Rhodamine solid complexes as fluorescence probes to monitor the dispersion of cyclodextrins in polymeric nanocomposites

Article abstract of DOI:10.1016/j.dyepig.2012.02.009

Rhodamines B and 6G have been used to evaluate the dispersion of β-Cyclodextrin in a thermoplastic matrix, poly(ethylene-co-vinyl acetate), by high energy ball milling. In a first stage, a study of the binding properties of β-Cyclodextrin with both fluorophores has been carried out, to determine which of them forms the most stable complex with the macrocycle, its topology and to check whether their fluorescence is kept after the milling process. Both systems have been fully characterized in the solid state (FTIR and XRD, TGA and fluorescence spectroscopy), and in solution (¹H NMR ROESY, steady state and time-resolved fluorescence spectroscopy). Then, nanocomposites based on the thermoplastic matrix and the cyclodextrin complexes have been cryomilled and processed in the form of thin films. Only Rhodamine B forms a complex stable enough to track the nanofiller dispersion within the polymer. This labeled cyclodextrin is uniformly dispersed throughout the matrix after the milling and film forming, yielding a blue-shifted and remarkably enhanced fluorescent response when compared to the same material prepared with the mixture of Rhodamine B and β-Cyclodextrin.

Full text of DOI:10.1016/j.dyepig.2012.02.009

Dyes and Pigments 94 (2012) 427e436
Contents lists available at SciVerse ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Rhodamine solid complexes as fluorescence probes to monitor the dispersion of
cyclodextrins in polymeric nanocomposites
a
b
c
a,
*
R. Serra-Gómez , G. Tardajos , J. González-Benito , G. González-Gaitano
a
Dpto de Química y Edafología, Universidad de Navarra, Facultad de Ciencias, 31080 Pamplona, Navarra, Spain
Dpto. Química-Física I, Universidad Complutense de Madrid, Spain
Dpto. C. Ing. Materiales Ing. Química, Universidad Carlos III de Madrid, Spain
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 23 November 2011
Received in revised form
Rhodamines B and 6G have been used to evaluate the dispersion of
matrix, poly(ethylene-co-vinyl acetate), by high energy ball milling. In a first stage, a study of the binding
properties of -Cyclodextrin with both fluorophores has been carried out, to determine which of them
b-Cyclodextrin in a thermoplastic
b
8
February 2012
forms the most stable complex with the macrocycle, its topology and to check whether their fluorescence
Accepted 10 February 2012
Available online 22 February 2012
is kept after the milling process. Both systems have been fully characterized in the solid state (FTIR and
1
XRD, TGA and fluorescence spectroscopy), and in solution ( H NMR ROESY, steady state and time-
resolved fluorescence spectroscopy). Then, nanocomposites based on the thermoplastic matrix and the
cyclodextrin complexes have been cryomilled and processed in the form of thin films. Only Rhodamine B
forms a complex stable enough to track the nanofiller dispersion within the polymer. This labeled
cyclodextrin is uniformly dispersed throughout the matrix after the milling and film forming, yielding
a blue-shifted and remarkably enhanced fluorescent response when compared to the same material
Keywords:
Cyclodextrins
Rhodamine
Polymeric nanocomposite
High energy ball milling
Fluorescence probe
NMR
prepared with the mixture of Rhodamine B and
b
-Cyclodextrin.
Ó 2012 Elsevier Ltd. All rights reserved.
1
. Introduction
mechanisms of chain scission and sample oxidation, HEBM has been
successfully used to obtain polymer blends with improved
mechanical properties and polymer nanocomposites with a real
dispersion of nanoparticles [6e10].
Cyclodextrins (CDs) are cyclic oligosaccharides composed of 6,
7 or 8
D-glucopyranose rings termed a, b and gCD respectively.
Polymer nanocomposites are generally built by homogeneous
dispersion of a nanoscopic filler (nanofiller) into the polymeric
matrix [1]. Because of the large surface to volume ratio of nano-
particles, the interphase formed between the nanoparticles and the
polymer constitutes a greater fraction of the whole material than in
common composites, even with small amounts of the nanofiller
Their sizes (ca. 1 nm) fall within the lower limit of the nanometric
scale so they can be considered as a limiting type of nanofiller.
CDs are shaped like truncated cones, with a hydrophobic cavity
and a hydrophilic exterior. The precise number of hydroxyl
groups according to the number of glucose units makes it
possible to establish strong interactions with certain polymeric
matrices. CDs can also be easily modified with other functional
groups which allow modulating such interactions according to
the nature of the matrix, or can be grafted to common nanofillers
(nanoparticles, layers,.) for improving the properties of the
material. The ability to form inclusion complexes with organic
molecules inside the cavity, e.g. a monomer for its further poly-
merization [11] also represents an added value to produce
a material with features different from those of the single
constituents of the nanocomposite. Yet, in spite of their exclusive
properties the use of CDs in the field of nanocomposites is still
very limited [12e14].
(
less than 5% by weight), a feature which has important conse-
quences on the final properties of the material [2]. Obtaining
a perfectly homogeneous dispersion is critical for a nanocomposite
to have the properties that are expected. However, regardless of the
method used for the mixing, this issue becomes more critical the
smaller the nanofiller is. Recently, high energy ball milling (HEBM),
a conventional method used for synthesis and processing of inor-
ganic materials [3e5], has been revealed as a new way of processing
thermoplastic matrix nanocomposite materials, not only due to its
potential results in terms of nanoparticle dispersion, but also from
an economical and clean point of view. Although the mechanical
action may have adverse effects of wearing on solid polymers by
*
Corresponding author. Tel.: þ34 948 425 600x6315.
