9866 Langmuir 2006, 22, 9866-9873

Role of Surfactants in the Sorption of the Whitening Agent Tinopal

CBS onto Viscose Fibers: A Fluorescence Spectroscopy Study
Eduardo T. Iamazaki and Teresa D. Z. Atvars*
Instituto de Quı́mica, UniVersidade Estadual de Campinas, Caixa Postal 6154,
13084-971 Campinas, SP, Brazil

ReceiVed May 9, 2006. In Final Form: September 4, 2006

In the present work, we studied the role of an anionic surfactant, sodium dodecyl sulfate, and a cationic surfactant,
dodecyltrimethylammonium chloride, in the sorption of 4,4′-distyrylbiphenyl sodium sulfonate (Tinopal CBS) onto
modified cellulose fibers. Fluorescence spectroscopy was used to quantify the amount of sorbed Tinopal CBS on the
fiber surface. Differences in the spectral properties and the efficiency of sorption of the whitener/surfactant/fiber
system are explained in terms of electrostatic interactions. Our results also show that the sorption efficiency is greater
for solutions containing cationic surfactants only below the critical micelle concentration, while anionic surfactants
show a smooth influence on the sorption process.

Introduction In general, the whitening of textiles by FWA is obtained by

Fluorescent whitening agents (FWAs), also called optical
brighteners, are fluorescent organic compounds with a natural
affinity for various substrates. These agents are used to obtain
whiter and brighter fabrics, paper, and plastics. They absorb in
the UV region and emit in the visible spectral region.1-3 FWAs,
in general, are stilbene-type agents that have been extensively
used in the textile and detergent industries. Optical brighteners
possess one or two stilbene double bonds and are commercially
available in the E or E,E form because the other isomers exhibit
low or a lack of fluorescence.4,5
Among textile fibers, regenerated cellulose is widely em-
ployed.6 Cellulose is a syndiotactic polyacetal of glucose whose
macromolecules are maintained together by intermolecular
hydrogen bonds, which explains its high tendency to crystallize
and its ability to form fibrillar strands.7,8 Nevertheless, regenerated
cellulose fibers are a category of fibers whose properties depend
on their chemical composition, molar mass, degree of polym-
erization, and supermolecular arrangement. In addition, the degree
of crystallinity and the way the crystallites are oriented compared
with those of the original cellulose can be strongly modified.9
Cellulose fibers are a semicrystalline material, and thus, under
the usual conditions neither water nor aqueous solutions of
dyestuffs penetrate into the crystalline regions of these fibers.10
* To whom correspondence should be addressed. E-mail:
tatvars@iqm.unicamp.br.
(1) Leaver, I. H.; Milligan, B. Dyes Pigm. 1984, 5, 109-144.
(2) Canonica, S.; Kramer, J. B.; Reiss, D.; Gygax, H. EnViron. Sci. Technol.
1997, 31, 1754-1760.
(3) Kramer, J. B.; Canonica, S.; Hoigné, J.; Kaschig, J. EnViron. Sci. Technol.
1996, 30, 2227-2234.
(4) Kumar, G. S.; Neckers, D. C. Chem. ReV. 1989, 89, 1915-1925.
(5) Blanco, M.; Jimenez, L.; Valverde, I. EnViron. Toxicol. Chem. 2001, 20,
2193-2197.
(6) Encyclopedia of Polymer Science and Engineering; Mark, H. F., et al.,
Editorial Board; Kroschwitz, J. I., Editor-in-Chief; John Wiley & Sons: New
York, 1985; Vol. 6.
(7) Krässig, H. A. Cellulose: Structure, Accessibility and ReactiVity; Polymer
Monographs, Vol. 11; Gordon and Breach Science Publishers: Yverdon,
Switzerland, 1993.
(8) Kreze, T.; Strnad, S.; Stana-Kleinschek, K.; Ribitsch, V. Mater. Res.
InnoVations 2001, 4, 107-114.
(9) Cook, G. Handbook of Textile Fibres, 5th ed.; Merrow: Durham, U.K.,
1984.
(10) Encyclopedia of Polymer Science and Engineering; Mark, H. F., Editorial
Board; Kroschwitz, J. I., Editor-in-Chief; John Wiley & Sons: New York, 1985;
Vol. 3.
dipping the fibers into a hot bath. For sorption on the fiber surface
a strong FWA/surface interaction is necessary, which should
involve van der Waals, hydrogen-bonding, and hydrophobic
interactions as well as electrostatic forces.11-13 In the detergent
industry, FWA and surfactants are universal ingredients. There-
fore, when present, the surfactant should also play an important
role in the FWA/surface sorption process.
The interaction between surfactants and charged surfaces or
polymer chains has been a subject of considerable interest, with
potential applications in several areas.14-21 In particular, sorption
by surfactant interaction on an oppositely charged surface is
often a cooperative process inducing aggregation of the surfactant
at the solid/liquid interface.22-28 This phenomenon, initiated at
concentrations well below the bulk critical micelle concentration,
cmc, could be pictured as the formation of two-dimensional
micelles tied to the surface. The sorption results from both the
electrostatic interaction between the surface and the ionic head
of the surfactant and the lateral hydrophobic interactions between
the hydrocarbon chains to bring about reorganization and
aggregation of the adsorbed surfactant molecules on the substrate
(11) Mazeau, K.; Vergelati, C. Langmuir 2002, 18, 1919-1927.
(12) Timofei, S.; Schmidt, W.; Kurunczi, L. Simon, Z. Dyes Pigm. 2000, 47,
5-16.
(13) Rojas, O. J.; Ernstsson, M.; Neumann, R. D.; Claesson, P. M. J. Phys.
Chem. B 2000, 104, 10032-10042.
(14) Yan, Y. G.; Bein, T. Chem. Mater. 1993, 5, 905-907.
(15) Hayakawa, K.; Mouri, Y.; Maeda, T.; Satake, I.; Sato, M. Colloid Polym.
Sci. 2000, 278, 553-558.
(16) Okahata, Y.; Shimizu, A. Langmuir 1989, 5, 954-959.
(17) Wu, J.; Harwell, J. H.; ORear, E. A. J. Phys. Chem. 1987, 91, 623-634.
(18) Boufi, S.; Gandini, A. Cellulose 2001, 8, 303-312.
(19) Dekany, I.; Farkas, A.; Regdon, I.; Klumpp, E.; Narres, H. D.; Schwuger,
M. J. Colloid Polym. Sci. 1996, 274, 981-988.
(20) Wolfe, T. A.; Demirel, T.; Baumann, E. R. Clays Clay Miner. 1985, 33,
301-311.
(21) Esumi, K.; Toyoda, H.; Goino, M.; Suhara, T.; Fukui, H. Langmuir 1998,
14, 199-203.
(22) Harwell, J. H.; Hoskins, J. C.; Schechter, R. S.; Wade, W. H. Langmuir
1985, 1, 251-262.
(23) Fan, A.; Somasundaran, P.; Turro, N. J. Colloids Surf., A 1999, 146,
397-403.
(24) Bohmer, M. R.; Koopal, L. K. Langmuir 1992, 8, 2649-2659.
(25) Goloub, T. P.; Koopal, L. K. Langmuir 1997, 13, 673-681.
(26) Scamehorn, J. F.; Schechter, R. S.; Wade, W. H. J. Colloid Interface Sci.
1982, 85, 463-478.
(27) Chandar, P.; Somasundaran, P.; Turro, N. J. J. Colloid Interface Sci.
1987, 117, 31-46.
(28) Somasundaran, P.; Fuersten, D. W. J. Phys. Chem. 1966, 70, 90-96.

