Available online at www.sciencedirect.com

Colloids and Surfaces B: Biointerfaces 62 (2008) 83–90

Characterization of adsorption properties of extracellular polymeric

substances (EPS) extracted from sludge

Guo-Ping Sheng, Meng-Lin Zhang, Han-Qing Yu

∗

School of Chemistry, University of Science & Technology of China, Hefei, Anhui 230026, China

Received 27 May 2007; received in revised form 17 September 2007; accepted 17 September 2007

Available online 1 October 2007

Abstract

Extracellular polymeric substances (EPS) of sludge play an important role in the adsorption of organic pollutants in wastewater biological

treatments. Experiments were conducted to characterize the adsorption properties of EPS extracted from aerobic sludge (AE-EPS) and anaerobic
sludge (AN-EPS) using a dye-probing method in this study. A model cationic dye, Toluidine blue (TB), was used as the dye probe. The adsorption
of dye onto EPS to produce a dye–EPS complex would cause a change in the solution absorbance, attributed to the difference between the visible
spectra of the dye and dye–EPS complex. From the change in the absorbance, the equilibrium absorption capability of EPS could be evaluated.
Results indicate that Langmuir adsorption isotherm was able to adequately describe the adsorption equilibrium of TB onto both EPS at various pH
values. From the Langmuir adsorption isotherm, the maximum binding capabilities were calculated to be 1.9 and 2.5 mmol/g EPS for AE-EPS at
pH 7.0 and 11.0, and 1.6 and 1.9 mmol/g EPS for AN-EPS at pH 7.0 and 11.0, respectively. The first-order rate constants were calculated to be
0.033 and 0.35 min

−1

for AE-EPS at pH 7.0 and 11.0, and 0.069 and 0.18 min

−1

for AN-EPS at pH 7.0 and 11.0, respectively. The results of the

present study demonstrated that the dye-probing method was appropriate for investigating the adsorption process of EPS in aqueous solution.
© 2007 Elsevier B.V. All rights reserved.

Keywords: Adsorption; Dye; Extracellular polymeric substances (EPS); Langmuir; Toluidine blue

1. Introduction

In biological wastewater treatment, extracellular polymeric

substances (EPS) are produced by the microorganisms in aerobic
and anaerobic sludge when organic materials in wastewater are
consumed. EPS are a complex high molecular-weight mixture
of polymers excreted by microorganisms, produced from cell
lysis and hydrolysis, and adsorbed organic matter from wastew-
ater

[1]

. EPS are a major component of microbial aggregates

for keeping them together in a three-dimensional matrix due to
bridging with multivalent cations and hydrophobic interaction

[2]

. They are involved in the formation of microbial aggregates,

microbial adhesion to surfaces, bio-flocculation, settleability
and stability

[1,3–6]

. The bacterial cells also produced more

EPS to protect themselves against unfavorable conditions, such
as in the presence of toxic substances

[7]

. With the application

of numerous innovative analytical instruments, a large amount

∗

Corresponding author. Fax: [phone].

E-mail address:

[e-mail]

(H.-Q. Yu).

of information about the structural and functional properties
of EPS and their environmental behavior has been obtained.
EPS contain large quantities of negatively charged functional
groups and have strong capabilities to adsorb heavy metals and
organic pollutant

[1,7–9]

. However, few attempts have been

made to characterize the adsorption characteristics of EPS. This
is because EPS are dissolvable in water and cannot be separated
with the adsorbed substances through centrifugation, which is
commonly used in adsorption experiments for separating sor-
bent and adsorbed substances. In some studies polarography
method has been employed to evaluate the adsorption of heavy
metals onto EPS

[10]

. Equilibrium dialysis method

[11,12]

or

precipitation of EPS with dose of ice cold organic solvent

[13]

has also been used to separate EPS and the adsorbed substances.
However, these methods are time-consuming and inconvenient.
Furthermore, little information can be obtained concerning the
adsorption characteristics of organic substances onto EPS. A
new approach is highly desirable to explore the adsorption char-
acteristics of EPS.

