5208
Langmuir 1997, 13, 5208-5210
Notes
On the Carbon Dioxide and Benzene
Adsorption on Activated Carbons To Study
Their Micropore Structure
C. Moreno-Castilla, J. Rivera-Utrilla,*
F. Carrasco-Marı́n, and M. V. López-Ramón
Departamento de Quı́mica Inorgánica, Grupo de
Investigación en Carbones, Facultad de Ciencias,
Universidad de Granada, 18071 Granada, Spain
Received March 3, 1997. In Final Form: June 18, 1997
Introduction
In a previous paper1 it has been shown that the
micropore volumes obtained from the Dubinin-Astakhov
equation applied to CO2 adsorption isotherms at 273 K on
activated carbons with a medium to high degree of
activation are affected by the adsorption on the mesopore
surface. In these cases the real micropore volume and its
distribution and dispersion, as well as the mesopore
surface area of the activated carbons, can be obtained
from the benzene adsorption isotherms by applying the
Dubinin-Stoeckli equation2-8 corrected with the Dubinin-Zaverina equation.4,8-10
The activated carbons studied in the above paper were
prepared by steam activation of a Spanish bituminous
coal with an ash content of 11.5%. This mineral matter
present in the coal can play an important role in the
activation process by increasing the rate of the carbon
steam reaction.
The aim of the present work is to study the porous
texture from the benzene isotherms of a series of activated
carbons obtained by steam activation of the same parent
coal as that used in a previous study.1 In this work,
however, the parent coal was demineralized by froth
flotation prior to being activated. The textural characteristics of these activated carbons will be compared with
those corresponding to the activated carbons from the
untreated parent coal in order to discover the effect of the
mineral matter in coals on the porosity of the resulting
activated carbons.
Experimental Section
A bituminous coal (A) from Puertollano (Spain) was demineralized by froth flotation (sample AF) in a glass column with
a porous plate at the bottom to create the bubbles. The frother
and collector used were diacetone alcohol and kerosene, respectively. The particle size used was 1.0-1.4 mm. Experimental
details of the flotation process and the chemical characteristics
of the floated coal were reported elsewhere.11
(1) Moreno-Castilla, C.; Carrasco-Marı́n, F.; López-Ramón, M. V.
Langmuir 1995, 11, 247.
(2) Dubinin, M. M. Dokl. Akad. Nauk SSSR 1984, 275, 1442.
(3) Kadlec, O. In Characterization of Porous Solids; Gregg, S. G.,
Sing, K. S. W., Stoeckli, F., Eds.; Society of Chemical Industry: London,
1979; p 13.
(4) Dubinin, M. M. Carbon 1985, 23, 373.
(5) Dubinin, M. M. Carbon 1989, 27, 457.
(6) Dubinin, M. M.; Polyakov, N. S.; Kataeva, L. I. Carbon 1991, 29,
481.
(7) Stoeckli, F.; Huguenin, D.; Greppi, A. J. Chem. Soc., Faraday
Trans. 1993, 89, 2055.
(8) Polyakov, N. S.; Dubinin, M. M.; Kataeva, L. I.; Petuhova, G. A.
Pure Appl. Chem. 1993, 65, 2189.
(9) Dubinin, M. M.; Izotova, T. M.; Kadlec, O.; Krainova, O. L. Izv.
AN SSSR, Ser. Khim. 1975, 1232.
(10) Dubinin, M. M.; Degtyarev, M. V.; Nikolaev, K. M.; Polyakov,
N. S. Izv. Akad. Nauk SSSR, Ser. Khim. 1989, 7, 1463.
(11) Rivera-Utrilla, J.; López-Ramón, M. V.; Carrasco-Marı́n, F.;
Maldonado-Hódar, F. J.; Moreno-Castilla, C. Carbon 1996, 34, 917.
S0743-7463(97)00233-3 CCC: $14.00
The proximate analysis of the parent coal (wt %, dry basis) is
VM ) 21.5, ash ) 11.5, and Cfixed ) 67.0. After its demineralization, the ash content of this coal was 4.0%. As in the parent
coal, the main minerals detected by FTIR and XRD experiments
in the demineralized coal were illite, kaolinite, and quartz.
Three activated carbons were obtained from the demineralized
coal. The coal was pyrolized at 1123 K (10 K min-1) in N2 for
1 h (sample AFP) and then activated in steam at 1113 K for
different periods of time (2.5, 5, and 10 h). Experimental details
of this process were reported elsewhere.11,12 These activated
carbons will be referred to in the text as AFP followed by a number
that corresponds to the percentage burn-off.
