Chemical Engineering Journal 203 (2012) 9–18

Contents lists available at SciVerse ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier.com/locate/cej

Modified oil palm leaves adsorbent with enhanced hydrophobicity for crude
oil removal
S.M. Sidik a, A.A. Jalil b,⇑, S. Triwahyono a, S.H. Adam b, M.A.H. Satar b, B.H. Hameed c
a
Ibnu Sina Institute for Fundamental Science Studies, Faculty of Science, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Johor, Malaysia
b
Institute of Hydrogen Economy, Faculty of Chemical Engineering, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Johor, Malaysia
c
School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia, 14300 Nibong Tebal, Penang, Malaysia

h i g h l i g h t s g r a p h i c a l a b s t r a c t

" A new low-cost biodegradable oil

adsorbent was developed from oil
palm leaves (OPLs).
" Lauric acid enhanced the
hydrophobicity and adsorption
ability of OPL.
" The adsorption was controlled by
film-diffusion mechanism.
" Maximum adsorption capacity was
1176 ± 12.92 mg g1 at optimized
conditions.
" The OPLLA has the potential of
removing crude oil from oily water
mixture.

a r t i c l e i n f o a b s t r a c t

Article history: The removal of crude oil from water by lauric acid (LA) modified oil palm leaves (OPLsLA) was investigated
Received 22 April 2012 by batch adsorption after varying pH (2–11), contact time (10–60 min), adsorbent dosage (0–52 g L1),
Received in revised form 28 June 2012 initial oil concentration (0–6400 mg L1) and temperature (303–323 K). The modification significantly
Accepted 28 June 2012
increased the hydrophobicity of the adsorbent, thus creating OPLLA with much better adsorption capacity
Available online 6 July 2012
for crude oil removal. The results gave the maximum adsorption capacity of 1176 ± 12.92 mg g1 at
303 K. The significant uptake of crude oil from water was proven by FT-IR and FE-SEM analyses. The iso-
Keywords:
therms studies revealed that the experimental data agrees with the Freundlich isotherm model. The
Low-cost adsorbent
Oil spill
pseudo-second-order kinetics model fitted well the experimental results. Boyd’s and Reichenberg’s equa-
Hydrophobicity tion on adsorption dynamic revealed that the adsorption was controlled by internal transport mechanism
Kinetics and film-diffusion was the major mode of adsorption. The prepared adsorbent showed the potential to
Isotherms use as a low-cost adsorbent in oil-spill clean-up.
Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction amount of petroleum energy source [2]. Therefore, it is essential

to collect and clean the oil promptly after a spillage.
Massive oil spills have occurred frequently as a result of human Advanced removal and recovery of oil by oil sorbent materials is
mistakes and carelessness, deliberate acts (i.e. war and vandalism) of great interest from economical and ecological standpoints and
and natural disasters (i.e. earthquakes and hurricanes) [1]. These therefore, various materials have been applied to this end [3].
incidents resulted in a great deal of damage to the human health These studies almost exclusively use either inorganic mineral
quality, serious environmental pollution and loss in the huge materials, organic synthetic product or organic natural materials
as the adsorbent [4]. Recently, the use of various types of natural
organic adsorbents [2,5–11] is particularly interesting because of
⇑ Corresponding author. Tel.: +60 7 5535581; fax: +60 7 5536165.
their greater adsorption capacities, higher biodegradability and
E-mail address: [email protected] (A.A. Jalil).

1385-8947/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cej.2012.06.132
10 S.M. Sidik et al. / Chemical Engineering Journal 203 (2012) 9–18

