Materials Science and Engineering C 37 (2014) 177183

Contents lists available at ScienceDirect

Materials Science and Engineering C

journal homepage: www.elsevier.com/locate/msec

Synthesis of an organic conductive porous material using starch aerogels

as template for chronic invasive electrodes
Ricardo Starbird a,, Carlos A. Garca-Gonzlez b, Irina Smirnova b,
Wolfgang H. Krautschneider c, Wolfgang Bauhofer a
a
b
c

Institute of Optical and Electronic Materials, Hamburg University of Technology, Hamburg, 21073, Germany
Institute of Thermal Separation Processes, Hamburg University of Technology, Hamburg, 21073, Germany
Institute of Nanoelectronics, Hamburg University of Technology, Hamburg, Germany

a r t i c l e

i n f o

Article history:
Received 30 January 2013
Received in revised form 10 December 2013
Accepted 24 December 2013
Available online 7 January 2014
Keywords:
Aerogel
Conductive polymer
Structuration
Extracellular matrix scaffold
Supercritical CO2 drying

a b s t r a c t
We report the development of an organic conducting mesoporous material, as coat for invasive electrodes, by a
novel methodology based on the use of starch aerogel as template. The poly(3,4-ethylenedioxythiophene)
(PEDOT) aerogel was synthesized by polymerization of 3,4-ethylenedioxythiophene within a saturated starch
aerogel with iron (III) p-toluenesulfonate (oxidizing agent) and subsequent removal of the polysaccharide template, followed by supercritical CO2 drying. The chemical structure and oxidation state of the resulting material
were studied by Raman spectroscopy. The morphology and surface properties of the obtained nanoporous material were investigated by scanning electron microscopy (SEM), micro computed tomography (CT) and nitrogen
adsorptiondesorption techniques. The composition and thermal behaviour were evaluated by energy dispersive
spectroscopy (EDS) and thermogravimetric analysis (TGA) respectively. A preliminary biocompatibility test veried the non-cytotoxic effects of the PEDOT aerogel. The large surface area and wide pore size distribution of the
PEDOT conductive aerogel, along with its electrical properties, enable it to be used as extracellular matrix scaffold
for biomedical applications.
2014 Elsevier B.V. All rights reserved.

1. Introduction
Long-term invasive electrodes require low electrochemical impedance and biocompatibility [1]. Poly(3,4-ethylenedioxythiophene)
(PEDOT), as coating layer on invasive electrodes, is used for electrical
stimulation of nerve cells and recording of brain impulses since this
polymer provides lower impedance (Z) and higher charge injection
(Qinj) than bare metal electrodes [25]. These electrochemical properties of the PEDOT and its simple processability can be applied to a
broad range of applications [6,7]. Besides, cytocompatible scaffolds are
recommended to improve the biological response [8], because those
materials emulate the topological and microstructural characteristics
of the extracellular matrix. Extracellular scaffolds require properties
such as a high degree of porosity, high surface-volume ratio, high degree
of pore interconnectivity and appropriate pore size [8]. PEDOT aerogel
as extracellular scaffold is thereby a strategy to improve the biological
and electrical properties of the invasive electrode interface.
Aerogels are highly porous solid nanomaterials with unique characteristics, such as extremely low densities, large open pores and high
inner surface areas that provide interesting physical properties [9,10].
They can be prepared from molecular precursors of inorganic and organic origin by solgel processing and subsequently by supercritical
Corresponding author at: Eissendorfer strasse 38, 21073, Hamburg, Germany.
E-mail address: ricardo.starbird@tu-harburg.de (R. Starbird).
0928-4931/$ see front matter 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.msec.2013.12.032

uid drying to remove the solvents in the wet gels [7,1014].

Templating has been traditionally used as a processing technique for
the engineering of materials with controlled structure and morphology.
The use of aerogels as a template gives the possibility to fabricate tailormade nanostructured materials.
Starch aerogel microspheres in the range of 200400 m, obtained after supercritical drying of emulsion-templated starch gels, are
promising templates for new materials in the biomedical eld
[13,15]. Microspherical starch aerogel template system has previously been used by Tang et al. as a negative template for the preparation of 3D-interconnected TiO2 monoliths with hierarchical mesoand macroporosity [16]. The aerogel template was then removed via
thermal treatment to decompose the starch [16]. Therefore, this
approach was restricted to the preparation of monolithic materials
with higher thermal stability than the template. In this work, a breakthrough technique using aerogel templating for thermally sensitive materials in the form of microspheres is herein developed.
We proposed the use of starch aerogels as a positive template for
PEDOT aerogels due to its high interaction with heavy metals. Carbohydrates (e.g. starch) and derivatives form metal ion complexes with high
interaction between the oxygen donor atoms present in the carbohydrates and the metal ions [18]. Iron (III) salts (Fig. 1a) are commonly
used in the polymerization of 3,4-ethylenedioxythiophene (EDOT) as
oxidizing agent, so that iron (III) salts can oxidise EDOT onto the aerogel
surface [19]. Although, the anion in the iron (III) salt does not participate

178

R. Starbird et al. / Materials Science and Engineering C 37 (2014) 177183

(a)
S

Ethanol, 40 oC

O OO S

S
O

Fe(OTs)3

O
O

Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH

(DSMZ).

