Content uploaded by B. Ya. Venhryn
Author content
All content in this area was uploaded by B. Ya. Venhryn on Jan 17, 2015
Content may be subject to copyright.
28 28
Volume 57
Issue 1
September 2012
Pages 2837
International Scientific Journal
published monthly by the
World Academy of Materials
and Manufacturing Engineering
© Copyright by International OCSCO World Press. All rights reserved. 2012
Changes in the fractal and electronic
structures of activated carbons produced
by ultrasonic radiation and the effect
on their performance in supercapacitors
B.Ya. Venhryn a,*, I.I. Grygorchak a, Z.A. Stotsko a, Yu.O. Kulyk b, S.I. Mudry b,
V.V. Strelchuk c, S.I. Budzulyak c, G.I. Dovbeshko d, O.M. Fesenko d
a Lviv Polytechnic National University, 12 St. Bandera Street, Lviv 79013, Ukraine
b Ivan Franko National University of Lviv, 8 Kyrylo and Mefodiy Street, Lviv 79005, Ukraine
c V. Lashkaryov Institute of Semiconductor Physics, National Academy of Science of Ukraine,
41, pr. Nauky, Kyiv, 03028, Ukraine
d Institute of Physics, National Academy of Science of Ukraine, 46, pr. Nauky, Kyiv 03028, Ukraine
* Corresponding email address: venhryn_b@ukr.net
Received 09.07.2012; published in revised form 01.09.2012
ABSTRACT
Purpose: Effect of ultrasonic irradiation on change of electron structure as well as fractal one of
activated carbons and motivation that these changes are most responsible for the improvement of
functional parameters in supercapacitors, were the aim of this paper.
Design/methodology/approach: Experimental studies were carried out by means of impedance
spectroscopy, cyclic voltammetry, XRay diffraction, small angle XRay diffraction, XRay photoelectron
spectroscopy, IRspectroscopy, MicroRaman spectroscopy and galvanostatic cycling methods.
Findings: Ultrasonic modification of carbon is effective method to increase the specific capacitance as
well as power of carbonbased supercapacitors. Changes of parameters of double electric layer are tightly
related with change of fractal dimension and allow increasing the percolate mobility of charge carries.
Research limitations/implications: This research is a complete and accomplished work.
Practical implications: Carbon materials, modified by ultrasonic irradiation, can be used as
promising electrode materials in energy storage devices of new generation.
Originality/value: This work is of urgent importance for studying of physical and chemical processes
in energy storage systems. It is shown that method of ultrasonic irradiation is highly effective for
modification of carbonbased materials as electrodes in supercapacitors.
Keywords: Ultrasonic radiation; Xray small angle scattering; Nanocluster; Nanoporous carbon;
Supercapacitor
Reference to this paper should be given in the following way:
B.Ya. Venhryn, I.I. Grygorchak, Z.A. Stotsko, Yu.O. Kulyk, S.I. Mudry, V.V. Strelchuk, S.I. Budzulyak,
G.I. Dovbeshko, O.M. Fesenko, Changes in the fractal and electronic structures of activated carbons
produced by ultrasonic radiation and the effect on their performance in supercapacitors, Archives of
Materials Science and Engineering 57/1 (2012) 2837.
PROPERTIES
1. Introduction
Nowadays, the development of electric vehicles production and
alternative energetics makes demands on creation of highly powerful
systems of accumulation and storage of energy. Significant efforts in
this direction though and allowed to increase the voltage to
4.54.7 V, but their specific capacitance remains low.
On that reason it is of high importance to improve the
properties of functional materials, which determine the working
parameters of mentioned above systems. Such problem is general
for various materials, especially with nanoscale structure and is
extensively studied in present time, for instance in [1, 2].
Overcoming this problem can be achieved using super
capacitors with caacitive or pseudocapacitive energy storage
mechanism. However, the key to success in this endeavor is to
ensure the combination of optimal porous structure with
corresponding electronic structure of the material that provides
unblocking of Helmholtz capacitance by capacitance of spatial
charge region in solids. Commonly used for this purpose methods
of chemical modification of the porous structure [35] do not give
the desired simultaneous changes of the electronic structure. For
this reason, in this paper, we try to solve the mentioned above
problem, since a significant number of works on improvement of
active materials of supercapacitors unreasonably associated only
with the modification of their porous structure [69] or connecting
to the surface of certain functional redox groups [1012] without
adequate attention to providing the necessary electronic structure.
We hope that more effectiveness of energy storage systems can be
reached by means of ultrasonic treatment that is the motivation of
investigation on influence of ultrasonic irradiation on super
capacitor parameters.
2. Backgrounds and experimental
It is clear that electric double layer (EDL) is the main
functional element in physicalchemical processes, occurring in
supercapacitors. Obviously that concept approach on this problem
is supposed to be related with structure features of EDL, created
at the electrolytenonmetallic solid phase interface (Fig. 1a). This
interface supposes the presence in this circuit (Fig. 1b) the
capacitance of volume charge region (VCR) of solid phase – ɋSC
region, which can block the Helmholtz capacitance.
Fig. 1. Electric double layer model for nonmetallic electrodes – a)
and equivalent electric circuit – b)
According to circuit, which is shown in Fig. 1b, the total
capacitance of EDL equals:
11 11
H
SC G
CC CC
, (1)
where ɋɇ – Helmholtz capacitance. Commonly the Gouy
Chapman capacitance of diffusive layer in electrolyte ɋG,
significantly prevails the capacitance of dense part of EDL, which
is known as Helmholtz layer. Noted problem of blocking in fact is
absent for metallic electrodes, whereas it is actual for carbon
graphite ones due to significant value of Debye screening radius.
It is clear that unblocking of Helmholtz capacitance promotes the
increase of ɋSC, which is proportional to density of states for
delocalized charge transfers at Fermilevel N(EF) in accordance to
well known relation [13]:
^`
1
2
00SC SC F
C e NE HH , (2)
where
H
SC – relative permittivity of volume charge region,
H
0 –
permittivity for vacuum, e0 – charge of electron.
Just increase of ɋSC should be the result of modification, but
not alone. On other hand, the higher concentration of charge
carries causes the higher degree of charge screening that is the
main reason of more density in EDL, increasing in such way the
capacitance of Helmholtz layer.
For this purpose we propose a new mechanism of ultrasonic
modification, main idea of which is in changing of fractal
structure with mediate effect on electron structure that occurs due
to cumulative action [14] on the nanoporous structure of carbon,
as well as cavitation processes and direct ultrasonic influence on
subsystem of impurities [15].
We have used the carbon with nanoporous structure
(Sact = 467 m2/g, maximum of pore diameter distribution function
d = 1.5 nm). The ultrasonic irradiation was carried out by means
of ultrasonic disperser in aqueous medium at frequency 22 kHz.
Both aqueous (30% KOH) and nonaqueos (1 Ɇ (C2H5)4NBF4 in
acetonenitryl) solutions were used as electrolyte systems for
molecular energy storage devices.
In order to study the porous structure of materials under
investigation we have used the method of XRay small angle
scattering. Investigation was carried out with using DRON 3
powder diffractometer (CuKD radiation,
O
=1.5418 Å). The single
crystal of Ge was used to obtain the monochromatized radiation
by reflecting from (111)  planes. In order to restrict the parasitic
scattering from crystalmonochromator and reduce scattering
phone the special slit units were installed before the sample and
detector which have ±4 mm shift in perpendicular to initial beam
direction. Using of Ge perfect single crystal and focusing system
for initial and diffracted beam allowed us to measure the small
angle XRay spectra starting from wavevector s = 0.01 Å1. The
slit of 0.1 mm width, that corresponds the volume resolution
'
(2ș) = 0.03ɨ was installed before detector. Scattered intensity
was recorded in scanning regime within angle interval 0.25  4.00ɨ
with step 0.05ɨ and exposition 100 s. In case of investigation in
wide angle range the slit of 1.00 mm width was installed before
detector. Peak positions in diffraction patterns have been deter
mined with accuracy

12% whereas other structure parameters
such as size of nanoparticles, fractal dimension value and their
29
READING DIRECT: www.archivesmse.org
1. Introduction
Nowadays, the development of electric vehicles production and
alternative energetics makes demands on creation of highly powerful
systems of accumulation and storage of energy. Significant efforts in
this direction though and allowed to increase the voltage to
4.54.7 V, but their specific capacitance remains low.
On that reason it is of high importance to improve the
properties of functional materials, which determine the working
parameters of mentioned above systems. Such problem is general
for various materials, especially with nanoscale structure and is
extensively studied in present time, for instance in [1, 2].
Overcoming this problem can be achieved using super
capacitors with caacitive or pseudocapacitive energy storage
mechanism. However, the key to success in this endeavor is to
ensure the combination of optimal porous structure with
corresponding electronic structure of the material that provides
unblocking of Helmholtz capacitance by capacitance of spatial
charge region in solids. Commonly used for this purpose methods
of chemical modification of the porous structure [35] do not give
the desired simultaneous changes of the electronic structure. For
this reason, in this paper, we try to solve the mentioned above
problem, since a significant number of works on improvement of
active materials of supercapacitors unreasonably associated only
with the modification of their porous structure [69] or connecting
to the surface of certain functional redox groups [1012] without
adequate attention to providing the necessary electronic structure.
