Preparation of high surface area carbons from Corncob with KOH etching
plus CO
2
gasification for the adsorption of dyes and phenols from water
Ru-Ling Tseng
a,b,∗, Szu-Kung Tseng
a, Feng-Chin Wu
caGraduate Institute of Environmental Engineering, National Taiwan University, Taiwan
bDepartment of Safety, Health and Environmental Engineering, National United University, Miao-Li 360, Taiwan cDepartment of Chemical Engineering, National United University, Miao-Li 360, Taiwan
Received 26 May 2005; received in revised form 15 December 2005; accepted 22 December 2005 Available online 7 February 2006
Abstract
Carbonaceous adsorbents with controllable surface area and the microporous volume ratio (Vmicro/Vpore) were activated from carbonized corncobs
(i.e., char) with KOH etching plus CO2gasification in this work. Activated carbons derived from KOH/char ratio equal to 1 and CO2gasification
time from 0 to 60 min exhibited BET surface area increasing from 1071 to 1991 m2g−1and the V
micro/Vporevalues decreasing rapidly from 0.805
to 0.565. And those derived from KOH/char ratio of 4 and CO2gasification time from 0 to 30 min exhibited high BET surface area from 2402
to 2844 m2g−1. Scanning electron microscopic (SEM) results revealed that violent reactions took place on the surfaces of honeycombed holes in
these carbons when the KOH/char ratio was equal to 1 and CO2gasification was used. The adsorption of three dyes (MB, BB1, and AB74) and
three phenols (phenol, 4-CP, and 2,4-CP) from water on all activated carbons at 30◦C were investigated. Adsorption kinetics was in agreement with the Elovich equation, and the values of the Elovich parameter (1/b) of the carbons with different CO2gasification time were compared. The
equilibrium isotherms were in agreement with the Langmuir equation, and they were used for comparing the amounts of adsorption corresponding to the monolayer coverage of the different carbons.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Activated carbons; KOH etching plus CO2gasification; Corncob; Pore properties; Adsorption
1. Introduction
In general, the physicochemical properties of activated car-bons (ACs) have been found to strongly depend on the activation process as well as the nature of the raw material. Moreover, an understanding of the influence of activation variables on the physicochemical properties of ACs is very important in devel-oping the structure of carbons in both physical and chemical activation processes[1]. The high adsorption capacity of ACs is attributed to surface area, pore volume, and porosity. These char-acteristics are functions of the type of raw material employed as well as the method of activation[2]. This is especially important in the development of micro- and mesopores, which are related to the adsorption capability and ability of carbons to various types of chemicals either from gas or liquid. Accordingly, recent
research has focused on the development of ACs with desired
∗Corresponding author. Tel.: +886 37 381775; fax: +886 37 333187.
E-mail address: [email protected] (R.-L. Tseng).
pore structures as well as with new application possibilities in different fields[3].
Besides, activated carbons can also be applied to surperca-pacitors[4–7], catalyst supports of fuel cells[8], safe storage of large quantities of CH4[9]or H2, and to the field of
biomedi-cal engineering. Activated carbons used in these fields not only have high specific surface area but also high ratio of mesopores and macropores providing the main transport channels for adsor-bates.
In previous studies, peach stone was activated with H3PO4
with a weight ratio of peach stone/H3PO4equal to 1:0.91 to form
activated carbon, which was then under CO2gasification for 9 h.
Activated carbon with a total pore volume of 2 cm3g−1and a mesoporous volume ratio of 67.5% was obtained[10]. Coconut shell was activated with ZnCl2with a weight ratio of coconut
shell/ZnCl2ratio equal to 1:1. The obtained activated carbon was
further under 6 h of CO2gasification. Activated carbon with a
BET surface area of 2634 m3g−1 and a mesoporous volume
ratio of 62% was obtained[11]. When coconut shell was acti-vated with KOH with a weight ratio of coconut shell/KOH equal
0927-7757/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfa.2005.12.042
Nomenclature
Ce solute concentration in the aqueous phase at equi-librium (mol m−3)
Ct solute concentration in the aqueous phase at time
t (mol m−3)
C0 initial solute concentration in the aqueous phase (mol m−3)
Dp average diameter of pores (nm)
KL Langmuir constant defined in Eq.(6)(m3mol)
qe amount of adsorption at equilibrium (mol kg−1)
qt amount of adsorption at time t (mol kg−1)
qmon amount of adsorption corresponding to mono-layer overage (mol kg−1)
Smicro micropore surface area (m2/g)
Sp total BET surface area (m2/g)
t adsorption time (min)
V volume of the solution (m3)
Vmicro micropore volume (cm3/g)
Vpore total pore volume (cm3/g)
W amount of dry adsorbents used (kg)
ρb bulk density (kg/m3)
to 1:0.5 and was further under 6 h CO2gasification, activated
car-bon with a BET surface area of 2390 m2g−1and a mesoporous volume ratio of 47% was obtained [12]. The abovementioned studies revealed that CO2gasification did enhance high
meso-pore volume ratio as well as surface area in activated carbons. Not much research on the preparation and characterization of ACs derived from corncob has been reported although corncob is a cheap and abundant agricultural waste of no economical value
[13]. In addition, corncob-derived ACs have been proven to be highly porous and rich in mesopores, exhibiting high adsorption capacity to methylene blue and Pb2+ions[13]. However, ZnCl2
activation was reported to be a very suitable process for the preparation of corncob-derived ACs with an essentially micro-porous structure[14].
