Nanoscale conformational ordering in polyanilines investigated by saxs and afm
Understanding the adsorption mechanisms in nanostructured polymer films has become crucial for their use in technological applications, since film properties vary considerably with the experimental conditions utilized for film fabrication. In this paper, we employ small-angle X-ray
scattering (SAXS) to investigate solutions of polyanilines and correlate the chain conformations with morphological features of the nanostructured films obtained with atomic force microscopy (AFM). It is shown that aggregates formed already in solution affect the film morphology; in
particular, at early stages of adsorption film morphology appears entirely governed by the chain conformation in solution and adsorption of aggregates. We also use SAXS data for modeling poly(o-ethoxyaniline) (POEA) particle shape through an ab initio procedure based on simulated
annealing using the dummy atom model (DAM), which is then compared to the morphological features of POEA films fabricated with distinct pHs and doping acids. Interestingly, when the derivative POEA is doped with p-toluene sulfonic acid (TSA), the resulting films exhibit a fibrillar morphology—seen with atomic force microscopy and transmission electron microscopy—that is consistent with the cylindrical shape inferred from the SAXS data. This is in contrast with the globular morphology observed for POEA films doped with other acids.
Published on: Mar 3, 2016
Transcripts - Nanoscale conformational ordering in polyanilines investigated by saxs and afm
This article was published in an Elsevier journal. The attached copy
is furnished to the author for non-commercial research and
education use, including for instruction at the author’s institution,
sharing with colleagues and providing to institution administration.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
Author's personal copy
Journal of Colloid and Interface Science 316 (2007) 376–387
Nanoscale conformational ordering in polyanilines investigated by
SAXS and AFM
Fabio L. Leite a,b,∗
, Mario de Oliveira Neto b
, Leonardo G. Paterno c
, Michel R.M. Ballestero b
Igor Polikarpov b
, Yvonne P. Mascarenhas b
, Paulo S.P. Herrmann a
, Luiz H.C. Mattoso a
Osvaldo N. Oliveira Jr. b
a National Nanotechnology Laboratory for Agriculture (LNNA), Embrapa Agricultural Instrumentation (EMBRAPA), Rua XV de Novembro, 1452, P.O. Box 741,
CEP 13560-970, São Carlos-SP, Brazil
b Instituto de Física de São Carlos, USP, P.O. Box 369, CEP 13560-970, São Carlos-SP, Brazil
c Departamento de Engenharia de Sistemas Eletrônicos, EPUSP, 05508-900, São Paulo-SP, Brazil
Received 16 July 2007; accepted 28 August 2007
Available online 4 September 2007
Understanding the adsorption mechanisms in nanostructured polymer ﬁlms has become crucial for their use in technological applications,
since ﬁlm properties vary considerably with the experimental conditions utilized for ﬁlm fabrication. In this paper, we employ small-angle X-ray
scattering (SAXS) to investigate solutions of polyanilines and correlate the chain conformations with morphological features of the nanostructured
ﬁlms obtained with atomic force microscopy (AFM). It is shown that aggregates formed already in solution affect the ﬁlm morphology; in
particular, at early stages of adsorption ﬁlm morphology appears entirely governed by the chain conformation in solution and adsorption of
aggregates. We also use SAXS data for modeling poly(o-ethoxyaniline) (POEA) particle shape through an ab initio procedure based on simulated
annealing using the dummy atom model (DAM), which is then compared to the morphological features of POEA ﬁlms fabricated with distinct
pHs and doping acids. Interestingly, when the derivative POEA is doped with p-toluene sulfonic acid (TSA), the resulting ﬁlms exhibit a ﬁbrillar
morphology—seen with atomic force microscopy and transmission electron microscopy—that is consistent with the cylindrical shape inferred
from the SAXS data. This is in contrast with the globular morphology observed for POEA ﬁlms doped with other acids.
© 2007 Elsevier Inc. All rights reserved.
Keywords: Self-assembly; SAXS; AFM; TEM; Nanostructures; Thin ﬁlms; PANI; Adsorption and conformation
The discovery and development of new semiconducting
polymers have brought great promise for a number of techno-
logical applications [1,2] and posed new challenges in terms
of understanding fundamental properties of organic materials.
A key feature of these polymers is the possibility of alter-
ing their electrical and optical properties with small changes
in composition or even in the experimental procedures to pro-
duce the samples . Indeed, the incorporation of a substituent
group in a polymer such as polyaniline or polythiophene in-
duces considerable changes in the physicochemical properties,
* Corresponding author. Fax: +55 16 33725958.
E-mail address: firstname.lastname@example.org (F.L. Leite).
including solubility in organic solvents or in aqueous solu-
tions. Samples are normally produced in the form of ﬁlms,
which then allows another avenue to pursue in obtaining spe-
ciﬁc properties. If ﬁlms are fabricated with techniques such as
the Langmuir–Blodgett (LB) [4,5] or layer-by-layer (LbL) [6,7]
methods, for instance, the properties may be controlled at the
molecular level. One such example was demonstrated in LB
ﬁlms of polyaniline and a ruthenium complex , in which the
intimate contact between the molecules of the two components
led to electrical and electrochemical properties that differed
completely from those obtained for cast or spin-coated ﬁlms
of the same materials. The reason for these differences was elu-
cidated with Raman spectroscopy measurements, where PANI
was found to exhibit enhanced oxidation induced by the inti-
mate contact with the ruthenium complex .
