ARTICLE IN PRESS

Bioresource Technology xxx (2005) xxxxxx

Adsorption characteristics of malachite green on activated carbon derived from rice husks produced by chemicalthermal process
I.A. Rahman *, B. Saad, S. Shaidan, E.S. Sya Rizal
School of Chemical Sciences, Universiti Sains Malaysia, 11800 Penang, Malaysia Received 20 November 2003; received in revised form 13 December 2004; accepted 26 December 2004

Abstract Phosphoric acid (H3PO4) and sodium hydroxide (NaOH) treated rice husks, followed by carbonization in a owing nitrogen were used to study the adsorption of malachite green (MG) in aqueous solution. The eect of adsorption on contact time, concentration of MG and adsorbent dosage of the samples treated or carbonized at dierent temperatures were investigated. The results reveal that the optimum carbonization temperature is 500 C in order to obtain adsorption capacity that is comparable to the commercial activated carbon for the husks treated by H3PO4. It is interesting to note that MG adsorbed preferably on carbon-rich than on silica rich-sites. It is found that the behaviour of H3PO4 treated absorbent followed both the Langmuir and Freundlich models while NaOH treated best tted to only the Langmuir model. 2005 Elsevier Ltd. All rights reserved.
Keywords: Adsorption; Malachite green; Activated rice husks

1. Introduction Many countries have given prominence to the development of aquaculture as a long term strategy in providing sucient source of protein to their ever growing population. The raising of large numbers of sh, prawns and others in conned space as in modern aquaculture practice necessitates the use of an extensive range of chemicals for the prevention and treatment of diseases, thus posing as a source of water pollution. The used of malachite green (triphenylmethane) in aquaculture industry has a long history, dating back to 1933, when it was rst introduced. The compound, originally extensively used for the dyeing of textiles, has since found wide applications in the aquaculture industry as it is relatively cheap, eective and easy to obtain. Its use in the aquaculture practice in many countries, including
*

Corresponding author. Fax: +60 4 6574854. E-mail address: arismail@usm.my (I.A. Rahman).

Malaysia has not been regulated. Many adverse eects from the consumption of the dye due to its carcinogenic, genotoxic, mutagenic and teratogenic properties in animal studies have been reported (Sandra et al., 1999). Adsorption on activated carbon has long been recognized to be one of the most eective methods for the removal of pesticides and organic compounds from aqueous solutions. Agriculture wastes such as oil palm nut shell (Chan et al., 1980), rice husks (Rahman et al., 2000), olive-waste cakes (Bcaoui et al., 2001), acorns and olive seeds (La, 2001), corncobs (ElHendawy et al., 2001), coconut shells or palm seeds (Hu and Srinivasan, 2001), and guava seeds (Rahman and Saad, 2003) have been used as low-cost adsorbents. These materials were activated either by chemicals, steam, gas or their combinations. The use of adsorbents derived from rice husk to adsorb some phenolic compounds (Mbui et al., 2002) and the herbicide paraquat (Rahman et al., in press) had been reported. Relatively little work had been reported on the absorption of

0960-8524/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2004.12.015

ARTICLE IN PRESS
2 I.A. Rahman et al. / Bioresource Technology xxx (2005) xxxxxx

malachite green from aqueous solution despite its widespread use. The adsorption behaviour of malachite green, formalin, chloramine-T and oxytetracycline onto coal-based activated carbon was studied by Aitcheson et al. (2000). The removal of malachite green by agriculture wastes was only reported by Guo et al. (2003) and Garg et al. (2003). In the present work, we report our evaluation on the feasibility of using rice husks pretreated with H3PO4 and NaOH for the removal of malachite green from aqueous solution. The inuence of adsorbent loading and contact time were also studied.