cost effectiveness compared to the inorganic and synthetic organic sufficient agitation. Following this, hexane with the same volume
adsorbents that are normally used. However, abundant and locally of water was added to the beaker and agitated for 3 min. The mix-
available adsorbents with a good oil-sorption capacity are still in ture was then let to stand for 5 min for the separation of the two
need. immiscible phases. The quantity of the adsorbent transferred to
The local oil-palm industry is estimated to generate 30 million the organic phase was determined by filtration and subsequent
tons of a lignocellulosic biomass per year in the form of trunks, drying and weighing. The results are expressed in term of the per-
fronds, empty fruit bunches and leaves [12]. These wastes are cent of adsorbent transferred into the organic phase. The values of
not being utilized effectively; land filling and open burning are the percent proportion used as the estimations of the degree of
common practices to eliminate these oil-palm residues, and these hydrophobicity of the adsorbent were calculated based on the fol-
methods can cause pollution that adversely impacts the ecosystem lowing equation:
[13]. Therefore, finding uses for these abundant biomasses, espe-  
Weight of adsorbent in hexane
cially on a large scale, would be profitable from both an environ- HDð%Þ ¼  100 ð2Þ
mental and an economic point of views. Original weight
Therefore, herein we chose oil-palm leaves (OPLs) as a model
adsorbent and studied their potential for removing crude oil from 2.3. Characterization of crude oil and OPL
water. OPL was reported to consist of 47.7% holocellulose, 44.53%
a-cellulose and 27.35% lignin [14]. Due to the availability of the spe- The hydrocarbon content in the crude oil before and after 7 days
cific functional groups such as hydroxyl (AOH) groups in the OPL, of exposure under sunlight was analyzed using a gas chromato-
the surface characteristics of the OPL can be modified to improve graph equipped with a mass spectrometer detector (GC–MS, Hew-
its adsorption properties. Several studies have been conducted for lett Packard HP 5890A-Series II, USA). The infrared spectra of the
the surface modifications of agricultural waste/byproduct [15,16], OPL sample were obtained using a Fourier-transform infrared spec-
but the use of less expensive fatty acids as the modification agent trometer (FT-IR, Spectrum GX, Perkin Elmer, USA). The samples
is still rare. were prepared as KBr pellets and scanned over the range of 400–
In the present study, the influence of surface modification on 4000 cm1 to identify the functional groups that were responsible
adsorption performance were evaluated for the removal of crude for adsorption. The surface morphology of the OPL before and after
oil under various experimental conditions of pH, adsorbent dosage, adsorption was examined using a field-emission scanning electron
contact time, initial oil concentration and temperature. A few at- microscope (FE-SEM, JEOL JSM 6710F). The surface area of the
tempts have been made in studying the engineering aspects of adsorbents was measured using surface analyzer (Quantachrome
the adsorption mechanism such as kinetics, equilibrium and ther- Autosorb-1 analyzer) by BET method.
modynamics of the adsorption of crude oil onto OPLLA.
2.4. Adsorption experiments
2. Material and methods
Adsorption experiments were performed by adding 1 g of OPL
2.1. Materials in a 50 mL conical flask containing 25 mL of a 2400 mg L1 oil mix-
ture. The pH of the working mixtures was adjusted to the desired
The mature OPL were collected from the Johor oil-palm planta- value with 0.1 M HCl or NaOH. The mixtures were prepared under
tion, Malaysia. The crude oil was obtained from petroleum plant constant stirring at a rate of 300 rpm at room temperature (303 K)
station, Kerteh, Malaysia and used as received. The oily-water mix- to reach equilibrium. The samples were then withdrawn at appro-
ture was prepared at the desired concentration using Milli-Q priate time intervals and centrifuged at 3500 rpm for 15 min. The
water. supernatant was filtered using a Millex-HN filter (Millipore,
0.45 lm), and the oil left at the top of the mixture was extracted
2.2. Preparation of adsorbent using n-hexane. The residual oil concentration was determined
using a double-beam UV/vis spectrophotometer (Thermo Scientific
The green-colored OPL were cut into small pieces to obtain suit- Genesys 10UV Scanning) at 280 nm. All experiments were per-
ably sized OPL for blending. The pieces were washed with water to formed in triplicate. Statistical analysis was carried out by compar-
remove any adhering substances and were then oven-dried at ing means of each crude oil removals using one-way analysis of
353 K. The pieces were then ground and sieved to a constant size variance.
of 355–500 lm and oven-dried at 373 K for 24 h to a constant At any time, t, the adsorption uptake of oil adsorbed (qt, mg/g)
weight before being stored in a plastic bottle. The uniformity of on OPL was calculated by the following mass-balance equation:
OPL particles in this size range led to a well-mix during the  
C0  Ct
adsorption. Adsorption uptake; qt ¼ V ð3Þ
m
Treated OPL was prepared by suspending 2 g of powdered OPL
in 200 mL of 1.0 M lauric acid solution. The mixture was stirred where Co and Ct (mg/L) are the liquid-phase concentrations of the
for 6 h at room temperature and then filtered. The resultant adsor- oil at the initial and at time t, respectively, V (L) is the volume of
bent sample was washed with n-hexane several times before being the mixture and m (g) is the mass of OPL used.
oven-dried overnight at 353 K. Weight percent gain (WPG) of the
OPL due to the modification with lauric acid was calculated accord- 3. Results and discussion
ing to:
  3.1. Characterizations of the adsorbent
Weight gain
WPGð%Þ ¼  100 ð1Þ
Original weight
3.1.1. Functional groups identification
The hydrophobicity degree (HD) of the OPL and OPLLA were The determination of the chemical structure of OPL before and
defined as the tendency of the materials to be removed from the after pretreatment, as well as before and after adsorption, could
water phase into a non-polar phase. In this experiment, 1.0 g of verify the possible functional groups involved in its binding mech-
adsorbent was placed in a beaker with 20 mL of water and with anism between lauric acid and oil. Fig. 1 shows the FTIR spectra of
 S.M. Sidik et al. / Chemical Engineering Journal 203 (2012) 9–18 11

Fig. 1. FTIR spectra of OPLuntreated, OPLLA and OPLLA after oil adsorption.

Fig. 2. FE-SEM images of (a) OPLuntreated, (b) OPLLA and (c) OPLLA after oil adsorption. (Magnification: 250).

OPLuntreated, OPLLA and OPLLA after oil adsorption. A strong band at The attachment of crude oil onto the surface of OPLLA was
3400 cm1 is attributed to the intra- and inter-molecular hydro- observed at 2970 and 2855 cm1; where the peaks become more
gen-bonded (OAH) stretching that occurs in cellulose. Two peaks intensified after the adsorption process. A well-pronounced trough
observed between 2970 and 2855 cm1 were assigned to the pres- at 1632 cm1 which is associated with the presence of C@C
ence of CAH asymmetric stretching of CH3 and CH2 groups [17]. stretching of aromatics of the crude oil was increased significantly
The presence of a peak observed at 1750 cm1 is attributed to after the crude oil adsorption. The adsorption of crude oil was fur-
the C@O stretching of the carbonyl group [18]. The vibration of ther evidenced by the presence of a new peak at 1400 cm1 after
C@C stretching mode was observed at 1632 cm1, may be attrib- the adsorption which represents the CAH bending of crude oil.
uted from aromatics of both OPL and the crude oil. The band at The CAH deformation band at 593 cm1 decreased in intensity
1400 cm1 may represent the CAH bending vibration mode [19]. after the adsorption, possibly because of the interaction of the lau-
The broad peaks at approximately 1103 and 1060 cm1 show the ric acid chain with the petroleum hydrocarbons.
CAOAC stretching vibration of lignin and the CAO stretching of
cellulose and hemicellulose, respectively [20].
The increase in the peak intensity of CH3 and CH2 groups after 3.1.2. Morphology identification
the treatment of OPL with lauric acid has suggested the successful The FE-SEM images of OPL before and after the pretreatment
impregnation of lauric acid into the OPL. This hypothesis was con- with lauric acid and also after oil sorption were showed in Fig. 2.
firmed with the presence of a new peak at 1750 cm1, correspond- The smoother surface of OPL became corroded and coarse with a
ing to the carbonyl group originated from the lauric acid chain. This highly porous texture after the pretreatment, suggesting the re-
result is in good agreement with the improvement of the hydropho- moval of waxy cuticle layer of the OPL during the modification,
bicity degree of the OPL from 16.3% to 84.7% after the treatment. In thus increased contact area for the adsorption of oil. Following
addition, it was supported with the increment in the weight of OPL crude oil adsorption, the pores of OPLLA were filled and adhered
after the treatment with WPG is about 4.62%. It is suggested; the with oil, which covered the surface of the OPLLA with a series of
increased in the hydrophobicity of OPL after modification with LA irregular oil cavities. The irregular surfaces implied the adsorption
will increase the affinity of OPLLA to absorb more crude oil. of crude oil onto the OPLLA.
12 S.M. Sidik et al. / Chemical Engineering Journal 203 (2012) 9–18

8 2.5
(a) OPLuntreated
OPL untr eated( ads )
(b) OP L u n t r e a t e d
OPLLA
OPL LA( ads ) OP L L A
7 OP L L A a f t e r o i l
OPLLA
OPL LA af teroiadsorption
after l ads or pti on( ads ) ads orpt ion
OPLuntreated
OPLuntreated(des) desorb 2.0
6 OPLLA(des)
OPLLA desorb

dV/dlogD, (x 104 cm3/g)

OPLLA(des)
Pore volume,(cm3/g)

OPLLA af ter adsorption desorbs

5 1.5

4
1.0
3

2
0.5

0.0
0 0 5 10 15 20 25 30 35 40 45 50
0 0.2 0.4 0.6 0.8 1
Relative pressure,(P/Po) Pore diameter, (nm)

Fig. 3. Nitrogen adsorption–desorption isotherms and pore size distributions of OPLuntreated, OPLLA and OPLLA after oil adsorption.