We hope that more effectiveness of energy storage systems can be
reached by means of ultrasonic treatment that is the motivation of
investigation on influence of ultrasonic irradiation on super
capacitor parameters.
2. Backgrounds and experimental
It is clear that electric double layer (EDL) is the main
functional element in physicalchemical processes, occurring in
supercapacitors. Obviously that concept approach on this problem
is supposed to be related with structure features of EDL, created
at the electrolytenonmetallic solid phase interface (Fig. 1a). This
interface supposes the presence in this circuit (Fig. 1b) the
capacitance of volume charge region (VCR) of solid phase – ɋSC
region, which can block the Helmholtz capacitance.
Fig. 1. Electric double layer model for nonmetallic electrodes – a)
and equivalent electric circuit – b)
According to circuit, which is shown in Fig. 1b, the total
capacitance of EDL equals:
11 11
H
SC G
CC CC
, (1)
where ɋɇ – Helmholtz capacitance. Commonly the Gouy
Chapman capacitance of diffusive layer in electrolyte ɋG,
significantly prevails the capacitance of dense part of EDL, which
is known as Helmholtz layer. Noted problem of blocking in fact is
absent for metallic electrodes, whereas it is actual for carbon
graphite ones due to significant value of Debye screening radius.
It is clear that unblocking of Helmholtz capacitance promotes the
increase of ɋSC, which is proportional to density of states for
delocalized charge transfers at Fermilevel N(EF) in accordance to
well known relation [13]:
^`
1
2
00SC SC F
C e NE HH , (2)
where
H
SC – relative permittivity of volume charge region,
H
0 –
permittivity for vacuum, e0 – charge of electron.
Just increase of ɋSC should be the result of modification, but
not alone. On other hand, the higher concentration of charge
carries causes the higher degree of charge screening that is the
main reason of more density in EDL, increasing in such way the
capacitance of Helmholtz layer.
For this purpose we propose a new mechanism of ultrasonic
modification, main idea of which is in changing of fractal
structure with mediate effect on electron structure that occurs due
to cumulative action [14] on the nanoporous structure of carbon,
as well as cavitation processes and direct ultrasonic influence on
subsystem of impurities [15].
We have used the carbon with nanoporous structure
(Sact = 467 m2/g, maximum of pore diameter distribution function
d = 1.5 nm). The ultrasonic irradiation was carried out by means
of ultrasonic disperser in aqueous medium at frequency 22 kHz.
Both aqueous (30% KOH) and nonaqueos (1 Ɇ (C2H5)4NBF4 in
acetonenitryl) solutions were used as electrolyte systems for
molecular energy storage devices.
In order to study the porous structure of materials under
investigation we have used the method of XRay small angle
scattering. Investigation was carried out with using DRON 3
powder diffractometer (CuKD radiation,
O
=1.5418 Å). The single
crystal of Ge was used to obtain the monochromatized radiation
by reflecting from (111)  planes. In order to restrict the parasitic
scattering from crystalmonochromator and reduce scattering
phone the special slit units were installed before the sample and
detector which have ±4 mm shift in perpendicular to initial beam
direction. Using of Ge perfect single crystal and focusing system
for initial and diffracted beam allowed us to measure the small
angle XRay spectra starting from wavevector s = 0.01 Å1. The
slit of 0.1 mm width, that corresponds the volume resolution
'
(2ș) = 0.03ɨ was installed before detector. Scattered intensity
was recorded in scanning regime within angle interval 0.25  4.00ɨ
with step 0.05ɨ and exposition 100 s. In case of investigation in
wide angle range the slit of 1.00 mm width was installed before
detector. Peak positions in diffraction patterns have been deter
mined with accuracy

12% whereas other structure parameters
such as size of nanoparticles, fractal dimension value and their
1. Introduction
2. Backgrounds and experimental
30 30
B.Ya. Venhryn, et. al.
Archives of Materials Science and Engineering
fraction in material have been estimated from intensity curves
with accuracy 35 %.
Electrochemical investigation of activated carbon was carried
out by means of threeelectrode cell with chlorinesilver reference
electrode. Impedance measurements were done within (102105 Hz)
frequency region with using “AUTOLAB” (“EɋO CHEMIE”
Holland) of measuring system, attached with FRA2 and GPES
software. Creation of impedance models was realized using the
ZView 2.3 (Scribner Associates) software. Cyclic voltamperograms
of electrochemical cells were recorded with scan rate 0.01 V/s.
Chargedischarge galvanostatic cycles were maintained by means
of electronic galvanostatic unit.
Electron structure of nanoporous carbon before and after
ultrasonic modification has been investigated by means of
Xray photoelectron spectroscopy with using Kratos Axis Ultra
Xray photoelectron spectrometer and combination scattering
spectroscopy by means of triple spectrometer JobinYvon/Horiba
T64000. The possible influence of surface functional groups was
checked by means of IRspectroscopy with using of IRFourier
spectrometer Bruker IFS 66.
MicroRaman measurements were carried out in the
backscattering geometry at room temperature using above
mentioned spectrometer and thermoelectricalcoled charge
coupled device (CCD) detector. Ar+/Kr+ laser (488.0 nm) was
used as excitation source. The laser power was changed in the
range of 0.25  25 mW. The samples were placed on a computer
controlled XY table with a displacement step of 0.1 µm. The
Olympus BX41 confocal optical microscope equipped with a
×100 (numerical aperture NA = 0.90) was used to focalize the
laser light on the sample and collect the scattered light to the
spectrometer. A 100 µm confocal diameter diaphragm was placed
at the back focal plane of the objective provided the lateral
submicron resolution of the measurements.
3. Results and discussion
The influence of ultrasonic waves on supercapacitor
parameters should be analyzed starting from diffraction data.
Diffractions patterns for carbon before and upon ultrasonic
treatment during 5, 10, 15 and 20 min show the similar features,
which are not observed for initial material. The intensity curves,
corresponding to such diffraction patterns and obtained when
treatment was done in volume of dispersive medium 50 ml are
shown in Fig. 2.
Wide diffraction maxima with untypical for crystalline
materials profile, indicate the existence of amorphous structure
for both initial and treated samples. This maximum corresponds to
interlayer distance in graphite structure. It is seen that principal
peak position shifts to large wave vector values when duration of
treatment is 5 and 10 min. With next treatment duration increase
to 15 and 20 min this parameter shifts in opposite direction and
attains the value, which is somewhat less than for initial material.
Such behavior means that in real space the interlayer distance
decreases at short treatment duration and then increases at more
long duration.
By means of DebyeScherrer formula the mean radius of
initial carbon nanoparticles has been determined from the half
width of principal peak:
02 cos
rO
E
T, (3)
where ȕ – half width of diffraction maximum, ș – half of
scattering angle.
012345678
I (s)
s, Å
1
5
4
3
2
1
100
50
0
0
0
0
0
Fig. 2. Diffraction patterns for initial carbon – (1) and after
ultrasonic treatment during 5 – (2), 10 – (3), 15 – (4) and 20 min –
(5) in volume of dispersive medium 50 ml
Small angle scattering curves for carbon upon ultrasonic
treatment during 5 min at different volumes of H2O are shown in
Fig. 3a. As is seen the linear dependence of scattered intensity is
observed over wide range of wave vector [s0, smax]
(smax = 0.327Å1). Thus I(s) dependence is a function I(s) ~ sn,
where n varies within interval 2<n<3, that indicates the fractal
structure of graphite. We can state that within above mentioned
range of wave vector Xray small angle scattering is formed from
carbon nanoclusters, which are volume fractals, created from
initial carbon nanoparticles. Low limit of wave vector interval so
allows determining the mean radius of fractal cluster:
0
f
R
s
S
. (4)
At the same time it is known [16] that fractal cluster density
depends on its size as:
3
00
D
f
f
f
R
r
§·
U U ¨¸
©¹ , (5)
where Df = n  fractal dimension,
U
0, ro  density and radius of
initial materials, respectively. Using the equations (4) and (5) we
obtain the formula for specific surface area:
3
00
33 D
f
ff f
f
R
SR Rr
§·
¨¸
UU
©¹ . (6)
Parameters of fractal structure, calculated by means of formula
(3)  (6) for carbon upon ultrasonic treatment are listed in Table 1.
Increasing of H2O volume during ultrasonic treatment leads to
sensible changes in small angle scattering spectra. Particularly so
shifts to larger values, that is related with decrease of fractal
cluster radius (Table 1). Some changes are observed also for
fractal dimension, namely its increase to n=2.74 for H2O volume
200 ml at 5 min irradiation duration.
3. Results and discussion
Table 1.
