Bioresource Technology 98 (2007) 145–149

Preparation and characterization of activated carbon from palm shell

by chemical activation with K2CO3
Donni Adinata, Wan Mohd Ashri Wan Daud *, Mohd Kheireddine Aroua
Department of Chemical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

Received 28 June 2005; received in revised form 30 October 2005; accepted 5 November 2005
Available online 27 December 2005

Abstract

Palm shell was used to prepare activated carbon using potassium carbonate (K2CO3) as activating agent. The influence of carboniza-
tion temperatures (600–1000 C) and impregnation ratios (0.5–2.0) of the prepared activated carbon on the pore development and yield
were investigated. Results showed that in all cases, increasing the carbonization temperature and impregnation ratio, the yield decreased,
while the adsorption of CO2 increased, progressively. Specific surface area of activated carbon was maximum about 1170 m2/g at 800 C
with activation duration of 2 h and at an impregnation ratio of 1.0.
 2005 Elsevier Ltd. All rights reserved.

Keywords: Activated carbon; Palm shell; Chemical activation; Pore development

1. Introduction row micropore distribution. Although it is a frequently

used to prepare activated carbons, the general mechanism
Activated carbon is a well known as porous material, of chemical activation is not well understood, and the
with large specific surface area, which is useful in adsorp- various interpretations found in literature underline the
tion of both gases and solutes from aqueous solution. process complexity. In general terms, chemical activation
Therefore, it has been widely used for separation of gases, by alkalis consists in solid–solid or solid–liquid reaction
recovery of solvents, removal of organic pollutants from involving the hydroxide reduction and carbon oxidation
drinking water and a catalyst support. As environmental to generate porosity. McKee (1983) studied the gasification
pollution is becoming a more serious problem, the need of graphite by alkali metal compounds and found that
for activated carbon is growing. It is a versatile adsorbent K2CO3 was reduced by carbon in an inert atmosphere
because of its good adsorption properties. Various materi- (Hayashi et al., 2002b; Okada et al., 2003).
als are used to produce activated carbon and some of the Other chemicals, which frequently used are ZnCl2,
most commonly used are agriculture wastes such as coco- H3PO4, ZnCl2, KOH and NaOH (Guo and Lua, 2003;
nut shell (Kirubakaran et al., 1991), pistachio shell (Abe Lua and Yang, 2004; Raymundo-Pinero et al., 2005). How-
et al., 1990), saw dust (Xiongzun et al., 1986), walnut shell ever, alkali hydroxides such as KOH and NaOH are haz-
(Khan et al., 1985), tropical wood (Maniatis and Nurmala, ardous, expensive and corrosive (Lillo-Rodenas et al.,
1992) and almond shell (Hayashi et al., 2000). 2004) and ZnCl2 is unfriendly the environment and create
Chemical activation has been shown as a efficient waste disposal problem (Guo and Lua, 2002). Thus, a more
method to obtain carbons with high surface area and nar- benign chemical is desired; K2CO3 is not a hazardous
chemical and not deleterious as it is frequently used for
food additives.
*
Corresponding author. Tel.: +60 379676897; fax: +60 379675371. Palm shell (also known as endocarp) is a cheap and
E-mail address: [email protected] (W.M.A. Wan Daud). abundant agricultural by-product in tropical countries like

0960-8524/$ - see front matter  2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2005.11.006
146 D. Adinata et al. / Bioresource Technology 98 (2007) 145–149

Malaysia and Indonesia. Palm shell has been successfully A typical carbonization run began by changing 100 g of
converted into a well-developed activated carbon by ther- impregnated sample in the reactor and heated up to the
mal activation (physical activation) and chemical activa- carbonization temperature in flowing stream of nitrogen
tion using with carbon dioxide (CO2) and H2PO3 (Guo (15 l/min) (Wan Daud and Wan Ali, 2004). The tempera-
and Lua, 2000, 2002). However, activated carbon prepared ture of reactor was increased at the rate of 10 C/min, until
from palm shell using K2CO3 activation for carbonaceous it reached the final carbonization temperature. The carbon-
precursors has not been thoroughly investigated. The aim ization temperature was varied from 600 to 1000 C with
of this work was to use K2CO3 as activating agent and activation duration of 2 h. After carbonization, the sample
study the influences of carbonization temperatures and was cooled down under nitrogen (N2) flow and was washed
impregnation ratios on pore development and yield. sequentially several times with hot water, and finally with
cold distilled water to remove residual chemicals. Then
2. Methods the sample was dried at 110 C.