10.1021/la061309k CCC: $33.50 © 2006 American Chemical Society

Published on Web 10/28/2006
Tinopal CBS Sorption onto Viscose Fibers
Figure 1. Chemical structure of 4,4′-distyrylbiphenyl sodium
sulfonate salt (Tinopal CBS).
surface.22,29 The driving force for the surfactant association is
the reduction of the contact area between the hydrocarbon
lyophobic tails and the surrounding water molecules undergoing
an increase of configurational entropy.
Fluorescence spectroscopy is a very sensitive technique with
applications in numerous studies including aggregation in
microheterogeneous systems such as surfactant and polymer
solutions,30-32 properties of polymer films,33-36 the polarity of
solvents, surfaces, and polymers,37,38 and several others.39 The
reason for this sensitivity is that molecules in the electronically
excited state interact differently with their environment than does
the electronic ground state because the charge density distributions
in both states are different. Therefore, emission from the excited
to the ground state depends on both the time correlation between
environmental relaxation and the excited-state emission lifetime
and the specific interaction between the medium and the excited-
state molecule.40-42
Tinopal CBS, 4,4′-bis(2-disulfonic acid styryl)biphenyl with
molecular formula C28H18O6S2Na2 (Figure 1), is a whitening
agent (optical brightener) with high affinities for various
substrates, absorbing in the UV region and emitting light in the
blue region.43,44 It is widely used in the detergent, dyeing, printing,
and paper-making industries. Tinopal CBS is mainly used in
industry to increase the whiteness of cellulose fiber and cotton
and extensively used in synthetic washing powder and liquid
detergents. The effects of brightening cotton by this consistency
product are 2.7 times bigger than those of the fluorescent
brightener diphenylethene double triazine.45
In an earlier report we showed that the fluorescence spectrum
of Tinopal CBS sorbed onto viscose fibers is strongly modified
when compared with the spectrum in aqueous solutions.46 We
(29) Rupprecht, H.; Gu, T. Colloid Polym. Sci. 1991, 269, 506-522.
(30) Neumann, M. G.; de Sena, G. L. Colloid Polym. Sci. 1997, 275, 648-654.
(31) Iamazaki, E. T.; Schmitt, C. C.; Neumann, M. G. Langmuir 2001, 17,
3486-3490.
(32) Iamazaki, E. T.; Schmitt, C. C.; Neumann, M. G. J. Colloid Interface Sci.
2003, 264, 490-495.
(33) Barja, B. C.; Chesta, C.; Atvars, T. D. Z.; Aramendia, P. F. J. Phys. Chem.
B 2005, 109, 16180-16187.
(34) Xu, J. Q.; Luo, C. P.; Atvars, T. D. Z.; Weiss, R. G. Res. Chem. Intermed.
2004, 30, 509-525.
(35) Christoff, M.; Yamaki, S. B.; de Oliveira, M. G.; Atvars, T. D. Z. J. Appl.
Polym. Sci. 2004, 92, 830-838.
(36) Prado, E. A.; Yamaki, S. B.; Atvars, T. D. Z.; Zimerman, O. E.; Weiss,
R. G. J. Phys. Chem. B 2000, 104, 5905-5914.
(37) Benrraou, M.; Bales, B. L.; Zana, R. J. Phys. Chem. B 2003, 107, 13432-
13440.
(38) Pandey, S.; Redden, R. A.; Hendricks, A. E.; Fletcher, K. A.; Palmer, C.
P. J. Colloid Interface Sci. 2003, 262, 579-587.
(39) Winnik, M. A. Photophysical and Photochemical Tools in Polymer
Science: Conformation, Dynamics, Morphology; NATO ASI Series C: Math-
ematical and Physical Sciences, Vol. 182; D. Reidel Publishing Co.: Dordrecht,
The Netherlands, 1986.
(40) Birks, J. B. Photophysics of Aromatic Molecules, 2nd ed.; Wiley-
Interscience: London, 1970.
(41) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Academic
Press: New York, 1999.
(42) Martins, T. D.; Yamaki, S. B.; Prado, E. A.; Atvars, T. D. Z. J. Photochem.
Photobiol., A 2003, 156, 91-103.
(43) Blanco, M.; Jimenez, L.; Valverde, I. Text. Res. J. 2001, 71, 130-134.
(44) Stana, K. K.; Pohar, C.; Ribitsch, V. Colloid Polym. Sci. 1995, 273,
1174-1178.
(45) Beijing Prina Chemical Industry Co. Ltd., Huangcun Daxing District,
Beijing, China; http://www.tradekey.com/product_view/id/32373.htm.
(46) Yamaki, S. B.; Barros, D. S.; Garcia, C. M.; Socoloski, P.; Oliveira, O.
N., Jr.; Atvars, T. D. Z. Langmuir 2005, 21, 5414-5420.
Langmuir, Vol. 22, No. 24, 2006 9867
