Short Communication

Biosorption of copper, cobalt and nickel by marine

green alga Ulva reticulata in a packed column
K. Vijayaraghavan
a
, J. Jegan
b
, K. Palanivelu
c
, M. Velan
a,
*
a
Department of Chemical Engineering, Anna University, Alagappa College of Technology, Chennai 600 025, India
b
Department of Chemical Engineering, Sri Venkateswara College of Engineering, Sriperambudur 602 105, India
c
Centre for Environmental Studies, Anna University, Chennai 600 025, India
Received 8 July 2004; received in revised form 10 December 2004; accepted 13 December 2004
Available online 28 January 2005
Abstract
Biosorption of copper, cobalt and nickel by marine green alga Ulva reticulata were investigated in a packed bed up-
ow column. The experiments were conducted to study the eect of important design parameters such as bed height and
ow rate. At a bed height of 25 cm, the metal-uptake capacity of U. reticulata for copper, cobalt and nickel was found
to be 56.3 0.24, 46.1 0.07 and 46.5 0.08 mg g
1
, respectively. The Bed Depth Service Time (BDST) model was
used to analyze the experimental data. The computed sorption capacity per unit bed volume (N
0
) was 2580, 2245
and 1911 mgl
1
for copper, cobalt and nickel, respectively. The rate constant (K
a
) was recorded as 0.063, 0.081 and
0.275 l mg
1
h
1
for copper, cobalt and nickel, respectively. In ow rate experiments, the results conrmed that the
metal uptake capacity and the metal removal eciency of U. reticulata decreased with increasing ow rate. The Thomas
model was used to t the column biosorption data at dierent ow rates and model constants were evaluated. The col-
umn regeneration studies were carried out for three sorptiondesorption cycles. The elutant used for the regeneration of
the biosorbent was 0.1 M CaCl
2
at pH 3 adjusted using HCl. For all the metal ions, a decreased breakthrough time and
an increased exhaustion time were observed as the regeneration cycles progressed, which also resulted in a broadened
mass transfer zone. The pH variations during both sorption and desorption process have been reported.
2005 Elsevier Ltd. All rights reserved.
Keywords: Heavy metals; Marine algae; Regeneration; Thomas model; Euent treatment
1. Introduction
Biosorption is the passive uptake of heavy metals from
aqueous solutions by biological materials (biosorbents).
Biosorption has been studied in various types of biomass
including marine algae (Davis et al., 2000), bacteria (Ve-
glio et al., 1997) and fungi (Kapoor et al., 1999). Marine
algae oer advantages for biosorption because their mac-
roscopic structures oer a convenient basis for the
production of biosorbent particles suitable for sorption
process applications (Vieira and Volesky, 2000).
Although there have been a number of studies on the
use of brown algae in heavy metal removal, limited stud-
ies have been directed towards the use of green marine
algae for this purpose. Among fewbatch studies on green
seaweeds, Kuyucak and Volesky (1989) observed that a
0045-6535/$ - see front matter 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2004.12.016
*
Corresponding author. Tel.: +91 44 22203506; fax: +91 44
22352642.
E-mail addresses: velan_1@yahoo.com, velan@annauniv.
edu (M. Velan).
Chemosphere 60 (2005) 419426
www.elsevier.com/locate/chemosphere
green alga Halimeda opuntia performed equally well in
cobalt biosorption along with one of the best-performed
brown seaweed A. nodosum at higher pH conditions. In
addition, Zeroual et al. (2003) successfully employed
green seaweed Ulva lactuca for mercury biosorption.
The practical application of heavy metal sorption is
most eectively carried out in a packed bed column, as
it eciently utilize the sorbent capacity and results in a
better quality of the euent (Volesky and Prasetyo,
1994). However, the utilization of biomaterials to ad-
sorb metal ions in column mode has not been reported
very often. Among few studies, Valdman et al. (2001)
used a mini-column of 0.5 cm ID and 1.6 cm in length
packed with Sargassum sp., which exhibited Cu and
Zn uptake capacities of 11.9 and 21.0 mg g
1
, respec-
tively. Yan and Viraraghavan (2001) employed a plastic