O S

O
O

2.2. Template preparation

O
n

(b)

Fig. 1. (a) Chemical polymerisation of 3,4-ethylenedioxythiophene (adapted from [17]),

the doped structure of PEDOT (bipolaron) enables the electronic and ionic conductivity.
(b) The PEDOT aerogel in DMSO demonstrates the structural cohesion of the resulting
material.

in the reaction, it inuences the composition of the PEDOT polymer

complex [20]. Hence, iron (III) p-toluenesulfonate (Fe(OTs)3) is selected
as an oxidizing agent because it is a common and well-established
method in the PEDOT oxidation [2123]. Moreover, an increase in the
doping level as well as conductivity of the PEDOT were reported when
p-toluenesulfonate was utilized as counterion in alcoholic media, compared with other organic solvents [21,22].
Zhang et al. rst reported the emulsion template synthesis of PEDOT
aerogels [7,11]. The authors attributed the absence of previous reports
on the organic conducting aerogels to the difculty in obtaining the conductive organic solgel during the aerogel preparation [7]. Our methodology overcomes the above-mentioned problem, since a well-studied
material regarding solgel chemistry (i.e. starch) is used to prepare
the aerogel template with dened properties. In this manner, it is feasible to avoid the conductive organic gelation step and it is possible to
control the properties of the nal material through the nanostructured
starch template. The conductive PEDOT aerogel preserves the highly porous characteristics of the starch aerogel template, due to the high interactions with the template during the reaction (Fig. 1b). Previously,
Abidian et al. [24,25] reported a signicant improvement in the electrical and biocompatible properties of chronic neural microelectrode
when nanostructured conductive polymers were used as interface.
The inherent electrical properties of the PEDOT, along with the porous
morphology acquired by the template structuration, endow the
PEDOT aerogel with proper characteristics to improve the biological
and electrical properties of the electrode interface. The potential application of this material is discussed in the development of invasive electrodes in biomedical science.

2. Experimental section
2.1. Materials
3,4-Ethylenedioxythiophene (EDOT) and iron (III) p-toluenesulfonate
hexahydrate Fe(OTs)3 6H2O (yellow powder) were purchased from
SigmaAldrich Company Ltd. Ethanol (99.8% purity), dimethylsulfoxide
(DMSO) and acetonitrile analytical-grade solvents were purchased
from Carl Roth Company. Native corn starch (Starch Amylo N-460)
was provided by Roquette. Vegetable (canola) oil was obtained
from domestic shops. CO 2 (N99.9 mol% purity) was supplied by
AGA Gas GmbH. All chemicals were used without further purication. The mouse broblast cell line NIH-3T3 was obtained from