2.1. Material and chemical reagent 3. Results and discussion

Palm shell obtained from Malaysia oil palm shell 3.1. Yield of activated carbon and chemical recovery
(MOPS) were dried, crushed and sieved to a particle size ratio of K2CO3
fraction of 1–2 mm. Potassium carbonate (K2CO3) (purity
99.9%; Fisher Scientific) was dissolved in distilled water to The proximate analyses of palm shell were as follows
prepare a saturated solution. (dry wt basis %): carbon 18.7, moisture 7.96, ash 1.1, vol-
atile 72.47, C 50.01, H 6.9, N 1.9, S 0, O 41, cellulose 29,
2.2. Raw material analysis halocellulose 47.7 and lignin 53.4 more than 100%. The
yield of activated carbon decreased as the activation
An elemental analysis was carried out using a Perkin temperature increased (Table 1). When chemical activation
Elmer CHNO/S Analyzer 2400. Lignin, cellulose and halo- with K2CO3 impregnation was used, increasing the carbon-
cellulose were determined using TAPPI method (T-13wd- ization temperature decreased the yield progressively due
74, T-17wd-70 and T-9m-54, respectively). to be release of volatile products as a result of intensifying
dehydration and elimination reaction; it also indicated that
2.3. Characterization of activated carbon char of the palm shell was gasified by K2CO3. McKee
(1983) studied the gasification of graphite by alkali metal
The characterization of activated carbon samples was and found that K2CO3 was reduced in an inert atmosphere
carried out using CO2 adsorption at 273 K using a by carbon (Hayashi et al., 2002b; Okada et al., 2003).
Micromeritics ASAP 2010 surface area analyzer. The sam- The presence of K2CO3 in the interior of the precursor
ple was placed inside the tube and the glass bulb was restricted the formation of tar as well as other liquids such
inserted inside the tube and stopper. The bulb was inserted as acetic acid and methanol by formation of cross-links, and
by slanting the sample tube almost to a horizontal position inhibited the shrinkage of the precursor particle by occupy-
and the bulb was allowed to gradually down the sample ing certain substantial volumes (Guo and Lua, 2003). The
tube. Before the experiment began, the adsorbents were results showed that the chemical recovery ratio of K2CO3
degassed (104 mmHg) at 393 K. The surface areas of the with selected conditions above was 0.99, 0.96, 0.86, 0.65
samples were measured based on Brunauer–Emmet–Teller and 0.25 for carbonization temperature of 600, 700, 800,
(BET) method and Dubinin–Radushkevich (DR) method 900 and 1000 C, respectively. This results supported previ-
was applied to calculate the micropore volume. Pore size ous findings that K2CO3 was reduced by carbon in inert
distribution (PSD) was obtained from Horvath–Kawazoe atmospheres, whereby char reacted with K2CO3 and then
(HK) analysis and solid density of activated carbon was the later was removed during gasification. Above 800 C,
measured by helium displacement with an ultrapycnometer carbon reduced the impregnated K2CO3 and was consumed
(AccuPyc 1330 pycnometer). through the formation of CO. Thus, the specific surface
area and the pore volume increased (Hayashi et al., 2002b).
2.4. Preparation of activated carbon As shown in Table 1, the yield was strongly affected by
the impregnation ratio. Increasing the impregnation ratio
Palm shells were mixed with saturated solution K2CO3 decreased yield and increased ‘‘burn-off’’ of palm shell
and kneaded. This mixture was then dried in an oven at (Gomez-Serrano et al., 2005). Higher K2CO3 concentra-
110 C for 24 h to prepare the impregnated sample. In this tions consistently yielded products with a much larger
work, 0.5, 0.75, 1.0, 1.5 and 2.0 impregnation ratios were external surface area, as well as large micropore volumes
used. The impregnation ratio is given by (Evans et al., 1999). The chemical recovery ratio of
K2CO3 was 0.15, 0.37, 0.86, 0.89, and 0.98 for the impreg-
ðweight of K2 CO3 in solutionÞ nation ratio of 0.5, 0.75, 1.0, 1.5, and 2.0, respectively,
impregnation ratio ¼
ðweight of palm shellÞ which showed high chemical recovery ratio of products
 D. Adinata et al. / Bioresource Technology 98 (2007) 145–149 147