In general, ACs with different pore size distributions usu-ally result from the differences in precursors and treatments
[10,15]. In our laboratory, a series of studies were conducted to prepare porous carbons from various wood wastes and fruit shells by physical activation method (steam) and chemical acti-vation method (KOH), in which these carbons were evaluated for the possibility of applications to industrial pollution control and supercapacitors[16–21].
According to previous report, high surface area microp-ore activated carbon of BET surface area of 2595 m2g−1 and Smicro/Sp ratio of 0.899 was obtained from corncob
acti-vated with KOH [22]. Furthermore, CO2 will be utilized to
develop ACs of high surface area and higher ratio mesopore for wider applications in this study. In this work comparisons were made for the physicochemical properties of the corncob carbon activated by KOH followed by CO2gasification for
dif-ferent time duration. The physical properties studied included the BET surface area, pore size distribution, and total pore
volume. The kinetics and equilibrium of the adsorption to methy-lene blue, basic brown 1, acid blue 74, 2,dichlorophenol, 4-chlorophenol, and phenol in aqueous media were systematically discussed.
2. Materials and methods
2.1. Preparation of carbons by KOH etching plus CO2 gasification
According to previous report[22], the ACs activated with KOH/char ratios of 0.5, 1.0, and 2 were classified as Type I, the activation reaction being surface activation and micropore etching. And those with KOH/char ratios equal to 3, 4, 6 were classified as Type II, the activation reactions being only micro-pore etching. The AC activated with KOH/char ratio equal to 1 (Type I) and that with KOH/char ratio equal to 4 (Type II) were selected to study the effects of the combined CO2
gasifica-tion in this research. The effects on both ACs from various CO2
gasification times were compared.
Activated chars were prepared from corncob by carboniza-tion in nitrogen. The apparatus and procedures were described in detail elsewhere [22]. These chars were well mixed with water and KOH in a stainless steel beaker with the weight ratios of KOH/char equal to 1 and 4. Water was evaporated at 130◦C for 24 h, and these dried mixtures consisting of chars and KOH (without KOH lost) were obtained. The dried mix-tures were placed in a sealed ceramic oven, heated at a rate of 10◦C/min to 780◦C, and kept at this temperature for 1 h. In the meantime, nitrogen gas flowed into the oven at a rate of 3× 10−3m3min−1. When the time was up, the nitrogen gas was shut off and CO2 immediately flowed into the oven at a
rate of 2× 10−3m3min−1. The total duration of nitrogen and CO2flows to the oven kept at 780◦C was 60 min. The activated
products were cooled to room temperature and washed with deionized water[22]. The samples were classified according to the KOH/char ratio and the CO2gasification time and denoted
as Cob1000, Cob1015, Cob1030, Cob1060, Cob4000, Cob4015, and Cob4030, respectively. The first three letters, Cob, represent the material, corncob; the forth digit represents the KOH/char ratio; and the last three digits represent the CO2gasification time
(minutes).
2.2. Measurements of physical properties
The yield was defined as the weight ratio of final carbons to the initial dried raw materials. The BET surface area of the car-bon (Sp) was obtained from the N2adsorption isotherm at 77 K
with a sorptiometer (Porous Materials, BET-202A). Prior to this measurement, the samples were first dried in an oven at 130◦C overnight and then, quickly placed in the sample tube. After that, the tube was heated at 230◦C and evacuated for 4 h until the pressure was less than 1.33× 10−4mbar. The total pore volume (Vpore) was deduced from the adsorption data based on the
man-ufacturer’s software, and the pore size distribution was derived from the BJH theory[23]. The micropore volume (Vmicro) and
Fig. 1. Adsorption/desorption isotherms of N2 at 77 K on activated carbons
derived from corncob using KOH with CO2gasification (carbons is Cob1000
(--), Cob1015 (--), Cob1030 (--), Cob1060 (-♦-), Cob4000 (-䊉-), Cob4015 (--), and Cob4030 (--), respectively).
method[24,25]. The surface area corresponding to the microp-ores (Smicro) was obtained from the difference between Spand Sext[26].
2.3. Procedures for adsorption experiments
Six solutes including acid blue74 (AB74), basic brown 1 (BB1), methylene blue (MB), 2,dichloropenol (2,DCP), 4-chloropenol (4-CP), and phenol are analytical reagent grade (Merck). Molecular weights are, respectively, 466.4, 419.4, 284.3, 163.0, 128.6, and 94.1 g mol−1. The molecular structures of MB, BB1, and AB74, and the characteristics of the phenols and dyes are from the previous study[27].