column of 1.27 cm ID and 40 cm in length packed with
immobilized M. rouxii biomass (4.5 g) and observed metal
removal capacities of 4.06, 3.76, 0.36 and 1.36 mg g
1
for Pb, Cd, Ni and Zn, respectively. Kaewsarn (2002)
used a glass column of 1 cm ID packed with 1g of Padina
sp. biomass, which exhibited Cu-biosorption capacities
ranged between 49.58 and 52.76 mg g
1
at dierent ini-
tial copper feed concentrations. Volesky et al. (2003) em-
ployed a column of 2.5 cm ID and 50 cm in length
packed with 38 g of dry Sargassum lipendula biomass,
which exhibited Cu-biosorption capacity of approxi-
mately 38 mg g
1
during seven regeneration cycles.
Ulva reticulata, a marine green alga, was the focus of the
present investigation to remove copper, cobalt and nickel
ions from aqueous solutions in a packed bed up-ow col-
umn. U. reticulata is a very common green alga found
throughout the Palk bay, and recent batch studies revealed
copper biosorption potential of the biomass (Vijayaragha-
van et al., 2004). In the present study, the eects of design
parameters, such as bed height and ow rate, on metal
biosorption have been examined. In addition, metal
biosorption behavior of U. reticulata in three consecutive
sorptiondesorption cycles has been investigated.
2. Materials and methods
2.1. Biomass
U. reticulata was collected from Mandapam (India)
and sun dried. Dry biomass was cut with a knife into
irregular shaped particles between 1 and 3 mm in size.
The biomass was then washed with distilled water, then
ltered and nally dried overnight at 60 C and subse-
quently used for sorption experiments.
2.2. Column
Continuous-ow sorption experiments were con-
ducted in a glass column with an internal diameter of
2 cm and 35 cm in length. At the top of the column,
an adjustable plunger was attached with a 0.5 mm stain-
less sieve. At the bottom of the column, a 0.5 mm stain-
less sieve was attached followed by glass wool. A 2 cm
high layer of glass beads (1.5 mm in diameter) was
placed at the column base in order to provide a uniform
inlet ow of the solution into the column.
2.3. Experimental procedure
Analytical grades of CuSO
4
5H
2
O, CoSO
4
7H
2
O,
NiSO
4
6H
2
O, HCl, NaOH, and CaCl
2
2H
2
O were
purchased from Ranbaxy Fine Chemicals Ltd., India.
Metal ion solutions were prepared by dissolving their
corresponding sulfate salt in distilled water. Based on
the preliminary study, the pH of the copper stock solu-
tion was adjusted to 5.5 by adding 0.1 M HCl or 0.1 M
NaOH (Vijayaraghavan et al., 2004). Using the same
chemicals, pH of cobalt and nickel stock solutions
was adjusted to 4 and 4.5, respectively (Vijayaraghavan
et al., in press). All the column studies were performed at
a constant temperature of 30 C.
Aknown quantity of U. reticulata biomass was placed
in the column to yield the desired bed height of the sor-
bent. Metal ion solution having an initial concentration
of 100 mgl
1
was pumped upward through the column
at a desired ow rate by a peristaltic pump (Miclins).
Samples were collected fromthe exit of the column at dif-
ferent time intervals and analyzed for metal ion concen-
tration using Atomic Absorption Spectrophotometer
(AAS 6VARIO; Analytik Jena, Germany). The pH of
the inuent and the euent were recorded. Operation
of the column was stopped when the euent metal con-
centration exceeded a value of 99.5 mgl
1
or higher.
After the column reached exhaustion, the loaded bio-
mass with metal ions was regenerated using 0.1 M CaCl
2
solution adjusted to pH 3 with conc. HCl (Vijayaragha-
van et al., 2004). The owrate was adjusted to 5 ml min
1
.
After elution, distilled water was used to wash the bed
until the pH in the wash euent stabilized near 7.0. The
regenerated bed was reused for next cycle. These cycles
of sorption followed by desorption were repeated for three
times to evaluate the biomass sorption capacity. To deter-
mine the weight loss after three cycles, the biomass was
washed with distilled water and dried at 60 C overnight.
All experiments were carried out in duplicates, and the
deviations were within 5%. For all graphical representa-
tions, the mean values were used. All statistical analyses
were made using MATLAB