Parameters of fractal structure for carbon after ultrasonic
treatment
Ultrasonic
treatment
duration, min
V (H2O),
ml
ro,
Å
Rf,
Å Df S,
m2/g
5
50
100
200
9.0
10.5
10.0
98
87
80
2.55
2.52
2.74
367
445
332
10
50
100
200
8.5
11.0
10.0
115
97
142
2.59
2.46
2.56
378
580
363
15
50
100
200
11.0
9.5
11.0
28
25
22
2.36
2.23
2.16
579
589
654
20
50
100
200
10.0
10.0
10.0
25
40
35
2.77
2.49
2.53
600
780
665
Increase of treatment duration to 10 min does not cause the
significant changes in scattered curves (Fig. 3b). Values of fractal
dimensions vary within range 2.46  2.59 whereas the radius
offractal clusters increases from 97 Å to 142 Å. It should be noted
also that feature of previous increase of Df with increasing of H2O
volume (Table 1) is another.
More significant changes of fractal structure are observed in
samples under investigation upon ultrasonic treatment during 15
(Fig. 3c) and 20 min (Fig. 3d). Scattering curves, represented
indouble logarithmic coordinates, reveal two regions with
different slope. This slope within [so, smax] range for 15 min
regime is equal to dimension of Euclidian space (n=3). It can be
supposed that within this angular interval the scattering curve is
formed by carbon nanoclusters, which do not reveal the fractal
structure and are rather related with dense packed agglomerate of
initial carbon nanoclusters. The slope of this curve within [smin, so]
range decreases and lies in interval 2<n<3 that, as was mentioned
above, is typical for scattering from fractal aggregates of volume
kind, whose elements are carbon clusters of Rf radius. It should be
noted also, that in this case the increase of H2O volume promotes
the decrease both of fractal dimension (from 2.36 to 2.16) and
size of carbon clusters (from 28 Å to 22 Å).
Special attention is paid to analysis of small angle scattering
spectra of materials upon ultrasonic treatment during 20 min. In
this case for [so, smax.] range the index n is about 4 indicating the
scattering from homogeneous spherical particles with smooth
surface, which is commonly interpreted by Porod law [17].
a) b)
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5
2
4
6
8
10
12
14
16
200
100
50
ln (s)
ln I(s)
ml H2O
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5
2
4
6
8
10
12
14
16
200
100
50
ln I(s)
ln (s)
ml H
2
O
c) d)
4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5
2
4
6
8
10
12
14
16
ml H
2
O
200
100
50
ln I(s)
ln (s)
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0
4
6
8
10
12
14
16
18
ml H
2
O
Initial
ln I(s)
200
100
50
ln (s)
Fig. 3. Dependence of Xray scattered intensity versus wave vector after ultrasonic treatment during 5 – a), 10 – b), 15 – c) and 20 min – d)
31
Changes in the fractal and electronic structures of activated carbons produced by ultrasonic radiation and the effect...
Volume 57 Issue 1 September 2012
fraction in material have been estimated from intensity curves
with accuracy 35 %.
Electrochemical investigation of activated carbon was carried
out by means of threeelectrode cell with chlorinesilver reference
electrode. Impedance measurements were done within (102105 Hz)
frequency region with using “AUTOLAB” (“EɋO CHEMIE”
Holland) of measuring system, attached with FRA2 and GPES
software. Creation of impedance models was realized using the
ZView 2.3 (Scribner Associates) software. Cyclic voltamperograms
of electrochemical cells were recorded with scan rate 0.01 V/s.
Chargedischarge galvanostatic cycles were maintained by means
of electronic galvanostatic unit.
Electron structure of nanoporous carbon before and after
ultrasonic modification has been investigated by means of
Xray photoelectron spectroscopy with using Kratos Axis Ultra
Xray photoelectron spectrometer and combination scattering
spectroscopy by means of triple spectrometer JobinYvon/Horiba
T64000. The possible influence of surface functional groups was
checked by means of IRspectroscopy with using of IRFourier
spectrometer Bruker IFS 66.
MicroRaman measurements were carried out in the
backscattering geometry at room temperature using above
mentioned spectrometer and thermoelectricalcoled charge
coupled device (CCD) detector. Ar+/Kr+ laser (488.0 nm) was
used as excitation source. The laser power was changed in the
range of 0.25  25 mW. The samples were placed on a computer
controlled XY table with a displacement step of 0.1 µm. The
Olympus BX41 confocal optical microscope equipped with a
×100 (numerical aperture NA = 0.90) was used to focalize the
laser light on the sample and collect the scattered light to the
spectrometer. A 100 µm confocal diameter diaphragm was placed
at the back focal plane of the objective provided the lateral
submicron resolution of the measurements.
3. Results and discussion
The influence of ultrasonic waves on supercapacitor
parameters should be analyzed starting from diffraction data.
Diffractions patterns for carbon before and upon ultrasonic
treatment during 5, 10, 15 and 20 min show the similar features,
which are not observed for initial material. The intensity curves,
corresponding to such diffraction patterns and obtained when
treatment was done in volume of dispersive medium 50 ml are
shown in Fig. 2.
Wide diffraction maxima with untypical for crystalline
materials profile, indicate the existence of amorphous structure
for both initial and treated samples. This maximum corresponds to
interlayer distance in graphite structure. It is seen that principal
peak position shifts to large wave vector values when duration of
treatment is 5 and 10 min. With next treatment duration increase
to 15 and 20 min this parameter shifts in opposite direction and
attains the value, which is somewhat less than for initial material.
Such behavior means that in real space the interlayer distance
decreases at short treatment duration and then increases at more
long duration.
By means of DebyeScherrer formula the mean radius of
initial carbon nanoparticles has been determined from the half
width of principal peak:
02 cos
rO
E
T, (3)
where ȕ – half width of diffraction maximum, ș – half of
scattering angle.
012345678
I (s)
s, Å
1
5
4
3
2
1
100
50
0
0
0
0
0
Fig. 2. Diffraction patterns for initial carbon – (1) and after
ultrasonic treatment during 5 – (2), 10 – (3), 15 – (4) and 20 min –
(5) in volume of dispersive medium 50 ml
Small angle scattering curves for carbon upon ultrasonic
treatment during 5 min at different volumes of H2O are shown in
Fig. 3a. As is seen the linear dependence of scattered intensity is
observed over wide range of wave vector [s0, smax]
(smax = 0.327Å1). Thus I(s) dependence is a function I(s) ~ sn,
where n varies within interval 2<n<3, that indicates the fractal
structure of graphite. We can state that within above mentioned
range of wave vector Xray small angle scattering is formed from
carbon nanoclusters, which are volume fractals, created from
initial carbon nanoparticles. Low limit of wave vector interval so
allows determining the mean radius of fractal cluster:
0
f
R
s
S
. (4)
At the same time it is known [16] that fractal cluster density
depends on its size as:
3
00
D
f
f
f
R
r
§·
U U ¨¸
©¹ , (5)
where Df = n  fractal dimension,
U
0, ro  density and radius of
initial materials, respectively. Using the equations (4) and (5) we
obtain the formula for specific surface area:
3
00
33 D
f
ff f
f
R
SR Rr
§·
¨¸
UU
©¹ . (6)
Parameters of fractal structure, calculated by means of formula
(3)  (6) for carbon upon ultrasonic treatment are listed in Table 1.
Increasing of H2O volume during ultrasonic treatment leads to
sensible changes in small angle scattering spectra. Particularly so
shifts to larger values, that is related with decrease of fractal
cluster radius (Table 1). Some changes are observed also for
fractal dimension, namely its increase to n=2.74 for H2O volume
200 ml at 5 min irradiation duration.
Table 1.
Parameters of fractal structure for carbon after ultrasonic
treatment
Ultrasonic
treatment
duration, min
V (H2O),
ml
ro,
Å
Rf,
Å Df S,
m2/g
5
50
100
200
9.0
10.5
10.0
98
87
80
2.55
2.52
2.74
367
445
332
10
50
100
200
8.5
11.0
10.0
115
97
142
2.59
2.46
2.56
378
580
363
15
50
100
200
11.0
9.5
11.0
28
25
22
2.36
2.23
2.16
579
589
654
20
50
100
200
10.0
10.0
10.0
25
40
35
2.77
2.49
2.53
600
780
665
Increase of treatment duration to 10 min does not cause the
significant changes in scattered curves (Fig. 3b). Values of fractal
dimensions vary within range 2.46  2.59 whereas the radius
offractal clusters increases from 97 Å to 142 Å. It should be noted
also that feature of previous increase of Df with increasing of H2O
volume (Table 1) is another.
More significant changes of fractal structure are observed in
samples under investigation upon ultrasonic treatment during 15
(Fig. 3c) and 20 min (Fig. 3d). Scattering curves, represented
indouble logarithmic coordinates, reveal two regions with
different slope. This slope within [so, smax] range for 15 min
regime is equal to dimension of Euclidian space (n=3). It can be
supposed that within this angular interval the scattering curve is
formed by carbon nanoclusters, which do not reveal the fractal
structure and are rather related with dense packed agglomerate of
initial carbon nanoclusters. The slope of this curve within [smin, so]
range decreases and lies in interval 2<n<3 that, as was mentioned
above, is typical for scattering from fractal aggregates of volume
kind, whose elements are carbon clusters of Rf radius. It should be
noted also, that in this case the increase of H2O volume promotes
the decrease both of fractal dimension (from 2.36 to 2.16) and
size of carbon clusters (from 28 Å to 22 Å).