The kinetic experiments were carried out in a Pyrex glass vessel 100 mm in inner diameter, 130 mm in height, and fitted with four glass baffles (10 mm in width). The aqueous solu-tion (0.6 dm3) with 0.3 g carbon powder was agitated at 500 rpm using a Cole-Parmer Servodyne agitator having a six-flat-blade impeller (12 mm high and 40 mm wide). When carbons were added to the vessel, adsorption time was recorded. The vessel was also immersed in a water bath controlled at 30◦C. During the experiment, aqueous samples (5 cm3) were taken from the solution and the concentrations were analyzed, and the concen-trations of solutes in the aqueous phase were determined with a Hitachi UV–vis spectrophotometer (U-2001). The amount of
Fig. 2. Pore size distribution of the activated carbons derived from corncob using KOH with CO2gasification.
adsorption at time t, qt(mol/kg), was calculated by qt= (C0− Ct)V
W (1)
where C0and Ctare the liquid concentrations (mol m−3) at the
beginning and time t, respectively, and W/V is the dose of dried carbons (kg m−3). The experiment error was mostly within 4%. In the adsorption equilibrium experiments, 0.1 g carbon was well dispersed in 0.1 dm3aqueous solution in a 0.25-dm3flask and stirred for 5 days in a water bath (Haake Model K-F3) at 30◦C. Preliminary tests showed that adsorption was complete after 3 days. After filtration with glass fibers, the concentrations were analyzed. Each experiment was repeated at least three times under identical conditions. The amount of adsorption at equilib-rium, qe(mol kg−1), was calculated according to Eq.(2): qe=(C0− Ce)V
W (2)
where Ceis the equilibrium liquid concentrations (mol m−3). 3. Results and discussion
3.1. Physical properties of activated carbon
Identifying the pore structure of adsorbents is an essential procedure before designing the adsorption process, which is commonly determined by the adsorption of inert gases[28,29].
Fig. 1shows the typical adsorption/desorption isotherms of N2
Table 1
Physical properties of carbons derived from corncob using KOH and CO2activation
Carbons KOH/char ratio (−) Gasification time (min) Sp(m2/g) Smicro/Sp(−) Vpore(cm3/g) Vmicro/Vpore(−) ρb(kg/m3)
Cob1000 1 0 1071 0.923 0.691 0.805 198 Cob1015 1 15 1383 0.886 0.875 0.721 157 Cob1030 1 30 1625 0.859 0.991 0.679 155 Cob1060 1 60 1991 0.790 1.270 0.565 108 Cob4000 4 0 2402 0.930 1.290 0.844 112 Cob4015 4 15 2510 0.905 1.369 0.796 92 Cob4030 4 30 2844 0.894 1.533 0.772 91
at 77 K for all ACs with different CO2gasification times. Two
different types of curves are found inFig. 1: one group of corn-cob ACs were prepared with KOH/char ratio equal to 1 (empty circles in the figure), and another group of ACs were prepared with KOH/char ratio equal to 4 (solid circles in the figure). Note that, when P/Povalues vary from 0 to 1.0, the initial adsorbed
volumes of the group with KOH/char ratio equal to 1 are almost the same (327–369 cm3g−1) at P/Po approaching zero, while
under specified P/Povalues, the adsorbed volume increases with
increased CO2gasification time. For the group with KOH/char
ratio equal to 4 the initial adsorbed volumes are almost the same (488 cm3g−1) at P/Poapproaching zero, while under specified P/Povalues, the adsorbed volume increases with increased CO2
gasification time, the same as the group with KOH/char ratio equal to 1. But, when the P/Povalue equaled to or was above
0.2, these three curves approach horizontal. All the above results indicated that developing the pore structure of corncob ACs strongly depended on the CO2gasification time.
Fig. 2shows a typical pore size distribution of all ACs acti-vated with KOH etching combined with CO2gasification. For
the group with KOH/char ratio equal to 1, one peak is visible. The peak is in the mesopore region with pore sizes between 3 and 5 nm. On the other hand, most pores of the AC group with a KOH/char equal to 4 are below 3 nm.Fig. 2shows that very wide pore distributions were produced in the activated carbons with KOH/char ratio equal to 1 because of the additional CO2
gasification, while microporous type was still maintained for the activated carbons with KOH/char ratio equal to 4 in spite of the additional CO2 gasification. It appears that the process of
KOH etching plus CO2gasification has two different reaction
mechanisms on carbon. These two graphs together with SEM observations are used to present and explain this result.