0021-9797/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
Author's personal copy
F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387 377
The large ability to alter polymer properties is obviously
advantageous to achieve a rich variety of features. In the lu-
minescent, semiconducting polymers such as poly(p-phenyl-
enevinylenes), the emission quantum efﬁciency, color and po-
larization can be changed by mere introduction of functional
groups in the polymer backbone . In cases where con-
ductivity is the property to be exploited, orders-of-magnitude
changes are observed by varying the dopant concentration or
pH [11–13]. There is a considerable disadvantage though in
this ease with which polymer properties vary, particularly with
regard to industrial applications. Reproducibility in the experi-
mental results for any given type of sample or measurement is
relatively poor, especially if one compares with the inorganic
semiconductors. In the speciﬁc case of nanostructured ﬁlms
from polyanilines, several studies have shown that the adsorp-
tion mechanisms and the ﬁlm properties are entirely altered by
a mere change in pH . Parent polyaniline (PANI) and its
derivatives have been widely used mainly due to their remark-
able electrochemical, optical, electrical and mechanical prop-
erties, and good environmental stability in the emeraldine base
(EB) and emeraldine salt (ES) states [15,16]. These properties
have been exploited in sensors [17,18] and other applications
In order to identify the parameters that most affect polyani-
line ﬁlm properties, the morphology of LbL ﬁlms has been
studied in detail [21–23]. The choice of the LbL method to
produce the ﬁlms was based on the unique features offered by
the method, which are basically the simplicity in the experi-
mental procedures and ability to control ﬁlm properties at the
molecular level [24–26]. In the LbL ﬁlms, polymer adsorption
is in most cases driven by ionic interactions, with oppositely
charged materials from aqueous solutions being deposited al-
ternately on a solid substrate. Secondary interactions—e.g., hy-
drogen bonding and van der Waals—may also be important
for adsorption , as it has been shown for polyanilines, for
which H-bonding contributes even when the molecules are pro-
tonated. The layer thickness can be controlled by altering the
interactions responsible for adsorption, which is carried out by
changing materials and preparation conditions such as pH, ion
dopant, ionic strength and concentration of the solutions.
It was clear from the morphological studies mentioned above
that ﬁlm properties also depended strongly on the conforma-
tion of the polymer in solution. Access to this type of infor-
mation is not straightforward because the polyaniline solutions
are polydisperse. Here we employ small angle X-ray scattering
(SAXS) to investigate the properties of parent PANI, poly(o-
methoxyaniline) (POMA) and poly(o-ethoxyaniline) (POEA)
in solution at various pHs and doping acids. SAXS is useful to
probe the material structure on a scale from 0.05 to 2000 Å [28,
29], as it may provide statistically meaningful measurements of
total volumes, surface areas and scattering centers . Knowl-
edge about the external surface structure of the polymer can be
obtained by measurements of the scattering intensity I(q) ver-
sus the scattering vector modulus. We then study the inﬂuence
from the conformation of polymers in solution on the properties
of nanostructured ﬁlms. Parameters such as roughness, aggre-
gation, and fractal dimension were used to interpret the data.
2. Experimental details
PANI, POMA, and POEA were chemically synthesized ac-
cording to the methods described in Ref.  (Fig. 1). Solutions
Fig. 1. Emeraldine base form of (a) PANI, (b) POMA, and (c) POEA.
Author's personal copy
378 F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387
were prepared by dissolving the dedoped polymers (EB) in wa-
ter and dimethyl acetamide (DMAc) with a concentration of
3 g L−1. The solutions were left stirring overnight and then ﬁl-
tered obtaining a ﬁnal concentration ≈2.5 gL−1. The polymers
in these solutions were doped with HCl, with the pH adjusted at
selected values between 1.5 and 10, as indicated. For the study
of different dopants, POEA was doped with one of the follow-
ing acids: p-toluenesulfonic acid (p-TSA), camphorsulfonic
acid (CSA), sulfanilic acid (SAA), and hydrochloric (HCl) acid.
The doping acids were all purchased from Sigma–Aldrich, and
used as received.
The SAXS experiments were conducted at the National Syn-
chrotron Light Laboratory (LNLS), Campinas, Brazil, using a
monochromatic X-ray beam (λ = 1.488 Å), which focuses the
beam horizontally, and a one dimensional position-sensitive X-
ray detector to record the scattering intensity. To perform the
SAXS measurement, the aqueous solution was placed in a ﬂat
cell and sealed. The SAXS curves were normalized with respect
to (i) the decreasing intensity of the coming synchrotron beam
and (ii) the sample absorption. The SAXS intensity produced
by the solvent (water and DMAc) was measured and subtracted
from the total scattering intensity before the analysis.
The scattered intensity was measured over the scattering
vectors, q = (4π/λ)sinθ, where 2θ is the total scattering an-
gle and λ is the wavelength generated from the rotating anode
source which was monochromatized by a crystal monochrome-
ter. In Guinier’s theory, the X-ray scattering intensity from the
sample (I) depends on the number of particles per unit volume
(Np), the electron density difference between particles and the
medium ( ρ), volume of the particle (v), radius of gyration
(Rg) and the scattered intensity of a single electron (Ie) .
(1)I(q) = Ie(q)Np( ρv)2
In a Guinier plot, lnI(q) vs q2(I(q) → I(0)), the slope of
the linear region allows Rg to be obtained . Since solutions
containing particles of different sizes (polydisperse) show sev-
eral linear regions in the Guinier plot, in order to generate low
resolution models we considered only the SAXS curve from
the last linear region. This region has a high minimum angle
and therefore information on the large distances in the mole-
cules is lost . This was performed as an attempt to include
the information due to the smallest particles in solution as if the
system were monodisperse (three-dimensional models).