E-mail address: gaitano@unav.es (G. González-Gaitano).
0143-7208/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.dyepig.2012.02.009

428
R. Serra-Gómez et al. / Dyes and Pigments 94 (2012) 427e436
A problem that arises when dispersing CDs in a polymer comes
of their interaction. In the solid state, however, the number of
investigations is limited, most of them related to the development
of solid state dye lasers [17,18] or to the use of CD-based resins to
adsorb organic dyes [19].
In this work we propose the use of the fluorescent properties of
Rhodamine B and Rhodamine 6G (hereafter RhB and Rh6G, Fig. 1a
from the reduced size of these macrocycles, which makes it difficult
to confirm that an adequate dispersion has taken place. AFM, SEM
or TEM microscopy can be used to this purpose, but these are
techniques which involve sample preparation processes that
require a consistent amount of time and effort, more focused to
a final product characterization and not convenient for regular
checks. In addition to that, the electron beam can melt the matrix
when thermoplastic polymers are being used.
A different approach may be to use fluorescence microscopy by
previously tagging the CD by inclusion of a fluorophore in the
cavity, provided it is stable enough. Rhodamine-based dyes present
unique fluorescent properties and are therefore used in a wide
variety of applications, ranging from applied chemistry and physics
to biochemistry or microbiology [8]. Rhodamines are known to
and b) and their ability to form complexes with bCD (Fig. 1c), to
produce a solid complex that acts as a fluorescent probe to eval-
uate the dispersion of these oligosaccharides into a thin-film
polymeric nanocomposite material prepared by HEBM. The poly-
meric matrix used here is ethylene vinyl acetate (EVA), a thermo-
plastic polymer with polar vinyl acetate groups and crystalline
ethylene domains. EVA copolymers are a good choice to combine
with these fluorophores, as they present optical homogeneity and
organic dyes usually show good compatibility with polymeric
matrices. In addition, their easy processability makes them
excellent for miniaturization and to be included in other
composite systems [20].
interact with
bCD in different ways, depending on the state of
aggregation of the fluorophore and its molecular form [15,16],
increasing or decreasing the fluorescence quantum yield as a result
Fig. 1. (a) Rhodamine B; (b) Rhodamine 6G; (c) Scheme of b-Cyclodextrin.

R. Serra-Gómez et al. / Dyes and Pigments 94 (2012) 427e436
429
In this research we have focused on the dispersion of the
macrocycle as the nanofiller in the polymer, leaving the grafting
of the CD to nanoparticles and further nanocomposite synthesis
for future projects. The first step is to evaluate that the
complexes between the fluorophore and the CD form and are
stable enough to endure the HEBM process, that they present
2.3. Characterization techniques
b
The crystalline structure of materials under study was charac-
terized by X-ray diffraction, XRD, on randomly oriented powder
preparations using a Bruker D8 Advance diffractometer with a X
Kristalloflex K760 X-Rays generator, with a copper anode emitting
a different fluorescent response when attached to the
b
CD than
typical X radiation K
angles were monitored from 2
in 2 ). Analysis of the XRD patterns was carried out with XRD
a
1
¼ 1.5417 Å at 40 kV and 30 mA. Diffraction
ꢀ
ꢀ
ꢀ
in free form and that they can be dispersed homogenously, so
they can be used as dispersion fluorescence probes in nano-
materials. For this purpose the systems have been previously
studied in solution to characterize their complex formation
characteristics (stability, temperature dependence, stoichiometry
and topology) and the fluorescent emission. Finally, the solid
state products are tested and combined with the polymer using
the HEBM method, to verify their adequate dispersion and
whether their properties are retained along the nanocomposite
processing.
q
¼ 2 e40 at a rate of 3 s/step (0.02
q
Wizard 2.4.11 software (Bruker GmbH). FTIR-ATR analyses of
powder and films were performed with a FTIR-ATR Nicolette Avatar
ꢂ1
360, using a resolution of 2 cm and averaging 32 scans. Spectral
analysis treatment was undertaken with the OMNIC E.S.P. v5.1
software (Nicolet). Thermogravimetric analysis, TGA, was carried
out in a TGA-SDTA 851 Mettler Toledo. Samples were subjected to
ꢀ
ꢀ
ꢀ
a heating program from 25 C to 600 C at a heating rate of 10 C/
min under a N atmosphere.
2
Fluorescence studies were undertaken using an Edinburgh
Instruments FLS920 spectrofluorimeter equipped with a 450 W
Xenon arc lamp. Samples were excited at 553 nm and the emission
recorded from 560 to 750 nm under constant stirring, averaging 5
scans with a 1 nm step and 0.1 s dwell time. Excitation and emission
slits were 2 nm and 3 nm, respectively. Measurements in solution
were carried out at 15 C, 25 C, 35 C and 45 C in a 10 mm path
length quartz cuvettes controlled by a Lauda Ecoline RE104 ther-
mostat. Each isotherm was repeated three times. For the solid
samples the powder or a portion of the film was sandwiched
between two quartz glasses in the instrument sample holder.