Table 1shows the pore properties of all ACs, including Sp, Smicro/Sp, Vpore, Vmicro/Vpore, and ρb. Note that the Sp values
of all ACs gradually increase with increased gasification time. These results are in agreement with those in other literatures
[11,12]. The Spvalues of the group with KOH/char ratio equal
Fig. 3. Micropore volume (Vmicro), mesopore volume (Vmeso), mean pore size
(Dp), and yield of activated carbons using KOH with CO2gasification with
different BET surface-areas: (a) KOH/char = 1 (䊉) (); KOH/char = 4 () (♦); (b) KOH/char = 1 () (); KOH/char = 4 () ().
to 1 (1071–1991 m2g−1) and the group with KOH/char ratio equal to 4 (2402–2844 m2g−1) are comparable to those of the ACs prepared from corncobs activated with steam[17], 50 wt% H3PO4[13], 200 wt% ZnCl2[14], 15 wt% KOH, and 37.5 wt%
K2CO3[2], having the Spvalues of 943, 960, 1563, 1806, and
1541 m2g−1, respectively. The raw materials in the above men-tioned studies were directly soaked in chemicals to activate them, which is different from the two step process of KOH etching plus CO2gasification for activated carbon preparations in this study.
This is why the Spvalues of the group with KOH/char ratio equal
to 4 of this study are higher than those in the other studies. In this paper BET surface area (1991 m2/g) of the activated carbon with KOH/char ratio equal to 1 plus 60 min of CO2gasification
is approximately equal to that (1976 m2/g) with KOH/char ratio equal to 3 without CO2gasification[22]. Less chemical dose is
required for the preparation of higher surface activated carbons. The fraction of micropore area, Smicro/Sp,of the group with
KOH/char ratio equal to 1 show that prolonged CO2gasification
Fig. 5. Test of the Elovich equation for the adsorption of dyes and phenols on the activated carbons using KOH combined with CO2gasification (a) phenol, (b)
4-CP, (c) 2,4-DCP, (d) MB, (e) BB1, and (f) AB74 (carbons is Cob1000 (--), Cob1015 (--), Cob1030 (--), Cob1060 (-♦-), Cob4000 (-䊉-), Cob4015 (--), and Cob4030 (--), respectively).
leads to a decrease in Smicro/Sp(0.923–0.790).In addition, the
fraction of mircropore volume, Vmicro/Vpore,of the group with
KOH/char ratio equal to 1 shows that prolonged CO2
gasifica-tion leads to a decrease in Vmicro/Vpore (0.805–0.565). Porous,
high surface area activated carbons are gradually, widely applied to the field like supercapacitors. KOH activation can only pro-duce microporous activated carbon, and the process of KOH plus CO2gasification can produce activated carbon of higher ratio
mesopores (Vmicro/Vpore= 0.565 of Cob1060), which improves
mass transfer within the activated carbon.
These results also indicate that CO2 gasification promotes
the formation of mesopores within the activated carbon, as can be seen in Fig. 2. Table 1 shows that Vpore values increase
with prolonged CO2gasification. The Vporevalues of the group
Table 2
Kinetic parameters for the adsorption of phenols and dyes at 30◦C
Solute Carbon 1/b (mol kg−1) t0(10−3h) r2(−)
Phenol Cob1000 0.529 8.79 0.987 Cob1015 0.488 3.69 0.994 Cob1030 0.429 3.02 0.997 Cob1060 0.417 1.99 0.998 Cob4000 0.525 1.21 0.989 Cob4015 0.370 3.24 0.980 Cob4030 0.412 0.78 0.981 4-CP Cob1000 0.719 6.29 0.991 Cob1015 0.599 2.18 0.994 Cob1030 0.695 2.99 0.992 Cob1060 0.584 1.01 0.988 Cob4000 0.889 2.70 0.995 Cob4015 0.733 1.68 0.997 Cob4030 0.779 2.49 0.993 2,4-DCP Cob1000 0.621 7.96 0.996 Cob1015 0.657 7.31 0.996 Cob1030 0.662 6.34 0.995 Cob1060 0.582 2.66 0.996 Cob4000 0.898 4.48 0.992 Cob4015 0.694 3.41 0.992 Cob4030 0.689 2.74 0.992 MB Cob1000 0.115 6.14 0.969 Cob1015 0.152 5.31 0.983 Cob1030 0.181 5.64 0.992 Cob1060 0.226 6.00 0.989 Cob4000 0.256 6.34 0.989 Cob4015 0.225 2.59 0.999 Cob4030 0.253 3.52 0.995 BB1 Cob1000 0.086 9.35 0.975 Cob1015 0.122 10.7 0.987 Cob1030 0.126 7.08 0.996 Cob1060 0.154 8.06 0.998 Cob4000 0.153 9.06 0.999 Cob4015 0.154 6.38 0.997 Cob4030 0.161 6.09 0.990 AB74 Cob1000 0.025 5.90 0.970 Cob1015 0.041 3.08 0.993 Cob1030 0.055 4.15 0.997 Cob1060 0.065 4.53 0.990 Cob4000 0.072 11.6 0.996 Cob4015 0.081 4.96 0.995 Cob4030 0.080 3.49 0.992
with KOH/char ratio equal to 4 (1.29–1.53 cm3g−1) are much higher than those prepared from corncobs in the other studies
[2,13,14,18]. This is attributed to the fact that the holes walls were not contracted or twisted and large Vporevalues were
cre-ated from the KOH etching in the char interior when the corncob was activated at high KOH/char values[22]. The last item in
Table 1 is bulk density (ρb). For the ACs of the group with KOH/char ratio equal to 1, theρbvalues significantly decreased from 198 to 108 kg m−3. For the group with KOH/char ratio equal to 4 theρbvalues decreased from 112 to 91 kg m−3 with-out significant variation. Theρbvalues of the ACs in this work are obviously lower than those prepared from lignocellulosic materials in other studies[13,30,31], probably due to the porous honeycomb structure of corncob ACs[22]. This unique struc-ture is believed to decrease the resistance of liquid phase mass transfer within GAC (granular activated carbon).