software.
2.4. Modeling and analysis of column data
The analysis of breakthrough curve was done using
BDST model. The BDST is a simple model for predict-
ing the relationship between bed height, Z, and service
time, t, in terms of process concentrations and adsorp-
420 K. Vijayaraghavan et al. / Chemosphere 60 (2005) 419426
tion parameters (Zulfadhly et al., 2001). Hutchins (1973)
proposed a linear relationship between bed height and
service time given by Eq. (1):
t
N
0
Z
C
0
m

1
K
a
C
0
ln
C
0
C
b
1

1
where C
0
is the initial metal ion concentration (mgl
1
);
C
b
is the breakthrough metal ion concentration (mgl
1
);
N
0
is the sorption capacity of bed (mgl
1
); m is the
linear velocity (cmh
1
) and K
a
is the rate constant
(l mg
1
h
1
).
Successful design of a column sorption process re-
quired prediction of the concentrationtime prole or
breakthrough curve for the euent (Yan and Viraragha-
van, 2001). Various mathematical models can be used
to describe xed bed adsorption. Among these, the Tho-
mas model is simple and widely used by several investi-
gators (Yan and Viraraghavan, 2001; Aksu and Go nen,
2003). The linearized form of Thomas model can be ex-
pressed as follows:
ln
C
0
C
1

k
Th
Q
0
M
F
k
Th
C
0
t 2
where k
Th
is the Thomas model constant (l mg
1
h
1
),
Q
0
is the maximum solid-phase concentration of solute
(mg g
1
), V is the throughput volume (l), F is the volu-
metric ow rate (l h
1
) and M is the sorbent mass (g).
The quantity of metal retained in the column
represented by the area above the breakthrough curve
(C versus t), is obtained through numerical integration
(Volesky et al., 2003). Dividing the metal mass adsorbed
(m
ad
) by the sorbent mass (M) leads to the uptake capa-
city (Q) of the biomass.
The breakthrough time (t
b
, the time at which metal
concentration in the euent reached 1 mgl
1
) and bed
exhaustion time (t
e
, the time at which metal concentra-
tion in the euent exceeded 99.5 mgl
1
) were used to
evaluate the mass transfer zone (Dt) given by Eq. (3):
Dt t
e
t
b
3
Total amount of metal sent to column (m
total
) can be cal-
culated from Eq. (4):
m
total

C
0
F t
e
1000
4
Total metal removal percent with respect to ow volume
can be calculated from Eq. (5):
Total metal removal %
m
ad
m
total
100 5
The metal mass desorbed (m
d
) can be calculated from
the elution curve (C versus t). The elution eciency
(E) can be calculated from Eq. (6):
E %
m
d
m
ad
100 6
3. Results and discussion
3.1. Eect of bed height
Breakthrough curves for biosorption of Cu, Co and
Ni by U. reticulata biomass at dierent bed heights are
shown in Fig. 1. In order to yield dierent bed heights,
3.57, 4.67 and 5.87 g of biomass were added to pro-
duce 15, 20 and 25 cm, respectively. Similar to copper,
Copper
0
20
40
60
80
100
0 4 8 12 16
Time (h)
C
o
p
p
e
r

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
0
20
40
60
80
100
C
o
b
a
l
t

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
0
20
40
60
80
100
N
i
c
k
e
l