Therefore, the main objective of this study was to develop

a dye-probing method for characterizing the adsorption

0927-7765/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:

10.1016/j.colsurfb.2007.09.024

84

G.-P. Sheng et al. / Colloids and Surfaces B: Biointerfaces 62 (2008) 83–90

Table 1
The physico-chemical characteristics of TB

Color index

CAS number

Structure

pK

a

Solubility (g/100 mL)

Water

Ethanol

52040

92319

11.8

3.82

0.57

characteristics between EPS and cationic dye. Since there are
a large number of negatively charged functional groups on the
EPS surface under neutral or alkali conditions, EPS could bind
with cationic dyes through the electrostatic interactions. Such
a binding would change the absorbance spectrum of the dye,
from which the adsorption characteristics of EPS would be eluci-
dated. The influences of different experimental parameters, such
as pH, adsorption time and EPS concentration, on the adsorption
were investigated. Moreover, the adsorption isotherm model and
kinetics were also explored. Information about such interactions
could be useful for understanding the adsorption characteristics
of the EPS from various origins.

2. Materials and methods

2.1. Sludge

The aerobic sludge was collected from an aeration tank in the

Wangxiaoying Municipal Wastewater Treatment Plant, Hefei,
China. The anaerobic methanogenic sludge was sampled from
a full-scale upflow anaerobic sludge blanket reactor treating
citrate-producing wastewater. This reactor was operated at a
hydraulic retention time of 12 h, pH of 6.9, and temperature
of 35

◦

C.

2.2. EPS extraction and chemical analysis

The EPS of the aerobic and anaerobic sludge (respectively

defined as AE-EPS and AN-EPS) were extracted using the cation
exchange resin (CER) technique (Dowex Marathon C, 20–50
mesh, sodium form, Fluka 91973)

[2,14]

. Both sludge samples

were harvested by centrifugation at 3000 rpm for 15 min, and
then the pellets were washed twice with 100 mM NaCl solution.
Later, the sludge pellets were re-suspended in NaCl solution
and the solution was transferred to an extraction beaker, fol-
lowed by the CER addition with a dosage of 60 g/g suspended

solids. These suspensions were stirred for 12 h at 200 rpm and
4

◦

C. After removing CER by settlement, the solutions were cen-

trifuged at 12,000 rpm and 4

◦

C for 30 min to remove remaining

sludge components. The supernatants were then filtered through
0.45-

␮m cellulose acetate membranes and used as the EPS frac-

tion for chemical analysis and adsorption experiments.

The content of carbohydrates was measured with the anthrone

method using glucose as a standard

[15]

, while the contents

of proteins and humic substances were determined using the
modified Lowry methods using chicken egg albumin and humic
acids as standards, respectively

[2]

. The total EPS content was

measured as the sum of these three substances. The total EPS
content was also measured as total organic carbon (TOC) using
a TOC analyzer (V

CPN

, Shimadzu Co., Japan). The suspended

solids and volatile suspended solids (VSS) of the sludge were
determined according to the Standard Methods

[16]

.

2.3. Chemicals

One model cationic dye, Toluidine blue (TB), was selected

as the dye probe in this study. It was purchased from Shang-
hai Chemical Co., China, and used without further purification.
The molecular formula of TB is C

15

H

16

ClN

3

S, and its physico-

chemical characteristics are shown in

Table 1

. This dye is a

phenothiazine class of water-soluble dye. At pH 7.0, almost
all of TB molecules are present in its ionic form, while at pH
11.0 about 86.3% of TB are in ionic form. It is stable and has
been widely used as a staining agent in biology, physiology and
medicine, as it can bind with the negatively charged groups on
the cell surface rapidly

[17]

.