Fluorescence lifetimes were measured with the same equipment
using as the radiation source a PDL800-B Picoquant pulse diode
driver and 455 nm and 500 nm diodes, with full width at half
maximum (FWHM) of 1600 and 1700 ps, respectively. The instru-
ment response was measured by using a Ludox 30% aqueous
suspension, purchased from Aldrich. Data treatment was per-
formed with FAST v3.0 software (Edinburgh Instruments). The films
were observed with an Olympus CH40 fluorescence microscope
equipped with a ColorView camera (Soft Imaging Systems).
2
. Materials and methods
2.1. Materials
Rhodamine B 99% pure (Basic Violet 10; C.I. 45170; 9-(2-
ꢀ
ꢀ
ꢀ
ꢀ
Carboxyphenyl)-3,6-bis(diethylamino)xanthylium chloride) and
Rhodamine 6G 99% pure (Basic Red 1, C.I. 45160, ethyl 2-(6-(eth-
ylamino)-3-(ethylimino)-2,7-dimethyl-3H-xanthen-9-yl)benzoate
monohydrochloride) were purchased from Acros Organics.
b
-Cyclodextrin was bought from Wacker Chemicals (Cavamax
Pharma, 99.5% pure). Polyethylene-co-vinyl acetate (composition
3
ꢀ
ꢀ
12% by weight in vinyl acetate, density 0.933 g/cm at 25 C, Vicat
ꢀ
temperature ASTM D 1525 ¼ 65 C and melting point 95 C), was
supplied by Sigma Aldrich.
2.2. Sample preparation
ꢂ3
RhB:
b
CD complex was prepared by mixing a RhB 2.5 ꢁ 10
M
aqueous solution with
b
CD at an equimolar ratio followed by
stirring for 10 min, poured into a crystallizer and placed in the
stove at 70 C until solvent evaporation. The resulting solid is
NMR experiments were performed in a Bruker Avance 700
Ultrashield (700 MHz). The samples were prepared in D O (99.990%
2
ꢀ
a dark red flaked product. The physical mixture of RhB and
b
CD
in deuterium purchased from Sigma Aldrich), with no buffers
added, using the HDO signal as the reference. Monodimensional
experiments were done by averaging 256 scans. ROESY experi-
ments were carried out on 32 scans with presaturation of the
solvent signal [21] by using the pulse sequence described in the
literature [22], with an optimal mixing time of 600 ms. Tempera-
was prepared by weighing the same amounts used for the
complex preparation and mixing them in a vortex shaker,
resulting in a greenish powder mixture. The same procedure was
carried out with Rh6G. In this case the solid mixture and the
compound prepared by crystallization did not show any visual
differences.
ꢀ
ture was set to 25 C in all cases.
A mixture of 5% by weight of RhB:bCD and EVA was subjected to
HEBM under cryogenic conditions (cryomilling) in a MM400
RETSCH miller. Stainless steel milling tools (a vessel of 50 mL of
capacity and one ball of 20 mm diameter) were used. The process
was carried according to the following protocol: 1 h of active cry-
omilling divided into 12 cycles of 5 min of milling at 25 Hz and
3. Results and discussion
3.1. Complexes in solution: stoichiometry, stability and structure
1
5 min of resting in liquid nitrogen. Another vessel endured the
As a first stage, the complexes between rhodamines and bCD
have been investigated. The analysis of the chemical shifts of the H
NMR signals of the complex in relation to the signals from the pure
1
same procedure with the physical mixture of RhB and
EVA at the same proportions (95% by weight of EVA).
b
CD and the
Thin films were prepared by hot pressing. The powder obtained
from the milling processes was deposited between two Teflon
plates and then pressed and heated in an oven at 150 C for 20 min.
RhB and
the extent of the complex formation. For the
undergoing the most important changes are the H
b
CD (Table 1 and Figs. 2 and 3) are the main indication of
CD, the protons
and H , i.e.,
b
ꢀ
6
5
After that, the prefilms obtained were cooled inside the oven down
those located at the mid-bottom inner side of the cavity and at the
narrower rim of the macrocycle, respectively. Less significant
ꢀ
2
to 40 C. A small portion of the prefilm (about 9 mm ) was then
sandwiched between the twoTeflon plates and clamped within two
changes were noted for the inner H
CD. Finally, tiny shifts are detected in the outer protons H
. All these resonances except that for H shift upfield. If we
consider now the guest molecule, RhB, the protons of the substit-
uents CH e and CH e (downfield) and the H (upfield) experience
3
, at the mid-upper part of the
ꢀ
stainless steel plates, introduced in a preheated oven at 150 C and
1
, H and
2
heated for 120 min. The film was slowly cooled down inside the
oven to room temperature, thus avoiding any thermal stress in the
sample.
H
4
4
3
2
D

430
R. Serra-Gómez et al. / Dyes and Pigments 94 (2012) 427e436
Table 1
between the methyl group and aromatic protons Ha and Hb of RhB
with the CD.
Strong ROE cross peaks arise between the H
3.76 ppm and 3.55 ppm) with the CH group of the RhB, that of H
being the most intense. In addition, a small peak arises between H
and CH
RhB is entering the cavity by the secondary hydroxyl rim. At
.62 ppm and 6.78 ppm two more interactions arise, corre-
sponding to the Ha and Hb from the RhB xanthene rings with the
of the CD. The intramolecular interaction of Ha and Hb with
the CH e can be ruled out, as no cross peaks arise in the ROESY
spectrum for the RhB alone in water, confirming the intermolec-
ular interaction with H
Changes in the chemical shifts of protons of bCD and RhB.