Starch aerogel (SA) microspheres were prepared according to the

literature [14]. Briey, oil:starch solution emulsions were prepared by
mixing a 15% (w/w) corn starch dispersion in water with the corresponding amount of vegetable oil to get a 2:1 volume phase ratio. The
resulting emulsion was then heated up to 120 C and pressurized at
0.10.2 MPa in a closed stirred autoclave (Berghof RHS 295 at
300 rpm). After 20 min, the pressure of the autoclave was released
and the temperature of the emulsion lowered up to 45 C using an
ice bath (cooling rate: 3 C/min) under agitation (marine type;
d = 40 mm; stirring: 1400 rpm). After centrifugation, particles were
separated from the oil phase, soaked in ethanol and placed in the refrigerator (4 C) for retrogradation for 48 h. After retrogradation, starch
particles were transferred to a fresh ethanol solution (i.e., second solvent exchange). Finally, the resulting starch alcogels were dried by extraction of the solvent with a continuous ow (24 Nl/min) of
supercritical carbon dioxide (scCO2) during 4 h at 40 C and
11.012.0 MPa.
2.3. EDOT polymerization
PEDOT aerogel (PA) was prepared adding ca. 40 mg. of aerogel
starch spheres to 250 L of one saturated solution of iron (III) ptoluenesulfonate in ethanol. Ultrasonic stirring was used until the starch
spheres were deposited in the bottom of the vial. The starch spheres acquired a yellowish appearance and this composite (SFeA) was stored for
24 h in solution and then washed with 10 mL of ethanol. Subsequently,
0.50 mmol of EDOT was dissolved in 750 L of ethanol and added to the
SFeA. This mixture was left to stand for 24 h at 40 C. In these conditions
a material with a constant blue colour was produced. The obtained
starch PEDOT aerogel (SPA) was washed again with 1 mL of ethanol
at least 5 times. Alternatively, in an attempt to study the effect of the
template during the synthesis, a sample of PEDOT was prepared in the
absence of the starch template using the same procedure as described
above.
The obtained starch PEDOT aerogel was washed with DMSO in order
to remove the starch template. The process was repeated until no white
precipitate was observed, when few drops of acetonitrile were added to
the DMSO. Finally, the PEDOT aerogel was washed and stored in acetonitrile (ca. 500 L). The resulting material was dried by extraction of the
acetonitrile with a continuous ow of supercritical carbon dioxide
(scCO2).
2.4. Aerogel characterization
PEDOT particles, starch aerogels and composites were studied by
scanning electron microscopy (SEM). Starch aerogels and composites
were characterizated by energy dispersive spectroscopy (EDS), thermal
gravimetric analysis (TGA), Raman spectroscopy, N2 adsorption
desorption analysis and helium pycnometry. The nal microstructure
of the PEDOT aerogel was studied by micro computed tomography
(CT). SEM was conducted in a ZEISS Supra 35 eld-emission-gun scanning electron microscope at 510 kV and samples for SEM were prepared by cutting some pieces of the resulting PEDOT-starch aerogels
and placing them onto carbon adhesive tape. Energy dispersive spectroscopy (EDS) was executed at 20 kV using an Oxford SDD detector.
TGA measurements were performed in a Shimadzu TGA-50 Thermogravimetric Analyzer. Each sample was heated at 10 C/min of heating rate
in a nitrogen atmosphere up to 600 C. Prior to thermal treatment, the
samples were dried in a vacuum oven at 80 C for 24 h. Raman
spectra were obtained and recorded with one excitation line from the
visible range (633 nm) using a multichannel Jobin-Yvon HR800

R. Starbird et al. / Materials Science and Engineering C 37 (2014) 177183

spectrophotometer connected to a CCD detector. Textural properties of

the aerogels were determined by low-temperature N2 adsorption
desorption analysis (Nova 3000e). Prior to measurements, samples
were dried under vacuum (b 1 mPa) at 80 C for 20 h. Specic surface
area (ABET) was determined by the BET (BrunauerEmmettTeller)
method. Pore volume (Vp) and mean pore diameter (dp) in the mesopore
and small macropore region (201000 ) were estimated using the BJH
(BarrettJoynerHalenda) method. Pore diameter ranges of microporosity (b 20 ), mesoporosity (20500 ) and macroporosity (N500 )
terms referred in the text are in accordance to the IUPAC classication
[26]. The skeletal densities of the starch aerogel template and the
PEDOT aerogel were measured by helium pycnometry (Multivolume
Pycnometer 1305, Micromeritics Inc.) and the bulk density was estimated by measuring the weight and volume of a representative number of
aerogel spheres. The micro computed tomography of the PEDOT aerogel
(CT) was realized in a CT 35 Scanco Medical at 45 kV and 176 A.
2.5. Conductive aerogel on invasive electrodes
Gold electrodes were prepared evaporating 5 nm of titanium and
then 100 nm of gold by electron evaporation beam on polyimide
lms. The metal evaporation was performed in a background of
0.66 mPa at a deposition rate of 3 /s. After the deposition, a positive
photoresist was used to pattern 1 mm2 electrode area.
The impedance spectroscopy (IS) measurements were executed by
an LCR-meter Agilent 4284A. The measurements were achieved using
a two electrode setup conguration in NaClKCl saline solution (7.4
pH). A stainless steel vessel was utilized as counter electrode applying
an excitation signal of 50 mV with a frequency range of 20 Hz to
1 MHz, in agreement with [27]. The data were recorded by a Keithley
4200 semiconductor characterization system. PEDOT aerogel was simply deposited on the gold electrode using DMSO to wet the gold electrode surface. Then, the DMSO in the electrodes was washed with
acetonitrile and dried at 105 C for 2 h.
2.6. In vitro experiments
A broblast cell line NIH-3T3 acquired from Deutsche Sammlung
von Mikroorganismen und Zellkulturen GmbH (DSMZ) was used to perform the study of cytotoxicity. Cells were grown as monolayers in
Dulbecco's modied eagle's medium (Life Technologies) supplemented
with 10% FCS (PAA) and antibiotic mixture (100 U/mL penicillin and
100 g/mL streptomycin, PAA). Trypsin/EDTA (PAA) was used to detach
the cells from the culture asks. Cells were cultured in a humidied
atmosphere containing 5% CO2 before the experiments were conducted.
A Neutral Red uptake assay was performed according to ISO
109935 norm. Four samples, 1 4 cm in size, were prepared for
each material. Culture medium and electrode substrate without
PEDOT were used as negative controls. SDS (sodium dodecylsulphate)
was used as a positive control as suggested in ISO993-5. Cells were seeded on 96 well plates for 24 h at 37 C prior to the assay (1 104 cells/
well).
The PEDOT aerogel was sterilised for 1 h in 70% ethanol, washed in
PBS and extracted at 37 C for 24 h in culture medium containing 5%
FCS and Penicillin/Streptomycin. Another extraction was performed by
autoclaving for 20 min at 121 C in distilled water. The extraction uids
were collected and completed with some culture medium, containing
5% FBS and Penicillin/Streptomycin. Finally, the extracts and positive
control media were added to the culture plates and incubated for
24 h. The culture medium was then replaced for a neutral red solution
(Sigma Aldrich, 4 mg/mL in culture medium) and incubated for 3 h at
37 C. After this period, the neutral red reagent was extracted with ethanol/water/glacial acetic acid (50/49/1 [v/v/v]) and the optical density
(OD) was measured with a plate reader (Victor multilabel counter,
Perkin Elmer) at 530 nm. The mean OD values obtained were compared
statistically to the negative control values using the KruskalWallis test