In the absence of interference effects, a Fourier transform
connects the normalized particle form factor (and hence I(q))
to the pair distance distribution function, p(r), the probability
of ﬁnding a pair of small elements at a distance r within the
entire volume of the scattering particle as 
This function provides information about the shape of the
scattering particle as well as its maximum dimension, Dmax,
accounted for a certain r value where p(r) goes to zero. More-
over, the particle radius of gyration, Rg, value is given by 
Rg for the polyaniline and its derivates in solution at distinct pH values (HCl)
Material pH Rg (Å)
Region I—aggregates (q2 0.0016)
POEA 1.5 Precipitate
3.0 112.9 51.9
5.0 60.8 37.6
10.0 77.2 45.9
POMA 1.5 Precipitate
3.0 77.1 56.8
5.0 47.9 31.1
10.0 65.7 42.9
PANI 2.5 Precipitate
3.0 87.4 59.7
5.0 94.4 48.6
10.0 52.8 41.2
Region II—smallest particles (q2 0.0016) RGuinier
POEA 3.0 39.2 ± 2.7 42.5 ± 0.6
5.0 31.5 ± 3.0 36.9 ± 0.8
10.0 28.8 ± 2.9 32.0 ± 0.8
POMA 3.0 39.1 ± 2.0 41.4 ± 0.6
5.0 29.8 ± 2.8 31.9 ± 0.3
10.0 26.2 ± 1.8 28.2 ± 0.3
PANI 3.0 33.3 ± 1.3 34.0 ± 0.3
5.0 30.3 ± 1.9 34.3 ± 0.4
10.0 31.2 ± 2.3 35.8 ± 1.4
0 p(r)r2 dr
In this work, we make use of the GNOM program  to
calculate p(r) and estimate the radii of gyration (RGNOM
g ) (see
Table 1). The smearing effect, caused by the length of the de-
tector window, was corrected and the distance distribution func-
tion p(r) was calculated. The dummy atom model (DAM) was
generated with the program DAMMIN, which employs simu-
lated annealing to obtain a model that minimizes the discrep-
ancy between the theoretical and experimental curves . For
each structure, several models were tried, with no symmetry
imposed, and an average DAM was calculated using the pro-
gram DAMAVER . The 3D-envelopes were visualized with
VMD software .
The nanostructured ﬁlms were prepared with POEA. At
pH 3, POEA is not fully protonated, which leads to a chem-
ical structure that is a mixture of emeraldine salt and base
forms. For POEA, a few minutes are sufﬁcient to form a stable
and continuous layer . Fractal dimension and ﬁlm rough-
ness were analyzed on ﬁlms obtained with immersion times
between 1 and 180 s at pH 10 (EB). To analyze the pH ef-
fect on ﬁlm morphology only the initial adsorption stage was
considered, and therefore the immersion time was only 3 s, for
solutions with pH varying from 3 to 10. The inﬂuence from
the dopant was studied with ﬁlms produced with POEA so-
lutions at pH 3, immersion time of 180 s and 4 dopant acids
(CSA, SAA, HCl and TSA). Substrates were prepared with
optical glass slides (1 × 10 × 30 mm) previously cleaned in
H2SO4/H2O2, 7:3 v/v solutions for 1 h, followed by extensive
washing in ultra-pure water. The slides were then immersed into
Author's personal copy
F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387 379
a H2O/H2O2/NH4OH 5:1:1 v/v solution for 40 min and again
washed with large amounts of ultra pure water. The growth of
the POEA layers was monitored by measuring the UV–vis ab-
sorption spectrum with a UV–vis spectrophotometer Shimadzu
The POEA solution was also transferred onto a transmis-
sion electron microscope (TEM) copper grid for TEM obser-
vation on a Philips CM 200 operating at 200 kV. Atomic force
microscopy (AFM) images were taken in a Topometrix micro-
scope, model Discoverer TMX 2010, using silicon nitride tips
(V shape) with spring constant of 0.06 N/m (nominal value).
All images were obtained in the contact mode at a scan rate of
2 Hz. The root-mean-square roughness, Rms, and the surface
fractal dimension were calculated using WSxM 4.0 software
from Nanotec Electronica S.L. (Copyright © November 2003)
and scanning probe image processor (SPIP) version 18.104.22.168
from Image Metrology A/S 2003.
3. Results and discussion
3.1. Polyanilines in solution investigated by SAXS
Processing parameters are known to affect ﬁlm morphology
to a great extent, and there is evidence that some differences
Fig. 2. lnI vs q2 curves for (a) POEA, (b) POMA, and (c) PANI in distinct pH values, which show two linear proﬁles, demarcated by the dashed line.
Author's personal copy
380 F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387
Fig. 2. (continued)
are already present in the solution used to produce the ﬁlms.
To elucidate this behavior, SAXS analysis was performed for
polyanilines in various types of solution. Fig. 2 shows Guinier
plots, lnI vs q2, for POMA, POEA, and PANI solutions at
distinct pHs, namely 3.0, 5.0, and 10.0, where considerable
heterogeneity is seen in all samples. With a Guinier plot one
is able to calculate Rg since qRg should be around 1, i.e.,
qmaxRg 1.5. Two requirements to apply Guinier’s law are that
the particles (scattering centers) should all have the same size
(monodisperse) and be far from each other to avoid interpar-
ticle scattering. Because neither of these conditions applies to
the polymer samples, the analysis based on the Guinier region
only gives an estimate of the scattering center sizes . Two
or three linear regions can be identiﬁed in Fig. 2, from which
upper and bottom limits of aggregation radii may be estimated
for the aggregated particle sizes.
Table 1 shows the Rg values calculated with the procedures
established in Ref.  where we used two regions of the curve,
referred to as regions I (q2 0.0016) and II (q2 0.0016).