3.2. Effect of hydrocarbon fraction in crude oil

120

To study the capability of OPLLA to adsorb petroleum hydrocar-

100 bons with various fractions, the oil that was exposed for 7 days to
Adsorption uptake, (mg g -1)

air was used and compared with fresh crude oil obtained from the
80 plant station (closed oil). As can be seen from Fig. 4, the adsorption
uptake of closed oil is slightly higher than that of the exposed oil,
60 which indicates that the OPLLA adsorbed closed oil more easily than
exposed oil. This result may be explained by the GC–MS chromato-
40 grams shown in Fig. 5. After being exposed for 7 days, the lighter
hydrocarbons evaporated and the viscosity of the oil increased,
which resulted in the deceleration of the oil diffusion into the
20
internal surface of the OPLLA. A similar observation has been re-
ported for the removal of oil by walnut-shell media [23]. However,
0 the lighter hydrocarbons of closed crude oil that adhered to the
closed exposed
surface of the OPLLA were expected to desorb more easily during
Fig. 4. Effect of hydrocarbon fraction of crude oil on adsorption onto OPLuntreated. centrifugation than the heavy hydrocarbons in exposed crude oil
(pH 7, 2400 mg L1 oil, 40 g L1 OPLLA, 60 min, 303 K). due to the weaker binding strength of crude oil to OPL [24]. Overall,
only minor differences were observed in both the adsorption up-
3.1.3. Textural properties identification take and the removal percentage of the closed and exposed oil onto
Nitrogen sorption measurements were conducted to further the OPLLA. Therefore, the exposed oil was used for further investi-
characterize the pore parameter of the adsorbents. The obtained gations to combine the advantages of applying the spilt-oil
nitrogen adsorption–desorption and pore size distribution are conditions.
shown in Fig. 3a and b, respectively. In accordance with the IUPAC
classification, all the adsorbents showed a typical type II adsorp-
tion–desorption isotherm, indicating the presence of mesopore 3.3. Effect of pretreatment of OPL
[21]. From the physical parameters obtained from the N2 adsorp-
tion isotherm, it is evident that the modification of OPLuntreated with Preliminary study was conducted by using OPLuntreated as an
lauric acid increased the surface area and pore volume of the OPLLA. adsorbent to remove crude oil from aqueous mixture. However,
Thus, possibly enhances the ability of the adsorbent for trapping oil the OPLuntreated was found to show poor efficiency for crude oil re-
and also facilitate the pore distribution during adsorption. It is moval with the adsorption uptake of 92.32 ± 2.54 mg g1. This was
noted, the OPLLA after oil adsorption has slightly decreased in sur- expected as raw agricultural waste usually exhibited low sorption
face area and pore volume due to the oil pore-blocking effect. capacity due to the presence of many hydroxyl groups on its sur-
The structural heterogeneity is generally characterized in terms face. The presents of hydroxyl groups on the surface of OPLuntreated
of pore size distribution [22]. The pore size distribution curves is hypothesized to repel the crude oil molecules in the adsorption
showed a few peak detected, with the sharpest peak occurred at medium, thus reduced the adsorption uptake [25].
pore diameter between 2–2.3 nm for all adsorbents. The average The effectiveness of lauric acid in increasing the affinity of OPL
pore sizes of OPLuntreated, OPLLA and OPLLA after oil adsorption were to adsorb oil was proven with the increase in the adsorption up-
3.87 nm, 4.93 nm and 5.86 nm, respectively. take of OPL up to 369.60 ± 3.43 mg g1. A proposed mechanism
In overall, the BET analysis shows no significant changes in the in Fig. 6 was used to explain the results. The introduction of the
surface area, pore volume and pore diameter for the OPLuntreated, lauric acid chain to the OPLuntreated, which is rich with hydroxyl
OPLLA and OPLLA after crude oil adsorption. Therefore, we sug- groups of cellulose, hemicelluloses and lignin, promotes an esteri-
gested that surface area gave less contribution for the crude oil fication process to occur [26]. The substitution of hydroxyl groups
adsorption compared to the hydrophobicity as mentioned before with the alkyl chain from the lauric acid has lead to the creation of
in Section 3.1.1. a non-polar layer on the OPLLA surface. The increase in the hydro-
 S.M. Sidik et al. / Chemical Engineering Journal 203 (2012) 9–18 13

(a) (b)

Fig. 5. GC–MS chromatograms of (a) closed and (b) exposed crude oil.

Hydrophilic head Hydrophobic tail

O O O
O
OH OH O OH
OH OH O
O OH
O
OH
OH
O
OH OH O

OPLLA OPLLA
O

OPL
OH O OH

O
OH OH O
O
OH O
OH
OH
OH
OH O

surface
OH O
O OH

O
OH
O
adsolubilization of O OH

O
OH
O

modification oil droplet

untreated formation of reverse oil droplet attached to the

adsorbent micelle hydrophobic layer

Fig. 6. Proposed mechanism of oil adsorption onto OPLLA.

phobicity degree and the contact area of the OPLLA was believed to
372
enhance and provided more superior sites for higher oil adsorption
uptake.
370
Adsorption uptake, (mg g -1)