Special attention is paid to analysis of small angle scattering
spectra of materials upon ultrasonic treatment during 20 min. In
this case for [so, smax.] range the index n is about 4 indicating the
scattering from homogeneous spherical particles with smooth
surface, which is commonly interpreted by Porod law [17].
a) b)
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5
2
4
6
8
10
12
14
16
200
100
50
ln (s)
ln I(s)
ml H2O
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5
2
4
6
8
10
12
14
16
200
100
50
ln I(s)
ln (s)
ml H
2
O
c) d)
4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5
2
4
6
8
10
12
14
16
ml H
2
O
200
100
50
ln I(s)
ln (s)
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0
4
6
8
10
12
14
16
18
ml H
2
O
Initial
ln I(s)
200
100
50
ln (s)
Fig. 3. Dependence of Xray scattered intensity versus wave vector after ultrasonic treatment during 5 – a), 10 – b), 15 – c) and 20 min – d)
32 32
B.Ya. Venhryn, et. al.
Archives of Materials Science and Engineering
At space scales R>RS (where RS is radius of surface fractal
clusters) the nanoparticles form the volume fractal aggregates.
The increase of H2O volume to 100 and then to 200 ml leads the
growth of n to 3.45 and 3.70, respectively. Since in this case
3<n<4 one can assert the presence of nanoparticles with unsmooth
surface and fractal dimension Ds=6  n. In this case the surface
area of fractal aggregates of this kind equals [16]:
2
00
4
D
s
s
R
Sr
r
§·
S ¨¸
©¹
. (7)
Since that fractal dimension DS>2 follows that these fractals
have larger value of surface area in comparison to particles with
smooth surface (DS=2). On other hand, the specific surface area
for fractal aggregates, created from clusters with unsmooth
surface equals:
2
0
30
0
3D
s
s
s
R
Sr
r
R
§·
¨¸
©¹
U. (8)
Calculated results are listed in Table 1 .Change of small angle
scattering parameters is accordance with results illustrated in
Fig. 2.
Therefore, one can assert that dominant factor of specific
surface influence is an ultrasonic treatment duration, which, first
of all, is related with the change of fractal structure of carbon
nanoclusters. On that reason we shall put aside the influence of
dispersion surrounding volume, suggesting that this problem will
be the subject of further studies. Instead of this problem, the main
attention we focus on understanding of influence of ultrasonic
treatment duration, for instance at volume of 50 ml of H2O, since
feature of S(t) change is correlated for investigated volumes
(Table 1).
Comparing the obtained characteristics for fractal structure
with data on galvanostatic “chargedischarge” cycles (Fig. 4), one
can see the inconsistency between S and C, whose non monotone
behavior most probably indicate the blockade of Helmholtz layer
capacitance by of VCR capacitance.
0 5 10 15 20
0
100
200
300
400
500
600
700
800
900
1000
1100
1200
130
0
0 100 200 300 400 500 600 700 800 900 1000 1100
0.0
0.2
0.4
0.6
0.8
1.0
2
1
C=52 F/g
U, V
t, s
0.0
0.2
0.4
0.6
0.8
1.0
C=92 F/g
2
S, m2/g, 4C, F/g
t, mi
n
1
Fig. 4. Specific surface (1) and specific capacitance of
galvanostatic discharge (2) as a function of ultrasonic irradiation
duration
In inset to this figure the “chargedischarge” curves reveal the
increase of specific capacitance from 52 F/g to 92 F/g after
ultrasonic modification during 20 min in 50 ml of H2O. The
motive for such assertion are the infrared spectra (Fig. 5), which
exclude the contribution to increase of specific capacitance after
ultrasonic modification related with pseudocapacitance from
surface functional groups, since the intensity of all absorption
band decreases without appearance of new ones. Thus, new states
of Hbonded C=O (1746, 1711 cm1) with less intensity have
appeared in the sample after ultrasonic treatment. In initial sample
the position of C=O has position at 1742 cm1. In the region of
16001700 cm1 we observed 2 peaks at 1632 cm1 (C=C and H2O
vibration) and 1660 cm1 (C=C) in initial sample. After ultrasonic
treatment the band at 1660 cm1 disappeared and other band
splited into 2 bands 1640 cm1 (C=C) and 1626 cm1 (H2O) with
less intensity. We suppose that a number of defects in rings,
decreases in the sample after ultrasonic treatment (e.g. for C=C
bonds in open ring). In the same time we have observed a wide
band in the range of 1300750 cm1, where the number of different
states of CC bond became more after treatment. This fact could
be explained by appearance a great number of short fragments of
graphite net with less defects and correlate with Raman data
presented lower. We have registered a decrease of CH2 and CH3
in the region of stretching vibration and increase of ratio of
CH2/CH3 in the sample after treatment, that is an evidence of
decrease of CH3 broken bonds in the graphite domens. After
ultrasonic treatment a number of broken states and sp3 states
became less.
4000 3500 3000 2500 2000 1500 1000 500
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
1800 1700 1600 1500 1400 1300 1200 1100 1000 900 800
0.34
0.36
0.38
0.40
0.42
0.44
0.46
0.48
0.50
2
1
Absorbance, cm
1
Transmittance
2
1
800
876
1106
1166
1317
1383
1553
1660
1742
1462
1585
1632
Transm ittance
Fig. 5. Infrared spectra of carbon: 1 – initial sample; 2 – after
20 min of ultrasonic modification in dispersive medium volume
50 ml
Data of impedance spectroscopy also are the one more
motivation of this. First of all it should be noted that Nyquist plots
for all parameters of modification have typical shape (Fig. 6),
which shows some distributed capacitance.
That, as well as necessary to account the capacitance of VCR,
need to use de Levie model [18], modified by series connection of
parallel RSCCSC chain, as it is shown in inset to Fig. 6, at
construction of adequate impedance model. Here RSC and CSC 
resistance and capacitance of VCR respectively. Results of
computer parametric identification are shown as histograms
(Fig. 7), where ɋɇ means the capacitance of Helmholtz layer.
Fig. 6. Nyquist plots of carbon: 1 – initial sample; 2 – after 5 min
of ultrasonic irradiation in 50 ml H2O; 3 – affter 20 min of
ultrasonic irradiation in 50 ml (inset to figure – the equivalent
electric circuit)
Fig. 7. Parameters correlation for double electric layer and values
of specific capacitance for galvanostatic discharge
It should be noted that ultrasonic irradiation leads to reduction
of CSC (nonmonotone with increasing both t and ɋ), indicating
according to (2) the decrease of density of states on Fermi level.
Direct experimental confirmation of this was obtained in our
studies by means of Xray emission spectroscopy. Spectrum of
valence band for initial activated carbon (Fig. 8a) shows the two
shoulder profile with significant reduction of intensity at reaching
of bonding energy to Fermi level. This intensity on Fermi level is
proportional to electron density and equals I(EF)=13630. After
ultrasonic treatment the shape of valence band is not changed
(Fig. 8b), but sensible reduction of intensity on Fermi level to
I(EF)=9300 occurs.
At the same time it can be seen that ultrasonic treatment also
caused the reduction of RSC with nonmonotone behavior, which
is unidentical to one for ɋSC, that does not correlate with
decreasing of N(EF). Therefore, the reduction of ɋSɋ and RSɋ
shows no relations with decreasing of measured specific discharge
capacitance C in galvanostatic “chargedischarge” cycles (Fig. 7).
This gives the motive to assert that most responsible for C is the
factor of time constant RSCCSC – chain (insert to Fig. 7), which in
fact means the following:
a) shunting of CSC by resistance of VCR region, so in this case
RSC notes the fact, that effective unblocking of Helmholtz
capacitance occurs, when the plate charging time is larger
than period of its oscillations RSC;
b) symmetrization of voltagecurrent characteristic for regions of
cathode and anode polarization.
a)
76543210123
13400
13600
13800
14000
14200
14400
Inten sity (a.u.)
Energy, eV
b)
765432101234
9000
9500
10000
10500
11000
Intensity (u.a.)
Energy, eV
Fig. 8. Intensity of valence band for a) initial carbon and after
ultrasonic treatment during 20 min in dispersive medium volume
50 ml b)
Therefore, values of measured capacitance correlate with
parameters of EDL in untypical way. Data, obtained for materials
under investigation allow supposing that in hierarchic aspect for
the same values of active surface areas the most effective
unblocking of Helmholtz capacitance is due to the low value of
resistance for VCR (insert to Fig. 7).
Proposed mechanism is most acceptable for organic electrolyte
solution (1 Ɇ (C2H5)4NBF4 in acetonitryle). Here is a slight
increase of specific capacitance for all parameters used at ultrasonic
irradiation. But maximum (20 %) value of C in comparison with
initial sample can be reached for duration of t=10 min and t=15 min
for volume of dispersion medium 50 ml and 100 ml, respectively.