A careful investigation of the Guinier plots reveals two linear
regimes: I—a short linear proﬁle at very low q values, and II—
a wider linear proﬁle. Since the q values probe the structural
features at various length scales, the larger value of the slope
of the linear region I (compared to II) indicates a larger size of
the scatterers, i.e., chain aggregates . In region I, Rg of the
aggregates increases with decreasing pH, since aggregation is
precluded by the charges incorporated upon adding acid to the
solution. However, at low pHs ( 2.5), the solution becomes
unstable and starts to precipitate. This explains why Rg of the
structures in Table 1 increases as the pH was decreased from 5.0
to 3.0. Considering data for other pHs, we conclude that the best
tradeoff between solubility and aggregation upon dedoping the
polymers is attained at pH ≈ 5.0. The values of Rg in region II
from Guinier analysis are comparable to those obtained from
the shape factors (RGNOM). Therefore, we may infer that in di-
lute solutions of conducting polymers, chain conformation can
be estimated by SAXS.
Fig. 3 shows the low-resolution particle shape for the
polyanilines molecules in solution, determined from the exper-
imental data using the ab initio procedure . A less-packed,
coiled structure is observed for pH 3, while at pH 10 blobs are
formed, which are consistent with the radii of gyration in Ta-
ble 1 (region II), and with the literature according to which a
more extended structure is obtained by increasing the doping
The conformation of polyaniline molecules in solution also
depends on the doping acid [45–47], which is illustrated here
for POEA doped in pH 3.0 with four acids, namely HCl, CSA,
SAA, and TSA. Fig. 4 shows low-resolution envelopes obtained
with the same procedures as for those in Fig. 3. A less-packed,
coiled conformation was observed for HCl and CSA, while for
SAA and TSA the conformation is rod-like and cylindrical,
respectively. Counterions of organic acids are strongly bound
to the polymeric chain because they are less solvated. Conse-
quently, the charges in the polymeric chains are more effec-
tively screened and the polymer appears almost like a neutral
polymer. In this situation the polymeric chains assume a rod-
like conformation which results in aggregates of greater radii
of gyration. On the other hand, ions easily solvated in water,
as is the case of chloride, are less effective in screening the
polyion charges, and therefore the polymeric chains assume a
more extended coil-like conformation due to the intra and in-
ter molecular electrostatic repulsion. The consequences for the
ﬁlm morphology of these distinct conformations will be dis-
cussed in the next section.
Another parameter that may be obtained from the SAXS data
is the fractal dimension  of the molecular structures in so-
lution. As I(q) follows a power law (Eq. (4)) , the fractal
Author's personal copy
F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387 381
Fig. 3. Average DAM for (a) POEA, (b) POMA, and (c) PANI in (i) pH 10.0, (ii) pH 5.0, and (iii) pH 3.0 (HCl).
dimension may be determined from the SAXS proﬁle, analyz-
ing the power-law proﬁle (q > 0.01 Å
)  shown in Fig. 5.
In this region, the scattering proﬁles do not depend on the shape
of the scattering units.
(4)I(q) ∝ q−α
where I is the scattered intensity, q is the modulus of the scat-
tering vector, and the exponent α is related to the fractal dimen-
sion of the scattering particles. The slope of the linear region in
logI(q) vs log(q) plot gives the exponent α, the dimensionality
of the scattering object.
Systems with fractal behavior are characterized by geomet-
ric auto-similarity in a given region, i.e., the structure is inde-
pendent of the observation scale size . The fractal dimen-
sion helps quantify properties such as mass (m) and surface
area (A), since a fractal object varies with the radius of gyra-
tion . The slope (α) of logI(q) vs logq curves in the linear
region was calculated, with linear regression analysis, leading
to fractal dimensions (Df). For α between 1 and 3, Df is de-
termined with Eq. (5) and the material is characterized as mass
f ) in a three-dimensional space. The so-called surface
f ) have α between 3 and 4, with the fractal dimension
estimated with Eq. (6) . Mass fractals are scattering centers
whose mass increases proportionally to the volume while sur-
face fractals are dense scattering centers associated with surface
roughness. In Eq. (6), if Df = 2.0 we obtain the well-known
Porod’s law I(q) ∝ q−4 for nonfractal structures with smooth
f = |α|,
f = −|α| + 6.
The slopes in the linear region were calculated as −2.7,
−2.4, and −2.6 for PANI-EB, POEA-EB, and POMA-EB
(pH 10.0), respectively, corresponding to mass fractals. Values
Author's personal copy
382 F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387
of α for PANI, POEA, and POMA at pH 3.0 were −1.7, −1.3,
and −1.4, respectively. Therefore, in solution all polyanilines
exhibited mass fractals. As we shall see with the AFM data
for nanostructured ﬁlms, surface fractals are observed (see later
Fig. 6). Typically, α = 2 for Gaussian chains, platelets or discs,
or sheet-like lamellar objects; α = 1 for rigid rods, and α = 3
Fig. 4. Dummy residues model for POEA doped with (a) CSA (Rg = 28.8 Å),
(b) HCl (Rg = 32.0 Å), (c) SAA (Rg = 36.4 Å), and (d) TSA (Rg = 36.9 Å)
for tight compact structures. The values of α for the q-range
used (0.01 < q < 0.1 Å) indicate that the polymers adopt a rela-
tively more compact blob-like structure when undoped (pH 10),
while in the doped state they exhibit Gaussian-like chain struc-
There are two aggregation regimes depending on the limiting
factor for the aggregation: diffusion-limited cluster aggrega-
tion (DLCA) and reaction-limited cluster aggregation (RLCA)
[53,54]. For DLCA, aggregation is led by cluster diffusion with
a fractal dimension Df ≈ 1.7–1.8. Particles under RLCA repel
each other, growing more-compact aggregates with a limit frac-
tal dimension of Df ≈ 2.0–2.1. A continuous transition between
both regimes has also been reported, e.g., charged systems as a
function of the range of the interactions [55,56]. Experiments
have recently shown that aggregates grown in a 50/50 mix-
ture of positive and negative particles show low-density frac-
tal structures compared to those obtained from the universal
regimes, Df ≈ 1.2–1.4 . For our case, only PANI-ES is
under the DLCA regime, the others showed low-density frac-
tal structures i.e., the inner structure of the clusters was indeed
fractal and the dimension characterizing it was lower than for
DLCA . In uncharged systems, POEA and POMA are un-
der the RLCA regime, while for PANI the fractal dimension is
higher than the typical values for diffusive aggregation.