The dependence of Vmicro, Vmeso, Dp, and yield of AC on Sp
is shown inFig. 3a and b.Fig. 3a shows Vmicroas well as the
rela-tionship between Spand Vext. Note that Vmicroand Vmesoincrease
proportionally with the Spvalues. But when the Spvalues are
in the range of 1071–1991 m2g−1(the group with KOH/char equal to 1), the slope of Vmeso to Sp (4.47× 10−4cm3m−2)
is larger than that of Vmicro to Sp(1.75× 10−4cm3m−2). This
proves that Vmeso of ACs of the group with KOH/char ratio
equal to1 increases swiftly with CO2 gasification time while Vmicro increases slowly, thus facilitating the development of mesoporous carbons. When the Sp values are in the range of
2402–2844 m2g−1(the group with KOH/char equal to 4), both
Vmesoand Vmicro increase, but the Vmicro values are about four
time the Vmeso values. These carbons belong to microporous
carbons.
The Dp (average diameter of pores, 4Vpore/Sp) values of
ACs provide important data for various applications (adsorp-tion, supercapacitors, catalyst-support). They are related to the rate of mass transfer and effective surface area.Fig. 3b shows that although the yield might decrease, the Dpvalues are nearly
constant with the Spincreasing. But when the Spare in the range
of 1071–1991 m2g−1(the group with KOH/char ratio equal to 1), the average Dpis 2.53 nm, larger than that of 2.16 nm when
the Sp are in the range of 2402–2844 m2g−1(the group with
KOH/char ratio equal to 4).
Yield is related to the economics of activated carbon manufac-ture.Fig. 3b shows that yields of the group with KOH/char ratio equal to 1 (from 22.3% to 6.4%) decreased higher than those of the group with KOH/char ratio equal to 4 (from 18.1% to13.7%).
In the case of the group with KOH/char ratio equal to 1, the dis-tortion of surfaces caused by CO2gasification was obvious (it
was discovered that the oven temperature rose rapidly right after the introduction of CO2). However, the group with KOH/char
ratio equal to 4 the surface were not significantly changed. This is because the activation reactions were the surface activation and micropore etching for the group with KOH/char ratio equal to 1. Thus, exothermic reactions between CO2and Cfof carbons
occurred after the introduction of CO2, and caused the CO
pro-duction and temperature rise. However, the CO2effect was not
significant for the group with KOH/char ratio equal to 4. This is because large amount of KOH enveloped the carbons, thus lowered the reactions between CO2and carbons.
3.2. SEM observations
Typical SEM photographs of ACs activated with KOH etch-ing plus CO2 gasification are shown in Fig. 4a and b. The
group with KOH/char ratio equal to 1 (Cob1000) are highly porous with honeycomb shaped, irregular, cottony holes, and with contracted and twisted walls (Fig. 3b in literature[22]).
Fig. 7. Adsorption isotherms of phenols and dyes at 30◦C on the activated carbons using KOH combined with CO2gasification (a) phenol, (b) 4-CP, (c) MB, and
(d) BB1 (carbons is Cob1000 (--), Cob1015 (--), Cob1030 (--), Cob1060 (-♦-), Cob4000 (-䊉-), Cob4015 (--), and Cob4030 (--), respectively). The solid curves were calculated with the Langmuir equation.
After being gasified with CO2for 60 min (Cob1060), the cottony
outsides of holes are transformed into plain surfaces (Fig. 4a). On the other hand, the group with KOH/char ratio equal to 4 (Cob4000) had a regular arrangement of the porous honeycombs with well-arranged holes and with thick and smooth walls with-out a cottony structure (Fig. 3e in literature[22]). Changes in holes appearances of Cob4030 are not significant (Fig. 4b).