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
Cobalt
0 5 10 15 20 25
Time (h)
0 5 10 15 20 25
Time (h)
Nickel
Fig. 1. Breakthrough curves for biosorption of copper, cobalt
and nickel by Ulva reticulata at dierent bed heights (ow
rate = 5 ml min
1
; initial metal concentration = 100 mgl
1
,
copper inuent pH = 5.5, cobalt inuent pH = 4.0, nickel
inuent pH = 4.5, temperature = 30 C). Bed height: (r)
15 cm, (j) 20 cm, (m) 25 cm.
K. Vijayaraghavan et al. / Chemosphere 60 (2005) 419426 421
the uptake of nickel also increased with an increase in
bed height from 15 to 25 cm. However, cobalt uptake
capacity of biomass remained relatively constant at dif-
ferent bed heights examined. For all the metal ions, the
breakthrough and exhaustion time increased with an in-
crease in bed height as more binding sites were available
for sorption, also resulted in a broadened mass transfer
zone (Table 1). The slope of the S-curve from t
b
to t
e
de-
creased as the bed height increased from 15 to 25 cm,
indicating the breakthrough curve becomes steeper as
the bed height decreased.
The BDST model is based on physically measuring
the capacity of the bed at dierent breakthrough values.
The column service time was selected as time when the
euent metal ion concentration reached 1 mgl
1
. For
all the metal ions, the plot of service time against bed
height at a ow rate of 5 ml min
1
(graph not presented)
was linear indicating the validity of BDST model. The
sorption capacity of the bed per unit bed volume, N
0
,
was calculated from the slope of BDST plot, assuming
initial concentration, C
0
, and linear velocity, m, as con-
stant during the column operation. The rate constant,
K
a
, calculated from the intercept of BDST plot, charac-
terizes the rate of solute transfer from the uid phase to
the solid phase. The computed N
0
was 2580, 2245 and
1911 mgl
1
for copper, cobalt and nickel, respectively.
The rate constant K
a
was recorded as 0.063, 0.081 and
0.275 l mg
1
h
1
for copper, cobalt and nickel, respec-
tively. The BDST model parameters can be useful to
scale up the process for other ow rates without further
experimental run.
3.2. Eect of ow rate
The eect of ow rate on the biosorption of copper,
cobalt and nickel ions was studied by changing the ow
rates from 5 to 20 ml min
1
at a xed bed height of
25 cm. Breakthrough curves for biosorption of Cu, Co
and Ni on the U. reticulata biomass at dierent ow
rates are shown in Fig. 2. In general, for all the metal
ions, the breakthrough curves became steeper and the
breakthrough time decreased with increasing ow rate
(Table 1). This behavior may be due to insucient resi-
dence time of the solute in the column and diusion lim-
itations of the solute into the pores of the sorbent at
higher ow rates (Ko et al., 2000). Even though more
shortened mass transfer zone (usually preferable) was
observed at the highest ow rate, the total metal removal
percentage (a reective of system performance) and the
metal uptake were actually observed maximum at the
lowest ow rate (5 ml min
1
).
The column data were tted to the Thomas model to
determine the model constant (k
Th
) and maximum solid-
phase concentration of solute (Q
0
) at dierent ow rates.
It is clear from Fig. 2 that the model gave a good t of
the experimental data at all ow rates examined for cop-
per with high correlation coecients greater than 0.993.
For cobalt and nickel, good ts were observed at higher
ow rates (10 and 20 ml min
1
). At the lowest ow rate
(5 ml min
1
), the predicted curve slightly deviated from
the experimental data. Table 2 illustrates the model con-
stants k
Th
and Q
0
along with correlation coecients. In
general as the ow rate increased, the values of k
Th
also
Table 1
Column data and parameters obtained at dierent bed heights and ow rates
Metal Bed height
(cm)
Flow rate
(ml min
1
)
Uptake
(mg g
1
)
t
b
(h)
t
e
(h)
Dt
(h)
(dc/dt)
a
(mgl
1
h
1
)
Total
removal (%)
Copper
b
15 5 51.6 0.21 3.4 9.4 6.0 17.95 65.5
20 5 51.9 1.06 4.5 12.9 8.4 12.72 63.4
25 5 56.3 0.24 7.3 15.1 7.8 14.69 72.6
25 10 53.8 0.45 3.9 8.3 4.4 24.98 63.7
25 20 53.7 0.69 1.8 4.1 2.3 46.16 64.7
Cobalt
c