Just at these parameters the RSCCSC – factor and reduction of
Helmholtz capacitance are minimum. It should be noted here that in
this case, contrary to aqueous solution of potassium hydroxide, CH
for all parameters of ultrasonic irradiation decreases, that at taking
33
Changes in the fractal and electronic structures of activated carbons produced by ultrasonic radiation and the effect...
Volume 57 Issue 1 September 2012
At space scales R>RS (where RS is radius of surface fractal
clusters) the nanoparticles form the volume fractal aggregates.
The increase of H2O volume to 100 and then to 200 ml leads the
growth of n to 3.45 and 3.70, respectively. Since in this case
3<n<4 one can assert the presence of nanoparticles with unsmooth
surface and fractal dimension Ds=6  n. In this case the surface
area of fractal aggregates of this kind equals [16]:
2
00
4
D
s
s
R
Sr
r
§·
S ¨¸
©¹
. (7)
Since that fractal dimension DS>2 follows that these fractals
have larger value of surface area in comparison to particles with
smooth surface (DS=2). On other hand, the specific surface area
for fractal aggregates, created from clusters with unsmooth
surface equals:
2
0
30
0
3D
s
s
s
R
Sr
r
R
§·
¨¸
©¹
U. (8)
Calculated results are listed in Table 1 .Change of small angle
scattering parameters is accordance with results illustrated in
Fig. 2.
Therefore, one can assert that dominant factor of specific
surface influence is an ultrasonic treatment duration, which, first
of all, is related with the change of fractal structure of carbon
nanoclusters. On that reason we shall put aside the influence of
dispersion surrounding volume, suggesting that this problem will
be the subject of further studies. Instead of this problem, the main
attention we focus on understanding of influence of ultrasonic
treatment duration, for instance at volume of 50 ml of H2O, since
feature of S(t) change is correlated for investigated volumes
(Table 1).
Comparing the obtained characteristics for fractal structure
with data on galvanostatic “chargedischarge” cycles (Fig. 4), one
can see the inconsistency between S and C, whose non monotone
behavior most probably indicate the blockade of Helmholtz layer
capacitance by of VCR capacitance.
0 5 10 15 20
0
100
200
300
400
500
600
700
800
900
1000
1100
1200
130
0
0 100 200 300 400 500 600 700 800 900 1000 1100
0.0
0.2
0.4
0.6
0.8
1.0
2
1
C=52 F/g
U, V
t, s
0.0
0.2
0.4
0.6
0.8
1.0
C=92 F/g
2
S, m2/g, 4C, F/g
t, mi
n
1
Fig. 4. Specific surface (1) and specific capacitance of
galvanostatic discharge (2) as a function of ultrasonic irradiation
duration
In inset to this figure the “chargedischarge” curves reveal the
increase of specific capacitance from 52 F/g to 92 F/g after
ultrasonic modification during 20 min in 50 ml of H2O. The
motive for such assertion are the infrared spectra (Fig. 5), which
exclude the contribution to increase of specific capacitance after
ultrasonic modification related with pseudocapacitance from
surface functional groups, since the intensity of all absorption
band decreases without appearance of new ones. Thus, new states
of Hbonded C=O (1746, 1711 cm1) with less intensity have
appeared in the sample after ultrasonic treatment. In initial sample
the position of C=O has position at 1742 cm1. In the region of
16001700 cm1 we observed 2 peaks at 1632 cm1 (C=C and H2O
vibration) and 1660 cm1 (C=C) in initial sample. After ultrasonic
treatment the band at 1660 cm1 disappeared and other band
splited into 2 bands 1640 cm1 (C=C) and 1626 cm1 (H2O) with
less intensity. We suppose that a number of defects in rings,
decreases in the sample after ultrasonic treatment (e.g. for C=C
bonds in open ring). In the same time we have observed a wide
band in the range of 1300750 cm1, where the number of different
states of CC bond became more after treatment. This fact could
be explained by appearance a great number of short fragments of
graphite net with less defects and correlate with Raman data
presented lower. We have registered a decrease of CH2 and CH3
in the region of stretching vibration and increase of ratio of
CH2/CH3 in the sample after treatment, that is an evidence of
decrease of CH3 broken bonds in the graphite domens. After
ultrasonic treatment a number of broken states and sp3 states
became less.
4000 3500 3000 2500 2000 1500 1000 500
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
1800 1700 1600 1500 1400 1300 1200 1100 1000 900 800
0.34
0.36
0.38
0.40
0.42
0.44
0.46
0.48
0.50
2
1
Absorbance, cm
1
Transmittance
2
1
800
876
1106
1166
1317
1383
1553
1660
1742
1462
1585
1632
Transm ittance
Fig. 5. Infrared spectra of carbon: 1 – initial sample; 2 – after
20 min of ultrasonic modification in dispersive medium volume
50 ml
Data of impedance spectroscopy also are the one more
motivation of this. First of all it should be noted that Nyquist plots
for all parameters of modification have typical shape (Fig. 6),
which shows some distributed capacitance.
That, as well as necessary to account the capacitance of VCR,
need to use de Levie model [18], modified by series connection of
parallel RSCCSC chain, as it is shown in inset to Fig. 6, at
construction of adequate impedance model. Here RSC and CSC 
resistance and capacitance of VCR respectively. Results of
computer parametric identification are shown as histograms
(Fig. 7), where ɋɇ means the capacitance of Helmholtz layer.
Fig. 6. Nyquist plots of carbon: 1 – initial sample; 2 – after 5 min
of ultrasonic irradiation in 50 ml H2O; 3 – affter 20 min of
ultrasonic irradiation in 50 ml (inset to figure – the equivalent
electric circuit)
Fig. 7. Parameters correlation for double electric layer and values
of specific capacitance for galvanostatic discharge
It should be noted that ultrasonic irradiation leads to reduction
of CSC (nonmonotone with increasing both t and ɋ), indicating
according to (2) the decrease of density of states on Fermi level.
Direct experimental confirmation of this was obtained in our
studies by means of Xray emission spectroscopy. Spectrum of
valence band for initial activated carbon (Fig. 8a) shows the two
shoulder profile with significant reduction of intensity at reaching
of bonding energy to Fermi level. This intensity on Fermi level is
proportional to electron density and equals I(EF)=13630. After
ultrasonic treatment the shape of valence band is not changed
(Fig. 8b), but sensible reduction of intensity on Fermi level to
I(EF)=9300 occurs.
At the same time it can be seen that ultrasonic treatment also
caused the reduction of RSC with nonmonotone behavior, which
is unidentical to one for ɋSC, that does not correlate with
decreasing of N(EF). Therefore, the reduction of ɋSɋ and RSɋ
shows no relations with decreasing of measured specific discharge
capacitance C in galvanostatic “chargedischarge” cycles (Fig. 7).
This gives the motive to assert that most responsible for C is the
factor of time constant RSCCSC – chain (insert to Fig. 7), which in
fact means the following:
a) shunting of CSC by resistance of VCR region, so in this case
RSC notes the fact, that effective unblocking of Helmholtz
capacitance occurs, when the plate charging time is larger
than period of its oscillations RSC;
b) symmetrization of voltagecurrent characteristic for regions of
cathode and anode polarization.
a)
76543210123
13400
13600
13800
14000
14200
14400
Inten sity (a.u.)
Energy, eV
b)
765432101234
9000
9500
10000
10500
11000
Intensity (u.a.)
Energy, eV
Fig. 8. Intensity of valence band for a) initial carbon and after
ultrasonic treatment during 20 min in dispersive medium volume
50 ml b)
Therefore, values of measured capacitance correlate with
parameters of EDL in untypical way. Data, obtained for materials
under investigation allow supposing that in hierarchic aspect for
the same values of active surface areas the most effective
unblocking of Helmholtz capacitance is due to the low value of
resistance for VCR (insert to Fig. 7).
Proposed mechanism is most acceptable for organic electrolyte
solution (1 Ɇ (C2H5)4NBF4 in acetonitryle). Here is a slight
increase of specific capacitance for all parameters used at ultrasonic
irradiation. But maximum (20 %) value of C in comparison with
initial sample can be reached for duration of t=10 min and t=15 min
for volume of dispersion medium 50 ml and 100 ml, respectively.
Just at these parameters the RSCCSC – factor and reduction of
Helmholtz capacitance are minimum. It should be noted here that in
this case, contrary to aqueous solution of potassium hydroxide, CH
for all parameters of ultrasonic irradiation decreases, that at taking
34 34
B.Ya. Venhryn, et. al.
Archives of Materials Science and Engineering
into account the data on S increasing, indicates the shift of
maximum in poroes distribution function to smaller values that is
confirmed by data of precision porometry.
Taking into account that genesis of porosity depends on kind
of raw after activation carbonization; one can suppose that
ultrasonic treatment is effective also for carbon materials of
another origin. Really, for activated carbon, obtained from apricot
stones after 5 min of irradiation in 100 ml of dispersion medium,
about 30 % increase of discharge galvanostatic specific
capacitance for 1 Ɇ (C2H5)4NBF4 in acetonitryl occurs.