3.2. Film morphology
The inﬂuence from the state of aggregation, doping and con-
formation in solution on the ﬁlm properties has been mentioned
in previous studies [59,60]. The question that then arises is
whether a direct relationship can be established between the
conformations in solution and in the solid ﬁlm. The LbL tech-
nique allows ﬁlm preparation within short-time intervals, and
therefore it is important to probe the kinetics of polymer ad-
sorption to optimize the time of ﬁlm fabrication. For POEA,
a few minutes are sufﬁcient to form a stable and continuous
layer . We observed the ﬁlm morphology of POEA ﬁlms
Fig. 5. logI vs logq typical curve in the power-law region for POEA-ES (pH 3.0).
Author's personal copy
F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387 383
Fig. 6. AFM images (10 × 10 µm) of POEA-EB (pH 10.0) for immersion times of (a) 1 s (Rs = 69 ± 25 Å; Df = 2.39), (b) 3 s (Rs = 74 ± 30 Å; Df = 2.25), (c) 5 s
(Rs = 96 ± 30 Å; Df = 2.34), and (d) 180 s (Df = 2.23). Rt is 200 Å, calculated by SEM.
at very early stages of adsorption with the same solutions used
in SAXS experiments. The images are shown in Fig. 6, for ad-
sorption times of 1, 3, 5, and 180 s. The size of the globules
was determined from the images, but after correcting for the ef-
fects from a tip having similar size as the particles [61–63]. The
correction was made using a geometric relation:
s is the true radius of the particle, Rt is the tip radius,
and L is the apparent size.
The radii of the globules in the ﬁlm imaged in Fig. 6 were
69 ± 25, 74 ± 30, and 86 ± 30 Å, for 1, 3, and 5 s, respectively.
We did not calculate the size for 180 s because further aggrega-
tion takes place and no comparison can be made with POEA
conformation in solutions (see below). These radii are com-
pared with the aggregate size in solutions. To estimate the latter,
we assumed that the particles in solution are spherical, thus al-
lowing Rs (radius of globules) to be calculated using :
(8)Rs = (5/3)1/2
The radii of gyration (GNOM) of the smallest particles for
POEA-EB was 42.5 Å (Table 1), which leads to a Rs = 54 Å
using Eq. (6), in agreement with the particle size in the AFM
images for short adsorption times (ca. 1 s, Fig. 6a). For larger
aggregates, i.e., 50 Rg 77 Å, the radius of the globules
(Rs) obtained with Eq. (6) was 65 Rs 99 Å, consistent
with the particle size in the AFM images for an adsorption time
of 3 s (Fig. 6b). Therefore, at short immersion times (t 3 s)
the smallest POEA aggregates (DAM) adsorb on the substrate,
followed by further aggregation. In addition, within 3 s of ad-
sorption nucleation appears to be complete, followed by growth
of globules due to incorporation of additional polymer chains.
For longer periods of adsorption, diffusion along the surface
causes the ﬁlm morphology to become ﬂatter ; indeed,
a less rough ﬁlm is formed at the end of 180 s of immersion
(Fig. 6d). As expected, the same applies to POMA and PANI,
since the solutions are similar (emeraldine base). We also cal-
culated the fractal dimension for the POEA ﬁlms, according to
the methods used in Ref. . The values obtained varied from
2.23 to 2.39, and therefore all ﬁlms presented surface fractals
(Df > 2.2), i.e., rough with uniformly dense structures, in con-
trast to those values for solutions which were characteristic of
The effect from pH was analyzed by producing POEA ﬁlms
adsorbed with an immersion time of 3 s at several pHs, with the
AFM images shown in Fig. 7. In general, the ﬁlm morphology
is globular, with globules ranging from 200 to 1000 Å in di-
Author's personal copy
384 F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387
Fig. 7. AFM images of POEA at 3 s on (a) pH 3.0, (b) pH 5.0, and (c) pH 10.0 (5 × 5 µm).
ameter for pH 3 and from 600 to 2500 Å for pH 10. The larger
diameter for the higher pH is due to aggregation in solution (see
Table 1), which changes diffusion and the adsorption rates on
the surface, leading to large irregularities in the ﬁlms. At pH 3
(doped), the interactions between chains and with the surface
are less signiﬁcant, favoring formation of defects (small tun-
nels) and less condensed packing of polymer chains with less
aggregates (Fig. 7a). In this model, charged polyelectrolytes
form thin ﬁlms as the charged molecules adsorbed create a po-
tential barrier that prevents more molecules from adsorbing,
thus yielding ﬂat, ultra-thin ﬁlms .