also showed that electrostatic forces play important roles in the
sorption process by comparing the emission on fiber surfaces
with the fluorescence spectrum in self-assembled layer-by-layer
films. From the Tinopal CBS chemical structure we can see that
it should interact electrostatically with the cellulose surface
through the sulfonic groups and via dipole-induced dipole
interactions involving the polar groups of the polyacetal moieties
of the cellulose fibers and the polarizable extended conjugated
π-electron system of Tinopal. Therefore, using the intrinsic
photophysical property of a specific FWA and the ability of
fluorescence spectroscopy to sense properties of the environment,
we extended here our study of the FWA sorption process on the
surface of modified cellulose fibers to include the presence of
surfactants. Here the role of surfactants in the efficiency of the
sorption process as well as the influence of surfactants on the
fiber/FWA interactions was evaluated by fluorescence spec-
troscopy. Two types of surfactants were employed, dodecylt-
rimethylammonium chloride (cationic) and sodium dodecyl
sulfate (anionic), and experiments were performed both below
and above their critical micelle concentration (cmc). The aim of
our study is to obtain an explanation of the performance already
observed for the simultaneous use of fluorescent brighteners and
surfactants in detergents and how these dyestuffs are sorbed in
textile fibers.
Experimental Details
Chemicals. Viscose fibers (from Vicunha Textile S.A., Brazil),
4,4′-distyrylbiphenylsulfonic acid sodium salt (Tinopal CBS, Sigma-
Aldrich), sodium dodecyl sulfate (SDS; Sigma, 98%), and dode-
cyltrimethylammonium chloride (DTAC) (Fluka, 97%) were used
as received. Pyrene (Sigma-Aldrich) was recrystallized twice from
ethanol before use. Milli-Q-purified water was used throughout.
UV-Vis Spectroscopy. The absorption spectra of Tinopal CBS
were determined using a Hewlett-Packard-8452A diode array
spectrophotometer linked to an IBM PC. All samples were examined
at room temperature (25 ( 1.0 °C).
Steady-State Fluorescence Spectroscopy. Steady-state fluo-
rescence spectra were recorded using an ISS PCI photon spectro-
fluorimeter. The pyrene spectrum in aqueous solution was obtained
by excitation at λexc ) 334 nm, with the emission being recorded
from λem ) 340 nm to λem ) 550 nm. Tinopal CBS emission spectra
in aqueous solutions and sorbed on the surface of viscose fibers
were recorded from λem ) 370 nm to λem ) 600 nm with λexc ) 348
nm. Excitation spectra were recorded from λexc ) 270 nm to λexc
) 410 nm with λem ) 430 nm. All samples were examined at room
temperature (25 ( 1.0 °C). For solutions a square 1 cm quartz cuvette
was employed. All the spectra of Tinopal sorbed were recorded
using dried fibers. They were dried for several days in an oven under
dynamic vacuum at 35 °C. Emission and excitation spectra were
collected for a bunch of fibers supported between two quartz slices.
The measurements were collected in backface mode. When in
solutions, before spectral recording, all fibers were removed from
the Tinopal solutions. When the remaining solution exhibited some
turbidity, measurements were only started after the complete
decantation of the scattering particles with the sample maintained
in a sealed flask.
Surfactant aqueous solutions below and above the SDS and DTAC
cmc values containing Tinopal were prepared, and the fluorescence
spectra were recorded. The Tinopal fluorescence intensity at λem )
410 nm (5.8 × 10-6 mol L-1) as a function of the surfactant
concentration was plotted, and the surfactant cmc was determined.
For comparison, we also employed aqueous solutions of the surfactant
containing pyrene, a well-known fluorophore for this purpose.31,47,48
Here we transferred a sufficient amount of a methanolic pyrene
(47) Iamazaki, E. T.; de Britto, D.; Schmitt, C. C.; Campana, S. P.; Neumann,
M. G. Colloid Polym. Sci. 2004, 283, 33-40.
(48) Neumann, M. G.; Schmitt, C. C.; Iamazaki, E. T. Carbohydr. Res. 2003,
338, 1109-1113.
9868 Langmuir, Vol. 22, No. 24, 2006 Iamazaki and AtVars