Based on the SEM observations, the following hypotheses are proposed. When char is soaked in KOH, it should be a large amount of KOH surrounding it completely (including the interior of the holes). Activation does not happen on the char surface, resulting in the absence of a cottony structure. Accord-ingly, the char maintains its original structure, i.e., holes are not twisted or deformed because of the contraction of the walls. However, KOH soaking through the interior of the holes favors the development of micropores, resulting in an increase in the number of micropores[32]. Because a large quantity of KOH surrounds these holes, changes of these holes are not signifi-cant after 30 min of CO2gasification. In contrast, when char is
soaked in a relatively small amount of KOH, most of the KOH seeped deeply into the interior of char. If this is the case, the exterior surface of the holes should undergo violent activation, thus creating irregular cottony features on the surface of the holes. Because only a small amount of KOH protects the hole surfaces, the cottony outsides of the holes are gasified into plain surfaces.
3.3. Adsorption kinetics
Porous, high surface area activated carbons gradually have wider applications to fields such as supercapacitors. KOH acti-vation can only produce microporous activated carbon, and the process of KOH etching plus CO2gasification can produce
acti-vated carbon of higher ratio mesopores (Vmicro/Vpore= 0.565
of Cob1060), which improves mass transfer within the inte-rior of the activated carbon. The section of adsorption kinet-ics in this paper proved this fact. The Elovich equation was adopted to examine the mechanism of the adsorption process.
According to the literature[16], the Elovich equation after arrangement can be expressed as:
qt= 1 b ln(ab) + 1 b lnt (3)
where a and b are constants for any experiment. The validity of Eq.(3)is checked by the linear plot of qtversus ln t (Fig. 5). The
modeled results well agree with the measured ones as shown inFig. 5.Table 2 lists the results. The fit is quite good under the time ranges studied (correlation coefficient, r2> 0.970), and agrees with the assumption that t t0.Fig. 6(a) shows that the
1/b values of all dye solutes (MB, BB1, and AB74) on the ACs of the group with KOH/char ratio equal to 1 increase with the increased CO2gasification time. Their slopes are 1.82× 10−3 Table 3
Parameters of the Langmuir equation and covered area for the adsorption of phenols and dyes at 30◦C
Solute Adsorbent Langmuir Covered
qmon(mol kg−1) KL(m3mol−1) r2(−) Sc(m2g−1) Sc/Sp(−) Avg. Sc/Sp(−) (%)
Phenol Cob1000 2.48 5.10 0.997 653 0.610 0.476± 15.4 Cob1015 2.56 5.31 0.997 674 0.487 Cob1030 2.74 4.97 0.996 721 0.444 Cob1060 2.73 8.28 0.999 719 0.361 Cob4000 3.83 2.34 0.994 1008 0.420 0.464± 13.1 Cob4015 4.20 1.52 0.997 1105 0.416 Cob4030 4.50 1.81 0.997 1184 0.555 4-CP Cob1000 2.26 9.54 1.000 656 0.613 0.579± 5.2 Cob1015 2.74 8.44 1.000 795 0.575 Cob1030 3.39 5.60 0.998 983 0.605 Cob1060 3.59 5.69 0.998 1041 0.523 Cob4000 4.69 5.32 0.999 1362 0.567 0.658± 9.2 Cob4015 6.25 3.03 0.989 1813 0.722 Cob4030 6.71 3.63 0.984 1946 0.684 MB Cob1000 0.44 54.6 1.000 167 0.156 0.137± 6.4 Cob1015 1.24 70.4 0.999 182 0.132 Cob1030 1.48 118 1.000 218 0.134 Cob1060 1.71 76.9 1.000 251 0.126 Cob4000 2.49 60.4 1.000 366 0.152 0.153± 0.6 Cob4015 2.59 216 1.000 381 0.152 Cob4030 2.98 273 1.000 438 0.154 BBl Cob1000 1.80 11.8 0.996 319 0.298 0.305± 4.5 Cob1015 2.22 19.3 0.996 393 0.284 Cob1030 2.88 20.9 1.000 510 0.314 Cob1060 3.63 21.7 0.999 643 0.323 Cob4000 3.16 10.6 0.993 560 0.233 0.248± 4.0 Cob4015 3.72 22.8 0.998 659 0.263 Cob4030 3.98 15.9 0.999 705 0.248
(MB), 1.04× 10−3(BB1), and 6.51× 10−4(mol kg−1)(min)−1 (AB74).Fig. 6(b) shows that the 1/b values of all dye solutes on the ACs of the group with KOH/char ratio equal to 4 are unchanged with respect to the CO2gasification time, with the
average 1/b values of 0.245 (MB), 0.156 (BB1), and 0.078 (mol kg−1)(AB74), higher than those of the ACs of the group with KOH/char ratio equal to 1, meaning higher mass transfer. Besides, increased CO2gasification time has little effect on the
1/b values of all phenols solutes on the ACs of both groups. Here, it was proved that increased CO2gasification time was beneficial
to the mass transfer of larger molecular materials, such as dyes, for the ACs of the group with KOH/char ratio equal to 1. And the ACs of the group with KOH/char ratio equal to 4 all could maintain good mass transfer in the range of the CO2gasification
time studied.
