Chemosphere 54 (2004) 987–995

www.elsevier.com/locate/chemosphere

Biosorption of heavy metals using rice milling by-products.

Characterisation and application for removal of
metals from aqueous effluents
C
esar Ricardo Teixeira Tarley, Marco Aur
elio Zezzi Arruda *

Department of Analytical Chemistry, Institute of Chemistry, State University of Campinas––UNICAMP,

P.O. Box 6154, Campinas 13084-971, S~ao Paulo, Brazil
Received 19 September 2002; received in revised form 1 September 2003; accepted 1 September 2003

Abstract

The morphological characteristics as well as chemical composition of rice husks were evaluated by different tech-
niques such as spectroscopy and thermogravimetry. The material, which is considered a by-product obtained from rice
milling, was then investigated as a potential decontaminant of toxic heavy metals present in laboratory effluents. Studies
using glass columns were carried out at room temperature employing 100 ml of synthetic solutions containing Cd(II)
and Pb(II) at 100 mg l
1 in order to study the effects of pH, flow rate and particle size on Cd(II) and Pb(II) adsorption.
After establishing the optimised conditions, the potentiality of rice husks for removing Cd(II) and Pb(II) ions from 100
ml of laboratory effluent, presenting concentrations before treatment of 22 and 12 mg l
1 , respectively, was evaluated.
The ability to take up other metals species, such as Al(III), Cu(II) and Zn(II), present in this effluent was also studied.
According to the data obtained, under the optimised conditions (pH ¼ 4.0, flow rate of 8.0 ml min
1 and 6355 lm rice
husk particle size), 30 g of husks were necessary to attain the permissible limits for effluent release, as recommend by the
EPA, for those species evolved in this work (Al, Cd, Cu, Pb and Zn).