stock solution to a flask under a stream of nitrogen, after which a

certain amount of surfactant aqueous solution was added. The final
pyrene concentration was 5.0 × 10-7 mol L-1. Solutions were allowed
to equilibrate for at least 4 h prior to fluorescence measurements,
with λexc ) 334 nm; emission was recorded from λem ) 340 nm to
λem ) 550 nm. The vibronic ratio I1/I3 of the pyrene emission at
λem(1) ) 373 nm and λem(3) ) 384 nm was plotted versus the
surfactant concentration.
Quantification of the Tinopal CBS Sorbed onto the Viscose
Fibers. A 10 mL sample of Tinopal CBS aqueous solution (5.8 ×
10-6 mol L-1) was added to a flask containing approximately 0.3000
g of fiber. The flask was wrapped in aluminum foil and left in the
absence of light for 20 min. After this period the fibers were dried
in an oven under dynamic vacuum, at T ) 35 °C, for approximately
24 h, and then they were stored in a desiccator. The amount of
adsorbed Tinopal was determined by the difference between the
concentration that remains in solution and the initial concentration
of the solution. Concentrations were determined by fluorescence
spectroscopy using a calibration curve. This protocol was repeated
for aqueous solutions with and without surfactants (SDS and DTAC),
where the concentrations of surfactants were either below or above
the corresponding cmc.
Turbidy Measurements. The turbidity of solutions of Tinopal
CBS after the sorption process was determined using a model TAM-
300A turbidimeter from ALEMMAR.
ζ-Potential. A commercial ζ-potential analyzer (PALS ζ potential
analyzer, version 3.12, from Brookhaven Instruments Corp.) was
used to measure the electrophoretic mobility of Tinopal in aqueous
solutions, cellulose fibers immersed in water, and Tinopal aqueous
solutions with immersed fibers, 10 consecutive determinations were
taken for each sample at room temperature, and the data analyzed Figure 2. (a) Normalized absorption (s), excitation (0) and emission
are the average of these determinations. The ζ-potential was calculated (4) spectra of Tinopal CBS in aqueous solution (5.8 × 10-6 mol
from the electrophoretic mobility data using the Smoluchowski L-1) (λexc ) 350 nm). (b) Calibration curve (λem ) 410 nm) for the
Tinopal CBS fluorescence intensity as a function of the concentration
equation.
in aqueous solution.
Scanning Electron Microscopy. A JEOL JSM-6360LV scanning
electron microscope was used. Fibers were sputtered with gold/
palladium (80/20) alloy in a Bal-Tec MED 020 MCS 010 before (Figure 3a) with a profile similar to that for aqueous solutions
scanning. Images were recorded with an accelerating voltage of 20 without surfactant. However, it is noteworthy that the emission
kV. The image is shown in Figure 1 in the Supporting Information. intensities decrease with the increased surfactant concentration.
No excimer or aggregate emissions were observed, similar to the
Results and Discussion concentrated aqueous solutions. From the sigmoid plot of the
Photophysics of Tinopal CBS in Aqueous Solutions and fluorescence intensity versus surfactant concentration we de-
Sorbed onto Fibers. Electronic absorption and fluorescence termined the SDS cmc.
spectra of Tinopal CBS were studied in several systems: dilute Distinctly, the profile of the fluorescence spectrum of Tinopal
aqueous solutions, aqueous solutions as a function of the surfactant in aqueous DTAC solution shows remarkable differences when
concentration, and sorbed onto fiber surfaces. compared with that for pure aqueous solutions or aqueous SDS
Absorption spectra always show a characteristic absorption solutions (Figure 3b). The emission spectrum shows two well-
band centered at λabs ) 348 nm coincident with the excitation resolved vibronic peaks at λem ) 408 nm and λem ) 429 nm and
band λexc ) 350 nm (Figure 2a). The fluorescence spectrum of a shoulder (λem ) 465 nm) at the red edge. As we can see, there
Tinopal in aqueous solution exhibits a broad band centered at is a smaller Stokes shift for this system if compared with the
λem ) 430 nm, with a blue-edge shoulder around 415-420 nm. aqueous solutions without surfactant. One possible explanation
Fluorescence and excitation/absorption spectra are mirror images, for this result is that there is a hindrance for the biphenyl rotation
with a Stokes shift of 5062 cm-1 according to the peak maxima. when the molecule is interacting with the surfactant, and
This Stokes shift is remarkably larger for an aromatic molecule consequently, the number of molecular conformers is reduced.
with a rigid structure. There are some possible explanations for In the system fiber/Tinopal CBS/DTAC the spectrum is similar
this larger Stokes shift in solution, which should be attributed to that reported for Tinopal CBS sorbed onto cellulose fibers but
to different conformations of this molecule associated with the not similar to the emission spectrum of CBS in the LBL film due
rotation for the biphenyl moieties with very low activation energy to a significantly red shifted band.46 Furthermore, we also observe
in nonviscous solutions. A linear plot of the fluorescence intensity that there is a significant increase of the fluorescence intensity
at λem ) 415 nm versus the concentration of a Tinopal CBS from the addition of DTAC. In a previous report we explained
aqueous solution is only obtained for concentrations lower than that the differences in spectral profiles could be attributed to a
6.0 × 10-6 mol L-1 (Figure 2b), and this will be the concentration confinement of the Tinopal molecules in sorption sites, producing
range to be used for quantitative analytical purposes. No excimer a quasi-linear emission spectrum. In agreement with this, the
or aggregated emissions were observed for higher concentrations emission from a confined environment is less accessible to
although the inner filter effect becomes predominant and the quenching processes, and therefore, increases of the emission
total intensity decreases. intensity should be expected.40
Fluorescence spectra of Tinopal CBS in aqueous solutions Values of the cmc of DTAC and SDS observed in the sorption
containing SDS show a broad band centered at λem ) 430 nm process using Tinopal as a fluorophore are in agreement with
Tinopal CBS Sorption onto Viscose Fibers Langmuir, Vol. 22, No. 24, 2006 9869