Table 1
The pore characteristics and properties of activated carbons prepared from palm shell at different carbonization temperature and impregnation ratio
Activation conditionsa Yield (%) Variation (%) Solid density (g/cm3) Variation (%) BET (m2/g) Variation (%)
600–2–1 26.34 1.6 1.7659 0.4 319 1.3
700–2–1 22.24 1.4 2.0872 0.5 425 0.9
800–2–1 18.86 1.1 2.3567 0.1 1170 1.4
900–2–1 16.79 1.0 2.5341 1.6 544 1.9
1000–2–1 12.82 1.9 2.7341 1.1 339 1.0
800–2–0.5 27.84 2.2 2.7159 1.0 248 2.0
800–2–0.75 25.24 0.7 2.4561 1.2 476 1.3
800–2–1.5 14.79 1.8 1.9322 1.1 540 0.8
800–2–2.0 11.85 0.6 1.7113 0.5 332 1.1
a
Note: a–b–c denote carbonization temperature (C)–during activation (h)–impregnation ratio of K2CO3.

subjected to high impregnation ratio. Hot water washing Adsorption capacity of activated carbon largely depends
was, thus, very effective in reducing K2CO3, which has been on the amount of micropores that are present in the acti-
found to be related to formation of atomic K. Thus, atomic vated carbon. Hence, its measurement can be directly
K may intercalate and expand the inter layers of adjacent related to quantity of micropore volume in the sample.
hexagonal network planes consisting of C atoms, enhanc- The micropore volume was estimated from the adsorption
ing pore formation though hexagonal planes that are not of carbon dioxide at 273 K. The results showed that micro-
well developed as in graphite. K2CO3 was not detected pore volume of activated carbon prepared with selected
on the surface, indicating that the washing of this sample conditions above was 0.23, 0.50, 0.57, 0.48, and 0.15 cm3/
was effective to remove K from the surface on the products g for the carbonization temperature of 600, 700, 800, 900,
(Okada et al., 2003). and 1000 C with activation duration of 2 h at an impreg-
nation ratio of 1.0, respectively. The micropore volume
3.2. Pore development and characteristic of increased with carbonization temperature from 600 to
activated carbon 800 C and then between 800 and 1000 C, the micropore
volume of activated carbon slightly decreased, because of
The results showed that the specific surface areas enlargement surface of the micropores to mesopores. When
increased with the increase in carbonization temperature impregnation ratio was 0.5, 0.75, 1.0, 1.5, and 2.0, micro-
from 600 to 800 C and decreased slightly at 800– pore volume of activated carbon was 0.13, 0.28, 0.57,
1000 C. The maximum specific area obtained was 0.43, and 0.21 cm3/g, respectively. It showed that increas-
1170 m2/g. Therefore, it was deduced that K2CO3 was ing the impregnation ratio from 0.5 to 1.0 progressively
effective as activation reagent below 800 C. This progres- increased both the micropore volume. There was a pro-
sive carbonization temperature increased the C–K2CO3 nounced decreased in micropore volume when the ratio
reaction rate, resulting in increasing carbon ‘‘burn-off’’. was higher than 1.0 because rapid increase in mesopore
Concurrently, the volatiles from the samples continued volume increased and the total pore volume (micropore
to evolve with increasing carbonization temperature. The volume and mesopore volume) increased with increasing
develotilization process further developed the rudimentary impregnation ratio.
pore structure in the char, whereas the C–K2CO3 reaction The results obtained showed that increasing the impreg-
enhanced the existing pores and created new porosities nation ratio decreased the density of activated carbon
(Hayashi et al., 2002a). (Table 1). Higher impregnation ratio also tends to produce
The surface area of activated carbon increases with lower yield and density of activated carbon. When higher
increasing impregnation ratio from 0.5 to 1.0 and impregnation ratio was used, the weight losses were due
decreased slightly for ratios larger than 1. However, mes- to the increasing release of volatile products as result of
oporous surface area increased continually. The micropore intensifying dehydration and elimination reactions (Evans
surface areas were obtained by subtracting mesopore et al., 1999). By increasing the carbonization temperature
surface area from the corresponding BET surface area. the solid density increased progressively due to the release
Under the same condition when the impregnation ratio of low-molecular weight products (water, furan derivates
was 0.5, 0.75, 1.0, 1.5, and 2.0, the corresponding specific and laevoglucose) due to of dehydration and elimination
surface area of activated carbon was 248, 476, 1170, 540, reactions (Guo and Lua, 2002; Lua and Yang, 2004).
and 332 m2/g, respectively. There was a pronounced
decrease in micropore surface area when the ratio was 3.3. CO2 adsorption isotherm and pore size distribution of
higher than 1 due to enlargement of micropores to meso- activated carbon
pores. These results suggested that to obtain the optimal
value of surface area, the impregnation ratio should be Figs. 1 and 2 show the CO2 adsorption isotherms on
around 1.0. the prepared activated carbons at different carbonization
148 D. Adinata et al. / Bioresource Technology 98 (2007) 145–149