2.4. Equilibrium modeling

In this study, two classical adsorption models, i.e., Langmuir

and Freundlich adsorption isotherms, were employed to describe
the TB adsorption onto AE-EPS and AN-EPS. The Langmuir

Table 2
The main compositions of AE-EPS and AN-EPS

Type

Carbohydrates
(mg/g VSS)

Proteins
(mg/g VSS)

Humic substances
(mg/g VSS)

Total EPS

a

(mg/g VSS)

TOC (mg
C/g VSS)

Proteins/carbohydrates

AE-EPS

3.2

± 0.2

9.2

± 0.2

8.4

± 0.2

20.8

± 0.6

12.6

± 0.3

2.9

AN-EPS

5.6

± 0.2

8.0

± 0.4

14.2

± 0.4

27.9

± 1.0

13.3

± 0.6

1.4

a

Expressed as the sum of carbohydrates, proteins and humic substances.

G.-P. Sheng et al. / Colloids and Surfaces B: Biointerfaces 62 (2008) 83–90

85

adsorption isotherm is given as follows:

n

eq

=

K

L

N

L

[L]

eq

1

+ K

L

[L]

eq

(1)

where K

L

is the binding constant related to the affinity of the

binding sites; N

L

, the maximum amount of the dye per unit

weight of biomass to form a complete monolayer on the sur-
face bound, i.e., maximum binding capability of EPS; [L]

eq

,

the equilibrium concentration of the free dye; n

eq

, the amounts

of adsorbed dye per gram EPS at equilibrium, i.e., the average
binding number of dye onto EPS:

n

eq

=

[ML

n

]

eq

C

M

(2)

in which [ML

n

]

eq

is concentration of the dye bound with EPS at

equilibrium. C

M

represents the total analytical concentration of

EPS. From the non-linear regression between n

eq

and [L]

eq

, the

values of K

L

and N

L

could be estimated.

The essential features of the Langmuir isotherm can be

expressed in terms of a dimensionless constant separation factor
or equilibrium parameter, R

L

[18]

:

R

L

=

1

1

+ K

L

C

L

(3)

where C

L

is the initial dye concentration. The R

L

values within

the range of 0 < R

L

< 1 indicate favorable adsorption.

The Freundlich equation is given below:

n

eq

= K

F

[L]

1/n

F

eq

(4)

where K

F

(L/g) and n

F

are the Freundlich constants, accounting

for the adsorption density and adsorption intensity respectively.
The values of K

F

and n

F

can be estimated from the non-

regression between n

eq

and [L]

eq

.

2.5. Dye-probing analysis

The dye-probing method was used to evaluate the bind-

ing capability. With the absorption of dye to macromolecules
(e.g., EPS), the dye solution absorbance changed correspond-
ingly with the addition of macromolecules

[19]

. From such

a change, the values of n

eq

and [L]

eq

could be obtained. The

absorbance at a certain wavelength (A) was attributed to both
TB and TB–EPS complex, and it could be analyzed according
to Beer’s law.

For TBEPS mixture : A = ε

1

b[L] + ε

2

b[ML

n

]

(5)

For TB solution (without EPS addition) : A

0

= ε

1

bC

L

(6)

where b is the optical length, ε

1

, ε

2

are the extinction

coefficients of the free dye and bound dye, respec-
tively.

C

L

is

the

total

analytical

concentration

of

TB:

C

L

= [L] + [ML

n

]

(7)

Thus:

ΔA = A

0

− A = (ε

1

− ε

2

)b[ML

n

]

= Δεb[ML

n

]

(8)

Fig. 1. Effect of reaction time on the absorbance at 630 nm of the EPS–TB
complex solutions at (a) pH 7.0 and (b) pH 11.0.

The A value at a certain wavelength could be obtained from

the spectrum difference of the TB and TB–EPS solutions. The
values of ε

1

and ε

2

can be determined from the absorbance of

solutions in the absence and presence of EPS, thus ε can be
calculated.

Rearrangement of Eq.

(8)

gives:

[ML

n

]

=

A

bε

(9)

At equilibrium, substitution of A by A

eq

(the change of

absorbance at equilibrium) and introduction of Eq.

(9)

into Eqs.

(2)

and

(7)

give:

n

eq

=

A

eq

bεC

M

(10)

[L]

eq

=

C

L

− A

eq

bε

(11)

When the TB or EPS concentrations were altered, the values

of n

eq

and [L]

eq

would be changed correspondingly, and they

could be calculated from Eqs.