179

followed by multiple comparison test according to the Dunn's method

(p b 0.05). Culture medium was used as negative control.
3. Results and discussion
The EDOT oxidation was performed in ethanol, thus ethanol does
not collapse the starch aerogel structure but allows the dissolution of
the iron (III) salt. Under these conditions the starch aerogel (SA)
(Fig. 2a) changed the colour from white to the same yellow colour of
the iron (III) p-toluenesulfonate ethanolic solution (Fig. 2b). The yellowish colour of the particles did not disappear after several washes
with ethanol indicating the formation of a stable interaction between
the salt ions and the polysaccharide backbone, likely in the form of complexes. This mixture acquired the characteristic blue colour of the polymer after the EDOT addition (Fig. 2c). Then, the template was removed
by DMSO. The starch template is soluble in DMSO while PEDOT withstands without observable changes in the polar aprotic solvent (see
Fig. 1b). The colour of the material changed to dark blue (Fig. 2d) after
the scCO2 drying process. Alternatively, ne non-structured particles
were obtained during the polymerization of EDOT in the absence of
template.
3.1. Aerogel characterization
The chemical structure and the oxidation level of the starch PEDOT
aerogel and PEDOT aerogel were characterized by Raman spectroscopy.
The Raman spectra of the PEDOT samples (SPA and PA) (Fig. 3) present
two weak peaks at 1537 and 1367 cm1 associated with C=C asymmetric stretching and C=C thiophene ring stretching, respectively
[2830]. A middle band at 1256 cm 1 is attributed to the inter-ring
CC stretching, whereas the band of the deformation of the
oxyethylene ring is observed at 990 cm1 [2830]. Those signals
remain constant in both samples (SPA and PA). However, some differences are detected in the oxidation state of the samples around
1420 cm1 associated with the C=C symmetric stretching vibration
[2832]. The PA sample (Fig. 3, bottom) shows a strong band at
1427 cm 1 whereas the SPA sample (Fig. 3, top) exhibits a broader
band at 1434 cm1. The shift in the band around 1420 cm1 is related
to the formation of polaron and bipolaron by an oxidative doping in the
PEDOT chain [28,3032]. The SPA sample evidences a red shift and
broader band as a result of a higher oxidation states due to the presence
of the template. The lower oxidation state observed in the PA sample, as
a result of the template dissolution, indicates the reduction of the doping agent concentration in the polymer complex. The change in the

Fig. 2. (a) Starch aerogel (SA), (b) starch aerogel saturated with iron (III) ptoluenesulfonate (SFeA), (c) starch PEDOT aerogel (SPA) and (d) PEDOT aerogel (PA).

180

R. Starbird et al. / Materials Science and Engineering C 37 (2014) 177183

1427 cm-1

(a)
PA
SA
SPA
SFeA

SPA

PA

Weight retained (wt %)

Intensity (u.a)

100

80

60

40

20

0
100

500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700

200

250

300

Fig. 3. Raman spectra of PEDOT aerogel (bottom) and starch PEDOT aerogel (top). Dotted
line indicates the Raman shift related to C=C symmetric stretching at 1427 cm1.