3.4. Effect of pH 368

The initial pH of the mixture is a vital process parameter for 366

controlling the adsorption because the aqueous chemistry and sur-
face binding sites of the adsorbent are dependent on the pH value 364
of the mixture [27]. Herein, the effect of initial pH of the oil mix-
ture on the adsorption process was studied in the range of 2–12 362
(Fig. 7). The adsorption uptake at the mixture pH 2 until 9 in-
360
creased dramatically from 360.63 ± 0.27 to 370.18 ± 0.63 mg g1.
However, in the pH range of 9–2, the adsorption uptake was almost
358
the same as in the range of 370.18 ± 0.63 to 370.65 ± 0.56 mg g1. 0 2 4 6 8 10 12
This variation can be explained by the change in the amounts of pH
protons and hydroxyl ions present in the mixture.
Fig. 7. Effect of pH on adsorption of crude oil onto OPLLA. (2400 mg L1 exposed oil,
At low pH, huge amounts of protons were available in the mix-
40 g L1 OPLLA, 60 min, 303 K).
ture and they saturated the adsorbent sites. As a result, the cationic
properties of the adsorbent surface were increased [28]. This phe-
nomenon can greatly reduce the hydrophobic properties of the
adsorbent, thus severely affecting the performance of OPLLA for adjustment which resulted in the saponification process to occur
crude oil uptake. In addition, it was suggested that strong acidity [28]. When NaOH was added, the crude oil reacted or undergo
can cause coalescence and induce in the increase of crude oil drop- hydrolysis with the NaOH to produce glycerol and fatty acid salts,
let size [28]. These crude oil droplets in the hydrophobic layer are called soap. The resultant compound was more soluble in water
expected to experience severe stress and can be easily desorbed than in hexane. Therefore, when determining the final crude oil
from the adsorbent [26]. concentration of the treated sample, the crude oil concentration
The highest adsorption uptake was observed at high pH. How- was found to be lower at pH 9 and 12, since the crude oil was
ever, this did not correspond to the crude oil removal efficiency hydrolyzed in the aqueous mixture and was not being extracted
but the addition of sodium hydroxide (NaOH) for alkaline pH into the hexane.
14 S.M. Sidik et al. / Chemical Engineering Journal 203 (2012) 9–18

640 95 1400

90
540 1200
Adsorption uptake, (mg g -1)

85

Adsorption uptake, qe (mg g -1)

440 80 1000

Removal, (%)
75
340 800
70

240 65 600

60 Experimental
140 400
55 Langmuir
200 Freundlich
40 50
0 10 20 30 40 50 Temkin
Adsorbent dosage, (g L-1) 0
0 1000 2000 3000 4000
Fig. 8. Effect of adsorbent dosage on adsorption of oil onto OPLLA. (pH 7, Concentration, Ce (mg L-1)
2400 mg L1 exposed oil, 60 min, 303 K).
Fig. 9. Isotherm plots for crude oil adsorption onto OPLLA.

Thus, as shown in Fig. 7, the optimum pH for adsorption of

crude oil from aqueous mixture by OPLLA was at 7 and the adsorp- adsorption uptake of the crude oil due to the excess of the unsat-
tion uptake was 367.73 ± 0.34 mg g1. The condition seems to be urated adsorption sites.
advantageous for crude oil spill treatment since the pH of seawater In this study, three commonly used isotherms, the Langmuir
falls within the range of 6–9. Furthermore, this gives an advantage [30], the Freundlich [31] and the Temkin [32], were employed
to the treatment process whereby the final discharge after adsorp- and their non-linear form can be represented by the following
tion will have neutral pH. For this reason, no further adjustment of equations, respectively:
pH would be required after the adsorption before the effluent
qm K a C e
could be discharged. qe ¼ ð4Þ
1 þ K aCe

3.5. Effect of adsorbent dosage qe ¼ K F C 1=n ð5Þ

e

The effect of adsorbent dosage on the amount of crude oil re- qe ¼ B lnðAC e Þ ð6Þ
moved was studied by the application of dosage between 2 and
1
52 g L1. As can be seen in Fig. 8, the removal of crude oil was ob- where qe (mg g ) is the amount of oil adsorbed at the equilibrium
served to be dependent on the dosage of OPLLA. The rate of removal time, Ce (mg L1) is the equilibrium concentration of oil, KL (L mg1)
percentage increased instantaneously with increasing adsorbent and qm (mg g1) are the Langmuir adsorption constant and the the-
dosage from 2 to 52 g L1, and then gradually decreased until the oretical maximum adsorption capacity, respectively. The KF and n
equilibrium was achieved when the efficiency of crude oil removal are the Freundlich constants, which indicate the extent of the
was 91.00 ± 1.31%. In contrast, the adsorption uptake increased adsorption and the degree of nonlinearity between the solution
with decreasing OPLLA dosage, and the maximum sorption capacity concentration and the adsorption, respectively while A and B are
of 645.73 ± 6.65 mg g1 was attained when 2 g L1 of OPLLA was the Temkin constants. The isotherms fitting for Langmuir, Freund-
used. lich and Temkin of the adsorption of crude oil onto OPLLA were also
The increase in active sites led to a higher removal percentage shown in Fig. 9.
but lowered the crude oil uptake per unit of adsorbent. This is The applicability of the studied experimental data in fitting the
due to the greater number of active sites available on the surface equation isotherm models were determined by regression coeffi-
of the OPLLA for crude oil to be adsorbed at a higher the dosage cient, R2 and error function. Marquadt’s percent standard deviation
of OPLLA [29], thus leading to a higher interaction between crude was employed as the error function and is given as below:
oil particles and adsorbent. Meanwhile, the decrease in adsorption 0vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi2 1
u " #
Bu 1 X ðqe;meas  qe;cal Þ C
uptake is basically due to the higher unsaturated adsorption sites p

during adsorption reaction [30]. MPSD ¼ 100  @t A ð7Þ

p  n i¼1 qe;meas
i

3.6. Equilibrium isotherms where p is the number of experiments and n is the number of
parameters of a model equation. The adsorption capacity obtained
Adsorption isotherm is important to describe the interactions from the experimental data and calculated from the modeled equa-
between the solute and the adsorbent. For this purpose, the effect tion were denoted by qe and qe,calc, respectively. In the error estima-
of the initial concentration of oil adsorbed onto OPLLA was studied tion, the lower the MPSD value indicates better fit of the isotherm
at different initial oil concentrations ranging from 400 to equations [26].
6400 mg L1. As shown in Fig. 9, the adsorption uptake increased All of the coefficients and MPSD error values for these three iso-
with increasing initial concentration and reached its equilibrium therm models are listed in Table 1. The result shows a good fit of
stage at the initial concentration up to 5600 mg L1. This result the Langmuir isotherm to the experimental data with R2 value of
may be due to the saturation of the active sites on the OPLLA. A 0.967, suggesting the formation of monolayer coverage of crude
smaller number of hydrocarbon molecules at a lower concentra- oil onto the homogenous distribution of active sites on the OPLLA
tion have sufficient available adsorption sites to be adsorbed on surface [32]. However, the higher accuracy (R2 = 0.995) revealed
the OPLLA. Therefore, oil at lower concentrations resulted in a lower that the Freundlich isotherm model better described the adsorp-
 S.M. Sidik et al. / Chemical Engineering Journal 203 (2012) 9–18 15