All activated carbons were characterized by N2 and CO2
adsorption at 77 and 273 K, respectively, and by mercury
porosimetry, as explained in detail elsewhere.1,12,13 Some of the
most important parameters are summarized in Table 1, where
W0 and E0 are the micropore volume and the characteristic
adsorption energy, respectively, obtained by the application of
the Dubinin-Astakhov (DA) equation5,14,15 to the CO2 adsorption
data, and n is the exponent of this equation. Values of n between
1 and 4 are observed for most carbon adsorbents, with a value
of n > 2 for molecular sieve carbons or carbon adsorbents with
highly homogeneous, small micropores16-18 and values of n < 2
for strongly activated carbons and heterogeneous carbons. The
values of n and E0 in Table 1 decrease when the degree of
activation increases because this enhances the heterogeneity of
the micropores and their width.
SCO2 and SN2 in Table 1 correspond to the apparent surface
area measured with CO2 and N2, respectively, by application of
the DA equation to the data of the CO2 isotherm and the BET
equation to those corresponding to the N2 isotherm. The cross
sectional area for the N2 molecule was taken19 as 0.162 nm2 and
for CO2 as 0.187 nm2. The CO2 liquid density at 273 K was
taken19 as 1.03 g cm-3. V2 is the volume of pores with diameters
between 3.7 and 50 nm and V3 is the volume of pores with a
diameter greater than 50 nm. Both pore volumes were determined by mercury porosimetry.
The benzene isotherms on the pyrolized coal AFP and the
three activated carbons were obtained at 303 K in gravimetric
equipment following the method described elsewhere.20,21 Prior
to carrying out the adsorption runs, samples were outgassed
overnight at 383 K with a dynamic vacuum of 10-6 Torr.
Results and Discussion
Benzene adsorption isotherms for the AFP samples
obtained at 303 K are depicted in Figure 1. These
isotherms are close to type I of the BDDT classification.22
Thus, the initial part of the adsorption branch of the
isotherm represents micropore filling and the slope of the
plateau at medium and high relative pressure is due to
(12) López-Ramón, M. V.; Moreno-Castilla, C.; Rivera-Utrilla, J.;
Hidalgo-Alvarez, R. Carbon 1993, 31, 815.
(13) Carrasco-Marı́n, F.; López-Ramón, M. V.; Moreno-Castilla, C.
Langmuir 1993, 9, 2758.
(14) Dubinin, M. M.; Stoeckli, H. F. J. Colloid Interface Sci. 1980,
75, 34.
(15) Dubinin, M. M.; Astakhov, V. A. Adv. Chem. Ser. 1970, 102, 69.
(16) Dubinin, M. M.; Astakhov, V. A. Izv. Akad. Nauk SSSR, Ser.
Khim. 1971, 1, 5.
(17) Rand, B. J. Colloid Interface Sci. 1976, 56, 337.
(18) Kraehenbuehl, F.; Stoeckli, F.; Addoun, A.; Ehrburger, P.;
Donnet, J. B. Carbon 1986, 24, 483.
(19) Rodrı́guez-Reinoso, F.; Linares-Solano, A. In Chemistry and
Physics of Carbon; Thrower, P. A., Ed.; Marcel Dekker, Inc.: New York,
1989; Vol. 21, p 1.
(20) Domingo-Garcı́a, M.; Fernández-Morales, I.; López-Garzón, F.
J.; Moreno-Castilla, C. Langmuir 1991, 7, 339.
(21) Moreno-Castilla, C.; Carrasco-Marı́n, F.; Utrera-Hidalgo, E.;
Rivera-Utrilla, J. Langmuir 1993, 9, 1378.
(22) Gregg, S. J.; Sing, K. S. W. Adsorption Surface Area and Porosity,
2nd ed.; Academic Press: London, 1982.