PEDOT doping level demonstrates the close interaction between polymer and the starch template.
EDS analyses (see supplementary information) were used to corroborate the nature and qualitative composition of the obtained materials.
The starch aerogel sample shows only presence of carbon (0.28 keV)
and oxygen (0.52 keV) in its composition. SFeA sample evidences
sulphur (2.31 keV and 2.46 keV) and iron (0.62, 0.7, 6.4 and 7.1 keV)
signals as a result of the iron salt addition. Likewise, after the polymerization, an iron (0.62, 0.7, 6.4 and 7.1 keV) signal is observed in the
SPA sample even after several ethanol washes, indicating a high interaction starch-iron. Iron signal is absent only after the template dissolution
by DMSO in the PA sample. The PA sample shows carbon (0.3 keV) oxygen (0.5 keV) and sulphur (2.31 and 2.46 keV) signals, composition
expected for PEDOT.
TGA is a helpful technique to determine the thermal stability of materials as well as their compositional properties. TG analyses were performed, in order to conrm the interactions between the components
during the reaction. The initial decomposition temperature (onset)
and ash content are affected by the interactions of the reagents in the
different steps of the synthesis. In the case of the template, the thermal
depolymerization occurs when the temperature exceeds 280 C. The SA
sample thermogram illustrates two distinct regions, at 286 C and
320 C (Fig. 4a,b, dashed line). The initial phase at 286 C causes modication of the polymer structure due to a rapid breakdown of the molecular structure to form products as pyrodextrins [33,34]. A
carbonaceous residue remains (ca. 8% w/w) after all the volatile products have been driven off (Fig. 4a). It has been postulated that the transition observed at around 280 C is due to selective dehydration, the
polyhydroxyl groups elimination and the depolymerization [33,34].
The effect of the iron (III) in the starch template is observed at the
initial degradation temperature. The thermal decomposition of the
oxidant-template composite (SFeA) starts at 240 C (Fig. 4b, solid
line) and the composite generates 28% (w/w) of ashes (Fig. 4a, solid
line). The lower initial temperature and higher ash content can be associated with the interaction between the starch hydroxyl groups and iron
(III) ion from the Fe(OTs)3 salt, conrmed by EDS. The starch inter-chain
interactions are reduced by the iron (III) ions decreasing the temperature of decomposition [18] and yielding higher ash content than the
starch aerogel.
The SPA sample presents an initial thermal decomposition at around
260 C and degrades rapidly (Fig. 4b, dotted line). The thermal decomposition of the SPA composite takes place slowly between 290 and
450 C and producing ca. 20% (w/w) of ashes (Fig. 4a, dotted line).
The single thermal decomposition of SPA sample (Fig. 4b, dotted line)

350

400

450

500

550

600

T (C)

(b)
0.0

deriv. wt (wt %/min)

Wavenumber

150

(cm-1)

-0.5

-1.0

PA
SA
SPA
SFeA

-1.5

100

150

200

250

300

350

400

450

500

550

T (C)
Fig. 4. Thermogravimetry curves (a) and their derivatives (b) of starch aerogel (SA dashed
line), starch iron sample (SFeA solid line), starch PEDOT aerogel (SPA dotted line) and
PEDOT aerogel (PA dot-dashed line) in an atmosphere of owing dry nitrogen at
10 C/min heating rate.

rather than multi-thermal depolymerization of the composite components suggests intimate interations in the SPA sample.
The PEDOT aerogel shows analogous thermal behaviour (Fig. 4a,b,
dot-dashed line) to the SPA sample at a slightly lower temperature
(250 C). The dissolution of the template reduced the thermal stability
of the PA sample because the concentration of doping agent affects
the heat stability of the PEDOT [35,36]. However, comparable thermal
behavior has been described for PEDOT lms [37].
Raman spectroscopy and TGA analyses conrmed the high interaction between the oxidizing agent-template and templatepolymer
interactions during the PEDOT aerogel synthesis. This high interaction
is the key to enable that the conductive PEDOT aerogel retains the starch
aerogel morphology, even after the template dissolution.

Table 1
Physical properties of the starch template (SA) and the resulting PEDOT aerogel (PA).
Sample ID

Starch aerogel (SA)

PEDOT aerogel (PA)

ABET

Vp

dp

Skeletal
density

Bulk
density

(m2/g)

(cm3/g)

()

(g/cm3)

(g/cm3)

83 4
58 3

0.45 0.02
0.28 0.01

40 1
41 2

1.21 0.04
1.01 0.05

0.19 0.02
0.12 0.02

R. Starbird et al. / Materials Science and Engineering C 37 (2014) 177183

3.2. Aerogel physical properties

from the polymer complex might affect the skeletal density of the
PEDOT aerogel.
The textural properties of the resulting material is mainly dened by
the template. The samples are characterized by SEM and low temperature N2 adsorptiondesorption analysis to study the effect of this new
templating methodology in the nal properties of PEDOT aerogel.
Fig. 5 shows SEM images of PEDOT particles, starch aerogel, starchPEDOT and PEDOT aerogels, which offer a comparative study of the

(a)

Volume (cm3/g)