Table 1 droplets causing more interfacial area for the adsorption to happen
Coefficients and MPSD error values of the Langmuir, Freundlich and Temkin [36]. After 50 min, the adsorption rate was remained constant with
isotherms.
time, presumably due to the saturation of OPLLA surfaces with
Isotherm Parameters Values crude oil particles as well as the equilibrium between adsorption
Langmuir qm (mg g1) 1.95 and desorption process that occurred after the saturation [37].
KL (L mg1) 0.156 However, the experimental data were measured at 60 min to be
R2 0.967 sure that full equilibrium was attained. It is seen that nearly
MPSD 45.27
20 min is required for the adsorption to start at a very slow rate
Freundlich KF (mg g1) (L mg1)1/n 1.72 until the changes in adsorption uptake can be negligible relative
n 1.39
R2 0.995
to contact time needed. Therefore, under this effect of contact time,
MPSD 3.17 we chose 20 min as the optimum contact time to be used in further
Temkin A (1 g1) 0.007
study.
B 334 The study of adsorption kinetics is important in the treatment
R2 0.942 of aqueous effluent because it provides valuable information on
MPSD 21.42 the mechanism of the adsorption process. Two well-known models
of pseudo-first-order [38] and pseudo-second-order [39] were
used to find the possible rate-controlling steps involved in the
tion of crude oil onto OPLLA. In addition, the lowest MPSD value adsorption of crude oil, which can be expressed in the linear form
further confirmed the suitability of the Freundlich model in as the following equations, respectively:
describing the equilibrium data. These results demonstrated the k1 t
adsorption of crude oil onto OPLLA takes place through multilayer logðqe  qt Þ ¼ log qe  ð8Þ
2:303
adsorption process with the heterogeneous adsorption sites on
the solid’s surface [33]. The KF value was 1.72. In general, as the t 1 t
KF value increases, the adsorption capacity of the adsorbent for a ¼ þ ð9Þ
qt k2 q2e qe
given adsorbate also increases. According to literature, the adsorp-
tion is favorable when 1 < n < 10, and the higher the n value, the where qe (mg g1) and qt (mg g1) refer to the amount of oil ad-
stronger the adsorption intensity [34]. The value of n was greater sorbed at equilibrium and at any time, t (min), respectively, and
than unity (n = 1.39), which indicates that the oil is favorably ad- k1(g mg1 min1) and k2 (g mg1 min1) are the rate constant of
sorbed on OPLLA. In contrary, lower correlation coefficient of the the pseudo-first-order and pseudo-second-order adsorption,
Temkin model was observed (R2 = 0.942), which indicated unsatis- respectively. The linear regression coefficient, R2 and parameters
factory fitting between the experimental data and isotherm of kinetics models were calculated and listed in Table 2. The results
equation. reveal a good fit in pseudo-second order model with the R2 values
higher than 0.997 and the approximation of experimental qe,exp
and theoretical q2e,calc values at the chosen temperature compared
3.7. Adsorption kinetics
to the pseudo-second order model. Clearly, the adsorption behav-
iors of crude oil on OPLLA do not follow the pseudo-first-order kinet-
Time of contact between adsorbate and adsorbent is of great
ics model but predominantly fit the pseudo-second-order kinetics
importance in adsorption, because it depends on the nature of
model.
the system used [35]. From Fig. 10, it was observed that the crude
At the present time, the model of Boyd [40] is widely used for
oil uptake was rapid for the first 10 min, and thereafter proceeded
studying the mechanism of adsorption. Boyd model determines
at a slower rate and finally attained saturation. The initially high
whether the main resistance to mass transfer is in the thin film
rate of crude oil uptake may be attributed to the existence of bare
(boundary layer) surrounding the adsorbent particle, or in the
surface of active sites on OPLLA. It was noticed that as the time was
resistance to diffusion inside the pores. This model is expressed as:
prolonged from 10 to 40 min, the adsorption uptake was increased.
This is probably due to the breakage of crude oil droplets which 6 X
1
1
F ¼1 expðn2 Bt Þ ð10Þ
were enhanced and thus reduced the diameter of the crude oil p2 n¼1 n2
where F is the fractional attainment of equilibrium, at different
700 times, t, and Bt is a function of F.
303 K qe
600 F¼ ð11Þ
qt
Adsorption uptake, (mg g -1)

500 The values of Bt for each fraction adsorbed can be calculated by

313 K the integration of Eq. (6) proposed by Reichenberg [41] in the fol-
400 lowing approximations:
for F values >0.85,
300
323 K
Bt ¼ 0:4977  lnð1  FÞ ð12Þ
200 for F values <0.85,
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2 ffi!2
100 pffiffiffiffi pF
Bt ¼ p p ð13Þ
3
0
0 10 20 30 40 50 60
The linearity test of Bt vs. time were used to distinguish be-
Contact time, (min)
tween the film- and particle-diffusion controlled adsorption. If
Fig. 10. Effect of contact times of adsorption of crude oil onto OPLLA at different the plot is a straight line passing through the origin, the adsorption
adsorption temperatures. (pH 7, 2400 mg L1 exposed oil, 2 g L1 OPLLA). rate was governed by the particle diffusion; otherwise it was gov-
16 S.M. Sidik et al. / Chemical Engineering Journal 203 (2012) 9–18

Table 2
Coefficients of pseudo-first-order and pseudo-second-order adsorption kinetics models.

Temperature (K) qe.exp (mg g1) Pseudo-first order Pseudo-second order

2
k1  10 2
(g mg 1
min 1
) q1e.calc (mg g 1
) R k2  103 (g mg1 min1) q2e.calc (mg g1) R2
303 645.73 ± 6.65 10.02 488 0.986 0.900 667 0.998
313 467.58 ± 9.88 9.35 376 0.982 1.10 476 0.998
323 268.61 ± 6.53 8.36 200 0.975 1.70 278 0.997

5 0
(a) 0 (b) 10 20 30 40 50 60
4.5
4 -0.5

3.5
3 -1

log (1-F)
Bt

2.5
2 -1.5

1.5
1 303 K -2 303 K
0.5 313 K 313 K
323 K 323 K
0 -2.5
0 10 20 30 40 50 60
Time, (min) Time, (min)

Fig. 11. Boyd plots for the adsorption of crude oil onto OPLLA at different adsorption temperatures.