H
Free
d
(ppm)
Dd (ppm)
3 5
and H of the CD
(
3
5
b-Cyclodextrin
H
H
H
H
H
H
1
2
3
4
5
6
4.949
3.541
3.856
3.478
3.748
3.769
0.039
0.006
0.074
6
3
protons. These interactions imply unambiguously that
ꢂ0.015
6
0.155
0.101
Rhodamine B
H
b
5
H
H
H
H
H
H
H
H
H
H
CH3
CH2
A
1.115
3.455
6.624
6.772
6.786
7.100
7.313
7.608
7.608
7.834
ꢂ0.059
ꢂ0.079
ꢂ0.028
ꢂ0.019
ꢂ0.018
ꢂ0.013
0.068
ꢂ0.013
ꢂ0.013
ꢂ0.008
2
5
.
B1
B2
C
The fact that the interaction is taking place not in the central
carboxyphenyl ring, but in one of the diaminoethyl groups opens
the possibility of 2:1 complex, i.e. two CDs per guest. Job’s
continuous variation analysis [24] has been used to determine the
actual stoichiometry. For this purpose samples at different molar
D
E
F
G
ratios prepared with
b
CD and RhB concentrations ranging from 0 to
ꢂ4
1
5
ꢁ 10
2
M were prepared in D O and analyzed by H NMR
the larger shifts. These results prove unambiguously the inclusion
of the RhB in the CD, which involves the ethyl amino groups [23].
spectroscopy.
The Job plot for protons H
3
, H
2
of
bCD and CH
3
G D
, H and H of RhB
1
H ROESY experiments can provide more detailed information
is shown in Fig. 5, so a signal from inside the bCD cavity and one
about the inclusion mode through the intensity of the cross peaks
in the 2D spectrum, related to the closeness between protons of
host and guest. Fig. 4 shows expanded views of the correlation
from outside can be seen. The maximum change in the chemical
shift takes place when the molar ratio is 0.5, inferring that the
stoichiometry of the complex is 1:1.
Fig. 2. 1H NMR spectra in D
O of: (a) RhB; (b) bCD; (c) molar ratio 1:1.
2

R. Serra-Gómez et al. / Dyes and Pigments 94 (2012) 427e436
431
Fig. 3. 1H NMR spectra of different molar ratios RhB:
RhB range from 0 to 5 ꢁ 10 M.
b
CD. Concentrations of
bCD and
ꢂ4
As seen in Table 2, the protons of the carboxyphenyl ring also
shift, although no NOE signals are detected. A possible explanation
to the fact the stoichiometry sticks to 1:1 when having two dia-
Fig. 5. Job plot for selected protons of bCD and RhB.
minoethyl groups in the molecule, is that when the bCD approaches
one of the diaminoethyl branches, it might induce the movement of
the carboxyphenyl ring to the opposite side, as seen by the shift in
H
D
, and thus the molecule is not receptive to dock with another
b
CD
molecule in the other ring because of steric effects. Those same
experiments were performed on the Rh6Ge CD system, but neither
b
significant changes in the chemical shifts nor cross peaks in the
ROESY spectrum were detected. These results confirm that only RhB
forms a suitable stable inclusion complex with a 1:1 stoichiometry,
where the bCD enters the RhB by one of its diethylamine sides.
Fluorescence emission can be used to gather precise informa-
tion about the stability of the association. RhB presents a high
fluorescence quantum yield, but it easily aggregates forming dimers
and other species, especially in aqueous solution [25]. This process
manifests as a quenching in the emission attributed to the long
range dipoleedipole energy transfer from the monomer excited
state to the aggregates [16] and also in bathochromic shifts from
ꢂ3
5
87 nm to 650 nm in 10 M RhB solutions. Rhodamines and
b
CD
may interact in different ways depending on the fluorophore
concentration and its aggregation state. Thus, in dilute solutions of
RhB, the bCD causes a decrease in emission due to the formation of
the complex, less fluorescent than the free RhB. However, in
conditions at which RhB is in the form of aggregates and its fluo-
rescence quenched, the addition of
bCD produces the opposite
effect, yielding a fluorescent enhancement [26]. On the other hand,
Rh6G does not aggregate as easily as the RhB [27] because of the
bulkiness of the ester group in the phenyl ring and the methyl and
ethyl substituents, which hinder the molecule stacking. Conse-
quently, the presence of
Rh6G in aqueous solution.
The effect of successive additions of
b
CD enhances the emission in the case of
ꢂ6
b
CD to RhB 8 ꢁ 10 M is
shown in Fig. 6. Free RhB has its maximum fluorescent emission
at 582 nm, and experiences a slight blue shift of ca. 4 nm when
b
CD is added, along with a quenching in its fluorescence. Under
our instrumental conditions, the linear response of the
Table 2
Binding constants for the complex
b
CD:RhB ([RhB] ¼ 8 ꢁ 10ꢂ6 M).
ꢀ
ꢀ
ꢀ
ꢀ
15
C
25
C
35
C
45 C
D
H
DS
(J mol
ꢂ1
ꢂ1 ꢂ1
K )
(kJ mol
)
ꢂ3
K$10
L mol
5.1 ꢃ 0.5 4.8 ꢃ 0.2 4.1 ꢃ 0.2 3.1 ꢃ 0.2 ꢂ15 ꢃ 3 21 ꢃ 10
ꢂ1
Fig. 4. (a and b). Zoomed view of the ROESY spectrum of RhB:bCD (1:1).