(10)

and

(11)

. From the non-

linear regression between [L]

eq

and n

eq

using Eqs.

(1)

and

(3)

,

the Langmuir and Freundlich constants of both EPS could be
calculated.

In this study, a first-order model was used to fit the experi-

mental data. The first-order rate expression of Lagergren model
is expressed as follows

[20]

:

dn

dt

= k

1

(n

eq

− n)

(12)

86

G.-P. Sheng et al. / Colloids and Surfaces B: Biointerfaces 62 (2008) 83–90

Fig. 2. Effect of AE-EPS concentration on the spectra of complex solution at various pH. (a and b) pH 7.0; (c and d) pH 11.0. The concentration of TB added was
0.15 mM.

where k

1

is the first-order rate constant, n and n

eq

are the average

binding number of dye on EPS in the adsorption process and at
equilibrium, respectively.

The integrated form of Eq.

(11)

is:

n

n

eq

= 1 − e

−k

1

t

(13)

Introduction of Eq.

(10)

to Eq.

(13)

gives:

n

n

eq

=

A

A

eq

= 1 − e

−k

1

t

(14)

2.6. Adsorption experiments

The surface functional groups of EPS could be ionized and the

negative charges of EPS would depend highly on the solution
pH. In this study, to maintain the solution pH at a constant,
two phosphatic buffers (0.05 M, pH 7.0 and 11.0) were used.
TB solution at a pre-determined concentration (0–0.15 mM) of
1.0 mL and phosphatic buffer of 3.0 mL were added into a 10-mL
tube. The solution was then mixed, and 2-mL EPS samples were
added and mixed immediately using a vortex mixer. Thereafter,
the tube was placed at 25

◦

C for a pre-determined time before the

measurement of visible spectra using double distilled water as
reference. The visible spectra of TB–EPS mixture were scanned
from 400 to 800 nm using a double-beam spectrophotometer
(UV-4500, Shimadzu Co., Japan) with 1-cm cuvette. To obtain
the spectra of TB solution, 2-mL double distilled water was
added to the tube containing 1.0 mL TB solution and 3.0 mL
buffer, and then the mixture was scanned from 400 to 800 nm
against double distilled water after placed at 25

◦

C for a pre-

determined time. The spectrum difference was calculated by
subtracting the TB–EPS spectra from TB spectra at the same
TB concentration. The value of A

eq

at a certain wavelength

could be obtained from the curve of the spectrum difference.

3. Results

3.1. Composition of AE-EPS and AN-EPS

As listed in

Table 2

, carbohydrates, proteins and humic sub-

stances were the main components of both EPS. The component
variation implies that the content of EPS depended on the origins
of sludge. The total contents of AE-EPS and AN-EPS, expressed
as the sum of the contents of carbohydrates, proteins and humic
substances, were 20.8 and 27.9 mg/g VSS, respectively. The
ratios of proteins to carbohydrates were 2.9 for AE-EPS and
1.4 for AN-EPS, which were in accord with the results reported
previously

[1]

.

3.2. Effect of adsorption time on the absorbance of solution
at 630 nm

In order to know the TB adsorption equilibrium time to both

EPS, experiments were conducted to evaluate the TB adsorp-
tion rates by AE-EPS and AN-EPS at various pH values and
25

◦

C.

Fig. 1

illustrates the absorbance of TB and EPS mixture

at 630 nm as a function of reaction time. At pH 7.0, the solu-
tion absorbance at 630 nm would not change significantly after
90-min adsorption. However, the adsorption of TB onto both
EPS was a rapid process at pH 11.0. As shown in

Fig. 1

b, at

G.-P. Sheng et al. / Colloids and Surfaces B: Biointerfaces 62 (2008) 83–90

87

Fig. 3. Effect of AN-EPS concentration on the spectra of complex solution at various pH. (a and b) pH 7.0; (c and d) pH 11.0. The concentration of TB added was
0.15 mM.

pH 11.0, after 30 min the absorbance of the mixture solution
decreased to the minimum, and remained almost unchanged in
the subsequent 1 h. Thus, the adsorption equilibrium time was
set as 90 and 30 min in the subsequent tests at pH 7.0 and 11.0,
respectively.