2003 Elsevier Ltd. All rights reserved.

Keywords: Rice husks; Effluent treatment; Heavy metals

1. Introduction of these alternatives is chemical precipitation. However,

Effluent treatment is nowadays one of the most im-
portant targets for industry and other institutions,
mainly those where the effluent contains different con-
taminant species. Different alternatives for treating ef-
fluents are described in the literature, including chemical
precipitation, carbon adsorption, ion exchange, and
membrane separation process, among others (Matheic-
kal and Yu, 1999; Juang and Shiau, 2000; Lacour et al.,
2001; Yan and Viraraghavan, 2001). The most common
*
Corresponding author. Fax: +55-193-788-3023.
E-mail address: zezzi@iqm.unicamp.br (M.A. Zezzi Arr-
uda).
0045-6535/$ - see front matter
2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2003.09.001
some limitations in this process can be pointed out, such
as cost, low efficiency, labor-intensive operation, and
lack of selectivity of the precipitation process (Gaballah
and Kilbertus, 1998; Lee et al., 1998). New approaches
for removing metals based on biosorption using algae or
bacteria have been reported (Madrid et al., 1998; Dos
Santos and Lenzi, 2000). Although these materials pre-
sent binding sites, including amines, carboxyl and thiol
groups, their use for effluent treatment purposes may
not be suitable due to their poor natural abundance. On
the other hand, the use of other natural materials, such
as peanut shells, soybean hulls and corncobs, which are
available in large quantities, may present higher poten-
tials as inexpensive sorbents for effluent treatment
(Marshall et al., 1999; Wafwoyo et al., 1999; Vaughan
988 C.R. Teixeira Tarley, M.A. Zezzi Arruda / Chemosphere 54 (2004) 987–995
et al., 2001). In addition, although the use of these
materials is still not very common, in general, they
present good adsorption capacity (Dushenkov et al.,
1995). Bailey et al. (1999) have presented an interesting
review, which focuses on the potential of a wide variety
of low cost sorbents for heavy metals. According to
these authors, a low cost sorbent can be assumed if it
requires less prior processing, is abundant in nature, or
is either a by-product or waste material from another
industry. These materials could be alternatives for ex-
pensive treatment processes.
In this sense, various solid wastes can be used for
these purposes. In Brazil, rice husks are an interesting
candidate due to the annual rice production, which was
estimated as 12 million tons in 2002 (IBGE, 2002). Rice
husks, a by-product of rice processing, comprise 23% of
the rice grain, being considered a significant waste dis-
posal problem. Traditionally, rice husks have been used
in manufacturing block, employed in civil construction
as panels, or used by the rice industry itself as a source
of energy for boilers (Della et al., 2001). In addition, the
presence of compounds such as cellulose, hemi-cellulose,
lignin and silica in this material, with binding sites ca-
pable to take up metals also suggests the use of rice
husks as a possible low cost natural adsorbent for
metals.
The majority of applications are related to industrial
effluents (Atkinson et al., 1996; Khalid et al., 1999); the
use of natural adsorbents for removal of heavy metals
from laboratory effluents has not been described. In this
sense, the aim of this work was to characterise and use
rice husk materials for removing cadmium and lead
from laboratory effluents. These metals were chosen
based on their industrial applications and potential
pollution impact on the environment (Low et al., 2000).
2. Materials and methods
2.1. Reagents and Samples
Rice (Oryza sativa L) husks were kindly provided by
rice producers from Parapu~a city, S~ao Paulo state,
Brazil. To prepare the powdered husks, they were ini-
tially ground and homogenised using a food blender
with steel blades (Walita, S~ao Paulo, Brazil) for 10 min.
Particle sizes 6355 lm were obtained by passing the
milled material through a steel sieve. Afterwards, the
ground husks were stored in polyethylene bottles (high
density) and used without any other physical or chemi-
cal treatment. The experiments were carried out in glass
columns (30 · 3.4 cm) coupled to a peristaltic pump for
controlling the column eluate flow rate. Glass wool was
inserted in the column at the bottom to prevent loss of
husks during the effluent treatment.
All reagents were of analytical grade and deionised
water obtained from a Milli-Q system (Millipore, Bed-
ford, MA, USA) was used throughout. The Cd(II) and
Pb(II) solutions, at 100 mg l
1 concentrations were pre-
pared in distilled/deionised water using PbCl2 H2 O and
CdCl2 H2 O (Merck, Darmstadt, Germany).
The wastewater used was obtained from an analytical
chemistry laboratory. It was made up of residues from
sample preparation processes (different mineral acids
and oxidant agents), organic compounds used for ex-
traction of different metals in biological samples, slurry
samples (in general from biological samples) and, finally,
solutions of metals used for calibration procedures in
atomic absorption spectrometry.
2.2. Instruments and apparatus
The Cd(II) and Pb(II) concentrations were deter-
mined by using a Perkin–Elmer Model Analyst 300
flame atomic absorption spectrometer (FAAS) equipped
with deuterium lamp background correction. For mea-
suring other metals present in the effluent, a Perkin–
Elmer Model Optima DV 3000 inductively coupled
plasma optical emission spectrometer (ICP OES) was
also used. In addition, a Perkin–Elmer Model Analyst
600 electrothermal atomic absorption spectrometer
(ETAAS) equipped with Zeeman background correction
was used for measurements of Cd(II) and Pb(II) in
the rice husk samples. The operational conditions for
FAAS, ETAAS and ICP OES were those already es-
tablished by the manufacturers.
A Digimed DM20 pH-meter was employed for pH
measuring, while a QW-3000 microwave oven was used
for sample preparation, in order to evaluate the total
metal content in the rice husk samples.
2.3. Rice husk characterisation
The functional groups present in the rice husks were