4
32
R. Serra-Gómez et al. / Dyes and Pigments 94 (2012) 427e436
ꢂ6
fluorescence for the RhB ranges up to 8 ꢁ 10
M, free of
where i sums over all the concentrations of CD and l over all the
aggregation effects.
wavelength range. The input parameter is a vector that contains
the initial guess for the binding constants and , and the output is
As the complex stoichiometry for the
b
CD is 1:1, its formation
f
i
constant at a certain temperature can be expressed by the action
mass law as:
the estimation of the parameters with their error bounds, defined
as the confidence intervals corresponding to a significance level,
a
¼ 0.16. A weight factor,
difference between F0, , and the maximum value reached in the
binding, is introduced at each wavelength in Eq (5), to give a higher
statistical weight to that at which the changes in intensity are
l
u , taken as the absolute value of the
½
R : CDꢄ
l
K ¼
(1)
½
Rꢄ½CDꢄ
l
being R the free fluorophore, CD the free
b
CD and R:CD the complex.
higher [28]. The enthalpy and the entropy have been obtained from
the K dependence on the temperature through Van’t Hoff equation
and a weighted least-square method (Table 2).
The binding constants of RhB in water are relatively high, which
indicates a stable association. The increasing temperature produces
the diminution of K, as expected in an exothermic process. This
same trend has been obtained by measuring at lower concentra-
tions of RhB (data not shown). This stability with the temperature is
important, considering that the formation of the nanocomposites
requires conditions that, depending on the type of polymer, must
In the experiments, the concentration of R is kept constant, varying
that of CD. By combining the action mass law with the concentra-
tion mass balance for host and guest we get for [R] the following
quadratic equation:
ꢀ
ꢁ
1
K
R0
2
½
Rꢄ þ CD ꢂ R þ
½Rꢄ ꢂ K ¼ 0
(2)
0
0
The measured fluorescence at a certain wavelength, F
result of the contributions of the two fluorescent species, R and
R:CD, so thus
l
, is the
ꢀ
be higher than 120 C. As for the Rh6G:bCD system, it barely
experiences changes in the emission by adding CD, resulting in the
non-convergence of the fitting toward very low constants. This is
a confirmation that there is no significant complex formation, in
accordance with NMR data. It is worthy to mention that the
enthalpy is not too high for these types of complexes, being the
process controlled mainly by the entropy. According to ROESY data
and considering the bulkiness of RhB and the size of the cavity, the
R
R:CD
Fl ¼ F þ F
¼ a ½Rꢄ þ b ½R : CDꢄ
(3)
l
l
l
l
where a
l
and b
l
are constants related to the fluorescence quantum
yield and molar absorptivity of each fluorescent species at the
excitation wavelength, , and to experimental conditions (source
intensity, slit width and path length of the cell).
Dividing the above expression by F , i.e., the fluorescence
¼ a
in the absence of cyclodextrin, eq. (3) can be written as
l
0
l 0
R
ꢀ
ꢁ
F
F0
½R : CDꢄ
¼
½Rꢄ þ
f
(4)
R
0
l
where
f
¼ b /a
l
l
. The experimental data at a certain emission
wavelength can thus be fitted by a non-linear least-squares
procedure to the above equation, in which and K are left as
f
adjustable parameters. It is possible to improve the fitting using
a wider set of data by taking into account the emission measured at
each wavelength, and not only to a lmax. A multivariable analysis
can be performed by imposing the condition that the binding
constants are the same for each wavelength. The error function to
be minimized becomes
X Xꢂꢂ
ꢃ
ꢂ
ꢃ
ꢃ
cal
meas
2
E ¼
Fi=F0;l
ꢂ F =F
(5)
i
0;
l
l
i
ꢂ4
Fig. 7. Lifetime analysis: (a) fraction of the species found on 1 ꢁ10 M (solid symbols)
CD on the fluorescence of RhB 8 ꢁ 10ꢂ6 M in water.
and 1 ꢁ10 M (open symbols) of RhB upon addition of
ꢂ5
bCD; (b) Lifetime distributions.
Fig. 6. Effect of the addition of
b

R. Serra-Gómez et al. / Dyes and Pigments 94 (2012) 427e436
433
inclusion is shallow and only part of the guest is included.
Although the association constant is in good agreement with the
one stated by Liu et al. [15], the enthalpy and entropy values are
considerably different, as they obtain a non-expected positive
ꢂ1
ꢂ1 ꢂ1
ꢀ
D
H ¼ 40.8 kJ mol and
D
S ¼ 0.21 kJ mol
K
at 25 C. These
unusual values are reasoned in terms of the extra desolvation due
to the lactonization of the hydrated benzoate moiety at the
conditions of the experiment. It must be noticed that the constants
obtained by this method are apparent constants, as the equilibrium
between the cationic, lactone and zwitterionic form exists.
However, as stated by Mchedlov-Petrossyan et al. [29] the fraction
of the RhB molecules converting to the colorless lactone form in
water at our concentration is less than 1%, so in our case we can rule
out the lactonization effect mentioned before as being responsible
for the different values. The use of a phosphate buffer 0.1 M to
maintain the pH at 7.20 may be one of the reasons for this differ-
ence, as the ionic strength of the solution with this precise buffer is
considerable and it is well known the quenching effect of some
buffers on the fluorescence [30]. In our case, the pH has not been
controlled by addition of buffers and the low concentration of RhB
makes the ionic strength virtually zero.