3.3. Effect of EPS concentration on the visible spectra of
TB–EPS complex

The effect of the AE-EPS concentration on the changes of

visible spectra of TB and TB–EPS solution at various pH values
was recorded from 400 to 800 nm (

Fig. 2

). The TB concentration

was fixed at 0.15 mM. As the EPS concentration increased, the
extent of the changes of visible spectra between TB and TB–EPS
solution also increased, and leveled off thereafter (

Fig. 2

a and

c). The changes in absorbance at 630 nm (A

630

) were the most

significant. As shown in

Fig. 3

b and d, with an increase in EPS

concentration, the A

630

increased at initial; as AE-EPS concen-

tration exceeded 60 mg/L, the changes in A

630

became slight

and leveled off thereafter. Furthermore, the changes in A

630

for EPS were more significant at pH 11.0 than those at pH 7.0,

indicating the differences in the surface functional groups of
EPS at various pH values.

Fig. 3

shows the effect of the AN-EPS concentration on the

changes of visible spectra of TB and TB–EPS solution at various
pH values from 400 to 800 nm. The changing trend of visible
spectra for AN-EPS was the same as that for AE-EPS. With
an increase in AN-EPS concentration, the A

630

increased ini-

tially; as AN-EPS concentration exceeded 80 mg/L, the changes
in A

630

became slight and leveled off thereafter. Also, the

changes in A

630

at pH 7.0 were lower than those at pH 11.0.

Compared with AE-EPS, the changes in A

630

for AN-EPS

were lower at the same pH, suggesting the different functional
group content between AE-EPS and AN-EPS.

3.4. Equilibrium modeling

In this study, the Langmuir and Freundlich adsorption

isotherms were employed to describe the TB adsorption onto
AE-EPS and AN-EPS at various pH values (

Figs. 4 and 5

). The

Langmuir and Freundlich adsorption constants for both EPS
were calculated and are listed in

Table 3

. The high regression

Table 3
The adsorption parameters and standard deviations

pH

Langmuir constants

Freundlich constants

k

1

(min

−1

)

K

L

× 10

5

L/mol

N

L

(mmol/g)

R

L

R

2

K

F

(L/g)

n

F

R

2

AE-EPS

7

5.0 (0.6)

1.9 (0.1)

0.07 (0.01)

0.952

4.7 (0.7)

4.0 (0.5)

0.896

0.033 (0.001)

11

1.8 (0.3)

2.5 (0.2)

0.18 (0.03)

0.975

11.2 (1.6)

2.3 (0.2)

0.979

0.35 (0.01)

AN-EPS

7

2.4 (0.5)

1.6 (0.1)

0.14 (0.03)

0.903

5.3 (1.4)

2.9 (0.5)

0.826

0.069 (0.005)

11

2.3 (0.3)

1.9 (0.1)

0.15 (0.02)

0.957

8.5 (0.8)

2.4 (0.1)

0.938

0.18 (0.01)

88

G.-P. Sheng et al. / Colloids and Surfaces B: Biointerfaces 62 (2008) 83–90

Fig. 4. Langmuir and Frendlich adsorption isotherm of TB adsorption onto
AE-EPS at (a) pH 7.0 and (b) pH 11.0.

coefficients of determination indicate that it is feasible to use
the Langmuir model to describe the adsorption of TB to the
EPS at various pH values. It also suggests that the limited and
site-specific adsorption is associated with the bonding forces
between the positive charge in TB and negative charge in EPS.