15 5 45.2 0.23 2.9 13.7 10.8 9.90 39.4
20 5 45.8 0.01 4.3 15.6 11.3 8.19 40.4
25 5 46.1 0.07 5.3 22.1 16.8 5.60 40.7
25 10 42.0 0.17 2.8 9.7 6.9 12.97 40.5
25 20 37.9 0.2 1.3 4.2 2.9 30.53 40.3
Nickel
d
15 5 41.7 0.13 2.9 10.2 7.3 13.68 46.1
20 5 41.8 0.12 3.7 13.4 9.7 9.84 47.5
25 5 46.5 0.08 4.9 19.5 14.6 7.11 46.6
25 10 46.0 0.11 2.8 11.1 8.3 11.67 40.6
25 20 45.0 1.25 1.2 5.9 4.7 20.69 38.1
a
Slope of the breakthrough curve from t
b
to t
e
.
b
Conditions: inuent pH 5.5, temperature 30 C.
c
Conditions: inuent pH 4.0, temperature 30 C.
d
Conditions: inuent pH 4.5, temperature 30 C.
422 K. Vijayaraghavan et al. / Chemosphere 60 (2005) 419426
increased. It is evident, by comparing the experimental
Q values, that the model overestimated the metal uptake
capacity of U. reticulata at all ow rates. Similarly, Yan
et al. (2001) reported that Thomas model overestimated
Q
0
values of cadmium, lead and zinc biosorption by
immobilized Mucor rouxii.
3.3. Regeneration studies
In order to reduce the process costs, it is desirable to
reuse the biosorbent. The column regeneration studies
were carried out for three sorptiondesorption cycles at
5 ml min
1
. For all the metal ions, the column was
packed with 5.87 g of U. reticulata biomass yielding an
initial bed height of 25 cm and bed volume of 78.5 ml
with a packing density of 74.78 gl
1
. Table 3 summarizes
the breakthrough time, exhaustion time and column up-
take for all the cycles. At the end of third cycle, 5.31, 5.01
and 5.25 g of U. reticulata biomass was left in the column
after Cu, Co and Ni biosorption respectively.
As shown in Fig. 3, a decreased breakthrough time
and an increased exhaustion time were observed as the
regeneration cycles progressed, which also resulted in a
broadened mass transfer zone. This behavior is primar-
ily due to gradual deterioration of biomass because of
continuous usage. The overall performance of a packed
bed column is strongly related to the length and the
shape of the breakthrough curve. For copper biosorp-
tion, steep breakthrough curves were observed in all
the regeneration cycles indicating eective utilization
of the biosorbent material inside the column. Whereas
breakthrough proles obtained during cobalt and nickel
biosorption appeared to be more attened compared to
those obtained for copper biosorption, indicating a
wasteful long mass transfer zone inside the column.
The actual length of the bed was not constant during
the three regeneration cycles. It may be due to some sol-
uble constituents of the biomass dissolved during the
elution process. In spite of this weight loss, U. reticulata
retained its cobalt uptake capacity in all the three cycles
examined. Whereas copper and nickel uptake obtained
in the rst sorption cycle was never reached again in
any of the subsequent cycles. The uptake usually de-
pended on the previous elution step, since prolonged elu-
tion may destroy the binding sites or inadequate elution
may allow metal ions to remain in the sites.
The pH prole during the course of sorption process
was continuously monitored during the entire three
cycles (data not presented). In almost all the sorption
cycles, it was commonly observed that there was a sud-
den raise in euent pH in the early part of the process
(until the breakthrough occurred) and then pH tended
to decrease as the time progressed and eventually
reached the inuent pH. Ion exchange has been con-
rmed to be the major mechanism responsible in the
metal sequestering by algal biomass (Vieira and Volesky,
2000). Untreated marine algal biomass generally
contains light metal ions such as Na
+
, K
+
, Ca
2+
and
Mg
2+
. These ions were acquired from seawater and
bounded to the acidic functional groups of the alga.
When this native biomass exposed to metal solutions,
the release in light metal ions occurred and pH tended
to increase (Davis et al., 2003) and as the saturation of
bed proceeds the pH decreased.
For a systemof continuous operation to work success-
fully, the desorption process and agents must be eective
and should not cause much damage to the biosorbent.
Copper
0
20
40
60
80
100
0 4 8 12 16
Time (h)
C
o
p
p
e
r