ball-milled for 1 h into powder that passed through 75 lm mesh. The samples were referred to as PCP10 and PCP20, corresponding to the concentrations of acid used for their preparations. 2.3.2. Sodium hydroxide treated-carbonization process (NCP) About 10 g of dried rice husks were treated with 100 mL 10% sodium hydroxide at 30 and 100 C for 1 h in a paran bath. The sodium hydroxide used was an equivalent amount required to remove all silica present in the husks. The husks were then ltered and washed with distilled water until the ltrate was neutral. The treated husks were dried at 85 C in an oven and left overnight. The dried husk were ball-milled for 1 h into powder and sieved through 75 lm mesh after carbonization at 500 C for 30 min. The samples were coded as NCP30C and NCP100C corresponding to the appropriate treatment temperature. 2.4. Determination of silica Silica and carbonaceous material content of the sample was determined by heating at 700 C in an atmospheric furnace to constant mass. The residue and the mass losses were considered as silica and carbonaceous materials respectively. 2.5. Adsorption studies Malachite green (MG) of oxalate salt (AR grade, BDH) was used as adsorbate. An accurately weighed quantity of the salt was dissolved in distilled water to prepare 500 mg/L stock solution. A desired concentration for adsorption experiment was obtained by successive dilution. The calibration curve for adsorbate is linear from 1 to 20 mg/L (r2 = 0.9993) and the equation of the straight line is given by y = 0.1728x. One hundred millilitre of adsorbate was mixed with a quantity of adsorbent in a 250 mL ask and shaken using a mechanical shaker. It was next ltered through a lter paper, and the solution was analyzed spectrophotometerically at kmax = 617 nm at pH 56.

2. Methods 2.1. Raw material Dried and fresh rice husks were used as raw materials to produce the adsorbents. The rice husks were obtained from a rice mill in the Northern part of Peninsular Malaysia. Typical composition of the rice husks used is 16% silica, 14% lignin, 36% cellulose, 22% hemicelluloses, 3% extractive materials and 9% moisture (Rahman et al., 1997). Chemicals used were sodium hydroxide (99%, Merck) and phosphoric acid (85%, Baker). Aqueous solutions of the reagents with appropriate concentrations were prepared using distilled water. Commercial activated carbon was obtained from Sigma. 2.2. Facilities Weighing of samples were performed by using an analytical balance with precision 0.0001 g. Drying of samples was carried out in an electric oven equipped with fan (Heraeus Instrument D-6450). Carbonization was carried out in a Carbolite Furnace (CSF 1100) with facilities for nitrogen ow. The samples were milled into powder by using a ball-mill (Stoneware) with an alumina milling media. Quantication of malachite green for absorption studies was conducted using a UVVIS spectrophotometer (Hitachi, Model U2000). 2.3. Adsorbents 2.3.1. Phosphoric acid treated-carbonization process (PCP) The dried husks were impregnated overnight in phosphoric acid with concentrations of 10% and 20% in the mass ratio of 1:10. The husks were ltered and dried at 85 C in an oven overnight and subsequently carbonized under nitrogen atmosphere at temperatures ranging from 400 to 650 C, for 30 min. The carbonized husks were washed with distilled water until the ltrate was about pH 67. The samples were then dried at 85 C in an oven overnight and

3. Results and discussion 3.1. Characteristic of adsorbents Based on experimental observation, the reaction of phosphoric acid and organic constituents such as lignin was exothermic. Physical changes on rice husks were only observed when the husks were treated with 20% H3PO4. The husks become slightly darken and sticky and become more signicant on further increase in acid concentrations. Based on the above reasons and cost

ARTICLE IN PRESS
I.A. Rahman et al. / Bioresource Technology xxx (2005) xxxxxx 3

Fraction MG adsorbed (fMG)