Table 3
Thermodynamics parameters for the adsorption of oil onto OPLLA.

Temperature (K) qe (mg g1) DGo (kJ mol1) DHo (kJ mol1) DSo (kJ mol1 K1) Ea (kJ mol1)
303 1179 ± 12.92 0.804
313 706.83 ± 24.96 2.824 38.3 0.228 25.8
323 333.67 ± 17.00 5.372

Table 4
Comparison of oil adsorption capacity of various adsorbents.

Sorbent Adsorption capacity (g g1) pH Oil conc. (mg L1) Temperature (°C) Contact time (min) Ref.
Lauric acid treated oil palm leaves (OPLLA) 1.2 ± 0.12 7 5600 303 20 This study
Hydrophobized vermiculite with carnauba wax 1.1 ± 0.1 – 50 – 60 [5]
Hydrophobic aquaphyte – Salvinia sp. (HAS) 1.39 – 586 298 180 [6]
Hydrogel of chitosan based on polyacrylamide (HCP) 2.30 3 3000 298 200 [7]
Hydrophobic aerogel (HA) 2.80 – 1000 298 >180 [2]
Carbonized rice husks (CRH) 6.00 7.0 – 298 10 [8]
Black rice husk ash (BHRA) 6.22 – – 293 3 [9]
Oleic acid-grafted sawdust (OGSD) 6.00 5.19 8000 298 5 [10]
Recycled wool-based nonwoven material (RNWM) 11.50 – 80,000 – 10 [11]

erned by the film diffusion [42]. Fig. 11a shows the Boyd plot for tion of oil onto OPLLA was evaluated over the temperature range of
the crude oil adsorption on OPLLA. The plots for all the tempera- 303–323 K. As shown in Table 3, the adsorption uptake of the oil
tures are linear, but they do not pass the origin, indicating the increased with decreasing temperature, which indicates that the
film-diffusion-controlled mechanism. To reconfirm the above process was exothermic [44].
observation, log (1-F) vs. time was plotted at different tempera- The free energy change (DGo) was calculated by the following
tures and the straight lines deviating from the origin were obtained equation:
for all plots as shown in Fig. 11b. These observations further sup-
port the fact that the adsorption of crude oil onto OPLLA occurs DGO ¼ RT ln K C ð14Þ
via internal transport mechanism at all the temperatures studied where Kc is the equilibrium constant of the adsorption, which was
[43]. obtained from the following equation:
C e ðadsorbentÞ
3.8. Thermodynamics studies KC ¼ ð15Þ
C e ðsolutionÞ
Thermodynamics parameters were evaluated to confirm the where Ce (adsorbent) and Ce (mixture) are the equilibrium concen-
nature of the adsorption. The effect of temperature on the adsorp- trations of the oil molecules on the adsorbent and in mixture,
 S.M. Sidik et al. / Chemical Engineering Journal 203 (2012) 9–18 17

respectively. The values of the enthalpy change (DHo) and the en- 4. Conclusions
tropy change (DSo) could be determined from the van’t Hoff
equation: In conclusion, OPLLA shows the potential to be a crude-oil adsor-
bent in oily water. The advantage of using OPL as an adsorbent in
DSO DH O this study is that its economical and solving environmental prob-
ln K C ¼  ð16Þ
R RT lems due to the abundance of OPL in South East region, especially
The values of these thermodynamics parameters determined at Malaysia and Indonesia. The successful impregnation of lauric acid
three different temperatures are listed in Table 3. The increase in into OPL was supported by all the analysis studied. Pretreatment of
the DGo values with increasing temperature indicates a decrease the raw OPL by lauric acid was effectively enhanced its surface
in the feasibility and spontaneity of the adsorption at higher tem- hydrophobicity, thus increased the adsorption capacity. FTIR anal-
peratures [45]. The negative DHo value suggests that the crude oil ysis, BET analysis and FE-SEM micrographs have elucidated the sig-
adsorption may represent an exothermic process [42]. In general, nificant uptake of crude oil from the water onto OPLLA. The
the magnitude of the enthalpy is about 20–40 kJ mol1 for physi- adsorption of crude oil onto OPLLA exhibits pseudo-second-order
sorption and 80–400 kJ mol1 for chemisorption [46]. Therefore, kinetics, and the adsorption was controlled by film-diffusion and
this adsorption of crude oil onto OPLLA may classify as physisorp- governed by the internal transport mechanism. All of the experi-
tion since its absolute magnitude of enthalpy is 38.3 kJ mol1, mental data fit well with the Freundlich model. The thermodynam-
which is in the physisorption range. The negative value of DSo sug- ics studies presented here prove the feasibility, spontaneity and
gests that there is a decrease in randomness at the solid–mixture exothermic nature of the adsorption, which is controlled by a phys-
interface during the adsorption of crude oil from water due to isorption process.
the highly ordered crude oil molecules in the hydrophobic layer
of OPLLA at adsorption equilibrium [47].
Acknowledgements
The activation energy, Ea, is defined as the minimum kinetics
energy needed by the sorbate molecules to react with the active
We gratefully acknowledged the financial support by the Re-
sites available on the surface of the adsorbent [48]. The value of
search Grant from the Ministry of Higher Education Malaysia
Ea can be obtained by the Arrhenius equation by calculating the
(Grant. No. 78326) and the award of National Science Fellowship
slope of the plot of ln k2 vs. 1/T:
(Siti Munirah Sidik). We are also thankful to the Hitachi Scholar-
 