As shown in

Table 3

, the Langmuir equilibrium constants

of the adsorption of TB onto the AE-EPS were 5.0

× 10

5

and 1.8

× 10

5

L/mol at pH 7.0 and 11.0, respectively, while

the corresponding values for AN-EPS were 2.4

× 10

5

and

2.3

× 10

5

L/mol at pH 7.0 and 11.0, respectively. However, the

maximum binding capabilities of both EPS at various pH values
were different. For AE-EPS, they were 1.9 and 2.5 mmol/g EPS
at pH 7.0 and 11.0, and 1.6 and 1.9 mmol/g EPS for AN-EPS
at pH 7.0 and 11.0, respectively. The calculated R

L

value was

found to be 0.07 and 0.18 for AE-EPS at pH 7.0 and 11.0, and
0.14 and 0.15 for AN-EPS at pH 7.0 and 11.0, respectively, at the
TB concentration of 0.15 mM. This indicates that the adsorption
of TB to EPS was favorable.

In the Freundlich expression, K

F

and n

F

are related respec-

tively with the adsorption density and intensity. As listed in

Table 3

, the values of n

F

for both EPS at various pH values

were close, indicating that the adsorption mechanisms of TB
onto both EPS were same, while the K

F

values for both EPS at

pH 11.0 were larger than those at pH 7.0, suggesting that the
binding capabilities of these two EPS depended on solution pH.
The difference in the K

F

values for both EPS also suggests the

differences in the binding capabilities and the surface functional
groups of both EPS.

3.5. Kinetic analysis on the adsorption of TB onto EPS

In this study, a first-order model was used to fit the experi-

mental data. As shown in

Fig. 6

, from a non-linear regression

between A/A

eq

and time t using Eq.

(14)

, the rate constant k

1

was calculated and found to depend on the pH. The values were
0.033 and 0.069 min

−1

for AE-EPS and AN-EPS at pH 7.0, and

0.35 and 0.18 min

−1

at pH 11.0, respectively (

Table 3

). The high

coefficients of determination (>0.9) indicate that the assumption
of first-order kinetic model was appropriate for describing the
adsorption of TB onto both EPS.

4. Discussion

4.1. Significance of the dye-probing approach

The pollutant removal process in a biological wastewater

treatment system involves flocculation, adsorption, and bio-
logical oxidation. Both aerobic and anaerobic sludge has a
significant ability to adsorb organic pollutants and heavy met-
als. Because of vast surface areas and plenty of functional
groups, EPS have a substantial effect on the adsorption abil-
ity of sludge and play an important role in the adsorption of
organic pollutants in wastewater biological treatments. It is
also generally recognized that bacterial EPS can accumulate
nutrients from wastewater. Such an accumulation may be an
adaptive mechanism which enables microbial communities to
optimize utilization of carbon and nutrient sources

[21]

. Thus,

Fig. 5. Langmuir and Frendlich adsorption isotherm of TB adsorption onto
AN-EPS at (a) pH 7.0 and (b) pH 11.0.

G.-P. Sheng et al. / Colloids and Surfaces B: Biointerfaces 62 (2008) 83–90

89

Fig. 6. First-order kinetic model for TB adsorption onto EPS at (a) pH 7.0 and
(b) pH 11.0.

the measurement of EPS adsorption characteristics is benefi-
cial for elucidating the mechanisms for biological wastewater
treatment and understanding the precise functions of sludge
EPS.

EPS are dissolved in aqueous solutions and cannot be read-

ily separated with the adsorbed substances, which makes it
difficult to determinate their binding capacity. In this study, a
dye-probing method was used for a rapid determination of EPS
adsorption characteristics for cationic dye. The main advantage
of this approach was that such characterization could be read-
ily performed without the separation of sorbent (i.e., EPS) from
solutions. The adsorptive capacity was determined based on the
difference between the visible spectra of small molecular dye

and dye–EPS complex. This approach used in this study was
proven to be appropriate for studying the adsorption process of
EPS in aqueous solution.