consideration, we choose only 10% and 20% acid for the activation process. The reaction of NaOH on rice husks were quite dierent to that of H3PO4. The husks remained intact when treated with 10% NaOH even at 100 C. However, the NaOH was able to remove most of the silica, leaving behind a porous structure (Rahman et al., in press). When both samples were carbonized at 400 C, mass loss remained constant after 30 min. Thus, it was assumed that the optimum time for carbonization was 30 min in order to eliminate most of the volatile organic constituents, leaving behind a carbonaceous materials which was considered as activated carbon. The yields ranged from 59% to 40%, depending on the treatment and carbonization temperature (Table 1). 3.2. Adsorption behaviour of adsorbents The adsorption behaviour of the adsorbents was studied by evaluating the fraction of MG adsorbed, fMG, calculated as fMG C i C f =C i where Ci is the initial concentration of MG solution in contact with a certain quantity of absorbent, and Cf is the MG concentration after adsorption at a certain contact time. 3.2.1. Adsorption behaviour of PCP adsorbent A series of adsorption studies on the eect of contact time, concentration and adsorbent doses was conducted on PCP samples. The results on adsorption behaviour of PCP10 and PCP20 samples, at dierent carbonization temperatures as compared to commercial activated carbon (CA) are shown in Figs. 13. Overall, PCP20 samples reach maximum contact time faster than PCP10 at all carbonization temperatures. The order of increasing contact times for both PCP10 and PCP20 can be

1.1 Fraction MG adsorbed (fMG) 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0 10 20 30 40 50 60 70
400C 500C 650C CA PCP10

(a)
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0 10 20

Contact time (min)

PCP20

400C 500C 650C CA

30

40

50

60

70

(b)

Contact time (min)

Fig. 1. Adsorption of MG on PCP as a function of contact time, (a) PCP10 and (b) PCP20. PCP = 0.1 g/100 mL, MG = 20 mg/L.

Fraction MG adsorbed (fMG)

1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0 20 40 60 80 100 120 Concentration (mg/L) 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0 20 40 60 80 100 120 Concentration (mg/L)
400C 500C 650C CA PCP20 400C 500C 650C CA PCP10

Table 1 Yields and composition of PCP and NCP rice husks based absorbents Sample carbonized/ treated temperature PCP10 400 C 500 C 650 C PCP20 400 C 500 C 650 C NCP 30 C 100 C Raw husks (no carbonization) Yield (%) 51.5 46.6 42.1 53.4 47.3 47.4 59.2 40.0 Silica (%) 40.4 47.1 68.7 46.5 48.3 75.2 21.2 7.6 18 Carbonaceous material (%) 59.6 52.9 31.3 53.5 51.7 24.8 78.8 92.4 82

(a)
Fraction MG adsorbed (fMG)

(b)

Fig. 2. Inuence of concentration on adsorption of MG on (a) PCP10 and (b) PCP20. PCP = 0.1 g/100 mL, shaking time = 30 min.

ARTICLE IN PRESS
4
1.2
PCP10

I.A. Rahman et al. / Bioresource Technology xxx (2005) xxxxxx

Fraction MG adsorbed (fMG)

1.0 0.8 0.6 0.4 0.2 0.0 0.0 0.2 0.4 0.6 Adsorbent (g) 1.2
PCP20 400C 500C 650C CA

bonized samples were blocked by decomposition products of organic constituents, thus inhabiting the accessibility of active sites for the adsorption (Rahman and Saad, 2003). It is clearly shown that 500 C is the optimum temperature to produce activated carbon from rice husks through H3PO4 treatment, and the resulting carbon showed comparable adsorption capacity to the commercial activated carbon. 3.2.2. Adsorption behaviour of NCP adsorbent The fact that the adsorption of MG on PCP samples is preferable on the carbon site prompted further investigation. Rice husks were treated with 10% NaOH solution at 30 C and 100 C in order to remove silica, followed by carbonization at 500 C. The samples was labeled as NCP30C and NCP100C, and contained 21% and 8% silica respectively for both treatment temperatures. Fig. 4 represents the adsorption capacity of MG on NCP samples at dierent contact times as compared to CA and raw husks that were not carbonized. It can be readily seen that fMG increases with increasing treatment temperature, although at a slower rate when compared to CA. The raw husks failed to adsorb MG to a satisfactory level due to the absence of large pores to accommodate the bulky organic group of MG. Similar trends were observed on the fMG verses concentrations and adsorbent doses (Figs. 5 and 6). The
1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0 20 40 Contact time (min) 60 80
Raw husks NCP 30C NCP 100C CA

0.8

1.0

1.2

(a)
Fraction MG adsorbed (fMG)