Figure 3. Fluorescence spectra (λexc ) 348 nm) of aqueous solutions

of Tinopal (5.8 × 10-6 mol L-1) in the presence of (a) SDS and (b)
DTAC.

those determined by plotting the vibronic ratio I1/I3 of the pyrene Figure 4. Ratio of the vibronic band intensities I1/I3 of the pyrene
fluorescence spectrum ([Py] ) 5.0 × 10-7 mol L-1) versus fluorescence spectrum versus the concentration of (a) DTAC and
surfactant concentrations in aqueous solutions, where I1 and I3 (b) SDS. [Py] ) 5.0 × 10-7 mol L-1. For I1 λem ) 372 nm, for I3
are intensities of the pyrene bands at 373 and 384 nm, respectively. λem ) 384 nm, and λexc ) 334 nm.
Using a sigmoid fit, the cmc of DTAC is 19.6 mmol L-1 (Figure
4a), and that of SDS is 7.0 mmol L-1 (Figure 4b), which are in
agreement with the literature (20.0 and 7.0 mmol L-1,
respectively).49-51
According to the spectral differences of the Tinopal emission
in aqueous solutions due to the surfactants, we conclude that
with SDS (anionic surfactant), where the spectra are similar to
those of the aqueous solutions, the interaction surfactant/Tinopal
is repulsive and Tinopal remains in the aqueous solution or, at
most, at the micelle/water interface (Figure 5a). On the other
hand, the spectrum and, consequently, the interaction Tinopal/
DTAC (cationic surfactant) are similar to those of the layer-
by-layer system, and Tinopal is similarly sorbed onto the viscose
surface. Tinopal molecules preferentially interact with the micelle
surface, which is positively charged (Figure 5b). Therefore, the
electrostatic surfactant/Tinopal interaction plays an important
role in Tinopal distribution in micellar aqueous solutions.
Figure 6 shows the excitation (λem ) 430 nm) and emission
(λexc ) 348 nm) spectra of Tinopal CBS sorbed onto viscose Figure 5. Model for the structure of micelles with Tinopal
fibers for several sorption intervals. As can be seen, the emission distribution in aqueous solutions with SDS (a) and DTAC (b)
surfactants.
spectrum is composed of well-resolved vibronic bands centered
at λem ) 415 nm and at λem ) 438 nm with a shoulder at the
red edge, similar to our observation of the emission spectrum of DTAC (Figure 3b). We attribute the higher energy to the 0-0
layer-by-layer films46 as well as the aqueous solution containing band and the lower energy band to the vibronic 0-1 progression.
As previous discussed, this result is explained by the confinement
(49) Rosen, M. J. Surfactants and Interfacial Phenomena, 2nd ed.; John Wiley of the Tinopal CBS molecules within cavities of the fiber surfaces,
& Sons: New York, 1989. producing a quasi-linear (shaper) emission spectrum. It is well-
(50) Tadros, F. Surfactants; Academic Press: London, 1984.
(51) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1977, 99, 2039- known that greater spectral resolution can be observed for
2044. fluorophores in solutions or in solid matrixes when the cages
9870 Langmuir, Vol. 22, No. 24, 2006
Figure 6. Excitation (λem ) 430 nm) and fluorescence (λexc ) 348
nm) spectra of Tinopal sorbed onto viscose fibers for different times
(initial Tinopal concentration 5.8 × 10-6 mol L-1). Data are depicted
in Table 1 in the Supporting Information.
where the molecule is located have dimensions similar to those
of the fluorophore molecules.40,52 From the spectral similarity
among distinct media, we may conclude that electrostatic
interaction plays an important role in the fiber/Tinopal, Tinopal/
DTAC, and Tinopal in a layer-by-layer film systems.
It has been reported that cellulose fibers, natural as well as
regenerated, have a crystalline/amorphous microfibrillar structure
where the fibrils consist of a succession of crystallites and
intermediate less ordered amorphous regions. Lateral tie mol-
ecules connect laterally adjacent amorphous regions.53,54 Although
they should be semicrystalline, X-ray diffraction patterns for our
samples showed a completely amorphous material (figure not
shown). In fact, we observed that the crystallinity degree of
these fibers is very low, less than what can be determined by
X-ray diffraction using conventional procedures. Regardless of
the crystallintiy degree, it is well-known that dyes and fluorescent
molecules do not penetrate within the crystals.55-59 They can be
sorbed within the amorphous phase as well in the interface
between the amorphous and the crystalline domains. Therefore,
independently where the sorption sites are located, which we