4.2. Maximum binding capabilities and strength of EPS

Both maximum binding capability and strength of EPS could

be characterized with this dye-probing method. From the Lang-
muir adsorption isotherm, the maximum binding capabilities of
both EPS were calculated. It was 1.9 and 2.5 mmol/g EPS for
AE-EPS at pH 7.0 and 11.0, respectively, while only 1.6 and
1.9 mmol/g EPS for AN-EPS at pH 7.0 and 11.0. This implies
that the binding capabilities of both EPS at pH 7.0 were much
less than those at pH 11.0, and that the binding capability of both
EPS depended on pH. Furthermore, the binding capabilities of
AE-EPS were higher than those of AN-EPS at both pH values,
suggesting the binding capability of AN-EPS was much less than
that of AE-EPS. The difference in the binding capability of EPS
at various pH values might be related to the content of the ionized
functional groups of EPS. There are many carboxyl (pK

a

= 4.8),

phosphoric (pK

a

= 6.0–7.0), sulfhydryl (pK

a

= 8.2) and pheno-

lic acidic (pK

a

= 9.4–9.8) groups in EPS

[22]

. At pH 7.0, the

carboxyl and phosphoric groups would be ionized, while at pH
11.0, all of these groups are mostly ionized

[10,22]

. Through

electrostatic interaction TB could be adsorbed to these ionized
functional groups of EPS. Thus, the potential adsorbed sites of
EPS at pH 7.0 would be less than those at pH 11.0, resulting in a
low binding capability at pH 7.0 for both EPS. It should be noted
that at pH 11.0, hydrolysis of EPS molecules may occur. In this
case, the disulfide bindings in glycoproteins could be broken

[1]

.

This might also increase the functional groups of EPS.

The difference in the chemical compositions of AE-EPS

and AN-EPS would result in the different binding capabil-
ities of EPS. For AE-EPS, the percentage compositions of
carbohydrates, proteins and humic substances were 15.4%,
44.2% and 40.4%, respectively, while for AN-EPS, they were
20.1%, 28.7% and 50.9%, respectively (

Table 2

). The percent-

age composition of proteins in AE-EPS was higher than that in
AN-EPS, while the contents of carbohydrates and humic sub-
stances in AE-EPS were lower than those in AN-EPS. This
might result in the higher binding capability of AE-EPS than
AN-EPS. This also implies proteins have more binding sites

Table 4
Cationic dye adsorption results in this work and literature

Dyes

Absorbents

Maximum capacity (mmol/g)

Reference

Toluidine blue

AE-EPS

1.9–2.5

Present study

AN-EPS

1.6–1.9

Auramine O

Poly (

␥-glutamic acid)

0.9

[20]

Rhodamine B

0.8

Safranin O

1.4

Astrazon brilliant red 4G

Granular activated carbon

1.4–1.9

[24]

Methylene Blue

Dead macro fungi

0.6–0.7

[25]

Rhodamine B

0.05–0.08

Toluidine blue

Fly ash

0.02

[26]

Methylene blue

Pyrophyllite

0.2

[27]

Toluidine blue

Modificated fuller’s earth

0.7–3.7

[28]

Crystal violet

Saw dust

0.8

[29]

90

G.-P. Sheng et al. / Colloids and Surfaces B: Biointerfaces 62 (2008) 83–90

than carbohydrates and humic substances, and demonstrates the
important role of proteins in the adsorption characteristics of
EPS

[23]

.

4.3. Comparison with other adsorbents for cationic dye
adsorption

Many efforts have been made to study the adsorption between

various cationic dyes and various adsorbents.

Table 4

lists the

maximum binding capabilities of some adsorbents for cationic
dye reported in literatures. Compared with other adsorbents, EPS
have a high binding capacity. For instance, EPS have a higher
binding capacity for TB adsorption than fly ash. Such a high
binding capacity of EPS is attributed to the plenty of functional
groups in EPS molecules.

Acknowledgements

The authors wish to thank the Natural Science Foundation of

China (20577048 and 50625825), the China Postdoctoral Sci-
ence Foundation (20060400206), and K. C. Wong Education
Foundation, Hong Kong for the partial support of this study.

References

[1] J. Wingender, T.R. Neu, H.C. Flemming, Microbial Extracellular Polymeric

Substances: Characterization, Structures and Function, Springer-Verlag,
Berlin Heidelberg, 1999.