© 1997 American Chemical Society
Notes
Langmuir, Vol. 13, No. 19, 1997 5209
Table 1. Textural Parameters of the Activated Carbon Series
sample
% ash
AFP-14
AFP-24
AFP-53
7.2
8.2
13.2
S N 2,
m2
g-1
447
634
1002
SCO2, m2 g-1
W0(CO2), cm3 g-1
E0(CO2), nm
n(CO2)
V2, cm3 g-1
V3, cm3 g-1
453
688
1085
0.172
0.261
0.412
24.7
19.9
15.2
2.16
1.70
1.49
0.03
0.05
0.14
0.05
0.06
0.10
Table 2. Parameters Obtained from the Dubinin-Stoeckli Equation Applied to the Benzene Adsorption Isotherms
effective parameters
sample
AFP
AFP-14
AFP-24
AFP-53
W0,
cm3
g-1
0.004
0.130
0.250
0.440
L0, nm
d, nm
1.53
1.16
1.18
1.49
0.199
0.207
0.216
0.243
real parameters
E0, kJ
mol-1
18.7
19.6
19.5
15.1
Sme,
m2
g-1
W0,
1
44
67
226
cm3
g-1
0.003
0.110
0.220
0.360
L0, nm
d, nm
E0, kJ mol-1
1.42
1.09
1.12
1.20
0.096
0.139
0.139
0.149
19.9
22.8
22.0
16.2
isotherms quantitatively but cannot be used to calculate
the micropore volume distribution with size. The experimental isotherm must, in this case, be corrected for
the adsorption on mesopores according to Dubinin,4 the
experimental values of the adsorption isotherms, aexp, are
composed of the adsorption in micropores, ami, and in
mesopores, ame:
aexp ) ami + ame
(1)
Equation 1 can be converted to
aexp ) ami + γSme
(2)
where γ is the adsorption of the vapor (benzene) at a given
temperature per unit mesoporous surface area of the
adsorbent and Sme is the surface area of the mesopores.
The value of γ was experimentally obtained by Dubinin
and Zaverina,4,8,24 for carbonaceous adsorbents from the
adsorption isotherm of benzene on a nonporous carbon
black, the results obtained were further extended for other
organic adsorptives, and this more general equation is
known as the modified Dubinin-Zaverina equation:
Figure 1. Benzene adsorption-desoprtion isotherms on
activated carbons at 303 K: (]) AFP; (0) AFP-14; (O) AFP-24;
(4) AFP-53.
multilayer adsorption on the nonmicroporous surface
(mesopores, macropores, and the external surface).
In this series of isotherms there is an opening of the
knee and an increase in the slope of the linear branch,
which indicates that there is a progressive development
of micro- and mesoporosity with the degree of activation.19
The adsorption-desorption isotherms of benzene on the
activated carbons show a H4 type hysteresis loop,22
characteristic of slit-shaped pores, with the adsorption
and desorption branches parallel.
The theory of volume filling of micropores (TVFM)
describes the porous structure of microporous activated
carbons.2-8,19,22,23 According to this theory, three parameters of the activated carbon micropore structure can be
obtained: the micropore volume, W0; the micropore width
at the distribution curve maximum, L0; and the dispersion,
d, that characterizes the micropore distribution range.
These three parameters can be determined from the
benzene adsorption isotherms by applying the DubininStoeckli (DS) equation.2-8,19,22,23
When the activated carbon has a sufficiently developed
mesopore surface, the contribution of the adsorption on
mesopores to the total adsorption is considerable. In this
case, the application of the DS equation to the experimental benzene adsorption isotherm yields the effective
values of the parameters W0, L0, and d. These effective
parameters4 can be used to describe the adsorption
(23) Bansal, R. C.; Donnet, J. B.; Stoeckli, F. Active Carbon; Marcel
Dekker, Inc.: New York, 1988.
γ)
A
8.14 × 10-4
mmol/m2
exp ν
6.35β
[
]
(3)
where ν is the millimolar volume of the adsorptive and
was taken as 0.0898 for benzene and 6.35 is the characteristic adsorption energy, in kJ mol-1, on a mesoporous
surface. By using eqs 1 and 2, the plot of aexp against γ
should be a straight line whose slope is the value of the
mesopore surface area, Sme.
The experimental benzene adsorption isotherms were
corrected at each adsorption point for adsorption on
mesopores according to eqs 1 and 2 using the values of
Sme. The DS equation was again applied to the corrected
adsorption isotherms obtained. From this calculation, the
real parameters W0, L0, and d of the micropore structure
of the activated carbons were determined.
The effective and real parameter values obtained from
the benzene adsorption isotherms are reported in Table
2 along with Sme values. The parameters were determined
from the DS equation by minimizing the residual sum of
squares by iteration.3,25
The micropore size distribution was obtained from the
real parameters W0, L0, and d by applying the normal
distribution equation:
[
]
(L0 - L)2
W0
dW
exp )
dL 2dx2π
8d2
(4)
(24) Dubinin, M. M.; Degtyarev, M. V.; Nikolaev, K. M.; Polyakov,
N. S. Izv. Akad. Nauk SSSR, Ser. Khim. 1989, 7, 1463.
(25) Salas-Peregrı́n, M. A.; Carrasco-Marı́n, F.; López-Garzón, F. J.;
Moreno-Castilla, C. Energy Fuels 1994, 8, 239.
