haidekker MA,
Theodorakis EA. Molecular rotors: synthesis and evaluation as viscosity sen-
sors. Tetrahedron 2010;66(14):2582e8.
[
[
30] Rumble C, Rich K, He G, Maroncelli M. CCVJ is not a simple rotor probe. J Phys
Chem a 2012;116(44):10786e92.
31] Nipper ME, Majd S, Mayer M, Lee JCM, Theodorakis EA, Haidekker MA.
Characterization of changes in the viscosity of lipid membranes with the
molecular rotor FCVJ. Biochim Biophys Acta Biomembr 2008;1778(4):
[42] Iasilli G, Battisti A, Tantussi F, Fuso F, Allegrini M, Ruggeri G, et al. Aggregation-
induced emission of tetraphenylethylene in styrene-based polymers. Macro-
mol Chem Phys 2014;215:499e506.
[43] Pucci A, Tirelli N, Ruggeri G, Ciardelli F. Absorption and emission dichroism of
polyethylene films with molecularly dispersed push-pull terthiophenes.
Macromol Chem Phys 2005;206(1):102e11.
1148e53.
[
[
[
32] Goh WL, Lee MY, Joseph TL, Quah ST, Brown CJ, Verma C, et al. Molecular
rotors as conditionally fluorescent labels for rapid detection of biomolecular
interactions. J Am Chem Soc 2014;136(17):6159e62.
33] Haidekker MA, Ling T, Anglo M, Stevens HY, Frangos JA, Theodorakis EA. New
fluorescent probes for the measurement of cell membrane viscosity. Chem
Biol 2001;8(2):123e31.
34] Maity D, Manna AK, Karthigeyan D, Kundu TK, Pati SK, Govindaraju T. Visible-
near-infrared and fluorescent copper sensors based on julolidine conjugates:
selective detection and fluorescence imaging in living cells. Chem Eur J
[44] Galli G, Martinelli E, Chiellini E, Ober CK, Glisenti A. Low surface energy
characteristics of mesophase-forming ABC and ACB triblock copolymers with
fluorinated B blocks. Mol Cryst Liq Cryst 2005;441(1):211e26.
[45] Yasani BR, Martinelli E, Galli G, Glisenti A, Mieszkin S, Callow ME, et al.
A comparison between different fouling-release elastomer coatings contain-
ing surface-active polymers. Biofouling 2014;30(4):387e99.
[46] Hunley MT, Harber A, Orlicki JA, Rawlett AM, Long TE. Effect of hyperbranched
surface-migrating additives on the electrospinning behavior of poly(methyl
methacrylate). Langmuir 2007;24(3):654e7.
2
011;17(40):11152e61. S/1eS/40.
[47] Martinelli E, Fantoni C, Galli G, Gallot B, Glisenti A. Low surface energy
properties of smectic fluorinated block copolymer/SEBS blends. Mol Cryst Liq
Cryst 2009;500(1):51e62.
[
[
35] Haidekker MA, Akers W, Lichlyter D, Brady TP, Theodorakis EA. Sensing of
flow and shear stress using fluorescent molecular rotors. Sens Lett
2
005;3(1e1):42e8.
[48] Miller-Chou BA, Koenig JL. A review of polymer dissolution. Prog Polym Sci
2003;28(8):1223e70.
[49] Lide DR. CRC handbook of chemistry and physics. 86th ed. Boca Raton, FL,
USA: CRC Press LLC; 2004.
36] Zhu D, Haidekker MA, Lee J-S, Won Y-Y, Lee JCM. Application of molecular
rotors to the determination of the molecular weight dependence of viscosity
in polymer melts. Macromolecules 2007;40(21):7730e2.
[
[
37] Jee A-Y, Bae E-H, Lee M-Y. Internal twisting dynamics of dicyanovinyljuloli-
dine in polymers. J Phys Chem B 2009;113(52):16508e12.
[50] Bernardo G, Vesely D. Equilibrium solubility of alcohols in polystyrene
attained by controlled diffusion. Eur Polym J 2007;43(3):938e48.
[51] Mark JE. Physical properties of polymers handbook. 2nd ed. New York:
Springer Science þ Business Media, LLC; 2007.
38] Xiao P, Frigoli M, Dumur F, Graff B, Gigmes D, Fouassier JP, et al. Julolidine or
fluorenone based push-pull dyes for polymerization upon soft polychromatic
visible light or green light. Macromolecules 2014;47(1):106e12.
39] Perry RH, Green DW, Maloney JO. Perry's chemical engineer's handbook. 6th
ed. New York: McGraw-Hill; 1984.
ꢁ
[52] Martin M, Cocero M, Mato F. Vapor-liquid equilibrium data at 25 C for six
[
[
binary systems containing methyl acetate or methanol, with dichloro-
methane, chloroform, or 1,2-trans-dichloroethylene. J Solut Chem 1991;20(1):
87e95.
40] Shirley DA. High-rResolution X-ray photoemission spectrum of the valence
bands of gold. Phys Rev B 1972;5(12):4709e14.