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
C
o
b
a
l
t

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
Cobalt
0
20
40
60
80
100
N
i
c
k
e
l

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
0
20
40
60
80
100
0 5 10 15 20 25
Time (h)
0 5 10 15 20 25
Time (h)
Nickel
Fig. 2. Comparison of the experimental and predicted break-
through curves obtained at dierent ow rates according to the
Thomas model (bed height = 25 cm; initial metal concentra-
tion = 100 mgl
1
, copper inuent pH = 5.5, cobalt inuent
pH = 4.0, nickel inuent pH = 4.5, temperature = 30 C). Flow
rate: (r) 5 ml min
1
; (j) 10 ml min
1
; (m) 20 ml min
1
. (- - - -)
Predicted using Thomas model.
K. Vijayaraghavan et al. / Chemosphere 60 (2005) 419426 423
The elution curves of copper, cobalt and nickel through
a bed of U. reticulata biomass by 0.1 M CaCl
2
(in HCl,
pH 3) in all the three cycles are shown in Fig. 4. In all the
cycles, the elution curves exhibited a similar trend; a
sharp increase in the beginning followed by a gradual
decrease. As shown in Table 3, very high elution ecien-
cies were obtained for all the three metal ions and the
shortened time for elution process resulted in highly con-
centrated metal solutions. Among the three metal ions,
the time required for cobalt elution process was higher
and it resulted in high biomass weight loss. The pH pro-
le during the course of the elution process nearly
showed a trend similar that of the elution curve. For
all the cycles, it was generally observed that there is an
increase in pH in initial stages followed by gradual de-
crease in pH (data not shown).
The total volume of copper, cobalt and nickel bear-
ing solutions treated during this regeneration study
was 15.54, 20.88 and 18.15 l (in three cycles) respec-
tively. A shortened breakthrough time and a broad-
ened mass transfer zone were observed as the cycles
progressed, indicating a loss of sorption performance.
However, U. reticulata maintained a good metal sorp-
tion capacity in all the cycles examined. This observa-
tion clearly pointed out that the loss of sorption
performance was not mainly due to biomass damage
but rather because of sorbing sites, whose accessibility
becomes dicult as the cycles progressed (Volesky
et al., 2003).
4. Conclusion
1. U. reticulata was found to be an eective biosorbent
for the removal of copper, cobalt and nickel from
aqueous solutions.
Table 3
Sorption process parameters for three sorptiondesorption cycles
Metal Cycle no. Uptake
(mg g
1
)
t
b
(h)
t
e
(h)
Dt
(h)
(dc/dt)
a
(mgl
1
h
1
)
Bed height
(cm)