[2] B. Frolund, R. Palmgren, K. Keiding, P.H. Nielsen, Water Res. 30 (1996)

1749.

[3] A. Saa, O. Teschke, J. Colloid Interf. Sci. B. Q. 304 (2006) 554.
[4] S. Tsuneda, J. Jung, H. Hayashi, H. Aikawa, A. Hirata, H. Sasaki, Colloid

Surf. B-Biointerf. 29 (2003) 181.

[5] G.P. Sheng, H.Q. Yu, X.Y. Li, Biotechnol. Bioeng. 93 (2006) 1095.

[6] B.Q. Liao, D.G. Allen, G.G. Leppard, I.G. Droppo, S.N. Liss, J. Colloid

Interf. Sci. 249 (2002) 372.

[7] G.P. Sheng, H.Q. Yu, Z.B. Yue, Appl. Microbiol. Biotechnol. 69 (2005)

216.

[8] V. Guine, L. Spadini, G. Sarret, M. Muris, C. Delolme, J.P. Gaudet, J.M.F.

Martins, Environ. Sci. Technol. 40 (2006) 1806.

[9] J.C. Finlayson, B. Liao, I.G. Droppo, G.G. Leppard, S.N. Liss, Water Sci.

Technol. 37 (1998) 353.

[10] G. Guibaud, S. Comte, F. Bordas, S. Dupuy, M. Baudu, Chemosphere 59

(2005) 629.

[11] D.L. Parker, J.E. Mihalick, J.L. Plude, M.J. Plude, T.P. Clark, L. Egan, J.J.

Flom, L.C. Rai, H.D. Kumar, J. Appl. Phycol. 12 (2000) 219.

[12] J.L. Geddie, I.W. Sutherland, J. Appl. Bacteriol. 74 (1993) 467.
[13] N. Singh, R.K. Asthana, S.P. Singh, World J. Microbiol. Biotechnol. 19

(2003) 851.

[14] G.P. Sheng, H.Q. Yu, Water Res. 40 (2006) 1233.
[15] K. Raunkjær, T. Hvitved-Jacobsen, P.H. Nielsen, Appl. Microbiol. Biotech-

nol. 28 (1994) 251.

[16] APHA, Standard Methods for the Examination of Water and Wastewater,

19th ed., American Public Health Association, USA, 1995.

[17] L.H. Goldberg, S.Q. Wang, A. Kimyai-Asadi, Dermatol. Surg. 33 (2007)

761.

[18] G. Mckay, H.S. Blair, J.R. Gardner, J. Appl. Polym. Sci. 27 (1982) 3043.
[19] G.P. Sheng, M.L. Zhang, H.Q. Yu, Anal. Chim. Acta. 592 (2007) 162.
[20] B.S. Inbaraj, J.T. Chien, G.H. Ho, J. Yang, B.H. Chen, Biochem. Eng. J.

31 (2006) 204.

[21] G.M. Wolfaardt, J.R. Lawrence, R.D. Robarts, E.D. Caldwell, Appl. Env-

iron. Microbiol. 61 (1995) 152.

[22] H. Liu, H.H.P. Fang, Biotechnol. Bioeng. 80 (2002) 806.
[23] H.C. Flemming, Water Sci. Technol. 32 (1995) 27.
[24] P.C.C. Faria, J.J.M. Orfao, M.F.R. Rereira, Water Res. 38 (2004) 2043.
[25] N.S. Maurya, A.K. Mittal, P. Cornel, E. Rother, Bioresour. Technol. 97

(2006) 512.

[26] R.Y. Talman, G. Atun, Colloid Surf. A: Physicochem. Eng. Aspects 281

(2006) 15.

[27] A. Gucek, S. Sener, S. Bilgen, M.A. Mamano, J. Colloid Interf. Sci. 286

(2005) 53.

[28] G. Hisarli, J. Colloid Interf. Sci. 281 (2005) 18.
[29] S. Chakraborty, S. De, S. DasGupta, J.K. Basu, Chemosphere 58 (2005)

1079.