3.4. Equilibrium adsorption
Fig. 7shows the typical equilibrium adsorption of (a) phe-nol, (b) 4-CP, (c) MB, and (d) BB1 at 30◦C on the ACs pre-pared under different CO2gasification time.The correlation of
isotherm data by theoretical or empirical equations is essential to practical operation. The widely used Langmuir equation is given as: Ce qe = 1 KLqmon + 1 qmon Ce (4)
where qmon is the amount of adsorption (in mol kg−1)
corre-sponding to complete monolayer coverage and KLis the
Lang-muir constant. Linear plots of (Ce/qe) against Cegive KL and qmon(not shown). In addition, the parameters (listed inTable 3) estimated from Fig. 7are reliable since the fittings for BB1, MB, 4-CP and phenol adsorption on all ACs in the concentration range of study are excellent (correlation coefficient, r2> 0.984).
Table 3shows that the adsorption capacity (qmon) for all
solu-tions increases with increased Spexcept BB1. For Cob4030, the
values of qmonfor BB1, MB, 4-CP, and phenol are 3.98, 2.98,
6.71 and 4.5 mol kg−1, respectively, which are larger than those obtained earlier in similar solute-adsorbent systems[17,33–35].
Fig. 8(a) shows that in case of the group with KOH/char equal to 1, qmonvalues increase with increased Spexcept phenol.Fig. 8(b)
shows that in the case of the group with KOH/char equal to 4, qmon values slowly increase with increased Spexcept 4-CP.
These indicate that increased CO2 gasification time increases
both Spand qmon.
3.5. Surface coverage
An adsorbate forms the monolayer coverage on the sur-face of AC. Based on the covered mass (qmon) obtained from
the Langmuir equation and the projected area of an adsorbate molecule, adsorbate coverage per unit gram of activated carbon (Sc, m2g−1) can be obtained using the following equation:
Sc=6.023 × 1023× Am× qmon
1000 (5)
Fig. 8. Adsorption capacity (qmon) of phenol, 4-CP, MB, and BB1 at 30◦C on
the activated carbons using KOH combined with CO2gasification.
where Amis the projected area of a molecule (defined in[27]).
In addition, the surface coverage is defined as the ratio of Scto
BET surface area of the AC (Sc/Sp), and the calculated results are listed inTable 3.The Sc/Spvalues of both groups of ACs do not largely fluctuate except those of phenol, and their average values are listed inTable 3. These averages are 0.579, 0.305, and 0.137 for 4-CP, BB1, and MB adsorbed on the group with KOH/char ratio equal to 1, respectively; and are 0.658, 0.248, and 0.153 for 4-CP, BB1, and MB adsorbed on the group with KOH/char ratio equal to 4, respectively. The error is less than 9.2%. This result proves that additional CO2gasification increases the
sur-face area of the activated carbon, while the activated carbon still maintaining the same unit area adsorption capability. This fully proves the value of the process of KOH etching plus CO2
gasification for preparing activated carbon.
In the adsorption described by the Langmuir model the pro-cess can be expressed as:
C∗+ A ↔ C∗(A) (6)
where A represents the adsorbate molecule in the liquid phase,
C*the available adsorptive site, and C*(A) the sites occupied by
those ACs (derived from different activated conditions) would have the same ratio of the sites occupied by adsorbate and the available adsorptive site. In this research, Sc/Spvalues are about the same for the adsorptions of phenols and dyes on ACs acti-vated with the same KOH/char ratio combined with different CO2gasification time. This is in agreement with the theory of
Langmuir model.
4. Conclusions
SEM observations revealed that the corncob-derived acti-vated carbons of the group with KOH/char equal ratio to 1 were highly porous with honeycomb shaped, cottony holes, which were transformed into plain surfaces after being gasified with CO2for 60 min. When the CO2gasification time of the group
with KOH/char ratio equal to 1 was increased from 0 to 60 min, their micropore ratios decreased rapidly from 0.805 to 0.565 and their BET surface areas rapidly increased from 1071 to 1991 m2g−1; while, in the case of the group with KOH/char ratio equal to 4 the micropore ratios decreased slowly from 0.844 to 0.772 and the BET surface area increased slowly from 2402 to 2844 m2g−1. The BET surface area (2844 m2/g) of cob4030 is one of the highest among the activated carbons prepared from the plant materials found so far. Based on their physical char-acteristics (Smicro/Sp, Vmicro/Vpore, Dp,ρb, and yield), “surface
activation” with CO2 gasification was clearly obvious during
the activation process for the preparation of the group with KOH/char equal to 1; while “surface activation” was not obvi-ous and a KOH etching process occurred for the group with KOH/char equal to 4. For the adsorption kinetics, data from the adsorption of solution onto activated carbons were suitably fitted to the Elovich equation. It was proved that ACs with prolonged CO2gasification time was beneficial to mass transfer of larger
molecular materials (such as dyes). For the adsorption equilib-rium, the qmon values of Cob4030 carbon are larger than those
obtained earlier in similar solute-adsorbent systems. Besides, the average surface coverage ratios for phenols and dyes are in agreement with the theory of Langmuir model. It was proved that the activated carbons activated with KOH etching plus CO2
gasification had good adsorption capability.