Total
removal (%)
Time for
elution (h)
Elution
eciency (%)
Copper
b
1 56.3 0.24 7.3 15.1 7.8 14.69 25.0 72.6 4.1 98.3
2 51.1 1.25 5.5 17.3 11.8 10.05 24.4 56.8 4.0 96.5
3 47.8 0.85 4.4 19.4 15.0 7.32 24.1 48.8 4.3 99.4
Cobalt
c
1 46.1 0.07 5.3 22.1 16.8 5.60 24.8 40.7 6.5 99.9
2 46.3 0.99 4.9 23.1 18.2 5.30 24.1 39.8 5.9 99.9
3 45.3 0.19 3.5 24.4 20.9 4.53 23.6 36.4 6.9 99.2
Nickel
d
1 46.5 0.08 4.9 19.5 14.6 7.11 24.9 46.6 4.4 98.5
2 41.5 1.84 4.5 19.7 15.2 6.76 24.5 39.9 5.0 99.5
3 42.0 2.05 3.8 21.3 17.5 5.84 23.9 39.9 4.6 98.0
a
Slope of the breakthrough curve from t
b
to t
e
.
b
Conditions: inuent pH 5.5, temperature 30 C, elutant 0.1 M CaCl
2
(pH 3).
c
Conditions: inuent pH 4.0, temperature 30 C, elutant 0.1 M CaCl
2
(pH 3).
d
Conditions: inuent pH 4.5, temperature 30 C, elutant 0.1 M CaCl
2
(pH 3).
Table 2
Parameters predicted from Thomas model at dierent ow rates
Metal Flow rate (ml min
1
) k
Th
(l mg
1
h
1
) Q
0
(mg g
1
) R
2
Copper
a
5 0.0111 55.7 0.993
10 0.0196 58.2 0.997
20 0.0313 59.5 0.996
Cobalt
b
5 0.0053 55.6 0.968
10 0.0099 54.1 0.986
20 0.0299 47.8 0.983
Nickel
c
5 0.0062 54.9 0.959
10 0.0076 54.7 0.991
20 0.0126 52.5 0.992
a
Conditions: inuent pH 5.5, temperature 30 C.
b
Conditions: inuent pH 4.0, temperature 30 C.
c
Conditions: inuent pH 4.5, temperature 30 C.
424 K. Vijayaraghavan et al. / Chemosphere 60 (2005) 419426
2. For all the metal ions examined, the increase in col-
umn bed height resulted in improved sorption perfor-
mance. The BDST model constants were determined
and proposed for the use in column design.
3. Due to mass transfer limitations and insucient time
for the solute inside the column, the increase in ow
rate decreased the sorption performance of U. reticu-
lata. The Thomas model was able to t the column
data at dierent ow rates and the model constants
were evaluated.
4. A successful biosorption process operation required
the multiple reuse of the sorbent, which greatly
reduce the process cost as well as decrease the depen-
dency of the process on continuous supply of the
sorbent. The sorption performance of U. reticulata
was evaluated in three sorptiondesorption cycles.
5. Thus, U. reticulata exhibited very high copper, cobalt
and nickel biosorption capacities in packed column
compared to most of the biosorbents reported in
the literature. This alga showed unique ability to
remove all three heavy metal ions and retain its
uptake capacity in three regeneration cycles. Thus,
Copper
0
20
40
60
80
100
0 5 10 15 20 25
Time (h)
0 5 10 15 20 25
Time (h)
0 5 10 15 20 25 30
Time (h)
C
o
p
p
e
r