characterised by a Fourier transform infrared (FT-IR)
1605 spectrometer, using KBr discs to prepare the rice
husk samples. The spectral range varied from 4000 to
400 cm
1 .
The X-ray diffraction spectrum was obtained by us-
ing a Shimadzu XRD 6000 X-ray diffractometer. Sam-
ples were exposed to X-ray (k ¼ 1:54060 A ) with the 2h
angle varying between 5 and 50. The applied voltage
and current were 40 kV and 30 mA, respectively. An
elemental analyser (Perkin–Elmer PE 2400) was used for
C, N and H determinations. The thermal behaviour of
rice husk was obtained by using a TGA 2950 thermo-
gravimetric analyser. About 10 mg of rice husk were
heated up to 950 C in oxidant atmosphere at 20
C min
1 temperature rate.
The morphological characteristics of rice husks were
evaluated by using a JEOL JMT-300 scanning electron
 C.R. Teixeira Tarley, M.A. Zezzi Arruda / Chemosphere 54 (2004) 987–995
microscope. The rice husk samples were covered with a
thin layer of gold and an electron acceleration voltage of
20 kV was applied.
The silicon concentration in rice husks was quantified
by using a Shimadzu EDX 700 X-ray fluorescence
spectrometer after incidence radiation varying from 15
to 50 kV. The calibration graph was made using in-
strumental standards. In the investigation of 13 C NMR,
a Varian INOVA nuclear magnetic resonance spec-
trometer was employed. The high-resolution spectrum
of the rice husk samples was obtained with a frequency
of 500 MHz using cross polarisation and magic angle
spinning (CP/MAS).
2.4. Optimisation of biosorption parameters using rice
husks
For these experiments 100 ml of a solution contain-
ing Cd(II) and Pb(II) at 100 mg l
1 concentrations was
passed through the packed column. The experimental
parameters affecting the bioaccumulation of Cd(II) and
Pb(II) species were examined.
The effect of pH on the ability of rice husks to adsorb
metal ions was investigated. For this purpose, the pH
values of the Cd(II) and Pb(II) solution were varied
from 2 to 6. This solution was percolated through the
glass column containing 2.0 g of rice husks ( 6355 lm)
at a 4 ml min
1 flow rate.
Sample flow rates also play an important role in the
retention of elements. For checking the effect of flow
rate on metal uptake, this parameter was varied (2, 4, 6
or 8 ml min
1 ).
In order to examine the effect of particle size on metal
adsorption, the particle diameters were studied in the
6355 lm and 355 < £ 6 600 lm ranges. In addition,
whole rice husks also were investigated.
2.5. Procedure to remove heavy metals from laboratory
effluent
Columns were packed with a known mass (2–30 g) of
rice husks with particle size 6355 lm. The effluent,
containing several elements and organic compounds,
was taken and the pH was adjusted to 4.0 with 10
mol l
1 sodium hydroxide. Initial, Cd(II) and Pb(II)
concentrations were determined and 100 ml of the so-
lution was passed through the column. By using a
peristaltic pump, the flow rate was adjusted to the de-
sired value. The initial concentrations of the metal ions
present in the solution and the final concentrations after
treatment were both determined by FAAS. The limiting
Cd(II) and Pb(II) concentrations (0.474 and 1.32 mg l
1 ,
respectively) in the treated effluent were defined ac-
cording to the Environmental Protection Agency (EPA,
2000). The pH values of the treated effluent were defined
according to the Brazilian legislation (CONAMA,
989
1986). Only after attaining these conditions was the ef-
fluent considered for discharge. In order to evaluate the
treatment efficiency for other metals, after establishing
the optimal conditions for Cd(II) and Pb(II) laboratory
effluent treatment, a new column was employed with the
same conditions for performing a screening related to
other metals present in the same effluent. Thus, the ini-
tial and final concentrations of these metals [Zn(II),
Cu(II) and Al(III)] were also determined.
3. Results and discussion
3.1. Rice husk characterisation
Like all vegetable biomass, rice husks are composed
of cellulose, hemi-cellulose and lignin (Juliano, 1985). In
this way, high values (%, m m
1 ) were obtained for
carbon (42.6 ± 0.2), nitrogen (5.2 ± 0.1) and hydrogen
(0.8 ± 0.3).
The 13 C NMR spectrum of rice husks is shown in
Fig. 1. As shown, the spectrum presents the lignin-
cellulose structure. The resonance lines were attributed
to repeating units of cellulose. The signal intensity was
associated to resonance lines of glucose, which present
different signals as follows: C-1 at 106 ppm, C-2, 3, 5 at
74 and 76 ppm; C-4 at 85 and 89 ppm: C-6 at 64 and 65
ppm. The presence of amorphous and crystalline cellu-
losic regions has usually been confirmed by the splitting
of the C-4 resonance lines (Solum et al., 1995). Related
to lignin and hemi-cellulose, the resonance lines were
attributed to typical functional groups. In this way, the
peaks were attributed to acetate groups present in the
hemi-cellulose structure. Signals at 22 ppm (methyl
carbon) and 175 ppm (carboxyl carbon) were observed
while others were not identified due to the superimposed
cellulose resonance lines. The resonance lines for lignin
were associated to a guaiacyl unit as shown in Fig. 1
(structural unit––left side). It was possible identify the
peak related to methoxy groups at 57 ppm. In addition,
it was also possible to observe some peaks ranging from
115 to 150 ppm due to aromatic carbons (Freitas et al.,
1997).
The FT-IR technique is an important tool to identify
some characteristic functional groups, which are capable
of adsorbing metal ions. According to Fig. 2, a broad
band between 3000 and 3750 cm
1 indicates the presence
of OH groups on the rice husk surface (Kamath and
Proctor, 1998). This stretching is due to both the silanol
groups (Si–OH) and adsorbed water on the rice husk
surface. The stretching of the OH groups bound to
methyl radicals presented a signal between 2940 and
2820 cm
1 . These groups are present on the lignin
structure (L
opez Pasquali and Herrera, 1997). The peaks
located at 1730 and 1640 cm
1 are characteristics of
carbonyl group stretching from aldehydes and ketones.
990 C.R. Teixeira Tarley, M.A. Zezzi Arruda / Chemosphere 54 (2004) 987–995