have not identified in this study, we should say that Tinopal is
not located within the crystallites (if they are present). In addition,
we also observed by electron scanning microscopy that the
topology of the fibers is very flat on the micrometer scale (Figure
1 in the Supporting Information).
According to our measurements of the ζ-potential, modified
cellulose fibers have a small negative surface charge (-6.30
mV) when immersed in deionized water, which is in agreement
with other reports.60 In a similar way, Tinopal in deionized water
also has a small negative charge (-6.14 mV) and in principle
should exhibit a repulsive interaction with those fibers. Nev-
ertheless, the ζ-potential for the system fiber immersed in an
aqueous solution of Tinopal is positive (7.53 mV), which
(52) Coltro, L.; Dibbern-Brunelli, D.; Elias, C. A.; Talhavini, M.; de Oliveira,
M. G.; Atvars, T. D. Z. J. Braz. Chem. Soc. 1995, 6, 127-133.
(53) Schurz, J. Lenzinger Ber. 1994, 74, 37-40.
(54) Schurz, J.; Lenz, J. Macromol. Symp. 1994, 83, 273-289.
(55) Xu, J. Q.; Luo, C. P.; Atvars, T. D. Z.; Weiss, R. G. Res. Chem. Intermed.
2004, 30, 509-525.
(56) Schurr, O.; Yamaki, S. B.; Wang, C. H.; Atvars, T. D. Z.; Weiss, R. G.
Macromolecules 2003, 36, 3485-3497.
(57) Talhavini. M.; Atvars, T. D. Z.; Cui, C.; Weiss, R. G. Polymer 1996, 37,
4365-4374.
(58) Jang, Y. T.; Phillips, P. J.; Thulstrup, E. W. Chem. Phys. Lett. 1982, 93,
66-73.
(59) Phillips, P. J. Chem. ReV. 1990, 90, 425-436.
(60) Alila, S.; Boufi, S.; Belgacem, M. N.; Beneventi, D. Langmuir 2005, 21,
8106-8113.
Iamazaki and AtVars
Figure 7. Mass of Tinopal CBS sorbed per gram of fiber as a
function of immersion time (10 mL of aqueous Tinopal solution
with a concentration of 5.8 × 10-6 mol L-1 and constant temperature).
demonstrates that for some reason the negative charges of both
species are more than compensated. At this moment we do not
have a good explanation for this behavior although there is some
indication that the counterions probably play an important role.
Nevertheless, what we are certainly observing is that the
photophysical behaviors of Tinopal in the presence of either
DTAC or the fiber are similar, indicating that electrostatic forces
are important.
The excitation spectrum of Tinopal CBS sorbed on viscose
fibers shows a broad band centered at λexc ) 371 nm (Figure 6).
From the determination of the amount of Tinopal sorbed per
gram of fiber (Figure 7), we should say that a plateau is obtained
for approximately 60 min of dipping. This means that under our
experimental conditions (amount of fiber, amount of solution,
solution concentration, and temperature) a sorption equilibrium
is reached after 60 min of dipping. The intensity of the excitation
band increases for longer dipping intervals, demonstrating that
there is an increase of the amount of sorbed Tinopal. On the
other hand, the plot of fluorescence intensity at λem ) 410 nm
versus dipping time in a 10 mL aqueous solution containing 5.8
× 10-6 mol L-1 Tinopal shows a plateau after 60 min following
an initial increase, corresponding to a mass of sorbed Tinopal
around 0.300 g for 1 g of fiber. The plot is shown in Figure 7,
and the data for Tinopal sorbed onto fibers (mass by mass) (Table
1 in the Supporting Information) were determined according to
eq 1,
[Tinopal]sb ) [Tinopal]i - [Tinopal]f (1)
and we defined 20 min as the time for the dipping fibers in all
of the other experiments reported in this work. [Tinopal]sb,
[Tinopal]i, and [Tinopal]f are the amounts in grams of Tinopal
sorbed, in the initial solution, and in the final solution, respectively.
[Tinopal]f is determined in solution using the calibration curve
depicted in Figure 2b. In these experiments fibers are dipped into
the 10 mL aqueous Tinopal solutions with a concentration of 5.8
× 10-6 mol L-1 for 20 min.
Role of Surfactant in the Fiber Properties. To evaluate the
role of surfactant in the Tinopal sorption process, we performed
the experiments of Tinopal/fiber sorption for aqueous solutions
below and above the cmc of SDS and DTAC. In the first place,
we observed that both surfactants SDS and DTAC in aqueous
solutions induce changes of the fiber properties, independently
of the presence of Tinopal: The aqueous solution where the
fibers were dipped immediately becomes turbid. No turbidity is
observed in the absence of surfactant, and all of the initial
Tinopal CBS Sorption onto Viscose Fibers Langmuir, Vol. 22, No. 24, 2006 9871