Acknowledgment
Financial support of this work by the National Science Coun-cil of the Republic of China under contract no. NSC 94-2214-E-239-001 are gratefully acknowledged.
References
[1] D. Lozano-Castello, M.A. Lillo-Rodenas, D. Cazorla-Amoros, A. Linares-Solano, Carbon 39 (2001) 741.
[2] W.T. Tsai, C.Y. Chang, S.Y. Wang, C.F. Chang, S.F. Chien, H.F. Sun, Resour. Conserv. Recycling 32 (2001) 43.
[3] T. Kyotani, Carbon 38 (2000) 269. [4] H. Shi, Electrochim. Acta 41 (1996) 1633.
[5] B.E. Conway, Electrochemical Supercapacitors, Kluwer-Plenum Publish-ing Co., New York, 1999.
[6] L. Bonnefoi, P. Simon, J.F. Fauvarque, C. Sarrazin, J.F. Sarrau, A. Dugast, J. Power Sources 80 (1999) 149.
[7] D. Qu, H. Shi, J. Power Sources 74 (1998) 99. [8] J.H. Park, O.O. Park, J. Power Sources 111 (2002) 185.
[9] J. Sun, M.J. Rostam-Abadi, A.A. Lizzio, Gas. Sep. Purif. 10 (1996) 91. [10] M. Molina-Sabio, M.T. Gonzalez, F. Rodriguez-Reinoso, A.
Sepulveda-Escribano, Carbon 34 (1996) 505.
[11] Z. Hu, M.P. Srinivasan, Microporous Mesoporous Mater. 43 (2001) 267. [12] Z. Hu, H. Guo, M.P. Srinivasan, N. Yaming, Sep. Purif. Technol. 31
(2003) 47.
[13] A.M. El-Hendawy A-N, S.E. Samra, B.S. Girgis, Colloids Surf. 180 (2001) 209.
[14] W.T. Tsai, C.Y. Chang, S.L. Lee, Bioresour. Technol. 64 (1998) 211. [15] K. Jurewicz, K. Babel, A. Ziolkowski, H.J. Wachowska, Phys. Chem.
Soilds 65 (2004) 269.
[16] R.L. Tseng, F.C. Wu, R.S. Juang, Carbon 41 (2003) 487.
[17] F.C. Wu, R.L. Tseng, R.S. Juang, Environ. Technol. 22 (2001) 205. [18] R.S. Juang, F.C. Wu, R.L. Tseng, Colloids Surf. 201 (2002) 191. [19] F.C. Wu, R.L. Tseng, R.S. Juang, J. Hazard. Mater. 69 (1999) 287. [20] F.C. Wu, R.L. Tseng, R.S. Juang, J. Colloid Interface Sci. 283 (2005)
49.
[21] F.C. Wu, R.L. Tseng, C.C. Hu, C.C. Wang, J. Power Sources 138 (2004) 351.
[22] R.L. Tseng, S.K. Tseng, J. Colloid Interface Sci. 287 (2005) 428. [23] E.P. Barrett, L.G. Joyner, P.P. Halenda, J. Am. Chem. Soc. 73 (1951)
373.
[24] J.H. de Boer, B.C. Lippens, B.G. Linsen, J.C.P. Broekhoff, A. van den Heuvel, T.J. Osiga, J. Colloid Interface Sci. 21 (1966) 405.
[25] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, Pure Appl. Chem. 57 (1985) 603.
[26] E.F. Sousa-Aguilar, A. Liebsch, B.C. Chaves, A.F. Costa, Microporous Mesoporous Mater. 25 (1998) 185.
[27] F.C. Wu, R.L. Tseng, C.C. Hu, Microporous Mesoporous Mater. 80 (2005) 95.
[28] M.S. E1-Geundi, Adsorp. Sci. Technol. 15 (1997) 777.
[29] D.M. Ruthven, Principles of Adsorption and Desorption Processes, Wiley, New York, 1984.
[30] C. Moreno-Castilla, F. Carrasco-Marin, M.V. Lopez-Ramon, M.A. Alvarez-Merino, Carbon 39 (2001) 1415.
[31] A. Ahmadpour, D.D. Do, Carbon 35 (1997) 1723.
[32] Z. Hu, M.P. Srinivasan, Microporous Mesoporous Mater. 27 (1999) 11. [33] C.T. Hsieh, H. Teng, Carbon 38 (2000) 863.
[34] F.C. Wu, R.L. Tseng, R.S. Juang, J. Environ. Sci. Health A34 (1999) 1753.