Fig. 1. 13 C NMR spectrum of rice husks. Peak labels refer to the typical building units associated with cellulose (C), lignin (L) and
acetate groups in hemi-cellulose (H).

10

12 9
8
Transmittance (%)

7
9
Mass (mg)

6
5
2920 1730 1511
4
6 3420 1640
3

1053 2
3
1

0 0
4000 3000 2000 1000 0 200 400 600 800 1000
Frequency (cm )
--1 Temperature ( o C)

Fig. 2. Fourier transform infrared (FT-IR) spectrum of rice Fig. 3. Thermogravimetric curve (TG) of rice husks. For de-
husks. The arrows indicate the maximum signal obtained. For tails, see text.
details, see text.

ond step of pyrolysis was obtained when the tempera-

These groups can be conjugated or non-conjugated to
aromatic rings (1640 and 1730 cm
1 , respectively). The
peaks associated with the stretching in aromatic rings
were verified at 1511 cm
1 while deformations related to
C–H and C–O bonds were observed from 1085 to 1030
cm
1 .
The overall mass loss during thermogravimetric an-
alyses can be divided into steps related to moisture,
hemi-cellulose, cellulose and lignin (Fig. 3). Thus, as one
can see, a mass loss of 0.43 mg by heating the rice husks
up to 250 C was verified due to the elimination of
moisture retained in this material. Afterwards, the sec-
ture was varied from 250 to 360 C. In this step, a higher
mass loss (5.8 mg) was observed. The behaviour of the
pyrolysis curve at this temperature indicates hemi-
cellulose and cellulose decomposition, as well as loss of
the remaining adsorbed water (Teng and Wei, 1998).
Lignin decomposition occurs in the 360–525 C range,
thus indicating that this structure presents higher sta-
bility than hemi-cellulose and cellulose. Finally, no mass
loss was detected when the temperature was increased to
950 C. This result indicates the presence of oxides
(mainly those of aluminium and silicon), which are
stable at higher temperatures.
 C.R. Teixeira Tarley, M.A. Zezzi Arruda / Chemosphere 54 (2004) 987–995 991

Table 1
1200 Metal concentrations in rice husks determined by ICP OES or
ETAAS (n ¼ 5)
1000
Metal Concentration (lg g
1 )
Intensity (cps)

800 Al 761 ± 136

Ca 1434 ± 286
Cda 0.03 ± 0.01
600
Cr 39.2 ± 5.26
Fe 179 ± 31
400
K 2572 ± 67
Mg 329 ± 54
200
Mn 196 ± 3
Na 961 ± 106
0 Ni 22.9 ± 1.19
0 10 20 30 40 50
Pba 2.1 ± 0.7
2θ ( o)
a
Determined by ETAAS (results in ng g
1 ).
Fig. 4. X-ray diffractograms of rice husks. For details, see text.

Pb(II) were observed, thus indicating the potentiality of

The X-ray diffraction patterns for rice husks are
shown in Fig. 4. A broad peak from 15 to 35 2h dif-
fraction angles indicates the presence of amorphous
silica (Kamath and Proctor, 1998). In fact, the result
obtained confirms those already reported in the litera-
ture for rice husk ash (Della et al., 2001).
Fig. 5 represents an X-ray fluorescence spectrum for
rice husks. The silicon concentration was approximately
4.0%, and its high amount in the rice husks represents
high SiO2 contents, so that its presence can result in a
material with both good mechanical resistance and high
ash content.
Quantitative analysis by ICP OES and ETAAS
shows the predominance of K, Ca, and Na (Table 1). In
addition, high concentrations of Fe, Mn and Al in rice
husks were observed, which can be explained due to the
favourable availability of these metals in plants, mainly
those cultivated in flooded soils, as in the specific case of
rice (Swarup, 1989). Low concentrations of Cd(II) and
0.18
1 tained. However, when the pH was increased to 4,
A
0.16
. higher values for metal absorption were achieved [4.23
A = Si Kα
0.14
.
B = P Kα
Intensity (cps)
rice husks for effluent treatment.
Scanning electron micrographs of rice husks are
shown in Fig. 6. The morphology of this material can
facilitate the adsorption of metals, due to the irregular
surface of the rice, thus makes possible the adsorption of
metals in different parts of this material. So, based on
the morphology, as well as on the fact that high amounts
of silica are concentrated on the external epidermis of
rice husks (Della et al., 2001), it can be concluded that
this material presents an adequate morphological profile
to retain metal ions.
3.2. Optimisation of biosorption parameters using rice
husks
The first parameter studied was the pH because it
plays an important role in adsorption processes (Low
et al., 2000), mainly if treatment efficiency is considered.
As shown in Fig. 7, at pH ¼ 2 only a low adsorption for
Pb(II) (0.20 mg g
1 ) and Cd(II) (0.06 mg g
1 ) was ob-
mg g
1 for Pb(II) and 1.42 mg g
1 for Cd(II)], thus in-
dicating better adsorption capacity of these metals re-