Figure 9. Normalized fluorescence spectra of the sorbed Tinopal

CBS on fiber surfaces after the fibers were dipped into an aqueous
solution in the presence of different concentrations of (a) SDS and
(b) DTAC (λexc ) 348 nm).

Therefore, the sorption of FWA onto the fiber surfaces in the

Figure 8. Turbidy values for aqueous solutions containing viscose
fibers and (a) SDS and (b) DTAC. presence of surfactants depends on the relative strength of several
distinct interaction components arising from distinct interactive
Table 1. Average Values of the Turbidity in Nephelometric species: Tinopal/water, Tinopal/fiber, Tinopal/surfactant, sur-
Turbidity Units (NTUs) of Aqueous Solutions after the Viscose factant/water, and surfactant/fiber. In conclusion, because the
Fibers Were Dipped (Average of Five Measurements) fluorescence spectrum of Tinopal/water is similar to that of
[SDS] turbidity [DTAC] turbidity Tinopal/SDS, and that of Tiponal/DTAC is similar to that of
(mmol L-1) (NTUs) (mmol L-1) (NTU) Tinopal/fiber, emission spectra can be a fingerprint for these
1.0 1.98 ( 0.25 5.0 2.43 ( 0.25 distinct interactions.
2.0 2.50 ( 0.25 10.0 3.68 ( 0.25 Role of Surfactant in the Tinopal CBS Sorption onto Viscose
4.0 11.67 ( 2.5 20.5 4.48 ( 0.25 Fibers. Figure 9 shows the normalized fluorescence spectra of
6.0 29.67 ( 2.5 30.0 4.13 ( 0.25
8.0 8.67 ( 2.5 Tinopal CBS sorbed onto the fiber surfaces after the fibers were
12.0 4.23 ( 0.25 dipped into aqueous solutions in the presence of several
concentrations of SDS and DTAC. The fluorescence spectral
surfactant solutions with the same concentration are completely profile is distinct both for surfactant/Tinopal in solution and for
transparent. It is noteworthy that although all solutions are Tinopal/fiber, for both surfactants.
translucent, the highest light scattering (Table 1, Figure 8) is Figure 9a shows the fluorescence spectra of Tinopal sorbed
observed for surfactant solutions with concentrations similar to onto the fiber surfaces after the fibers were dipped into SDS
the cmc, i.e., 6.0 and 20 mmol L-1 for fiber/Tinopal/SDS and aqueous solutions with several surfactant concentrations. As
for fiber/Tinopal/DTAC aqueous systems, respectively. observed in other cases, the spectrum is similar to that observed
Because of the absence of scattering in both the surfactant on fibers, being composed of two separated vibronic bands. The
aqueous solution, independent of the surfactant concentration, emission profile is practically independent of the amount of SDS
and aqueous solutions with dipped fibers without surfactant, we in solution, and the apparent decrease of the relative intensity
conclude that the turbidity results from some phase separation of the higher energy 0-0 band around λem ) 410 nm compared
processes involving the fiber/surfactant interaction. As previously with the 0-1 vibronic band for a 7.0 mmol L-1 concentration
indicated, there is a possibility of surfactant aggregation on the is probably due to scattering effects of the surfactant solutions
fiber surface even though the surfactant concentration is below into which the fibers were dipped. Assuming, as suggested from
its cmc.22-28 Furthermore, higher turbidity is observed for the the greater turbidity of the SDS solution with fibers, that the
anionic surfactant SDS compared with the cationic surfactant fiber surface in the presence of Tinopal becomes positive (as
DTAC, indicating that the micro phase separation is more efficient confirmed by the measurements of the ζ-potential), negatively
for this system. This suggests that the fiber surface interacts charged SDS should also undergo some sorption onto the fiber
more strongly with the anionic surfactant, favoring phase surface if the interaction SDS/water molecules is weaker.
separation. According to this result, we conclude that the viscose Measurements of the ζ-potential for the system fiber/Tinopal/
fiber surface has small negative surface charges. SDS depend on the SDS concentration. It is initially negative for
9872 Langmuir, Vol. 22, No. 24, 2006 Iamazaki and AtVars