1.0 0.8 0.6 0.4 0.2 0.0 0.0 0.2 0.4 0.6 Adsorbent (g) 0.8 1.0 1.2
400C 500C 650C CA

(b)

Fig. 3. Eect of adsorbent dose on the adsorption of MG for (a) PCP10 and (b) PCP20 samples. MG = 100 mg/L, shaking time = 30 min.

written as 500 C > 400 C > 650 C. Only samples carbonized at 500 C reached contact time that is comparable to the commercial carbon (CA) (Fig. 1). Fig. 2 shows that the fMG is 100% for CA up to 100 mg/L. The fMG decreases with increasing carbonization temperature for both PCP10 and PCP20. Again, the PCP10 and PCP20 that was carbonized at 500 C showed an improvement in the overall uptake capacity up to 40 mg/L and 60 mg/L respectively. The eect of adsorbent doses on the fMG is shown in Fig. 3. The doses increase as carbonized temperature decreases and H3PO4 concentration increases. Only three samples were able to remove 100% of 100 mg/L MG in 100 mL solution, i.e., PCP10 (500 C) = 0.8 g, PCP20 (500 C) = 0.6 g, and PCP20 (400 C) = 1.0 g, and CA required 0.5 g. The increase in adsorption capacity for the PCP10 and PCP20 carbonized at 500 C is most likely related to the maximum surface area created and the availability of more adsorption sites for the bulky organic group. Meanwhile the decreases in adsorption in the above samples on increasing carbonization temperatures is closely attributed to the increase in silica content (Table 1). The eect is more signicant when the silica content reaches >50% in the samples. The results indicate that MG adsorption is more pronounced on the carbon surface rather than on silica surface. The decreased adsorption of PCP10 and PCP20 carbonized at 400 C was believed to be due to the fact that some pores in the car-

Fraction MG adsorbed (fMG)

Fraction MG adsorbed (fMG)

Fig. 4. Adsorption of MG on NCP as a function of contact time.

1.2 1.0 0.8 0.6

Raw husks

0.4 0.2 0.0 0

NCP 30C NCP 100C CA

20

40

60

80

100

120

Concentration (mg/L)

Fig. 5. Inuence of concentration on adsorption of MG on NCP.

ARTICLE IN PRESS
I.A. Rahman et al. / Bioresource Technology xxx (2005) xxxxxx
1.2 1.0 0.8 0.6 0.4 0.2 0.0 0.0
Raw husks NCP 30C NCP100C CA

Fraction MG adsorbed (fMG)

treatment of wastewater from activities such as aquaculture that had been polluted with malachite green. The close proximity and abundance of source of rice husks to aquaculture sites, especially in tropical countries merit serious consideration for its implementation.

4. Conclusion The studies show that activated carbon produced from rice husks has good adsorption capacity that is comparable to commercial activated carbon. It was found that the maximum carbonization temperature is 500 C for eective adsorption. It is believed that the malachite green is preferably adsorbed on carbon surface as the presence of silica to a certain extent, can retard the adsorption capacity. Thus, rice husks based material can provide attractive alternative adsorbents to remove malachite green in wastewater from textile industries and aquaculture activities.

0.2

0.4

0.6 0.8 Absorbent (g)

1.0

1.2

Fig. 6. Eect of NCP dose on the adsorption of MG. MG = 100 mg/L.

results indicate that 0.1 g NCP100C was able to adsorb up to 40 mg/L MG in 100 mL and 0.6 g/100 mL required to remove 100 mg/L. This indicates that by removing silica, adsorption sites on carbon increase and this enhanced the adsorption capacity. 3.3. Adsorption isotherms A simple and eective representation of the adsorption behaviour of both of the PCP and NCP samples is by using Langmuir and Freundlich models. When the above results are tested against these two models, PCP is well described by the Langmuir and Freundlich models while the NCP is best tted to only the Langmuir model. Parameters for both models are presented in Table 2. On the whole, the results indicate the signicance of using rice husks based activated carbon for the

Acknowledgements The authors are grateful to the Universiti Sains Malaysia for nancial support to carry out this research work.