Ea 1 ship Foundation and Prof. Goto Masahiro for their supports and
ln k2 ¼ ln A  ð17Þ advices.
R T

where R (8.314 J mol1 K1) is the gas constant. The slope of the plot
of ln k2 vs. 1/T gives a value of the activation energy (Ea) of References
25.8 kJ mol1 (Table 3). This value is showing that the presence of
[1] T.T. Lim, X. Huang, Evaluation of kapok (Ceiba pentandra (L.) Gaertn) as a
low energy barrier to initiate the reaction and implied that the natural hollow hydrophobic–oleophilic fibrous sorbent for oil spill cleanup,
adsorption of crude oil onto OPLLA is a physical adsorption. The va- Chemosphere 66 (2007) 955–963.
lue is consistent with the fact that the magnitude of the activation [2] D. Wang, T. Silbaugh, R. Pfeffer, Y.S. Lin, Removal of emulsified oil from water
by inverse fluidization of hydrophobic aerogels, Powder Technol. 203 (2010)
energy for physical adsorption is usually between 5 and 40 kJ mol1 298–309.
[49]. These results were also consistent with the value of enthalpy [3] S. Syed, M.I. Alhazzaa, M. Asif, Treatment of oily water using hydrophobic
in Table 3, which indicates the crude oil adsorption on OPLLA was nano-silica, Chem. Eng. J. 167 (2011) 99–103.
[4] O.K. Karaksi, A. Moutsatsou, Surface modification of high calcium fly ash for its
took place via physical adsorption.
application in oil spill clean-up, Fuel 89 (2010) 3966–3970.
[5] U.G. Da Silva, M.A.F. Melo, A.F. de Silva, R.A. Farias, Adsorption of crude oil on
anhydrous and hydrophobized vermiculite, J. Colloid Interf. Sci. 260 (2003)
3.9. Comparison of oil adsorption capacity of OPLLA with various 302–304.
adsorbents [6] T.H. Ribeiro, J. Rubio, R.W. Smith, A dried hydrophobic aquaphyte as an oil
filter for oil/water emulsions, Spill Sci. Technol. Bull. 8 (2003) 483–489.
[7] H.H. Sokker, N.M. El-Sawy, M.A. Hassan, B.E. Al-Anadouli, Adsorption of crude
Table 4 lists the comparison of maximum adsorption capacity of
oil from aqueous solution by hydrogel of chitosan based polyacrylamide
crude oil on various adsorbents. It was observed that OPLLA is com- prepared by radiation induced graft polymerization, J. Hazard. Mater. 190
parable to that of hydrophobized vermiculite with carnauba wax (2011) 359–365.
[5] and hydrophobic aquaphyte (HAS) [6] but its adsorption capac- [8] D. Angelova, I. Usunov, S. Uzunova, A. Gigova, L. Minchev, Kinetics of oil and oil
products adsorption by carbonized rice husks, Chem. Eng. J. 172 (2011) 306–
ity was slightly lower than hydrogel of chitosan based on polyacryl- 311.
amide (HCP) [7] and hydrophobic aerogel (HA) [2]. Carbonized rice [9] L. Vlaev, P. Petkov, A. Dimitrov, S. Genieva, Cleanup of water polluted with
husks (CRH) [8], black rice husk ash (BHRA) [9] and oleic acid- crude oil or diesel fuel using rice husks ash, J. Taiwan Inst. Chem. Eng 42 (2011)
957–964.
grafted sawdust (OGSD) [10] gave five times higher adsorption [10] S.S. Banerjee, M.V. Joshi, R.V. Jayaram, Treatment of oil spill by sorption
capacity compared to OPLLA and recycle wool non-woven material technique using fatty acid grafted sawdust, Chemosphere 64 (2006) 1026–
(RWNM) [11] posses the highest adsorption capacity. Generally, it 1031.
[11] M.M. Radetic, D.M. Jocic, P.M. Jovancic, Z.L.J. Petrovic, H.F. Thomas, Recycled
could be dictated that the hydrophobicity and surface area play wool based non-woven material as an oil sorbent, Environ. Sci. Technol. 37
an important role in the enhancement of oil sorption. Modifications (2003) 1008–1012.
of the adsorbents by acrylamide (HCP), trimethylsilyl (HA) and fatty [12] S. Sumathi, S.P. Chai, R. Mohamed, Utilization of oil palm as a source of
renewable energy in Malaysia, J. Renew. Sust. Energy Rev. 12 (2008) 2404–
acid groups (OGSD) enhanced their hydrophobicity, thus allow 2421.
more oil to be attracted to their surface. The hydrophobicity could [13] R. Hashim, N. Saari, O. Sulaiman, T. Sugimoto, S. Hiziroglu, M. Sato, R. Tanaka,
be also improved by pyrolysis, which provides moisture and vola- Effect of particle geometry on the properties of binderless particleboard
manufactured from oil palm trunk, Mater. Des. 31 (2010) 4251–4257.
tile components on the surface of the adsorbent (CRH and BHRA).
[14] R. Hashim, W.N.A.W. Nadhari, O. Sulaiman, F. Kawamura, S. Hiziroglu, M. Sato,
On the other hand, higher surface area posses by certain adsorbents T. Sugimoto, T.G. Seng, R. Tanaka, Characterization of raw materials and
either by their natural structure (HAS) or modification (RWNM) manufactured binderless particleboard from oil palm biomass, Mater. Des. 32
were also responsible for increasing the capability of trapping oil. (2011) 246–254.
[15] J. Mao, S.W. Won, S.B. Choi, M.W. Lee, Y.S. Yun, Surface modification of the
The pyrolysis on the preparation of RWNM resulted in fiber split- Corynebacterium glutamicum biomass to increase carboxyl binding site for
ting and fibrillation for enhancement of the specific surface area. basic dyes molecules, Biochem. Eng. J. 49 (2009) 1–6.
18 S.M. Sidik et al. / Chemical Engineering Journal 203 (2012) 9–18