0.12
. C = Rh Lα (source)
D = K Kα
lated to the proposed adsorbent. When pH values of 5 or
0.10
. E = Ca K α 6 were tested, only slight variations on the results were
F = Ca KKβ observed, indicating that good adsorption capacity was
0.08
. G = Mn Kα
H = Fe Kα already obtained at pH ¼ 4. It also interesting to com-
0.06 ment that, at pH ¼ 4, the relative standard deviation
0.04
. C (RSD) of the results were lower than 3.5% for Pb(II) and
H
0.02
.
B E G 0.6% for Cd(II).
DF The adsorption profile as a function of solution pH
0.00
can be explained based on surface charge density, having
0 2 4 6 8 10 12 14 16 18
Energy (KeV)
functional groups such as hydroxyl, carboxyl, amino
and others. As the surface charge density decreases with
Fig. 5. X-ray fluorescence spectrum of rice husks. For details, an increase in the solution pH, the electrostatic repul-
see text. sion between the positively charged metal ions and the
992 C.R. Teixeira Tarley, M.A. Zezzi Arruda / Chemosphere 54 (2004) 987–995
Fig. 6. Scanning electron micrograph of rice husks (6355 lm)
with (a) 500 and (b) 5000 times of magnification. The bars in-
dicate the magnification, represented by the first number (in
lm). Other numbers indicate the acceleration voltage (kV), date
(day month
1 ) and the number of the micrograph, respectively.
5 Pb(II) are 1.69 and 2.33, respectively, thus corroborating
(mg metal/g adsorbent)
4
3
Pb
Cd
2
1 carried out, thus indicating that the metal removal was
0 husks.
2 3 4 5 6 With the pH now fixed at 4 and other parameters
pH
Fig. 7. Effect of pH on removal of the Pb(II) and Cd(II) by rice
husks (mass of rice husk: 2.0 g; flow rate: 4.0 ml min
1 ; particle
size: 6355 lm; initial concentration: 100 mg l
1 for each metal).
surfaces of the rice husks is reduced, which causes more
adsorption. According to Chen et al. (2002), the inter-
actions between the functional groups –COOH, OH,
and –C–O–R in the carbohydrate backbone and metal
ions have been confirmed by X-ray photoelectron spec-
troscopy and FT-IR. Also the presence of silanol groups
on the surface of the rice husks, from the silica contained
in the material, tends to form a surface presenting
weakly acid characteristics (Ka ¼ 10
6 –10
8 ) (Dugger
et al., 1964). Khalid et al. (1999) have already described
these kinds of ion exchange reactions that take place on
the surface of this material. The interaction model based
on an exchange reaction is the most common to describe
the adsorption processes on biosorbents. Some success-
ful modelling studies are available in the literature to
explain metal biosorption or interactions between metal
and organic ligands of various biosorbents (Schiewer
and Wong, 1999; Lee and Davis, 2001). Among these
models, surface complex formation is able to take ac-
count of the adsorption effects related to the valence
metals and ionic radius. The higher adsorption of Pb(II)
than Cd(II) on rice husks can be attributed by coordi-
nation sphere of these two metal ions. Cd(II) offers a
smaller coordination sphere than Pb(II) since Cd(II)
offers a smaller ionic radius than Pb(II), 10 300 and
13 200 A , respectively (Emsley, 1991). The coordination
sphere of Cd(II) embraces a smaller number of hy-
droxyls and carboxyls than Pb(II). Similar results have
been reported by Chen et al. (2002), in which it was
found that metal binding on calcium alginated-based
ion exchange resin follows a decreasing order of
Pb(II) > Cu(II) > Ca(II). In addition, the preference of
biosorbents for metals has been related to metal elec-
tronegativity (Matos and Arruda, 2003). The reported
effect is a stronger attraction due to the higher electro-
negativity. The electronegativity values for Cd(II) and
to the adsorption profile of these metals.
According to the results obtained, the initial pH for
effluent treatment was fixed at 4. Another important
point to be stressed is that an increase in the pH values
to 5 was detected after the effluent treatment. This value
is within the range established by CONAMA for effluent
release. In addition, there was not observed any pres-
ence of precipitate when the effluent treatment was
completely related to the adsorption process on rice
being the same as those already described in item 2.4,
only slight variations in the results were seen when the
flow rate was varied from 2 to 8 ml min
1 [3% for both
Cd(II) and Pb(II)]. This means that a fast kinetic process
is involved in the adsorption of Cd(II) and Pb(II) on this
material. In addition, this is an interesting result because
higher flow rates can be used, minimising the time and
costs involved in effluent treatment. In this way, 8
ml min
1 was fixed as the optimised condition.
 C.R. Teixeira Tarley, M.A. Zezzi Arruda / Chemosphere 54 (2004) 987–995 993