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
0
20
40
60
80
100
C
o
b
a
l
t

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
0
20
40
60
80
100
N
i
c
k
e
l

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
Cobalt
Nickel
Fig. 3. Breakthrough curves for biosorption of copper, cobalt
and nickel by Ulva reticulata during three regeneration cycles
(bed height = 25 cm; ow rate = 5 ml min
1
, initial metal con-
centration = 100 mgl
1
, copper inuent pH = 5.5, cobalt inu-
ent pH = 4.0, nickel inuent pH = 4.5, temperature = 30 C).
Sorption cycle: (r) cycle 1; (j) cycle 2; (m) cycle 3.
Copper
0
400
800
1200
1600
2000
0 1 2 3 4 5 6 7
Time (h)
0 1 2 3 4 5
Time (h)
0 1 2 3 4 5 6
Time (h)
C
o
p
p
e
r

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
0
400
800
1200
1600
2000
C
o
b
a
l
t

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
0
400
800
1200
1600
N
i
c
k
e
l

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
g

l
-
1
)
Cobalt
Nickel
Fig. 4. Elution curves for copper, cobalt and nickel columns
using 0.1 M CaCl
2
(pH 3, HCl) during three regeneration cycles
(ow rate = 5 ml min
1
, temperature = 30 C). Elution cycle:
(r) cycle 1; (j) cycle 2; (m) cycle 3.
K. Vijayaraghavan et al. / Chemosphere 60 (2005) 419426 425
high biosorption eciency of the alga, low biomass
cost (mainly transportation cost), less dependency
on the biomass (due to reuse) and highly ecient
elutant makes this process an eective, cheap and
alternative technique for metal-bearing industrial
euents.
References
Aksu, Z., Go nen, F., 2003. Biosorption of phenol by immobi-
lized activated sludge in a continuous packed bed: predic-
tion of breakthrough curves. Proc. Biochem. 39, 599613.
Davis, T.A., Volesky, B., Mucci, A., 2003. A review of the
biochemistry of heavy metal biosorption by brown algae.
Water Res. 37, 43114330.
Davis, T.A., Volesky, B., Vieira, R.H.S.F., 2000. Sargassum
seaweed as biosorbent for heavy metals. Water Res. 34,
42704278.
Hutchins, R.A., 1973. New method simplies design of
activated carbon systems. Chem. Eng. 80, 133138.
Kaewsarn, P., 2002. Biosorption of copper (II) from aqueous
solutions by pre-treated biomass of marine algae Padina sp.
Chemosphere 47, 10811085.
Kapoor, A., Viraraghavan, T., Cullimore, D.R., 1999. Removal
of heavy metals using the fungus Aspergillus niger. Biores.
Technol. 70, 95104.
Ko, D.C.K., Porter, J.F., McKay, G., 2000. Optimised corre-
lations for the xed-bed adsorption of metal ions on bone
char. Chem. Eng. Sci. 55, 58195829.
Kuyucak, N., Volesky, B., 1989. Accumulation of cobalt by
marine alga. Biotechnol. Bioeng. 33, 809814.
Valdman, E., Erijman, L., Pessoa, F.L.P., Leite, S.G.F., 2001.
Continuous biosorption of Cu and Zn by immobilized waste
biomass Sargassum sp. Proc. Biochem. 36, 869873.
Veglio, F., Beolchini, F., Gasbarro, A., 1997. Biosorption of
toxic metals: an equilibrium study using free cells of
Arthrobacter sp. Proc. Biochem. 32, 99105.
Vieira, R.H.S.F., Volesky, B., 2000. Biosorption: a solution to
pollution? Int. Microbiol. 3, 1724.
Vijayaraghavan, K., Jegan, J., Palanivelu, K., Velan, M., 2004.
Copper removal from aqueous solution by marine green
alga Ulva reticulata. Electron. J. Biotechnol. 7, 6171.
Vijayaraghavan, K., Jegan, J., Palanivelu, K., Velan, M., in
press. Biosorption of cobalt(II) on marine macroalga Ulva
reticulata. 57th Annual Session of Indian Institute of
Chemical Engineers, Chemcon-2004, Mumbai, India, 27
30 December 2004.
Volesky, B., Prasetyo, I., 1994. Cadmium removal in a
biosorption column. Biotechnol. Bioeng. 43, 10101015.
Volesky, B., Weber, J., Park, J.M., 2003. Continuous ow
metal biosorption in a regenerable Sargasssum column.
Water Res. 37, 297306.
Yan, G., Viraraghavan, T., 2001. Heavy metal removal in a
biosorption column by immobilized M. rouxii biomass.
Biores. Technol. 78, 243249.
Yan, G., Viraraghavan, T., Chen, M., 2001. A new model for
heavy metal removal in a biosorption column. Adsorp. Sci.
Technol. 19, 2543.
Zeroual, Y., Moutaouakkil, A., Dzairi, F.Z., Talbi, M., Chung,
P.U., Lee, K., Blaghen, M., 2003. Biosorption of mercury
from aqueous solution by Ulva lactuca biomass. Biores.
Technol. 90, 349351.
Zulfadhly, Z., Mashitah, M.D., Bhatia, S., 2001. Heavy metals
removal in xed-bed column by the macro fungus Pycnop-
orus sanguineus. Environ. Pollut. 112, 463470.
426 K. Vijayaraghavan et al. / Chemosphere 60 (2005) 419426