Fig. 9. X-ray diffractograms of the solid samples.
In order to check the state of aggregation of the RhB at different
concentrations and the effect the CD may have, fluorescence life-
time analyses were carried out upon these samples. In the exper-
reconvolution distribution analysis at 200 intervals between 0 and
5
0 ns. Free RhB has a lifetime response of 1.7 ꢃ 0.2 ns (Fig. 7), which
correlates with the values found in the reference, 1.52 ns with
emission at 400 nm and 1.68 ns at 560 nm [31]. Upon addition of
ꢂ4
ꢂ5
iments, RhB concentration was fixed at 1 ꢁ 10 , 1 ꢁ 10 and
ꢂ6
ꢂ3
1
ꢁ 10 M, and aliquots of a 5 ꢁ 10
M bCD stock solution were
bCD, a new mode appears with fluorescence lifetime of
added to gradually increase the bCD concentration and induce the
0
.75 ꢃ 0.04 ns, which is retained upon successive additions (Fig. 7),
complex formation. The decay curves were processed by
its fraction increasing with the
b
CD concentration, accordingly to
ꢂ4
the shift of the equilibrium. When RhB 1 ꢁ 10 M is tested at
500 nm in the absence of bCD, a considerable fraction appears at
approximately the same lifetime than the complex 0.97 ꢃ 0.03 ns,
which is attributable to the RhB aggregation effect at high
ꢂ5
concentrations. When RhB is at lower concentrations, from 1 ꢁ10
ꢂ7
to 1 ꢁ 10 M, and
bCD is absent, this fraction does not appear and
that corresponding to the free RhB is close to 100%. These results
indicate the aggregation limit is between 10 M and 10 M for
aqueous solutions, in good agreement with the literature values
ꢂ4
ꢂ5
[32].
3.2. Solid complexes
The thermogravimetric analysis of the powder samples (Fig. 8)
shows how the RhB presents at least a three-step decomposition
Fig. 8. TGA curves (top) and 1st derivative (bottom) of the RhB samples.
Fig. 10. Fluorescence response of the solid samples.

434
R. Serra-Gómez et al. / Dyes and Pigments 94 (2012) 427e436
mixture they lose shape and are hidden in the baseline. Taking into
account that the solid complex is formed from the aqueous solu-
tion, and with the previous data confirming that the Rh6G does not
form complex with the bCD, we must conclude that the precipitate
will be a bare physical mixture of both components.
X-Ray diffraction patterns (Fig. 9) corroborate the above results.
The analysis shows a decrease in the crystallinity, as seen by the
absence of peaks and the amorphous halo when the complex is
formed in comparison to the commercial RhB and the physical
mixture. The latter itself matches the sum of the RhB and the bCD
diffractograms. As expected, Rh6G shows no evidence of complex
formation.
Finally, solid state fluorescence measurements were also
recorded on the commercial RhB, the physical mixture RhB-
bCD
and the RhB: CD complex. The emission spectra show a 20 nm
b
blue-shift upon complex formation. Commercial RhB has an
emission peak at 690 nm, and the complex at 670 nm. The physical
mixture behaves exactly like the commercial RhB (Fig. 10).
3.3. Effect of milling
The last step of the nanocomposite preparation is the milling of
the fluorescent probe in its complex form with the polymeric
matrix ensuring that the inclusion is kept after the HEBM and that
the probe is homogeneously distributed throughout the matrix. It is
known that the severe mechanical conditions occurring in these
processes may break bonds or produce free radicals [9]. For this
reason, the bCD has been milled alone under the same conditions as
Fig. 11. Comparison of the fluorescence of films produced with RhB:
the physical mixture. Top: normalized spectra; bottom: as measured.
b
CD complex and
1
those of the films. According to H NMR data, no changes are
perceived in the spectrum after milling, which indicates that the
process does not alter the chemical structure of the macrocycle.
Regarding the nanocomposites, after the products were milled and
processed into thin films, the fluorescence response of the film
ꢀ
ꢀ
process, starting around 180 C and ending at 500 C, and so does
the physical mixture RhB- CD. However, as confirmed by the first
derivative trace, once the RhB: CD complex is formed the decom-
b
b
containing the RhB:bCD complex presented a four-fold fluores-
position process changes to a single step process, confirming that
the inclusion yields a product with a different thermal behavior.
This difference can be seen by the absence of the first minimum
cence enhancement (Fig. 11, bottom) compared to the physical
mixture and a blue shift of 40 nm (Fig. 11, top), which is in good
agreement with the results of the solid samples before milling. It is
worthy to mention that the trend is the same than in concentrated
solution of RhB, where the fluorophore is extensively aggregated
ꢀ
around 180 C and the shift of the degradation temperature to
ꢀ
2
57 C. When the same analysis is carried out with the Rh6G, only
a slight shift in the temperature can be seen, but there is no
evidence concerning the formation of a different product.