The AFM images also exhibit wormlike structures represent-
ing aggregates made of several blobs for pH 5.0, as illus-
trated in Fig. 6a. This aggregation occurs on the sample sur-
face, within very short times. Some of these structures could
already exist in solution, as the aggregates depicted in Table 1
indicate. A possible mechanism for the formation of the struc-
tures appearing in Fig. 7 is as follows. Chains collapse to form
blobs due to hydrophobic interactions and form a compact core,
which adsorb as a layer (Fig. 7b), whereas protonated polyani-
line adopts an extended conformation due to the Coulomb re-
pulsion and forms several tunnels (Fig. 7a). Thus, for pH 3.0,
POEA structures are more extended than for higher pH, both
on the surface and in solution. In summary, although details
of conformation may be lost due to adsorption and solvent
evaporation, features of the micro and nano-conformation are
preserved on the ﬁlms.
As observed in the SAXS experiments, the chemical struc-
ture of the dopant counter-ions has a strong inﬂuence on the
properties of POEA solutions, and the same inﬂuence can be
seen on the ﬁlm fabrication process. In order to investigate such
an effect, POEA ﬁlms were produced with solutions contain-
ing different doping acids. We could verify by UV–vis spec-
troscopy, whose results are shown in Fig. 8, that the amount
of POEA, taken as proportional to the absorption, depends on
the acid used for ﬁlm deposition. Although POEA is doped in
all ﬁlms as conﬁrmed by the presence of the polaronic band at
700–800 nm, the amount of adsorbed POEA on glass increased
in the following order: TSA > SAA > CSA > HCl, which is
consistent with Paterno and Mattoso  who observed higher
adsorption for TSA-doped POEA. The larger adsorption for
POEA doped with the bulkiest anions may be explained by the
low mobility and low solvation of these counter-ions, causing
a higher screening effect of the charges in POEA chains. This
Author's personal copy
F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387 385
Fig. 8. (a) UV–vis absorbance spectra of POEA ﬁlms (1 layer of 3 min) doped with different acids in pH 3.0 (C = 0.6 gL−1).
Fig. 9. (a) TEM micrograph of POEA doped with TSA and (b) AFM image of POEA in emeraldine salt state, doped with TSA.
would lead to a more compact conformation of the polymer,
thus resulting in a larger amount of material adsorbed [67–69].
The larger amount of deposited material for POEA doped
with TSA and SAA (Fig. 8) in comparison with another or-
ganic acid (CSA) can be explained by the large radii of gyration
in solution. The differences observed on the amount of POEA
adsorbed in ﬁlms prepared with different acids are, analogously
to the different radii of gyration, due to the different degrees of
association between the charged polymer chains and the coun-
terions from the doping acids. As mentioned before, the counte-
rions of organic acids should be less solvated in water, interact-
ing more strongly with the polymeric chains. The electrostatic
repulsions between polymeric chains are therefore minimized,
which allows the chains to assume a more compact conforma-
tion and reach the substrate surface. Both effects are believed
to increase the amount of POEA adsorbed. The opposite is
found when the counterions are effectively dissolved in water
and the polymeric chains are in a more extended conformation.
The stiffness of the chains is caused by strong intra molecular
electrostatic repulsion, which also hampers adsorption of poly-
meric chains onto the substrate. Consequently, smaller amounts
of POEA are expected to adsorb.
The ﬁlm morphology also depends on the doping acid, as
illustrated in the TEM micrograph of Fig. 9. The ﬁlms nor-
mally display a globular morphology, with larger globules and
higher roughness for those doped with inorganic acids, e.g.,
HCl. One exception is noted, though, for the ﬁlm of POEA
processed with TSA, for which a cylindrical morphology was
observed (Fig. 9a). These well-deﬁned cylinders may arise from
the drying process in ultra-high vacuum (UHV) and treatment
in an ultrasound bath. Electron diffraction analysis showed that
POEA cylinders are amorphous, analogously to that observed
for stretched amorphous polymers . This cylindrical mor-
phology is difﬁcult to visualize with AFM because resolution is
Author's personal copy
386 F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387
Fig. 10. Schematic view of the cylindrical brush conformation where the cylin-
ders contain POEA (TSA) chains.
lost due to the tip radius (≈50 nm). Still, a quasi-ﬁbrillar mor-
phology was observed in Fig. 9b.
The result in Fig. 9 should be expected on the basis of
the SAXS measurements discussed in Section 3.1. The mor-
phological structure at the surface can be attributed to the in-
trinsic “comb-like” molecular architecture of the POEA-TSA
system in solution, which forces the polymer chains to adopt
the conformation of cylindrical brushes, similar to those of
PANI(AMPSA)0.5 (Cnres) samples . This conformation is
caused by the steric overcrowding of toluenesulfonate anions
which remain close to POEA cation radical segments due to
their lower solubility in water. These features are depicted in
a proposed model for the morphology, shown in Fig. 10, in
which the distance between two neighboring layers corresponds
to the distance between two neighboring POEA chains. Consid-
ering (as an extreme case) that a fully extended POEA backbone
with 120 monomers units has ∼1002 Å  for the chain con-
tour length, we conclude that the objects appearing within the
lamellae in the picture of Fig. 10 are indeed small molecular
aggregates. Such features are similar to polyaniline nanotubes
[73,74], whose electronic properties are size dependent. This
opens the way for developing new materials for devices with
tailored characteristics. Some structures shown in Fig. 9 have
a diameter of a few nanometers, typical of nanotubes, but the
diameter may reach 500 nm. The cylindrical structures found
here are highly irregular in size as they were produced by drop-
casting, rather than using controlled ﬁlm fabrication processes
such as the arc-evaporation method or chemical vapor deposi-
tion (CVD) .