Table 2 ml min
1 and particle size of 6 355 lm). For the effluent
Effect of particle diameter on Pb(II) and Cd(II) removal using treatment the mass of the rice husks necessary to pro-
rice husks as adsorbent material mote a good retention of the metals was investigated.
Particle diameter Pb RSD Cd RSD The sorbent mass was only investigated at this point of
(mg g
1 ) (%) (mg g
1 ) (%) the experiments, due to the differences between the re-
Whole husks 1.4 4.2 0.5 7.4 lated effluent (real sample) and those of the standard
(all diameters) solutions utilised for the optimisation of effluent treat-
355 < £ < 600 lm 3.4 3.3 1.1 3.0 ment conditions.
6355 lm 4.0 0.6 1.2 7.8 Before treatment, metal concentrations in the labo-
The results are expressed in mg of metal per g of adsorbent. The ratory effluent were determined and the concentrations
experimental conditions utilised were pH ¼ 4 and flow rate at for Cd(II) and Pb(II) were 22.0 and 12.0 mg l
1 , re-
8 ml min
1 (n ¼ 3). spectively. Concentration values and final pH after
RSD ¼ relative standard deviation. treatment are shown in Table 3. These results indicate
that 30 g of rice husks are enough for retaining metals
when 100 ml of laboratory effluent was treated. The
Another important parameter in the adsorption
concentrations obtained after effluent treatment were
process is related to the particle size of the adsorbent
lower than the values established by EPA for Cd(II)
material. Some metals are more retained on smaller
(0.474 mg l
1 ) and Pb(II) (1.32 mg l
1 ). In addition, this
particles (Jansson-Charrier et al., 1996). Therefore, as
natural material also increases the pH value of the ef-
the contact surface between sorbent and the liquid phase
fluent (see Table 3), making it similar to that recom-
plays an important role in the adsorption phenomenon,
mended by CONAMA (pH values between 5 and 9) for
the effect of rice husk particle size on metal adsorption
effluent release. Puranik and Paknikar (1997) made
was studied using three particle size groups (6355 lm,
similar observations on Pb(II) and Zn(II) removal by
355 < £ < 600 lm and the whole husks). Whole ma-
Streptoverticillium cinnamoneum waste biomass, where
terial (30 g) was ground during 10 min, as already de-
the pH value increased from 4.0 to 6.5 during the ad-
fined (item 2.1). From this material, 6.0 g were obtained
sorption process.
with particle size 6 355 lm and 7.0 g with particle size
Excellent results were also achieved for other metals
between 355 and 600 lm. Other diameters were not
determined in the same effluent after treatment, even
tested because they were obtained only in small quan-
with the conditions specifically optimised for removing
tities (0.75 g for 675 lm particle size). As observed in
Cd(II) and Pb(II). Table 4 shows the values for Al(III),
Table 2, a decrease in rice husk particle size led to an
increase in metal adsorption. The other particle sizes
(355 < £ < 600 lm) or the whole material presented
poorer results, mainly for Pb(II). It is important to stress Table 3
Metal concentrations (mg l
1 ) and pH values after effluent
that larger particles with spherical shapes, in general,
treatment using rice husks
present higher external mass transfer than smaller par-
ticles. In this case, higher metal adsorption from these Mass (g) Cd Pb pH
particles is attributed to mass transport inside the sor- 2 22.1 ± 0.2 5.8 ± 0.2 4.30 ± 0.03
bent particles (Leusch and Volesky, 1995). On the other 4 20.1 ± 1.3 2.3 ± 0.3 4.40 ± 0.06
hand, when the main processes of metal adsorption onto 6 19.5 ± 0.5 0.7 ± 0.1 4.60 ± 0.07
solid adsorbents are those based on adsorption on the 8 17.2 ± 0.9 <LQa 4.70 ± 0.02
20 1.70 ± 0.02 <LQ 5.20 ± 0.06
particle surface, higher metal adsorption is verified on
30 0.30 ± 0.02 <LQ 5.20 ± 0.03
smaller particles, as observed on rice husks. This com-
ment allows concluding that Cd(II) and Pb(II) sorption The adjusted pH value before treatment was 4.0 ± 0.2 (n ¼ 3).
a
on rice husks is attributed to adsorption process on the Values below limit of quantification (<0.05 mg l
1 Cd and
<0.53 mg l
1 Pb).
particle surface, mainly those related to ion exchange or
surface complex formation. Similar results have been
reported for cadmium adsorption on chitin (Benguella
and Benaissa, 2002). Table 4
Metal concentrations (mg l
1 ) before (BT) and after (AT) ef-
fluent treatment and the maximum concentrations recom-
3.3. Metal removal from laboratory effluent mended by EPA for effluent release
Metal BT AT EPA
The pH value of the laboratory effluent (defined in
Section 2.1 and used as a real sample) was between 0 and Al 24.1 0.48 –
1. All conditions for laboratory effluent treatment were Cu 10.7 1.2 4.14
Zn 21.4 2.2 2.87
the same as already defined (pH ¼ 4; flow rate at 8
994 C.R. Teixeira Tarley, M.A. Zezzi Arruda / Chemosphere 54 (2004) 987–995
Cu(II) and Zn(II), before and after effluent treatment,
and those recommend by EPA.
4. Conclusions
The proposed material is an excellent alternative for
removing chemical species from effluents. The physical
characterisation of rice husks pointed out some proper-
ties such as the presence of functional groups (carboxyl,
silanol, etc.) that make adsorption processes possible,
as well as the morphological aspects of the material.
After the characterisation, the material was used for
removing Cd(II) and Pb(II) from laboratory effluent.
Under the optimised conditions, the results obtained for
both studied species (and the others which were also
determined) were below those required by the EPA,
producing a treated effluent, which could be discarded.
Only as an example, if one considers a monthly ef-
fluent generation of 2500 l, about 750 kg of rice husks
will be needed for treating this effluent under the con-
ditions established in this work, with minimal costs.
With the remarkable characteristics already men-
tioned for effluent treatment and the high generation of
rice husk residues, a waste product, in the world this
material should be seriously considered for removal of
metals from effluents.
Acknowledgements
The authors acknowledge the Fundacß~ao de Amparo
a Pesquisa do Estado de S~ao Paulo––FAPESP (Grant
Number 99/12124-7 and fellowship number 01/05614-0)
and the Conselho Nacional de Desenvolvimento
Cient
ıfico e Tecnol