2 mM SDS (-10.26 mV), the negative charge increases to -36.15

mV for an SDS concentration of 8 mM, and then there is a charge
inversion to +14.67 mV for 12 mM SDS. According to these
data, the charge inversion occurs well above the SDS cmc. Then,
sorption of Tinopal and anionic surfactant molecules are
concurrent processes depending on the interspecies interactions,
and it seems to us that both can be sorbed although both have
negative charges and the apparent surface charges of the fibers
immersed in water are also negative. Regardless of the mechanism
and step-by-step sorption processes, we can say from our data
that Tinopal undergoes interaction with the fibers because the
spectral profile is different from that in solutions or in the presence
of SDS.
The emission spectra of the dried Tinopal/fiber after the fibers
were dipped into DTAC aqueous solutions at concentrations
below its cmc (5.0 and 10.0 mmol L-1) are similar to those
already observed for Tinopal on fibers and for Tinopal in aqueous
DTAC solutions, although the relative intensity of the 0-0 band
is lower than that of the 0-1 band (Figure 9b). The higher intensity
of the 0-1 vibronic bands can be explained by the higher
concentration of Tinopal on the surface, leading to the appearance
of the inner filter effect, as already indicated. Nevertheless, for
the dried Tinopal/fiber after the fibers were dipped into aqueous
DTAC solutions containing concentrations above its cmc (20
and 30 mmol L-1), we observe a blue shift of the fluorescence
spectrum, with the 0-0 vibronic band occurring at λem ) 400
nm with higher relative intensity and the 0-1 band at λem ) 430
nm. We also observed that the signal is very weak. A weaker
signal indicates that smaller amounts of Tinopal were sorbed;
therefore, the inner filter effect is minimized, and the 0-0 band
intensity increases. Figure 10. Relative ratio of Tinopal sorbed onto fibers (mass by
We also performed measurements of the ζ-potential for mass) as a function of the concentration of the surfactants after 20
solutions of Tinopal/DTAC with immersed fibers. According to min of immersion: (a) SDS, (b) DTAC. Data are depicted in Tables
2 and 3 in the Supporting Information, respectively.
these measurements, these systems show negative charge for
DTAC concentrations lower than its cmc (-7.37 and -13.23 above the cmc the amount of Tinopal sorbed is lower than in the
mV for 5 and 10 mM, respectively), and thus, there is a charge presence of SDS, supporting our earlier conclusion that there is
inversion for DTAC concentrations higher than the cmc (13.27 a competitive distribution of Tinopal between micelles in the
and 16 mV for 20 and 30 mM, respectively). Therefore, again aqueous system and the fiber surface. This is also in agreement
the charge inversion does not occur at the cmc of the surfactant, with the lower emission spectrum of Tinopal on fibers observed
and because there is an initial negative charge, DTAC can also after they were dipped into a solution with DTAC above the
be sorbed onto the fiber surface. Again, regardless of the cmc.
mechanism and step-by-step sorption processes, we can say from From the experiments of sorption of Tinopal onto the fibers,
our data that Tinopal undergoes interaction with the fibers because in the presence and absence of the surfactants, performed under
the spectral profile is different from that in solutions or in the the same conditions (dipping time of 20 min, 10 mL of Tinopal
presence of DTAC. at a concentration of 5.8 × 10-6 mol L-1, and constant
Quantitative Analysis of Tinopal CBS Sorption onto Viscose temperature), we observed that the ratio between Tinopal weight
Fibers. The quantitative analysis of the amount of sorbed Tinopal and fiber weight, mTinopal/mf, is around 4.9 × 10-7 in the absence
was made by measuring the difference of the emission intensity of surfactant (Table 1), 4.3 × 10-7 in the presence of 7.5 mmol
at λem ) 410 nm for the aqueous solution before and after the L-1 SDS (Table 2 in the Supporting Information), and 13.5 ×
sorption. The calibration curve for the fluorescence signal is 10-7 in the presence of 5.0 mmol L-1 DTAC (Table 3 in the
shown in Figure 2b. Taking the initial and final concentrations, Supporting Information). Nevertheless, above the DTAC cmc,
we determined the amount of Tinopal sorbed, using the the sorbed ratio is 4.35 × 10-7, which is similar to that in solutions
preestablished amount of fiber (in grams). Data for Tinopal sorbed without surfactant and also in the presence of SDS. These data
onto fibers (mass by mass) in the presence of SDS and DTAC show that the efficiency is greater for solutions of cationic
(Tables 2 and 3 in the Supporting Information, respectively) surfactants only below their cmc, while anionic surfactant shows
were determined according to eq 1. a very smooth influence on the sorption process.
Figure 10 illustrates the dependence of the amount of sorbed
Tinopal CBS per gram of fiber as a function of different amounts Conclusions
of SDS (anionic) and DTAC (cationic) surfactants in the aqueous Using fluorescence spectroscopy of an FWA, we showed that
solutions. Comparing the two systems, we observed that the electrostatic interaction should play an important role in the
largest sorption efficiency occurred for DTAC solutions below sorption process on viscose fiber surfaces. We also showed that
its cmc, while for SDS anionic surfactant solutions a greater Tinopal CBS should be employed as a sensor for determination
efficiency occurs as the surfactant concentration increases, of the cmc in aqueous surfactant solutions, in a manner similar
reaching a plateau at the cmc. It is noteworthy that for DTAC to that of the pyrene emission spectra, independent of the
Tinopal CBS Sorption onto Viscose Fibers
surfactant type (cationic or anionic). Nevertheless, the Tinopal/
cationic and Tinopal/anionic surfactant interactions are distinct,
and the fluorescence spectrum can be a fingerprint of such distinct
interactions. From the fluorescence emission profile we conclude
that there is a partition of Tinopal between the micelles and the
aqueous phase in micellar solutions and for cationic micelles
Tinopal is preferentially located within the micelles. The
fluorescence spectrum of Tinopal in confined environments shows
greater vibronic resolution, either in micellar solution or in the
solid state (sorbed onto fibers and in layer-by-layer films).
The role of surfactant in the sorption process depends on several
parameters, including the charge of the surfactant molecules and
surfactant concentration. Surfactant also interacts with the fiber
surface, increasing the interface concentration of surfactant that
undergoes phase separation. Quantitative analysis of the Tinopal
sorption onto the surface of viscose fibers shows that the anionic
surfactant has almost no influence while the cationic surfactant
plays an important role in the sorption only below its cmc. This
is a consequence of the relative strength of interactions between
surfactant micelles and Tinopal and between Tinopal and the
fibers.
It is noteworthy that there is no apparent correlation between
the efficiency of the sorption process and the ζ-potential
determined for every system. We have observed that the amount
of Tinopal sorbed per gram of fiber is almost constant in the
Langmuir, Vol. 22, No. 24, 2006 9873
system with SDS although there is a charge inversion form
negative to positive when the SDS concentration is increased
from 1 to 12 mM. Furthermore, a greater amount of Tinopal is
sorbed in the presence of DTAC solutions where the ζ-potential
is negative. Understanding these data requires further systematic
studies.
Acknowledgment. We thank FAPESP, CNPq, CAPES, and
MCT/CNPq/IMMP for financial support and fellowships. We
also thank Professor Carol Collins for useful discussions.
Supporting Information Available: Table 1, concentrations
of Tinopal in aqueous solutions and the weight by weight ratio of Tinopal
sorbed onto fibers (mass by mass) as a function of time (10 mL of
aqueous Tinopal solution with a concentration of 5.8 × 10-6 mol L-1
and constant temperature); Table 2, concentrations of Tinopal in aqueous
solutions and the weight by weight ratio between Tinopal sorbed onto
fibers (mass by mass) in the presence of SDS (dipping time timmersion )
20 min, 10 mL of Tinopal at a concentration of 5.8 × 10-6 mol L-1 and
constant temperature); Table 3, concentrations of Tinopal in aqueous
solutions and the weight by weight ratio of Tinopal sorbed onto fibers
(mass by mass) in the presence of DTAC (dipping time timmersion ) 20
min, 10 mL of Tinopal at a concentration of 5.8 × 10-6 mol L-1 and
constant temperature); and Figure 1, electron scanning micrograph of
a fiber surface. This material is available free of charge via the Internet
at http://pubs.acs.org.
LA061309K