Concerning to physical nature of observed changes, we select
at least two aspects:
a) we deal with untypical for carbon materials phenomena
stimulated by ultrasonic irradiation, which are typical for
semiconductors [15]. Consequently, such ultrasonic treatment
can be used for redistribution of impurities and in result for
modification of impurity energetic topology;
b) strong dependence of mobility and may be also hydrophilic
nature from fractal structure.
Studies of microstructure also have been carried out by means
of scanning microscopy. Data of both quantitative and qualitative
analysis of carbon materials before and after ultrasonic treatment
in different regimes by means of JEOL JSM6490LV scanning
microscope is the direct confirmation first of these two aspects. It
was found that ultrasonic irradiation slightly changes the
dispersivity and morphology of particles. At the same time this
irradiation essentially decreases the number of impurities in near
surface layers of material (Fig. 9).
In Table 2 the distribution of chemical elements, which
material consists is presented. In other words we can control the
surface electron states, created by impurities as well as by native
defects by means of ultrasonic irradiation.
Table 2.
Distribution of chemical elements before and after ultrasonic
treatment of carbon material
Element
weight %
(±0.1 %)
atomic %
(±0.1 %)
initial
after
ultrasonic
treatment
initial
after
ultrasonic
treatment
C 91.5 95.0 94.3 96.3
O 6.5 4.7 5.0 3.6
Mg 0.2  0.1 
K
1.2  0.4 
Ca 0.6 0.3 0.2 0.1
Total 100.00 100.00 100.00 100.00
Another aspect of physical nature of observed changes means
that dominant contribution to RSC is not contribution from
delocalized electrons, but from mobility, which should be
attributed to kind of fractal structure. This is good illustrated in
Fig. 7, where values of CH and S are compared.
We have studied the short range ordering structures of the
charcoal in situ by means of confocal microRaman spectroscopy
and the effect of different ultrasonic exposure. Figure 10 shows
such corresponding Raman spectra of the carbon, measured with a
laser excitation wavelength of 488 nm (2.54 eV).
It was found that the Raman spectra of the untreated carbon
consist of three major groups of bands (Fig. 10, curve 1). The first
group includes two main intense firstorder Raman bands at about
1360 and 1580 cm1, which correspond to the D and G
vibrational mode of partially graphitized materials, respectively
[19, 20].
a)
b)
Fig. 9. Distribution of elements in initial carbon material a)
and after ultrasonic treatment b)
Owing to the amorphous character of the carbon with a main
sp2 hybridization character, Gaussian fitting functions have been
chosen for the analysis of Raman peaks. By fitting G and D
bands, ID/IG ratio was obtained to be equal 2.2, indicating a
substantial amount of distortion of bond lengths and angles [21].
In addition, the D2 band, which corresponds also to the disordered
induced phase, is observed at around 1620 cm1 (Fig. 11). A weak
peak at about 1210 cm1 was observed as a shoulder Dband (see
insert Fig. 10), the assignment of which is not straightforward.
M. Armandi et al. [22] assigned a band at 1210 cm1 to a CO
stretch of COH surface groups.
The second frequency region (at 19002200 cm1) consists
strongly asymmetric and wide band (we will refer to it as ‘‘C’’
peak [23]). This weak peak can be satisfactorily fitted by two
Gaussian lines centered at around 2100 and 1980 cm1 (Fig. 12a).
These two components of the C peak in the Raman spectra
confirm the formation of a carbon with a substantial presence of
sp linear structures among a sp2 hybridized disordered network
[23]. It is also reasonable to assume the presence of a large
quantity polyyne and polycumulene moieties coexisting in the
films, giving origin to the vibrational frequencies of the CC bond
in both type chains at 2100 and 1980 cm1, respectively. Because
Raman cross section of linear chains in the carbon amorphous
network is not known, an accurate quantitative determination of
the sp content with respect to sp2 is not possible by Raman
spectroscopy [24]. At the same time, we have observed the
complete disappearance of the two components of the C peak
after ultrasonic treatment (Fig. 12b).
1000 1500 2000 2500 3000
1200 1 500 180 0
200
300
400
500
1000
1500
2000
2500
2
Raman shift (cm1
1
2
Intensity (a.u.)
Raman shi ft ( cm1)
1
Fig. 10. Raman spectra of initial carbon (1) and after ultrasonic
treatment in 50 ml H2O during 5 min (2) at Ȝexc=488 nm. Insert
shows spectra of the D and G spectral regions (9001800 cm1)
which are normalized to intensity Dpeak
a)
1000 1200 1400 1600 1800
0
100
200
300
D1
Intensity (a.u.)
Raman shift (cm1)
D4
Initial (first order)
D2
D3
G
b)
1000 1200 1400 1600 1800
0
700
1400
2100
2800
D1
D4
D2
D3
G
Intensity (a.u.)
Raman shift (cm1)
Ultrasound (first order)
Fig. 11. First order microRaman spectra of carbon: a) before and
b) after ultrasonic modification. G, D1, D2 bands fitted by
Lorentz functions, D3, D4  Gaussian functions
a)
2000 2400 2800 3200
C bands
2D4
2D3
2D1
G+D1
Intensity (a.u.)
Raman shift (cm1)
2G
Initial (second order)
b)
2000 2400 2800 3200
2D4
2D1
G+D1
2D3
2D2
Intensity (a.u)
Raman shif t (cm1)
Ultrasound ( second order)
Fig. 12. Second order microRaman spectra of carbon: a) before
and b) after ultrasonic modification
Thus, during this process ultrasonic treatment can induce the
rearrangement and ordering of a local region of carbon
surrounding the polyyne and cumulene sp chains, thus forming
graphitic nanodomains whose presence modify the Raman
spectrum and, in particular, strong increase intensity of the D and
G bands.
The last group of peaks between around 2300 and 3500 cm1
is due to the secondorder mode of graphite (graphitized
materials) [25].
By the way the increase of maximum of bands ratio D1/D4 can
be interpreted as decrease of ionic impurities content that is
confirmed by elemental analysis (Table 2). It is clear that such
decrease should cause the decrease the scattering of current
carries on ionized impurities and changes if electron structure of
activated carbon (particularly N(E) too).
One of most interesting effects, which were revealed, is the
disappearing of C peak after ultrasonic treatment that is supposed
to be related with above mentioned of more ordering of carbon
bonds that results in abrupt intensity reduction for band that
corresponds to stretching modes of linear sphybridized carbon
structures.
A deeper understanding of the stability of sp carbon structures
and of their role in the nanostructured carbon network would
35
Changes in the fractal and electronic structures of activated carbons produced by ultrasonic radiation and the effect...
Volume 57 Issue 1 September 2012
into account the data on S increasing, indicates the shift of
maximum in poroes distribution function to smaller values that is
confirmed by data of precision porometry.
Taking into account that genesis of porosity depends on kind
of raw after activation carbonization; one can suppose that
ultrasonic treatment is effective also for carbon materials of
another origin. Really, for activated carbon, obtained from apricot
stones after 5 min of irradiation in 100 ml of dispersion medium,
about 30 % increase of discharge galvanostatic specific
capacitance for 1 Ɇ (C2H5)4NBF4 in acetonitryl occurs.
Concerning to physical nature of observed changes, we select
at least two aspects:
a) we deal with untypical for carbon materials phenomena
stimulated by ultrasonic irradiation, which are typical for
semiconductors [15]. Consequently, such ultrasonic treatment
can be used for redistribution of impurities and in result for
modification of impurity energetic topology;
b) strong dependence of mobility and may be also hydrophilic
nature from fractal structure.
Studies of microstructure also have been carried out by means
of scanning microscopy. Data of both quantitative and qualitative
analysis of carbon materials before and after ultrasonic treatment
in different regimes by means of JEOL JSM6490LV scanning
microscope is the direct confirmation first of these two aspects. It
was found that ultrasonic irradiation slightly changes the
dispersivity and morphology of particles. At the same time this
irradiation essentially decreases the number of impurities in near
surface layers of material (Fig. 9).
In Table 2 the distribution of chemical elements, which
material consists is presented. In other words we can control the
surface electron states, created by impurities as well as by native
defects by means of ultrasonic irradiation.
Table 2.
Distribution of chemical elements before and after ultrasonic
treatment of carbon material
Element
weight %
(±0.1 %)
atomic %
(±0.1 %)
initial
after
ultrasonic
treatment
initial
after
ultrasonic
treatment
C 91.5 95.0 94.3 96.3
O 6.5 4.7 5.0 3.6
Mg 0.2  0.1 
K
1.2  0.4 
Ca 0.6 0.3 0.2 0.1
Total 100.00 100.00 100.00 100.00
Another aspect of physical nature of observed changes means
that dominant contribution to RSC is not contribution from
delocalized electrons, but from mobility, which should be
attributed to kind of fractal structure. This is good illustrated in
Fig. 7, where values of CH and S are compared.