The physical properties of the obtained aerogels were studied in

order to corroborate the template effect in the nal PEDOT aerogel. In
the case of the bulk density of the template (Table 1), its value is in
the range for these kinds of materials [13]. PEDOT aerogel shows a
lower bulk density compared to the starch template with no signicant
volume shrinkage or structural collapse. The differences in the bulk densities are mainly attributed to the lower skeletal density of the PEDOT
aerogel (Table 1). However, it is important to point out that the
PEDOT aerogel shows a lower density (Table 1) than the value reported
previously for PEDOT-OTs thin lms (1.49 g/cm3) [38]. The EDS experiment showed the removal of the iron species during the dissolution of
the template. We assume that the elimination of those components

181

300

SA
PA

200

100

0
0.0

0.5

1.0

Relative pressure (P/Po)

(b)

0.5

0.6

Pore volume (cm3/g)

0.5

0.4

0.4
0.3
0.3
0.2
0.2
0.1
0.1
0.0

0.0
100

Cumulative pore volume (cm3/g)

SA

1000

Pore Diameter (Angstrom)

Pore volume (cm3/g)

0.3

PA

0.3

0.2
0.2

0.1
0.1
0.0

Cumulative pore volume (cm3/g)

(c)

0.0
100

1000

Pore Diameter (Angstrom)

Fig. 5. SEM images of (a) PEDOT particles obtained in absence of template, (b) starch
aerogels (SA), (c) starch-PEDOT aerogels (SPA), (d) PEDOT aerogels (PA) at 50 k
magnications.

Fig. 6. (a) Nitrogen adsorption/desorption analyses of starch aerogel microspheres (SA

sample, dashed line) and PEDOT aerogel microspheres (PA, solid line), (b) BJH-pore size
distribution of the template SA sample and (c) BJH-pore size distribution of the PA sample.

182

R. Starbird et al. / Materials Science and Engineering C 37 (2014) 177183

morphology of PEDOT and template at different stages of the synthesis.

PEDOT synthesized without the starch aerogel template (Fig. 5a) does
not exhibit porous structures or any regular nanostructure. In contrast,
a three-dimensional network with interconnected morphology is observed for the starch aerogel template, starch-PEDOT aerogel and
PEDOT aerogel (Fig. 5bd). The interconnected morphology of the
nal PEDOT aerogel was conrmed by the micro-tomography test
(see suplementary information). The Fig. 5(bd) shows the rich
mesoporosity with wide pore sizes present in the aerogels. Nitrogen
adsorption/desorption isotherms of starch and PEDOT aerogels show a
soft slope for relative pressures (P/P0) in the range 0.1 to 0.85 followed
by a high volume uptake at P/P0 N 0.85 related to the presence of
mesopores and macropores. According to the IUPAC classication, the
adsorption isotherms of both samples (Fig. 6a) are of type IV isotherm
with hysteresis loops, associated with mesoporous materials. The low
volume of gas adsorbed at very low pressures (P/P0 b 0.1) for both samples seems to indicate that microporosity has no signicant contribution
to the porosity of these samples. The isotherm of the starch aerogel presents higher amounts of nitrogen adsorbed at a relative pressure close
to unity than for the PEDOT aerogel, indicating higher textural properties for SA samples. BJH-pore size distributions show a broad pore
range in the mesopore and small macropore region for both samples.
The SA sample (Fig. 6b) shows predominantly unimodal distribution,
whereas PA (Fig. 6c) presents a bimodal distribution. The polymerization into the pores of the starch aerogel and the subsequent dissolution
of the walls may generate this bimodal distribution. The difference in
the shape of the isotherms between the template SA and PA was correlated with a higher population of medium and large mesopores
(100500 ) for the SA sample. High BET-specic surface area is obtained for the starch aerogel (ABET = 83 4 m2/g in Table 1). PEDOT
aerogel shows lower BET-specic surface area (30%) and BJH-pore volume (38%) than the starch aerogel (Table 1). Nevertheless, the mean
BJH-pore diameter (dp) is similar for starch and PEDOT aerogels. The
mesoporosity in the aerogel network is similar to those reported for
some aerogels where the aerogel networks are composed of interconnected walls or bers [9,10,16].

3.3. Incorporation of PEDOT aerogel on invasive electrodes

Impedance spectroscopy (IS) involves measuring the electrical impedance magnitude and phase angle obtained with sinusoidal voltage
(50 mV). The gold electrode coated with PEDOT aerogel presents a signicant reduction in impedance in the range of frequencies from 20 Hz
to 10 kHz, as seen in Fig. 7 (left). The electrodes coated with PEDOT
aerogel showed a signicant decrease in the impedance magnitude
ranging from one to two orders of magnitude at the lower frequency region (20500 Hz). No signicant decrease of the impedance magnitude
is observed at higher frequencies (N10 kHz), since this frequency region
mainly depends on the resistive contribution of the solution [39].