180 12
Volume Adsorbed (cm3/g) STP

Average variation of each line is < 0.2% Average variation of each line is < 0.2%
160
140 10

Pore Volume (cm3/g-nm)

120
100 8
80
60 6
40
20 4
0
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
2
Relative Pressure (P/P0)

Fig. 1. CO2 adsorption isotherm on the prepared activated carbon at 0

0.4 0.5 0.6 0.7 0.8 0.9 1
different carbonization temperatures: (m) 600 C, (j) 700 C, (d) 800 C,
Pore Diameter (nm)
(·) 900 C, (s) 1000 C.
Fig. 3. The influence of carbonization temperatures on the pore size
distribution of activated carbon: (·) 600 C (h) 700 C (d) 800 C, (D)
120 900 C, (s) 1000 C.
Average variation of each line is < 0.2%
Volume Adsorbed (cm3/g) STP

100

80 12 Average variation of each line is < 0.2%

PoreVolume (cm3/g-nm) 10
60
8
40
6

20 4

0 2
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
Relative Pressure (P/P0) 0
0.4 0.5 0.6 0.7 0.8 0.9 1
Fig. 2. Adsorption isotherm of CO2 on the prepared activated carbons at Pore Diameter (nm)
different impregnation ratios: (j) 0.5, (m) 0.75, (d) 1, (+) 1.5, (s) 2. Fig. 4. The influence of impregnation ratios on the pore size distribution
of activated carbon: (s) 0.5, (·) 0.75, (d) 1, (j) 1.5, (D) 2.

temperatures and impregnation ratios. The amount of CO2

adsorbed at P/P0 = 0.03 increased for activated carbons increased the total pore volume of activated carbon (Isma-
prepared at relatively higher carbonization temperature dji and Bhatia, 2000, 2001).
and impregnation ratio. The isotherm curves are of Type
I isotherms, which generally exhibit the presence of micro- 4. Conclusions
pores. Further increase in carbonization temperature and
impregnation ratio through activation process widened The activated carbon prepared by chemical activation
the pores as well as increased the volume of micro pore with K2CO3 attained a maximum value of 1170 m2/g at a
and total of pore volume. carbonization temperature of 800 C with activation
The structural heterogeneity of porous material is gener- duration of 2 h and at an impregnation ratio of 1.0. The
ally characterized in term of the pore size distribution. The increase in the carbonization temperature and impregna-
pore size distribution is closely related to both kinetic and tion ratio decreased yield and increased the product
equilibrium properties of porous materials used in indus- adsorption capacity of CO2. From the results of yield of
trial application. The pore size distribution of activated activated carbon and reagent recovery ratio, it was con-
carbon produced from palm shell at different carbonization cluded that the carbon involved in the palm shell char
temperatures and impregnation ratios are given in Figs. 3 was removed as CO by reduction of K2CO3 above
and 4. From these figures, it was obvious that the carbon- 800 C. It was also found that for 2 h activation, the
ization temperature and impregnation ratio had significant mesopore volume increased with the increase in impregna-
effect on the pore structure of activated carbon produced. tion ratio and carbonization temperature.
At low temperature and impregnation ratio, the pore struc-
ture of mainly consisted of micropore; however, with the Acknowledgement
increase of carbonization temperature and impregnation
ratio, the creation of micropore structure and widening Financial support for this work was provided by IRPA
of micropores to mesopores also increased and also Grant (08-02-03-0233-EA233), University Malaya.
 D. Adinata et al. / Bioresource Technology 98 (2007) 145–149 149