The combination of SAXS measurements of polyanilines in
solutions and AFM imaging for adsorbed, nanostructured ﬁlms
allowed us to infer that ﬁlm morphology for short adsorption
times is governed by aggregation in solution. Indeed, estimates
of the size of the globules in the ﬁlms led to similar values to
those of the size of the aggregates of the polymer molecules
in solution. At higher adsorption times, ﬁlm formation was ac-
companied by further aggregation until eventually a less rough
ﬁlm was formed at ca. 180 s of immersion. Thus, the AFM
images represent an “off print” of the solution conformation
when molecules are adsorbed on the substrate. Signiﬁcantly, the
shape of the aggregates in solution depended on the pH, as ex-
pected from the different degrees of doping.
SAXS offers an effective tool for determining the fractal di-
mension of aggregates of particles. The fractal dimension of
particles formed by diffusion-limited process (DLCA) (i.e., fast
aggregation) was in the range 1.4–1.8 for PANI-EB and its
derivatives. At low pH, the fractal dimensions ranged from 1.9
and 2.7. The POEA and POMA values agree with ﬁndings in
the literature, for which the fractal dimension is 2.1–2.2 for
reaction-limited aggregation (RLCA). With the SAXS experi-
ments we showed that the changes in conformation of polymer
molecules in solution can be successfully reconstructed by an
ab initio procedure.
Also investigated was the inﬂuence of the doping acid for
POEA. Interestingly, POEA doped with TSA exhibited cylin-
drical aggregates in solution, which was then manifested as a
cylindrical morphology in the adsorbed ﬁlms studied by TEM.
We demonstrated that the chains stacking can be described by
a lamellar-like arrangement of cylindrical structures. This ap-
proximation enabled us to extract information on structural pa-
rameters which can be useful in designing nanostructures such
as nanotubes and nanoparticles.
This work was supported by CNPq, IMMP, Fapesp, and
 R.J. Waltman, J. Bargon, Can. J. Chem. 64 (1986) 1433.
 T. Ahuja, I.A. Mir, D. Kumar, Rajesh, Biomaterials 28 (2007) 791.
 M. Jaiswal, R. Menon, Polym. Int. 55 (2006) 1371.
 M.J. Breton, Macromol. Sci. Rev. Macromol. Chem. Phys. C 21 (1981)
 O.V. Chechel, E.M. Nikolaev, Instrum. Exp. Tech. 67 (1991) 750.
 J.C. Huie, Smart Mater. Struct. 12 (2003) 264.
 J.S. Lindsey, New J. Chem. 15 (1991) 153.
 K. Wohnrath, J.R. Garcia, F.C. Nart, A.A. Batista, O.N. Oliveira Jr., Thin
Solid Films 402 (2002) 272.
 K. Wohnrath, C.J.L. Constantino, P.A. Antunes, P.M. dos Santos, A.A.
Batista, R.F. Aroca, O.N. Oliveira Jr., J. Phys. Chem. B 109 (2005) 4959.
 S.H. Yang, J.T. Chen, A.K. Li, C.H. Huang, K.B. Chen, B.R. Hsieh, C.S.
Hsu, Thin Solid Films 73 (2005) 477.
 R. Murugesan, E. Subramanian, Mater. Chem. Phys. 80 (2003) 731.
 Z. Ping, G.E. Nauer, H. Neugebauer, J. Theiner, A. Neckel, J. Chem. Soc.
Faraday Trans. 93 (1997) 121.
 A.A. Pud, M. Tabellout, A. Kassiba, A.A. Korzhenko, S.P. Rogalsky, G.S.
Shapoval, F. Houze, O. Schneegans, J.R. Emery, J. Mater. Sci. 36 (2001)
 E.C. Venancio, L.G. Paterno, C.E. Borato, A. Firmino, L.H.C. Mattoso,
J. Braz. Chem. Soc. 16 (2005) 558.
 E.M. Paul, A.J. Ricco, M.S. Wrighton, J. Phys. Chem. 89 (1985) 1441.
 J.J. Langer, Synth. Met. 36 (1990) 35.
 K. Xu, L.H. Zhu, H.Q. Tang, Electrochim. Acta 52 (2006) 723.
 N. Gupta, S. Sharma, I.A. Mir, D. Kumar, J. Sci. Ind. Res. 65 (2006) 549.
 D.H. Zhang, Y.Y. Wang, Mat. Sci. Eng. B 134 (2006) 9.
Author's personal copy
F.L. Leite et al. / Journal of Colloid and Interface Science 316 (2007) 376–387 387
 V.V.R. Sai, S. Mahajan, A.Q. Contractor, S. Mukherji, Anal. Chem. 78
 F.L. Leite, L.G. Paterno, C.E. Borato, P.S.P. Herrmann, O.N. Oliveira Jr.,
L.H.C. Mattoso, Polymer 45 (2005) 12503.
 J.M.G. Laranjeira, E.F. da Silva, W.M. de Azevedo, E.A. de Vasconcelos,
H.J. Khoury, R.A. Simão, C.A. Achete, Microelectron. J. 34 (2003) 511.
 V.V. Shevchenko, L.V. Yemelina, Y.A.L. Kogan, G.V. Gedrovich, V.I.
Savchenko, Synth. Met. 37 (1990) 69.
 G. Decher, J.D. Hong, Makromol. Chem. Makromol. Symp. 46 (1991)
 G. Decher, J.D. Hong, Ber. Bunsen-Ges. Phys. Chem. Phys. 95 (1991)
 G. Decher, J.D. Hong, J. Schmitt, Thin Solid Films 210 (1992) 831.
 W.B. Stockon, M.F. Rubner, Macromolecules 30 (1997) 2717.
 M. Campos, B. Bello Jr., Synth. Met. 60 (1993) 1.
 M. Ruokolainen, G.T. Brinke, O. Ikkala, Macromolecules 29 (1996) 3409.
 C.Q. Jin, M. Park, Synth. Met. 124 (2001) 443.