The PEDOT aerogel dramatically decreases the phase angles as

shown in Fig. 7 (right). The phase measured in saline solution evidences
phase angle transition from low to high for the gold electrode. The phase
angles were in the range of 80 to 70 at frequencies lower than
1000 Hz indicating that the gold electrode is primarily behaving as a capacitor. On the other hand, after the PEDOT aerogel attachment to the
gold electrode a phase angle transition around 20 is observed. The
obtained phase angle is mainly attributed to the inherent large open
area of the PEDOT aerogel. However, in order to characterize the electrical performance of the PEDOT aerogel electrode, further studies of the
aerogel interphase electrical phenomenon would be necessary. Nevertheless, this measurement demonstrates that the PEDOT aerogel preserves its intrinsic electrical conductivity in the form of extracellular
matrix scaffold which is promising for the design of highly efcient invasive electrodes.
3.4. Cytotoxicity test
PEDOT aerogel presented a high percentage of cell viability and there
is no evidence of adverse effects on the broblast cells. This preliminary
result indicates the absence or low concentrations of toxic substances in
the extractable of the samples. The extracts of the PEDOT aerogel
showed cellular viability higher than 90% relative to the negative control
and there are non-cytotoxic effects related to the morphology of the
aerogel.
4. Conclusions
Aerogel templating in the form of microspheres was developed in
this work for thermally sensitive materials. High interactions between
the template and polymer allowed the preparation of aerogels after supercritical CO2 drying. Thermal, spectrophotometrical and elemental
analyses have demonstrated the templateoxidant and the template
polymer interactions, resulting in the synthesis of a material with similar structural properties to the original template (starch aerogel). This
processing technique is novel and suitable for the PEDOT structuration
and may be extended to other organic polymers. The resulting PEDOT
aerogel shows large BET surface area (58 3 m2/g), a porous structure
with a broad pore size distribution and lower impedance than the gold
electrode. A preliminary biocompatibility test evidenced non-cytotoxic
effects caused by PEDOT aerogel. PEDOT aerogel, as extracellular matrix
scaffold, shows signicant potential for the development of novel and
more efcient electrode interfaces for long-term invasive electrodes.
Acknowledgments
The authors want to acknowledge Christiane Goepfert from the Cell
Service Center of Hamburg University of Technology for the help in
performing the biocompatibility evaluation. The authors want to
thank Roquette Frres (Dr. O. Husler) for the supply of the corn starch.

105

0
-10
-20

104

Phase ()

Log |Z| (ohms)

Gold electrode
PEDOT aerogel electrode

103

-30
-40
-50
-60
-70
Gold electrode
PEDOT aerogel electrode

-80
10

102

103

104

105

Log frequency (Hz)

106

-90

102

103

104

105

106

Log frequency (Hz)

Fig. 7. Impedance magnitude (left) and phase (right) measurements for 1 mm2 electrode area: Gold electrode (solid line); PEDOT aerogel electrode (dashed line).

R. Starbird et al. / Materials Science and Engineering C 37 (2014) 177183

C.A. Garca-Gonzlez acknowledges the Spanish Ministry of Education for the nancial support through a postdoctoral fellowship in the
frame of the National Program for Staff Mobility from the R&D&i National Plan 20082011.
Ricardo Starbird would like to thank the German Academic Exchange Service (DAAD), the Costa Rica Institute of Technology (ITCR)
and the Science and Technology Bureaus of Costa Rica (MICIT and
CONICIT) for the scholarship provided.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.msec.2013.12.032.
References
[1] H. Yoon, D.C. Deshpande, V. Ramachandran, V.K. Varadan, Nanotechnology 19
(2008).
[2] S.J. Wilks, S.M. Richardson-Burns, J.L. Hendricks, D.C. Martin, K.J. Otto, Front.
Neuroeng. 2 (2009) 7.
[3] J.Y. Yang, D.C. Martin, J. Mater. Res. 21 (2006) 11241132.
[4] X.T. Cui, D.D. Zhou, IEEE Trans. Neural Syst. Rehabil. Eng. 15 (2007) 502508.
[5] R. Starbird, W. Bauhofer, M. Meza-Cuevas, W.H. Krautschneider, Biomedical Engineering International Conference (BMEiCON), 2012, IEEE, 2012, pp. 15.
[6] Y. Li, D. Wang, X. Pei, Z. Shi, L. Wang, W. Zhang, J. Jin, Soft Matter 7 (2011)
26822686.
[7] X. Zhang, D. Chang, J. Liu, Y. Luo, J. Mater. Chem. 20 (2010) 50805085.
[8] A.-D. Bendrea, L. Cianga, I. Cianga, J. Biomater. Appl. 26 (2011) 384.
[9] A.C. Pierre, G.M. Pajonk, Chem. Rev. 102 (2002) 42434265.
[10] C.A. Garca-Gonzlez, M.C. Camino-Rey, M. Alnaief, C. Zetzl, I. Smirnova, J. Supercrit.
Fluids 66 (2012) 297306.
[11] Y. Xu, Z. Sui, B. Xu, H. Duan, X. Zhang, J. Mater. Chem. 22 (2012) 85798584.
[12] C.A. Garcia-Gonzles, E. Carenza, M. Zeng, I. Smirnova, A. Roig, RSC Adv. 2 (2012)
98169823.
[13] C.A. Garca-Gonzlez, M. Alnaief, I. Smirnova, Carbohydr. Polym. 86 (2011)
14251438.