References
Aitcheson, S.J., Arnett, J., Murray, K.R., Zhang, J., 2000. Removal of aquaculture therapeutants by carbon adsorption 1. Equilibrium adsorption behaviour of single compound. Aquaculture 183, 269 284. Bcaoui, A., Yaacoubi, A., Dahbi, A., Bennouna, C., Phan Tan Luu, R., Maldonado-Hodar, F.J., Rivera-Utrilla, J., Moreno-Castilla, C., 2001. Optimization of conditions for the preparation of activated carbons from olive-waste cakes. Carbon 39, 425 432. Chan, K.C., Goh, S.H., Toh, H.T., Tan, W.I., Tang, T.S., 1980. Activated carbon from palm (Elaeis guineensis) nut shells. Malaysian J. Sci. 6 (B), 131137. El-Hendawy, A.N.A., Samra, S.E., Girgis, B.S., 2001. Adsorption characteristics of activated carbons obtained from corncobs. Colloids Surface A: Physicochem. Eng. Aspects 180 (3), 209 221. Garg, V.K., Gupta, R., Yadav, A.B., Kumar, R., 2003. Dye removal from aqueous solution by adsorption on treated sawdust. Bioresource Technol. 89, 121124. Guo, Y., Yang, S., Fu, W., Qi, J., Li, R., Wang, Z., Xu, H., 2003. Adsorption of malachite green on micro- and mesoporous rice husk-based active carbon. Dyes and Pigments 56, 219229. Hu, Z., Srinivasan, M.P., 2001. Mesoporous high-surface-area activated carbon. Micropor. Mesopor. Mater. 34, 267275. La, W.K., 2001. Production of activated carbon from acorns and olive seeds. Biomass. Bioenerg. 20, 5762.

Table 2 Langmuir and Freundlich parameters tted to experimental data for malachite green, MG, adsorbed on PCP and NCP rice husks based activated carbon Sample carbonized/ treated temp. PCP10 400 C 500 C 650 C PCP20 400 C 500 C 650 C NCP 30 C 100 C Langmuir parameters r
2

Freundlich parameters r2 1/n KF

Xm (mg/g)

KL

0.9720 76.9231 0.1555 0.9881 0.4450 14.3946 0.9670 77.5194 1.4176 0.9667 0.2677 38.2296 0.9517 83.3333 0.0829 0.9936 0.5927 8.2794 0.9594 77.5194 0.4108 0.9791 0.2609 29.7167 0.9725 80.0000 0.7530 0.9637 0.3629 29.4306 0.9639 92.5920 0.1223 0.9927 0.5493 12.6970 0.9865 57.1430 0.8536 0.8701 0.2870 21.9837 0.9958 56.497 1.0114 0.8956 0.2146 26.9650

ARTICLE IN PRESS
6 I.A. Rahman et al. / Bioresource Technology xxx (2005) xxxxxx Rahman, I.A., Saad, B., 2003. Utilization of guava seeds as a source of activated carbon for removal of methylene blue from aqueous solution. Malaysian J. Chem. 5 (1), 814. Rahman, I.A., Sing, Y.Y., Bari, M.F., Saad, B., 2005. Adsorption of paraquat by treated and untreated rice husks studied using ow injection-spectrophotometry, in press. Sandra, J.C., Lonnie, R.B., Donna, F.K., Daniel, R.D., Louis, T., Frederick, A.B., 1999. Toxicity and metabolism of malachite green and leucomalachite green during short-term feeding to Fischer 344 rats and B6C3F1 mice. Chem-Biol. Interact. 122, 153170.

Mbui, D.N., Shiundu, P.M., Ndonye, R.M., Kamau, G.N., 2002. Adsorption and detection of some phenolic compounds by rice husk ash of Kenyan origin. J. Environ. Monitor 4, 978984. Rahman, I.A., Ismail, J., Osman, H., 1997. Eect of nitric acid digestion on organic materials and silica in rice husks. J. Mater. Chem. 7 (8), 1215. Rahman, I.A., Ismail, J., Osman, H., 2000. Studies on Zn(II) ion adsorption by rice husks digested with nitric acid. Malaysian J. Chem. 2 (1), 1215.