We have studied the short range ordering structures of the
charcoal in situ by means of confocal microRaman spectroscopy
and the effect of different ultrasonic exposure. Figure 10 shows
such corresponding Raman spectra of the carbon, measured with a
laser excitation wavelength of 488 nm (2.54 eV).
It was found that the Raman spectra of the untreated carbon
consist of three major groups of bands (Fig. 10, curve 1). The first
group includes two main intense firstorder Raman bands at about
1360 and 1580 cm1, which correspond to the D and G
vibrational mode of partially graphitized materials, respectively
[19, 20].
a)
b)
Fig. 9. Distribution of elements in initial carbon material a)
and after ultrasonic treatment b)
Owing to the amorphous character of the carbon with a main
sp2 hybridization character, Gaussian fitting functions have been
chosen for the analysis of Raman peaks. By fitting G and D
bands, ID/IG ratio was obtained to be equal 2.2, indicating a
substantial amount of distortion of bond lengths and angles [21].
In addition, the D2 band, which corresponds also to the disordered
induced phase, is observed at around 1620 cm1 (Fig. 11). A weak
peak at about 1210 cm1 was observed as a shoulder Dband (see
insert Fig. 10), the assignment of which is not straightforward.
M. Armandi et al. [22] assigned a band at 1210 cm1 to a CO
stretch of COH surface groups.
The second frequency region (at 19002200 cm1) consists
strongly asymmetric and wide band (we will refer to it as ‘‘C’’
peak [23]). This weak peak can be satisfactorily fitted by two
Gaussian lines centered at around 2100 and 1980 cm1 (Fig. 12a).
These two components of the C peak in the Raman spectra
confirm the formation of a carbon with a substantial presence of
sp linear structures among a sp2 hybridized disordered network
[23]. It is also reasonable to assume the presence of a large
quantity polyyne and polycumulene moieties coexisting in the
films, giving origin to the vibrational frequencies of the CC bond
in both type chains at 2100 and 1980 cm1, respectively. Because
Raman cross section of linear chains in the carbon amorphous
network is not known, an accurate quantitative determination of
the sp content with respect to sp2 is not possible by Raman
spectroscopy [24]. At the same time, we have observed the
complete disappearance of the two components of the C peak
after ultrasonic treatment (Fig. 12b).
1000 1500 2000 2500 3000
1200 1 500 180 0
200
300
400
500
1000
1500
2000
2500
2
Raman shift (cm1
1
2
Intensity (a.u.)
Raman shi ft ( cm1)
1
Fig. 10. Raman spectra of initial carbon (1) and after ultrasonic
treatment in 50 ml H2O during 5 min (2) at Ȝexc=488 nm. Insert
shows spectra of the D and G spectral regions (9001800 cm1)
which are normalized to intensity Dpeak
a)
1000 1200 1400 1600 1800
0
100
200
300
D1
Intensity (a.u.)
Raman shift (cm1)
D4
Initial (first order)
D2
D3
G
b)
1000 1200 1400 1600 1800
0
700
1400
2100
2800
D1
D4
D2
D3
G
Intensity (a.u.)
Raman shift (cm1)
Ultrasound (first order)
Fig. 11. First order microRaman spectra of carbon: a) before and
b) after ultrasonic modification. G, D1, D2 bands fitted by
Lorentz functions, D3, D4  Gaussian functions
a)
2000 2400 2800 3200
C bands
2D4
2D3
2D1
G+D1
Intensity (a.u.)
Raman shift (cm1)
2G
Initial (second order)
b)
2000 2400 2800 3200
2D4
2D1
G+D1
2D3
2D2
Intensity (a.u)
Raman shif t (cm1)
Ultrasound ( second order)
Fig. 12. Second order microRaman spectra of carbon: a) before
and b) after ultrasonic modification
Thus, during this process ultrasonic treatment can induce the
rearrangement and ordering of a local region of carbon
surrounding the polyyne and cumulene sp chains, thus forming
graphitic nanodomains whose presence modify the Raman
spectrum and, in particular, strong increase intensity of the D and
G bands.
The last group of peaks between around 2300 and 3500 cm1
is due to the secondorder mode of graphite (graphitized
materials) [25].
By the way the increase of maximum of bands ratio D1/D4 can
be interpreted as decrease of ionic impurities content that is
confirmed by elemental analysis (Table 2). It is clear that such
decrease should cause the decrease the scattering of current
carries on ionized impurities and changes if electron structure of
activated carbon (particularly N(E) too).
One of most interesting effects, which were revealed, is the
disappearing of C peak after ultrasonic treatment that is supposed
to be related with above mentioned of more ordering of carbon
bonds that results in abrupt intensity reduction for band that
corresponds to stretching modes of linear sphybridized carbon
structures.
A deeper understanding of the stability of sp carbon structures
and of their role in the nanostructured carbon network would
36 36
B.Ya. Venhryn, et. al.
Archives of Materials Science and Engineering
provide a new insight in the physics and chemistry to produce the
new forms of carbon with tailored structural and functional
properties.
The change of phonon spectrum and as results the change of
current carries scattering mechanism due to ultrasonic irradiation
beside the above mentioned factors is also confirmed by existence
of second order of Raman spectrum (2200  3700 cm1), which is
caused by collection overtones of 1350 and 1550 cmí1 modes and
their combination. By fitting of Lorentz curves to this spectrum
we have identified the five bands (Fig. 12) from which the
transformation of 2G bond to 2D2 bond follows. Unfortunately we
can pronounce only this fact, whose nature is not clarified at
present time.
4. Conclusions
1. Ultrasonic modification of carbon, proposed in this work is
suitable, cheap and effective method to increase the specific
capacitance as well as power of carbonbased supercapacitors.
2. Significant improvement of properties, determining the
practical use is found to be caused by essential reduction of
time constant RSCCSC – chain of VCR after ultrasonic
irradiation, in particular RSC.
3. Changes of parameters of double electric layer are tightly
related with change of fractal dimension which at such
parameters of ultrasonic treatment increase the percolate
mobility of charge carries.
4. This method allows also controlling successfully the
admixture and native defects distribution, existing on material
surface and which are responsible for the surface electron
states formation.
References
[1] L.A. DobrzaĔski, Report on the main areas of the materials
science and surface engineering own research, Journal of
Achievements in Materials and Manufacturing Engineering
49/2 (2011) 514549.
[2] L.A. DobrzaĔski, M. Pawlyta, A. Hudecki, Conceptual study
on a new generation of the highinnovative advanced porous
and composite nanostructural functional materials with
nanofibers, Journal of Achievements in Materials and
Manufacturing Engineering 49/2 (2011) 550565.
[3] ChiChang Hu, WenYar Li, JengYan Lin, The capacitive
characteristics of supercapacitors consisting of activated
carbon fabricpolyaniline composites in NaNO3, Journal of
Power Sources 137 (2004) 152157.
[4] FengChin Wu, RuLing Tseng, ChiChang Hu, ChenChing
Wang, Effects of pore structure and electrolyte on the
capacitive characteristics of steam and KOHactivated
carbons for supercapacitors, Journal of Power Sources
144 (2005) 302309.
[5] K. Kierzek, E. Frackowiak, G. Lota, G. Gryglewicz,
J. Machnikowski, Electrochemical capacitors based on
highly porous carbons prepared by KOH activation,
Electrochimica Acta 49 (2004) 515523.
[6] A.B. Fuertes, F. Pico, J.M. Rojo, Influence of pore structure
on electric doublelayer capacitance of template mesoporous
carbons, Journal of Power Sources 133 (2004) 329336.
[7] M.W. Verbrugge, P. Liu, S. Soukiazian, Activatedcarbon
electricdoublelayer capacitors: electrochemical charac
terization and adaptive algorithm implementation, Journal of
Power Sources 141 (2005) 369385.
[8] K. Rajendra Prasad, N. Munichandraiah, Electrochemical
studies of polyaniline in a gel polymer electrolyte, High
energy and high power characteristics of a solidstate redox
supercapacitor, Electrochemical and SolidState Letters
5 (2002) A271A274.
[9] A. Malinauskas, J. Malinauskiene, A. Ramanavicius,
Conducting polymerbased nanostructurized materials:
electrochemical aspects, Nanotechnology 16 (2005) R51R62.
[10] A. Nishino, A. Yoshida, I. Tanahashi, I. Tajima,
M. Yamashita, T. Muranada, H.Yoneda, Planar capacitors
with electrical double layer with polarizable electrodes made
of activated carbon fiber, National Technical Report
31 (1985) 318330.
[11] B.E. Conway, Electrochemical supercapacitors. Plenum
Publishing, New York, 1999.
[12] Songhun Yoon, Jinwoo Lee, Taeghwan Hyeon, Seung
M. Oh, Electric doublelayer capacitor performance of a
new mesoporous carbon, Journal of The Electrochemical
Society 147 (2000) 25072512.
[13] H. Gerischer, An interpretation of the double layer capacity
of graphite electrodes in relation to the density of states at
the Fermi level, The Journal of Physical Chemistry
89 (1985) 42494251.