183

[14] C.A. Garca-Gonzlez, J.J. Uy, M. Alnaief, I. Smirnova, Carbohydr. Polym. 88 (2012)
13781386.
[15] C. Garca-Gonzlez, I. Smirnova, J. Supercrit. Fluids 79 (2013) 152158.
[16] L.Q. Tang, W. Ni, H.Y. Zhao, Q. Xu, J.X. Jiao, Bioresources 4 (2009) 3848.
[17] G. Schopf, G. Kossmehl, Polythiophenes - Electrically Conductive Polymers,
1291997. 3145.
[18] B. Gyurcsik, L. Nagy, Coord. Chem. Rev. 203 (2000) 81149.
[19] A. Elschner, S. Kirchmeyer, W. Lvenich, U. Merker, K. Reuter, PEDOT Principles and
Aplications of an Intrinsically Conductive Polymer, CRC Press, 2010.
[20] S. Kirchmeyer, A. Elscher, K. Reuter, Counterions for PEDOT, PEDOT Principles and
Applications of an Intrinsically Conductive Polymer, CRC Press, 2010. 8389.
[21] T.Y. Kim, C.M. Park, J.E. Kim, K.S. Suh, Synth. Met. 149 (2005) 169174.
[22] T.Y. Kim, J.E. Kim, K.S. Suh, Polym. Int. 55 (2006) 8086.
[23] Y.-H. Ha, N. Nikolov, S.K. Pollack, J. Mastrangelo, B.D. Martin, R. Shashidhar, Adv.
Funct. Mater. 14 (2004) 615622.
[24] M.R. Abidian, J.M. Corey, D.R. Kipke, D.C. Martin, Small 6 (2010) 421429.
[25] M.R. Abidian, D.C. Martin, Biomaterials 29 (2008) 12731283.
[26] K. Singh, D. Everett, R. Haul, L. Moscou, R. Pierotti, J. Rouquerol, T. Siemieniewska,
Pure Appl. Chem. 57 (1985) 603619.
[27] B.E. Conway, J.O.M. Bockris, R. White, A. Lasia, Electrochemical Impedance Spectroscopy and its Applications, Modern Aspects of Electrochemistry, Springer, US, 2002.
143248.
[28] W.W. Chiu, J. Travas-Sejdic, R.P. Cooney, G.A. Bowmaker, J. Raman Spectrosc. 37
(2006) 13541361.
[29] F. Tran-Van, S. Garreau, G. Louarn, G. Froyer, C. Chevrot, J. Mater. Chem. 11 (2001)
13781382.
[30] S. Garreau, G. Louarn, J.P. Buisson, G. Froyer, S. Lefrant, Macromolecules 32 (1999)
68076812.
[31] M. Lapkowski, A. Pron, Synth. Met. 110 (2000) 7983.
[32] T.Y. Kim, T.H. Lee, J.E. Kim, R.M. Kasi, C.S.P. Sung, K.S. Suh, J. Polym. Sci. A Polym.
Chem. 46 (2008) 68726879.
[33] P. Aggarwal, D. Dollimore, Instrum. Sci. Technol. 27 (1999) 191197.
[34] Y.X. Xu, V. Miladinov, M.A. Hanna, Cereal Chem. 81 (2004) 735740.
[35] H. Zheng, Y. Jiang, J. Xu, Y. Yang, J. Wuhan Univ. Technol. Mater. Sci. Ed. 26 (2011)
422428.
[36] R. Kiebooms, A. Aleshin, K. Hutchison, F. Wudl, J. Phys. Chem. B 101 (1997)
1103711039.
[37] R. Kiebooms, A. Aleshin, K. Hutchison, F. Wudl, A. Heeger, Synthetic Metals International Conference on Science and Technology of Synthetic, 101, 1999, pp. 436437.
[38] K.E. Aasmundtveit, E.J. Samuelsen, L.A.A. Pettersson, O. Inganas, T. Johansson, R.
Feidenhans, Synth. Met. 101 (1999) 561564.
[39] S.F. Cogan, Annu. Rev. Biomed. Eng. 10 (2008) 275309.