 L.H.C. Mattoso, S.K. Manohar, A.G. McDiarmid, A.J. Epstein, J. Polym.
Sci. A 33 (1995) 1227.
 M.G. Han, S.K. Cho, S.G. Oh, S.S. Im, Synth. Met. 53 (2002) 126.
 G. Porod, Small Angle X-Ray Scattering, Academic Press, London, 1982.
 D.I. Svergun, H.J. Koch, Curr. Opin. Struct. Biol. 12 (2002) 654.
 M. Kotlarckyk, S.H. Chen, J. Chem. Phys. 79 (1983) 2461.
 O. Glatter, O. Kratky, Small Angle X-Ray Scattering, Academic Press,
 D.I. Svergun, J. Appl. Cryst. 24 (1991) 485.
 D.I. Svergun, Biophys. J. 76 (1999) 2879.
 M.B. Kozin, D.I. Svergun, J. Appl. Cryst. 34 (2001) 33.
 W. Humphrey, W. Dalke, K. Schulten, J. Mol. Graph. 14 (1996) 33.
 N. Rosa-Fox, L. Esquivias, A.F. Craievich, J. Zarzycki, J. Non-Cryst.
Solids 121 (1990) 211.
 D. Bagchi, R. Menon, Chem. Phys. Lett. 425 (2006) 114.
 A.V. Semenyuk, D.I. Svergun, J. Appl. Cryst. 24 (1991) 537.
 A.G. McDiarmid, E.J. Epstein, Synth. Met. 69 (1995) 85.
 Y. Haba, E. Segal, M. Narkis, G.I. Titelman, A. Siegmann, Synth. Met. 110
 M.G. Han, S.K. Cho, S.G. Oh, S.S. Im, Synth. Met. 126 (2002) 53.
 M. Sniechowski, D. Djurado, B. Dufour, P. Rannou, A. Pron, W. Luzny,
Synth. Met. 143 (2004) 163.
 A. Eftekhari, M. Kazemzad, M. Keyanpour-Rad, Polym. J. 38 (2006) 781.
 D.W. Schaefer, K.D. Keefer, Phys. Rev. Lett. 53 (1984) 1383.
 D.W. Shaefer, Science 243 (1989) 1025.
 H. Boukari, L.S. Lin, M.T. Harris, Chem. Mater. 9 (1997) 2376.
 S. Neves, C. Polo Fonseca, Electrochem. Commun. 3 (2001) 36.
 D.A. Weitz, M. Oliveira, Phys. Rev. Lett. 52 (1984) 1433.
 W.C.K. Poon, M.D. Haw, Adv. Colloid Interface Sci. 73 (1997) 71.
 D. Asnagui, M. Carpineti, M. Giglio, M. Sozzi, Phys. Rev. A 45 (1992)
 A.Y. Kim, J.C. Berg, Langmuir 12 (2000) 2001.
 A.Y. Kim, K.D. Hauch, J.C. Berg, J.E. Martin, R.A. Anderson, J. Colloid
Interface Sci. 260 (2003) 149.
 A.M. Puertas, A. Fernandez-Barbero, F. de Lãs Nieves, J. Colloid Interface
Sci. 265 (2003) 36.
 A. Eftekhari, M. Kazemzad, M. Keyanpour-Rad, Polymer J. 38 (2006)
 W. Wu, J.Y. Huang, S.J. Jia, T. Kowalewski, K. Matyjaszewski, T. Pakula,
A. Gitsas, G. Floudas, Langmuir 21 (2005) 9721.
 G.R. Bushell, G.S. Watson, S.A. Holt, S. Myhra, J. Microsc. 180 (1995)
 C. Bustamante, J. Vesenka, C.L. Tang, W. Rees, M. Guthold, R. Keller,
Biochemistry 31 (1996) 22.
 T. Thundat, X.Y. Zheng, S.L. Sharp, D.P. Allison, R.J. Warmack, D.C. Joy,
T.L. Ferrell, Scanning Microsc. 6 (1992) 903.
 M.G. Han, S.K. Cho, S.G. Oh, S.S. Im, Synth. Met. 126 (2002) 53.
 L.G. Paterno, L.H.C. Mattoso, Polymer 42 (2001) 5239.
 L.G. Paterno, L.H.C. Mattoso, J. Appl. Polym. Sci. 83 (2002) 1309.
 M. Reghu, Y. Cao, D. Moses, A.J. Heeger, Synth. Met. 55–57 (1993) 5020.
 M. Angelopoulos, N. Patel, R. Saraf, Synth. Met. 55–57 (1993) 1552.
 K.J. Neoh, E.T. Tang, K.L. Tan, Polymer 35 (1994) 2899.
 L.E. Alexander, X-Ray Diffraction Methods in Polymer Science, Wiley–
Interscience, New York, 1969.
 M. Tiitu, N. Volk, M. Torkkeli, R. Serimaa, G.T. Brinke, O. Ikkala, Macro-
molecules 37 (2004) 7364.
 F.L. Leite, C.E. Borato, W.T.L. da Silva, P.S.P. Herrmann, O.N. Oliveira
Jr., L.H.C. Mattoso, Microsc. Microanal. 13 (2007) 304.
 H. Xiai, H. Sze, O. Chan, C. Xiao, D. Cheng, Nanotechnology 15 (2004)
 Z. Wei, M. Wan, T. Lin, L. Dai, Adv. Mater. 15 (2003) 136.
 M. José-Yacaman, M. Miki-Yoshida, L. Rendón, J.G. Santiesteban, Appl.
Phys. Lett. 62 (1993) 657.