References Ismadji, S., Bhatia, S.K., 2001. A modified pore filling isotherm for
liquid phase adsorption in activated carbon. Langmuir 17,
Abe, I., Tatsumoto, H., Ikuta, N., Kawafune, I., 1990. Preparation of 1488–1498.
activated carbon from pisthachio nut shell. Chem. Exp. 5, 177–180. Khan, A., Singh, H., Bhatia, A.K., 1985. Activated carbon from walnut
Evans, M.J.B., Halliop, E., MacDonald, J.A.F., 1999. The production of shells. Res. Ind. 30, 13–16.
chemically-activated carbon. Carbon 37, 269–274. Kirubakaran, C.J., Krishnaiah, Seshadri, S.K., 1991. Experimental study
Gomez-Serrano, V., Cuerda-Correa, E.M., Fernandez-Gonzalez, M.C., of the production of activated carbon from coconut shells in fluidized
Alexandre-Franco, M.F., Macias-Garcia, 2005. Preparation of acti- bed reactor. Ind. Eng. Chem. Res 27, 2411–2416.
vated carbons from chestnut wood by phosphoric acid-chemical Lillo-Rodenas, M.A., Juan-Juan, J., Cazorla-Amoros, D., Linares-
activation. Study of microporosity and fractal dimension. Mater. Lett. Solano, A., 2004. About reactions occurring during chemical activa-
59, 846–853. tion with hydroxides. Carbon 42, 1371–1375.
Guo, J., Lua, A.C., 2000. Characterization of adsorbent prepared from oil Lua, A.C., Yang, T., 2004. Effect of activation temperature on the textural
palm shell by CO2 activation for removal of gaseous pollutants. Mater. and chemical properties of potassium hydroxide activated carbon
Lett. 55, 334–339. prepared from pistachio-nut shell. J. Colloid Interface Sci. 274, 594–
Guo, J., Lua, A.C., 2002. Textural and chemical characterization of 601.
adsorbent prepared from palm shell by potassium hydroxide impreg- Maniatis, K., Nurmala, M., 1992. Activated carbon production from
nation at different stages. J. Colloid Interface Sci. 254, 227–233. biomass. Biomass Energy Ind. Environ. 274, 1034–1308.
Guo, J., Lua, A.C., 2003. Textural and chemical properties of adsorbent McKee, D.W., 1983. Mechanisms of the alkali metal catalyzed gasification
prepared from palm shell by phosphoric acid activation. Mater. Chem. of carbon. Fuel 62, 170–175.
Phys. 80, 114–119. Okada, K., Yamamoto, N., Kameshima, Y., Yasumori, A., 2003. Porous
Hayashi, J., Kazehaya, A., Muroyama, K., Watkinson, A.P., 2000. properties of activated carbons from waste newspaper prepared by
Preparation of activated carbon from lignin by chemical activation. chemical and physical activation. J. Colloid Interface Sci. 262, 179–
Carbon 32, 1873–1878. 193.
Hayashi, J., Horikawa, T., Muroyama, K., Gomes, V.G., 2002a. Raymundo-Pinero, E., Azais, P., Cacciaguerra, T., Cazorla-Amoros, D.,
Activated carbon from chickpea husk by chemical activation with Linares-Solano, A., Beguin, F., 2005. KOH and NaOH activation
K2CO3: preparation and characterization. Micropor. Mesopor. mechanisms of multiwalled carbon nanotubes with different structural
Mater. 55, 63–68. organization. Carbon 43, 786–795.
Hayashi, J., Horikawa, T., Takeda, I., Muroyama, K., Ani, F.N., 2002b. Wan Daud, W.M.A., Wan Ali, W.S., 2004. Comparison on pore
Preparing activated carbon from various nutshells by chemical development of activated carbon produced from palm shell and
activation with K2CO3. Carbon 40, 2381–2386. coconut shell. Bioresour. Technol. 93, 63–69.
Ismadji, S., Bhatia, S.K., 2000. Investigation of network connectivity in Xiongzun, Z., Famnao, Z., Lie, L., Qingrong, L., 1986. A new technique
activated carbons by liquid phase adsorption. Langmuir 16, 9303– to produce activated carbon (from saw dust of any humidity) by zinc
9313. chloride method. J. Nanjing Inst. Forest. 1, 19–30.