Confirming the results revealed above, FTIR-ATR analysis of the
and the addition of
rescent response, as explained above. All this confirms that the
RhB: CD complex can be used as a probe to monitor the dispersion
of the oligosaccharide alone or attached to other nanostructure, as
nanoparticles, as RhB behaves differently when bound to the CD
than when it is free in the matrix and that the complex is capable of
enduring the extreme conditions of the HEBM process without
losing its properties.
bCD produces an enhancement on the fluo-
b
compounds shows that the RhB-bCD system presents important
differences in the spectra compared to the physical mixture,
b
whereas the Rh6B-
b
CD spectrum cannot be distinguished from the
ꢂ1
mixture. The strong band at 1587 cm of RhB is assigned to the
aromatic ring CeC vibrations and remains when forming the
complex (data not shown). The rest of the RhB peaks in the
aromatic region persist in the complex, while in the physical
The films are shown in Fig. 12. The one containing the complex
presents a uniform and homogeneous appearance in contrast to the
Fig. 12. Film formed with the complex (left) and with the physical mixture (right).

R. Serra-Gómez et al. / Dyes and Pigments 94 (2012) 427e436
435
Fig. 13. 20ꢁ Fluorescent microscope image of the RhB:
b
CD complex film (left) and physical mixture (Right). Scale bar 200
mm.
l
ex ¼ 510 nm.
Fig. 14. 100ꢁ Microscope image of the RhB:bCD complex film (left) and physical mixture (Right). Scale bar 20 mm.
film with the simple physical mixture, also prepared by HEBM. The
latter presents a darker color with white spots corresponding to
phase, as stated by the disappearance of the endothermic peak at
200 C in TGA, characteristic of the RhB, and to the blue shift and
ꢀ
macroscopic domains of aggregated
a crystalline structure, as XRD experiments have shown, it seems
that in order to achieve a good dispersion of the CD in the solid
b
CD. As the complex lacks
emission enhancement observed by fluorescence. On the other
hand, Rh6G does not form complexes either in aqueous solution or
in the solid phase.
b
phase it is necessary to break its crystalline arrangement and
convert it to an amorphous form.
The RhB complex mixed by cryo HEBM with the polymer, EVA,
after the subsequent film production is homogeneously scattered
through the matrix, undergoing a four-fold enhancement in its
fluorescence, that is not observed with the physical mixture. The
Images at different scales of the films with the RhB:bCD
complex and the physical mixture were taken with a fluorescence
microscope. As it can be seen in Figs. 13 and 14, using the same
diaphragm and magnification conditions (20ꢁ and 100ꢁ), the
use of a fluorescent complex with
breaking the crystalline structure of the
b
CD has thus the double effect of
CD by forming an amor-
b
RhB:bCD film clearly shows an enhanced fluorescent response in
phous phase that makes possible the proper dispersion of the
macrocycle and to enhance the “visibility” of the macrocycle. These
results, apart from eliminating the need of using more sophisti-
cated techniques as SEM or AFM, are important for subsequent
investigations of nanocomposites based in CDs, either as nanofillers
by themselves or attached to other nanostructures.
relation to the other when excited at 510 nm. In addition to that,
the complex produces an excellent dispersion throughout the
matrix, whereas the mixture presents local domains of RhB at the
bottom section and other sectors that are richer in
bCD, where
there is no fluorescence response at all resulting in a dark image.
4
. Conclusions
Acknowledgments
The association between
bCD and RhB and Rh6G has been
Authors acknowledge the financial support of the projects
MAT2010-16815 and CTQ2010-18564 from the Ministerio de
Ciencia e Innovación, as well as the PhD grant for Rafael Serra
provided by the “Asociación de Amigos de la Universidad de
Navarra”.
1
studied both in solution and solid state by different techniques. H
NMR results combined with fluorescence spectroscopy confirm the
formation of a stable complex of RhB of 1:1 stoichiometry, with
a binding constant of the order of 10
temperatures. The mode of inclusion has been elucidated with the
aid of ROESY spectra, proving that the RhB enters the CD by any of
the ethylammonium substituents toward the wider rim of the
macrocycle, leaving exposed to the solvent the moiety of the RhB
that bears the carboxylic group. Analysis of the solid products
makes clear that the complex of RhB retains its stability in the solid
4
ꢂ1
M
on a wide range of
b
References
[
1] Hussain F, Hojjati M, Okamoto M, Russell E. Review article: polymer-matrix
nanocomposites, processing, manufacturing and application: an overview.
J Comp Mat 2006;40:1511e75.

436
R. Serra-Gómez et al. / Dyes and Pigments 94 (2012) 427e436
[
[
[
2] Zhao R, Torley P, Halley PJ. Emerging biodegradable materials: starch- and
protein-based bio-nanocomposites. J Mater Sci 2008;43:3058e71.
3] Suryanarayana C. Mechanical alloying and milling. Prog Mater Sci 2001;
[17] O’Conell RM, Saito TT. Plastics for high-power laser applications e a review.
Opt Eng 1983;22(4):393e9.
[18] Peterson OG, Snavely BB. Stimulated emission from flashlamp-excited organic
dyes in polmethyl methacrylate. Appl Phys Lett 1968;12(7):238e40.
[19] Crini G. Kinetic and equilibrium studies on the removal of cationic dyes from
aqueous solution by adsorption onto a cyclodextrin polymer. Dyes Pigm 2008;
77:415e26.
46(1e2):1e184.
4] Padella F, Incocciati E, Nannetti CA, Colella C, Casadio S, Magini M. Mechan-
ically activated low temperature synthesis of Sr doped lanthanum manganite.
Mechanically alloyed. J Metastable Nanocryst Mater Part 1 1998;269(2):
1
05e10.