5210 Langmuir, Vol. 13, No. 19, 1997
Notes
Figure 2. Micropore size distribution obtained from benzene
adsorption data.
Table 3. Real Parameters Obtained from the
Dubinin-Stoeckli Equation Applied to the Benzene
Adsorption Isotherms
sample
% ash
Sme, m2 g-1
W0, cm3 g-1
L0, nm
d, nm
AP-16
AP-30
AP-48
CP-40
CP-55
13.6
16.7
24.7
1.7
2.2
58
94
183
124
208
0.13
0.20
0.23
0.26
0.31
0.71
1.14
1.37
1.11
1.34
0.107
0.115
0.136
0.116
0.139
Micropore size distributions of the AFP activated carbon
series are shown in Figure 2.
Data in Table 2 indicate that the values of the real
parameters are lower than their corresponding values
obtained for the effective parameters, which is due to the
development of the mesopore area (Sme) that, in the
activated carbon series, increases with the percentage of
burn-off, varying from 44 m2‚g-1 for the activated carbon
whose burn-off is 14% to 226 m2‚g-1 for that with a burnoff equal to 53%.
As shown in Table 2 and Figure 2, the micropore volume,
W0, and the mean micropore width increase with activation. In all cases, the value of SCO2 is slightly higher than
SN2 (Table 1), which indicates that the value of W0(CO2)
obtained by applying the DA equation to the CO2 isotherm
is increased due to the fact that this also accounts for the
CO2 adsorbed on the mesopore surface. Thus, the real
parameter W0 obtained from benzene (Table 2) is, in all
cases, lower than the W0(CO2) values (Table 1). According
to the micropore size distribution of the activated carbons
(Figure 2) and the minimum kinetic diameters of the
adsorbates (CO2 ) 0.33 nm; N2 ) 0.36 nm; benzene ) 0.37
nm), it is clear that the microporosity of these activated
carbons is accessible to the three molecules of the
adsorbates.
Table 3 shows the real parameters obtained with
benzene for the other two series of activated carbons
prepared both from the untreated parent coal (series AP)
and from the parent coal demineralized by acid treatments
with HCl and HF at 333 K (series CP).1,11 The results
listed in this table have been discussed in a previous
paper,1 and they are given in this work to compare them
with those obtained with the series AFP.
Both the mesopore surface area, Sme, and the real
micropore volume, W0, obtained from benzene adsorption,
of the three series of activated carbons, as a function of
their percentage burn-off, are depicted in Figures 3 and
4, respectively. It is interesting to note that the mesopore
surface area linearly increases with the degree of activation or percentage BO of the activated carbon and is
independent of the demineralization pretreatment given
to the parent coal prior to preparation of the activated
carbon. These results indicate, therefore, that the mineral
Figure 3. Relationship between the mesoporous surface area
of the activated carbons and their percentage burn-off: (O) AFP
series; (0) AP series; (4) CP series.
Figure 4. Micropore volume from benzene isotherm (real
parameter) as a function of percentage burn-off: (O) AFP series;
(0) AP series; (4) CP series.
matter present in the parent coal did not affect mesoporosity development. Nevertheless, microporosity development was influenced by the above pretreatment (Figure
4). Thus, for a given BO value, the micropore volume
decreases following the order AFP series > CP series >
AP series. Hence, treating the parent coal by froth
flotation as a pretreatment to steam activation gives
activated carbons with ash contents lower than those from
the AP series and a microporosity more developed than
those from the AP and CP series. Both characteristics
are of capital importance from the viewpoint of the
industrial potential of these activated carbons.
Conclusions
The micropore volume of the activated carbons obtained
from the Dubinin-Astakhov equation applied to CO2
adsorption at 273 K is affected by adsorption on external
surface. Thus, to determine the real parameters of the
micropore structure, it was neccessary to apply the
Dubinin-Stoeckli equation to the benzene adsorption
isotherms at 303 K corrected by the Dubinin-Zaverina
approach.
The demineralization treatments given to the parent
coal prior to its activation affected microporosity development, whereas the mesopore surface area was independent
of these treatments and linearly increased with the
percentage BO.
Acknowledgment. We thank DGICYT (Project PB940754) for financial support.
LA970233C