ogico (CNPq) for financial support
and fellowships. We are grateful to Prof. Carol H.
Collins for language assistance.
References
Atkinson, B.W., Bux, F., Kasan, H.C., 1996. Bioremediation of
metal-contaminated industrial effluents using waste sludges.
Water Sci. Technol. 34, 9–15.
Bailey, S.E., Olin, T.J., Bricka, R.M., Adrian, D.D., 1999. A
review of potentially low-cost sorbents for heavy metals.
Water Res. 33, 2469–2479.
Benguella, B., Benaissa, H., 2002. Cadmium removal from
aqueous solutions by chitin: kinetic and equilibrium studies.
Water Res. 36, 2463–2474.
Chen, J.P., Hong, L., Wu, S., Wang, L., 2002. Elucidation of
interactions between metal ions and Ca alginate-based on
ion-exchange resin by spectroscopic analysis and modelling
simulation. Langmuir 18, 9413–9421.
CONAMA, 1986. National Council of Environment. Resolu-
tion number 20, June 18.
Della, V.P., Kuhn, I., Hotza, D., 2001. Caracterizacß~ao de cinza
de casca de arroz para uso como mat
eria-prima na
fabricacß~ao de refrat
arios de s
ılica. Quim. Nova 24, 778–782.
Dos Santos, M.C., Lenzi, E., 2000. The use of aquatic
macrophytes (Eichhornia crassipes) as a biological filter in
the treatment of lead contaminated effluents. Environ.
Technol. 21, 615–622.
Dugger, D.L., Stanton, J.H., Irby, B.N., McDonnell, B.L.,
Cummings, W.W., Maatman, R.W., 1964. Exchange of 20
metal ions with weakly acidic silanol group of silica gel. J.
Phys. Chem,––US 68, 757–760.
Dushenkov, V., Kumar, P.B.A.N., Motto, H., Raskin, I., 1995.
Rhizofiltration: the use of plants to remove heavy metals
from aqueous streams. Environ. Sci. Technol. 29, 1239–
1245.
Emsley, J., 1991. The Elements, second ed. Oxford University
Press, New York.
EPA, 2000. Environmental Protection Agency. Code of federal
regulations. Chapter I. Part 437. Subpart A––Metals
treatment and recovery, Section 437.11.
Freitas, J.C.C., Cunha, A.G., Emmerich, F.G., 1997. Physical
and chemical properties of a Brazilian peat char as function
of HTT. Fuel 73, 229–232.
Gaballah, I., Kilbertus, G., 1998. Recovery of heavy metal ions
through decontamination of synthetic solutions and indus-
trial effluents using modified barks. J. Geochem. Explor. 62,
241–286.
IBGE, 2002. Brazilian Institute of Geography and Statistics,
Research Division––DIPEQ/SC. Systematic Survey of Pro-
duction Agriculturist, Florian
opolis.
Jansson-Charrier, M., Guibal, E., Roussy, J., Surjous, R., Le
Cloirec, P., 1996. Dynamic removal of uranium by chitosan:
influence of operating parameters. Water Sci. Technol. 34,
169–177.
Juang, R.S., Shiau, R.C., 2000. Metal removal from aqueous
solutions using chitosan-enhanced membrane filtration. J.
Membrane Sci. 165, 159–167.
Juliano, B.O., 1985. Rice Chemistry and Technology. American
Association of Cereal Chemistry, New York.
Kamath, S.R., Proctor, A., 1998. Silica gel from rice hull ash:
preparation and characterization. Cereal Chem. 75, 484–
487.
Khalid, N., Rahman, A., Ahmad, S., Toheed, A., Ahmed, J.,
1999. Adsorption behavior of rice husks for the decontam-
ination of chromium from industrial effluents. J. Radioan.
Nucl. Chem. 240, 775–781.
Lacour, S., Bollinger, J.C., Serpaud, B., Chantron, P., Arcos,
R., 2001. Removal of heavy metals in industrial wastewaters
by ion-exchanger grafted textiles. Anal. Chim. Acta 428,
121–132.
Lee, S.M., Davis, A.P., 2001. Removal of Cu(II) and Cd(II)
from aqueous solution by seafood processing waste sludge.
Water Res. 35, 534–540.
Lee, S.H., Jung, C.H., Chung, H., Lee, M.Y., Yang, J., 1998.
Removal of heavy metals from aqueous solution by apple
residues. Process Biochem. 33, 205–211.
Leusch, A., Volesky, B., 1995. The influence of film diffusion on
cadmium biosorption by marine biomass. J. Biotechnol. 43,
1–10.
L