[16] G.F. Fanta, T.P. Abbort, R.C. Burr, W.M. Doane, Ion exchange reactions of [33] K. Gobi, M.D. Mashitah, V.M. Vadivelu, Adsorptive removal of Methylene Blue
quaternary ammonium halides with wheat straw. Preparations of oil- using novel adsorbent from palm oil mill effluent waste activated sludge:
adsorbents, Carbohydr. Polym. 7 (1987) 97–109. equilibrium, thermodynamics and kinetics studies, Chem. Eng. J. 171 (2011)
[17] C. Teas, S. Kalligeros, F. Zanikos, S. Stournas, E. Lois, G. Anastopoulos, 1246–1252.
Investigation of the effectiveness of absorbent materials in oil spills clean [34] Q.-S. Liu, T. Zheng, P. Wang, J.-P. Jiang, N. Li, Adsorption isotherm, kinetics and
up, Desalination 140 (2001) 259–264. mechanism studies of some substituted phenols on activated carbon fibers,
[18] A.B.P. Marin, J.F. Ortuno, M.I. Aguilar, V.S. Meseguer, J. Saez, M. Llorens, Use of Chem. Eng. J. 157 (2010) 348–356.
chemical modification to determine the binding of Cd(II), Zn(II) and Cr(III) ions [35] Z. Baysal, E. Cinar, Y. Bulut, H. Alkan, M. Dogru, Equilibrium and
by orange waste, Biochem. Eng. J. 53 (2010) 2–6. thermodynamics studies on biosorption of Pb(II) onto Candida albicans
[19] H. Yang, R. Yan, H. Chen, D.H. Lee, C. Zheng, Characteristics of hemicelluloses, biomass, J. Hazard. Mater. 161 (2009) 62–67.
cellulose, and lignin pyrolysis, Fuel 86 (2007) 1781–1788. [36] S. Micheal, K. Heike, S. Helmar, Adsorption kinetics of emulsifiers at oil-water
[20] K.K. Pandey, A.J. Pitman, FTIR studies of the changes in wood chemistry interfaces and their effect on mechanical emulsification, Chem. Eng. Process.
following decay by brown-rot and white-rot fungi, Int. Biodeterior. Biodegrad. 33 (1994) 307–311.
52 (2003) 151–160. [37] M.E. Argun, S. Dursun, M. Karatas, M. Guru, Activation of pine cone using
[21] K.Y. Foo, B.H. Hameed, Textural porosity, surface chemistry and adsorptive Fenton oxidation for Cd(II) and Pb(II) removal, Bioresour. Technol. 99 (2008)
properties of durian shell derived activated carbon prepared by microwave 8691–8698.
assisted NaOH activation, Chem. Eng. J. 184 (2012) 57–65. [38] S. Lagergren, B.K. Svenska, Zur theorie der sogennantan adsorption geloester
[22] A. Shahsavand, M.N. Shahrak, Reliable prediction of pore size distribution for stoffe, Veternskapsakad Handl. 24 (1898) 1–39.
nano-sized adsorbents with minimum information requirements, Chem. Eng. [39] Y.S. Ho, G. McKay, Pseudo-second order model for sorption processes, Process
J. 171 (2011) 69–80. Biochem. 34 (1999) 451–465.
[23] P. Schatzerg, Investigations of sorbents for removing oil spills from water. US [40] G.E. Boyd, A.W. Adamson, L.S. Myers Jr., The exchange adsorption of ions from
Coast Guard Report. No. 724110.1/2/1, US Coast Guard Headquaters, aqueous solution by organic zeolites. Part II. Kinetics, J. Am. Chem. Soc. 69
Washington, DC, 1971. (1947) 2836–2848.
[24] Q.F. Wei, R.R. Mather, A.F. Fortheringham, Oil removal from used sorbents [41] D. Reichenberg, Properties of ion exchange resins in relation to their structure.
using a biosurfactant, Bioresour. Technol. 96 (2005) 331–334. Part III. Kinetics of exchange, J. Am. Chem. Soc. 75 (1953) 589–598.
[25] R. Gong, Y. Jin, F. Chen, Z. Liu, Enhanced malachite green removal from [42] B.H. Hameed, M.I. El-Khaiary, Malachite green adsorption by rattan sawdust:
aqueous solution by citric acid modified rice straw, J. Hazard. Mater. 137 isotherm, kinetics and mechanism modeling, J. Hazard. Mater. 162 (2009)
(2006) 865–870. 344–350.
[26] S. Ibrahim, S. Wang, H.M. Ang, Removal of emulsified oil from oily wastewater [43] R. Ahmad, R. Kumar, Adsorption studies of hazardous malachite green onto
using agricultural waste barley straw, Biochem. Eng. J. 49 (2010) 78–83. treated ginger waste, J. Environ. Manage. 91 (2010) 1032–1038.
[27] J.Y. Farah, N.S. El-Gendy, L.A. Farahat, Biosorption of astrazone blue basic dye [44] A.A. Jalil, S. Triwahyono, S.H. Adam, N.D. Rahim, M.A.A. Aziz, N.H.H. Hairom,
from an aqueous solution using dried biomass of Baker’s yeast, J. Hazard. N.A.M. Razali, M.A.Z. Abidin, M.K.A. Mohamadiah, J. Hazard. Mater. 181 (2010)
Mater. 148 (2007) 402–408. 755–762.
[28] A.L. Ahmad, S. Bhatia, N. Ibrahim, S. Sumathi, Adsorption of residual oil from [45] M. Auta, B.H. Hameed, Preparation of waste tea activated carbon using
palm oil mill effluent using rubber powder, Braz. J. Chem. Eng. 22 (2005) 371– potassium acetate as an activating agent for adsorption of Acid Blue 25 dye,
379. Chem. Eng. J. 171 (2011) 502–509.
[29] V.O. Arief, K. Trilestari, J. Sunarso, N. Indraswati, S. Ismadji, Recent progress on [46] M.A.Z. Abidin, A.A. Jalil, S. Triwahyono, S.H. Adam, N.H.N. Kamarudin, Recovery
biosorption of heavy metals from liquid using low cost biosorbents: of gold(III) from an aqueous solution onto a Durio zibethinus husk, Biochem.
characterization, biosorption parameters and mechanism studies, CLEAN-Soil Eng. J. 54 (2011) 124–131.
Air Water 36 (2008) 937–962. [47] D. Bera, D. Lahiri, A. Nag, Kinetics study on bleaching of edible oil using charred
[30] W.S. Wan Ngah, M.A.K.M. Hanafiah, Biosorption of copper ions from dilute sawdust as a new adsorbent, J. Food Eng. 65 (2004) 33–36.
aqueous solution on base treatedrubber (Hevea brasiliensis) leaves powder: [48] L.G. Wade, Organic Chemistry, Pearson Education International, United States
kinetics, isotherm, and biosorption mechanism, J. Environ. Sci. 20 (2008) of America, 2006.
1168–1176. [49] Q. Li, L. Chai, Z. Yang, Q. Wang, Kinetics and thermodynamics of Pb(II)
[31] H.M.F. Freundlich, Über die adsorption in lösungen, Z. Phys. Chem. 57 (1906) adsorption onto modified spent grain from aqueous solutions, Appl. Surf. Sci.
385–470. 255 (2009) 4298–4303.
[32] B.H. Hameed, Removal of cationic dye from aqueous solution using jackfruit
peels as non-conventional low-cost adsorbent, J. Hazard. Mater. 162 (2009)
344–350.