[20] López Arbeloa F, López Arbeloa T, López Arbeloa I, Costela A, Garcia-Moreno I,
Figuera JM, et al. Relations between photophysical and lasing properties of
rhodamines in solid polymeric matrices. Appl Phys B 1997;64:651e7.
[21] Hwang TL, Shaka AJ. Water suppression that works. Excitation sculpting using
arbitrary wave-forms and pulsed-field gradients. J Magn Reson Ser A 1995;
112(2):275e9.
[
[
5] Rowlands SA, Hall AK, McCormick PG, Street R, Hart RJ, Ebell GF, et al.
Destruction of toxic materials. Nature 1994;367(6460):223.
6] Castrillo PD, Olmos D, Amador DR, González-Benito J. Real dispersion of iso-
lated fumed silica nanoparticles in highly filled PMMA prepared by high
energy ball milling. J Colloid Interface Sci 2007;308(2):318e24.
[
[
[
7] Olmos D, Dominguez C, Castrillo PD, González-Benito J. Crystallization and
[22] Bax A, Davis DG. Practical aspects of two-dimensional transverse NOE spec-
troscopy. J Magn Reson Ser A 1985;63(1):207e13.
[23] Bernini A, Spiga O, Ciutti A, Scarselli M, Bottoni G, Mascagni P, et al. Studies of
the inclusion complex between
b-cyclodextrin and paroxetine. Eur J Phar Sci
2004;22:445e50.
[24] Job P. Formation and stability of inorganic complexes in solution. Annali Chim
Appl 1928;9:113e203.
[25] Mchedlov-Petrosyan NO, Kholin YV. Aggregation of rhodamine B in water.
Russ J Appl Chem 2004;77(3):414e22.
final morphology of HDPE: effect of the high energy ball milling and the
presence of TiO
2
nanoparticles. Polymer 2009;50(7):1732e42.
8] González-Benito J, González-Gaitano G. Interfacial conformations and
molecular structure of PMMA in PMMA/silica nanocomposites. Effect of high-
energy ball milling. Macromolecules 2008;41(13):4777e85.
9] Shaw WJD. Current understanding of mechanically alloyed polymers. Mater
Sci Forum 1998;269(2):19e29.
[
[
10] Padella F, Magini M, Incocciati EU. Patent N 0963825 B1, Bulletin; 2003/
956, 57.
3
[26] Degani Y, Willner I. Lasing of rhodamine B in aqueous solution containing
11] González-Gaitano G, González-Benito J. Pseudorotaxanes of cyclodextrin and
diglycidyl ether of bisphenol A as precursors of new intramolecularly rein-
forced epoxy-based thermosets. Supramol Chem 2008;20:335e44.
b
-cyclodextrin. Chem Phys Lett 1984;104(5):496e9.
[27] Politzer IR, Crago KT, Hampton T, Joseph J. Effect of -cyclodextrin on the fluo-
b
rescence, absorption and lasing of rhodamine 6g, rhodamine b and fluorescein
disodium salt in aqueous solutions. Chem Phys Lett 1989;159(2e3):258e62.
[28] Sainz-Rozas PR, Isasi JR, González-Gaitano G. Spectral and photophysical
properties of 2-dibenzofuranol and its inclusion complexes with cyclodex-
trins. J Photoch Photobio A Chem 2005;173:319e27.
[29] Mchedlov-Petrossyan NO, Kakhtik VI, Bezugliy VD. Dissociation, tautomerism
and electroreduction of xanthene and sulfonephthalein dyes in N,N-
dimethylformamide and other solvents. J Phys Org Chem 2003;16:380e97.
[30] Schulman SG. Acidebase chemistry of excited singlet states. In: Wehry EL,
editor. Modern fluorescence spectroscopy, vol. 2. London: Heyden and Sons
Ltd.; 1976. p. 239e74.
[
[
[
[
[
12] Xu M, Wu S, Zeng F, Yu C. Cyclodextrin supramolecular complex as a water-
soluble ratiometric sensor for ferric ion sensing. Langmuir 2010;26(6):
4529e34.
13] Patel K, Angelos S, Dichtel WR, Coscun A, Yang YW, Zink JI, et al. Enzyme-
responsive snap-top covered silica nanocontainers. J Amer Chem Soc 2008;
130:2382e3.
14] Wu J, Gao C. Click chemistry approach to rhodamine B-capped polyrotaxanes
and their unique fluorescent properties. Macromol Chem Phys 2009;210:
1697e708.
15] Liu Y, Chen Y, Li B, Wada T, Inoue Y. Cooperative multipoint recognition of
0
0
organic dyes by bis(
b
-cyclodextrins) with 2,2 -bipyridine-4,4 -dicarboxy
[31] Magde D, Rojas GE, Seybold P. Solvent dependence of the fluorescence life-
times of xanthene dyes. Photochem Photobiol 1999;70:737e44.
[32] Ferreira JAB, Costa SMB. Non-radiative decay in rhodamines: role of 1:1 and
tethers. Chem Eur J 2001;7:2528e35.
16] Saenger WR, Jacob J, Gessler K, Steiner T, Hoffmann D, Sanbe H, et al.
Structures of the common cyclodextrins and their larger analogues e beyond
the doughnut. Chem Rev 1998;98(5):1787e802.
1:2 molecular complexation with b-ciclodextrin. J Photoch Photobio A Chem
2005;173:309e18.