opez Pasquali, C.E., Herrera, H., 1997. Pyrolysis of lignin and
IR analysis of residues. Thermochim. Acta 293, 39–46.
 C.R. Teixeira Tarley, M.A. Zezzi Arruda / Chemosphere 54 (2004) 987–995
Low, K.S., Lee, C.K., Liew, S.C., 2000. Sorption of cadmium
and lead from aqueous solutions by spent grain. Process
Biochem. 36, 59–64.
Madrid, Y., Barrio-Cordoba, M.E., C
amara, C., 1998. Bio-
sorption of antimony and chromium species by Spirulina
platensis and Phaseolus. Applications to bioextract anti-
mony and chromium from natural and industrial waters.
Analyst 123, 1593–1598.
Marshall, W.E., Wartelle, L.H., Boler, D.E., Johns, M.M.,
Toles, C.A., 1999. Enhanced metal adsorption by soybean
hulls modified with citric acid. Bioresour. Technol. 69, 263–
268.
Matheickal, J.T., Yu, Q., 1999. Biosorption of lead(II) and
copper(II) from aqueous solutions by pre-treated biomass of
Australian marine algae. Bioresour. Technol. 69, 223–229.
Matos, G.D., Arruda, M.A.Z., 2003. Vermicompost as natural
adsorbent for removing metal ions from laboratory efflu-
ents. Process Biochem. 39, 81–88.
Puranik, P.R., Paknikar, K.M., 1997. Biosorption of lead and
zinc from solutions using Streptoverticillium cinnamoneum
waste biomass. J. Biotechnol. 55, 113–124.
995
Schiewer, S., Wong, M.H., 1999. Metal binding stoichiometry
and isotherm choice in biosorption. Environ. Sci. Technol.
33, 3821–3828.
Solum, M.S., Pugmire, R.J., Jagtoyen, M., Derbyshire, F.,
1995. Evolution of carbon structure in chemically activated
wood. Carbon 33, 1247–1254.
Swarup, A., 1989. Transformation and availability of iron and
manganese in submerged sodic soils in relation yield and
nutrition of rice. Fert. News 34, 21–23.
Teng, H., Wei, Y., 1998. Thermogravimetric studies on the
kinetics of rice hull pyrolysis and the influence of water
treatment. Indian Eng. Chem. Res. 37, 3806–3811.
Vaughan, T., Seo, C.W., Marshall, W.E., 2001. Removal of
selected metal ions from aqueous solution using modified
corncobs. Bioresour. Technol. 78, 133–139.
Wafwoyo, W., Seo, C.W., Marshall, W.E., 1999. Utilization of
peanut shells as adsorbents for selected metals. J. Chem.
Technol. Biotechnol. 74, 1117–1121.
Yan, G., Viraraghavan, T., 2001. Heavy metal removal in a
biosorption column by immobilized M. rouxii biomass.
Bioresour. Technol. 78, 243–249.