[14] B.A. Agranat, M.N. Dubrovin, N.N. Khavskii, G.I. Eskin,
Fundamentals of physics and technologies of ultrasound,
Vysshaya Shkola, Moskva, 1987.
[15] I.V. Ostrovskii, A.B. Nadtochii, AA. Podolyan, Ultrasound
stimulated lowtemperature redistribution of impurities in
silicon, Fizika i tekhnika poluprovodnikov 36 (2002) 389391.
[16] I.V. Zolotukhin, Yu.E. Kalinin, V.I. Loginova, Solidstate
fractal sructures, International Scientific Journal for
Alternative Energy and Ecology 9 (2005) 5666.
[17] G. Porod, Xray low angle scattering of dense colloid
systems, Part I, KolloidZeitschrift 124 (1951) 83114.
[18] R. de Levie, On porous electrodes in electrolyte solutions,
Electrochimica Acta 8 (1963) 751780.
[19] Jaeyun Kim, Jinwoo Lee, Taeghwan Hyeon, Direct
synthesis of uniform mesoporous carbons from the
carbonization of assynthesized silica/triblock copolymer
nanocomposites, Carbon 42 (2004) 27112719.
[20] A.B. Fuertes, S. Alvarez, Graphitic mesoporous carbons
synthesised through mesostructured silica templates, Carbon
42 (2004) 30493055.
[21] J. Robertson, Hard amorphous (diamondlike) carbons,
Progress in Solid State Chemistry 21 (1991) 199333.
References
4. Conclusions
[22] M. Armandi, B. Bonelli, I. Bottero, C. Otero Areán,
E. Garrone, Synthesis and characterization of ordered porous
carbons with potential applications as hydrogen storage
media, Microporous and Mesoporous Materials 103 (2007)
150157.
[23] L. Ravagnan, F. Siviero, C. Lenardi, P. Piseri, E. Barborini,
P. Milani, Clusterbeam deposition and in situ characteri
zation of carbynerich carbon films, Physical Review Letters
89 (2002) 285506.
[24] A.C. Ferrari, J. Robertson, Interpretation of Raman spectra
of disordered and amorphous carbon, Physical Review B
61 (2000) 1409514107.
[25] A.C. Ferrari, J. Robertson, Resonant Raman spectroscopy of
disordered, amorphous, and diamondlike carbon, Physical
Review B 64 (2001) 075414.
37
Changes in the fractal and electronic structures of activated carbons produced by ultrasonic radiation and the effect...
Volume 57 Issue 1 September 2012
provide a new insight in the physics and chemistry to produce the
new forms of carbon with tailored structural and functional
properties.
The change of phonon spectrum and as results the change of
current carries scattering mechanism due to ultrasonic irradiation
beside the above mentioned factors is also confirmed by existence
of second order of Raman spectrum (2200  3700 cm1), which is
caused by collection overtones of 1350 and 1550 cmí1 modes and
their combination. By fitting of Lorentz curves to this spectrum
we have identified the five bands (Fig. 12) from which the
transformation of 2G bond to 2D2 bond follows. Unfortunately we
can pronounce only this fact, whose nature is not clarified at
present time.
4. Conclusions
1. Ultrasonic modification of carbon, proposed in this work is
suitable, cheap and effective method to increase the specific
capacitance as well as power of carbonbased supercapacitors.
2. Significant improvement of properties, determining the
practical use is found to be caused by essential reduction of
time constant RSCCSC – chain of VCR after ultrasonic
irradiation, in particular RSC.
3. Changes of parameters of double electric layer are tightly
related with change of fractal dimension which at such
parameters of ultrasonic treatment increase the percolate
mobility of charge carries.
4. This method allows also controlling successfully the
admixture and native defects distribution, existing on material
surface and which are responsible for the surface electron
states formation.
References
[1] L.A. DobrzaĔski, Report on the main areas of the materials
science and surface engineering own research, Journal of
Achievements in Materials and Manufacturing Engineering
49/2 (2011) 514549.
[2] L.A. DobrzaĔski, M. Pawlyta, A. Hudecki, Conceptual study
on a new generation of the highinnovative advanced porous
and composite nanostructural functional materials with
nanofibers, Journal of Achievements in Materials and
Manufacturing Engineering 49/2 (2011) 550565.
[3] ChiChang Hu, WenYar Li, JengYan Lin, The capacitive
characteristics of supercapacitors consisting of activated
carbon fabricpolyaniline composites in NaNO3, Journal of
Power Sources 137 (2004) 152157.
[4] FengChin Wu, RuLing Tseng, ChiChang Hu, ChenChing
Wang, Effects of pore structure and electrolyte on the
capacitive characteristics of steam and KOHactivated
carbons for supercapacitors, Journal of Power Sources
144 (2005) 302309.
[5] K. Kierzek, E. Frackowiak, G. Lota, G. Gryglewicz,
J. Machnikowski, Electrochemical capacitors based on
highly porous carbons prepared by KOH activation,
Electrochimica Acta 49 (2004) 515523.
[6] A.B. Fuertes, F. Pico, J.M. Rojo, Influence of pore structure
on electric doublelayer capacitance of template mesoporous
carbons, Journal of Power Sources 133 (2004) 329336.
[7] M.W. Verbrugge, P. Liu, S. Soukiazian, Activatedcarbon
electricdoublelayer capacitors: electrochemical charac
terization and adaptive algorithm implementation, Journal of
Power Sources 141 (2005) 369385.
[8] K. Rajendra Prasad, N. Munichandraiah, Electrochemical
studies of polyaniline in a gel polymer electrolyte, High
energy and high power characteristics of a solidstate redox
supercapacitor, Electrochemical and SolidState Letters
5 (2002) A271A274.
[9] A. Malinauskas, J. Malinauskiene, A. Ramanavicius,
Conducting polymerbased nanostructurized materials:
electrochemical aspects, Nanotechnology 16 (2005) R51R62.
[10] A. Nishino, A. Yoshida, I. Tanahashi, I. Tajima,
M. Yamashita, T. Muranada, H.Yoneda, Planar capacitors
with electrical double layer with polarizable electrodes made
of activated carbon fiber, National Technical Report
31 (1985) 318330.
[11] B.E. Conway, Electrochemical supercapacitors. Plenum
Publishing, New York, 1999.
[12] Songhun Yoon, Jinwoo Lee, Taeghwan Hyeon, Seung
M. Oh, Electric doublelayer capacitor performance of a
new mesoporous carbon, Journal of The Electrochemical
Society 147 (2000) 25072512.
[13] H. Gerischer, An interpretation of the double layer capacity
of graphite electrodes in relation to the density of states at
the Fermi level, The Journal of Physical Chemistry
89 (1985) 42494251.
[14] B.A. Agranat, M.N. Dubrovin, N.N. Khavskii, G.I. Eskin,
Fundamentals of physics and technologies of ultrasound,
Vysshaya Shkola, Moskva, 1987.
[15] I.V. Ostrovskii, A.B. Nadtochii, AA. Podolyan, Ultrasound
stimulated lowtemperature redistribution of impurities in
silicon, Fizika i tekhnika poluprovodnikov 36 (2002) 389391.
[16] I.V. Zolotukhin, Yu.E. Kalinin, V.I. Loginova, Solidstate
fractal sructures, International Scientific Journal for
Alternative Energy and Ecology 9 (2005) 5666.
[17] G. Porod, Xray low angle scattering of dense colloid
systems, Part I, KolloidZeitschrift 124 (1951) 83114.
[18] R. de Levie, On porous electrodes in electrolyte solutions,
Electrochimica Acta 8 (1963) 751780.
[19] Jaeyun Kim, Jinwoo Lee, Taeghwan Hyeon, Direct
synthesis of uniform mesoporous carbons from the
carbonization of assynthesized silica/triblock copolymer
nanocomposites, Carbon 42 (2004) 27112719.
[20] A.B. Fuertes, S. Alvarez, Graphitic mesoporous carbons
synthesised through mesostructured silica templates, Carbon
42 (2004) 30493055.
[21] J. Robertson, Hard amorphous (diamondlike) carbons,
Progress in Solid State Chemistry 21 (1991) 199333.
[22] M. Armandi, B. Bonelli, I. Bottero, C. Otero Areán,
E. Garrone, Synthesis and characterization of ordered porous
carbons with potential applications as hydrogen storage
media, Microporous and Mesoporous Materials 103 (2007)
150157.
[23] L. Ravagnan, F. Siviero, C. Lenardi, P. Piseri, E. Barborini,
P. Milani, Clusterbeam deposition and in situ characteri
zation of carbynerich carbon films, Physical Review Letters
89 (2002) 285506.
[24] A.C. Ferrari, J. Robertson, Interpretation of Raman spectra
of disordered and amorphous carbon, Physical Review B
61 (2000) 1409514107.
[25] A.C. Ferrari, J. Robertson, Resonant Raman spectroscopy of
disordered, amorphous, and diamondlike carbon, Physical
Review B 64 (2001) 075414.