Journal of Colloid and Interface Science 304 (2006) 18

www.elsevier.com/locate/jcis
Thermosensitive-polymer-coated magnetic nanoparticles:
Adsorption and desorption of Bovine Serum Albumin
N. Shamim, L. Hong, K. Hidajat, M.S. Uddin

Department of Chemical and Biomolecular Engineering, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260
Received 7 May 2006; accepted 24 August 2006
Available online 28 September 2006
Abstract
Adsorption and desorption behavior of Bovine SerumAlbumin (BSA) on surface-modied magnetic nanoparticles covered with thermosensitive
polymer (PNIPAM) was investigated as a function of temperature, pH, and ionic strength. Functionalization of surface-modied magnetic particles
was performed by seed polymerization using N-isopropylacrylamide (PNIPAM) as the main monomer. Characterization of these particles was
carried out using transmission electron micrography (TEM), and vibrating sample magnetometry (VSM). The adsorption results exhibited both
pH and temperature sensitivity. The results showed that the temperature effect on adsorption/desorption behavior was mainly dependent on the
properties of the particles surface. The effect of pH was also investigated and it was observed that a smaller amount of protein was adsorbed
at higher pH because of the electrostatic repulsive force between protein molecules and latex particles. The maximum amount of protein was
adsorbed near the isoelectric point of BSA. Desorption results showed that more protein was desorbed when adsorption was done at lower
temperatures and desorption efciency was found to be higher than 80%.
2006 Elsevier Inc. All rights reserved.
Keywords: Thermosensitive magnetic particles; N-isopropylacrylamide; Adsorption; Desorption; Isoelectric point; Bovine Serum Albumin
1. Introduction
In recent years, much interest has been focused on the de-
sign of smart and intelligent polymeric materials for techno-
logical applications and fundamental studies. These materi-
als can respond with shape and volume changes to small ex-
ternal stimuli, such as temperature, pH, ionic strength, and
magnetic elds. Among these intelligent polymeric materials,
N-isopropylacrylamide is the most widely studied thermosen-
sitive polymer. Poly(N-isopropylacrylamide) (PNIPAM) has a
lower critical solution temperature (LCST) of 32

C in water,
which is very close to room temperature. It changes from hy-
drophilic below the LCST to hydrophobic above it, due to the
reversible formation and cleavage of hydrogen bonds between
the amide groups and the surrounding water molecules [1].
This reversible thermosensitivity can be used in biomedical
and biological applications such as enzyme immobilization,
*
Corresponding author.
E-mail address: cheshahb@nus.edu.sg (M.S. Uddin).
cell sorting [25], protein adsorption and purication [6,7], and
drug delivery [8]. In addition, such thermosensitive polymers
may provide a nondenaturing environment for the immobilized
protein, which is particularly suitable for immunoassays and
protein concentration and purication. The main advantage of
using smart polymeric particles in bioseparation is that the sep-
aration process can be operated without large changes in envi-
ronmental factors. Thus, mild conditions for biomolecules can
be maintained during the entire separation process.
Smart polymeric lattices can be used as absorbents for pro-
tein separation because they have a large surface area and their
surface properties, such as hydrophobicity and hydrophilic-
ity, can be varied quite easily. Ding et al. [9] and Elaissari
and Bourrel [10] studied adsorption and desorption of pro-
tein on sub-micrometer-sized thermosensitive-polymer-coated
magnetic particles. Suitable desorption agents selected to re-
lease the adsorbed protein from magnetic nanoparticles were
acidic conditions [11], alkaline conditions [12], ionic strength
(salt solution) [1315], and temperature [10,16].
Some work has been published on the adsorption and de-
sorption of protein on sub-micrometer-sized thermosensitive-
0021-9797/$ see front matter 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcis.2006.08.047
2 N. Shamim et al. / Journal of Colloid and Interface Science 304 (2006) 18
polymer-coated particles, but only limited work has been pub-
lished on adsorption and desorption of proteins on nanosized
surface-modied magnetic particles. In this work we prepared
thermosensitive-(PNIPAM) polymer-coated magnetic particles
and these particles were used as a support for adsorbing Bovine
Serum Albumin (BSA) under different conditions of tempera-
ture, pH, and ionic strength. Desorption of BSA was also stud-
ied as a function of temperature and ionic strength.
2. Materials and methods
2.1. Materials
The chemicals used in this study and their sources were
as follows: iron(II) chloride tetrahydrate (99% purity): Fisher
(USA). Iron(III) chloride hexahydrate (98%): Nacalai Tesque
(Japan). Ammonium hydroxide (25%): Merck (USA). Thio-
diglycolic acid (98%), 4-vinylaniline (97%), N-isopropylacryl-
amide (97%), N,N-methylene bis-acrylamide, potassium per-
sulfate (99%), 2-mercaptoethanol (98%), methacrylic acid
(99%) and Bovine Serum Albumin (BSA): Sigma-Aldrich
(USA).
2.2. Methods
2.2.1. Preparation of surface-modied nanomagnetic particles
Magnetic particles were prepared by a chemical co-precipi-
tation method in an inert environment. Complete precipitation
of Fe
3
O
4
was achieved under alkaline condition and the molar
ratio was maintained at Fe
2+
:Fe
3+
= 1:2. In a typical synthe-
sis to obtain 1 g of Fe
3
O
4
precipitate, 0.86 g of FeCl
2
4H
2
O
and 2.35 g of FeCl
3
6H
2
O were dissolved under N
2
in 40 ml
of deaerated Milli-Q water with vigorous stirring at a speed
of 1000 rpm. As the solution was heated to 80

C, 100 mg of
thiodiglycolic acid was added, followed by 5 ml of NH
4
OH.
Further, thiodiglycolic acid (TDGA) was added to the suspen-
sion in ve 0.2-g amounts over a period of 5 min. The exper-
iment was continued for 30 min at 80

C. 4-Vinylaniline was
used as a secondary surfactant. In a typical procedure, 1 g of
the fresh precipitate obtained from the previous preparation was
combined with 20 ml of Milli-Q water and the mixture was
heated to about 50

C under vigorous stirring. About 0.2 ml of

4-vinylanile was added dropwise using a syringe. The experi-
ment was carried out for another 30 min. The stable water-based
suspension was then cooled to room temperature and washed
using Milli-Q water. The precipitates were isolated from the
solvent by magnetic decantation. The washing-decantation pro-
cedure was repeated ve times to eliminate free electrolyte and
remove the excess surfactants used for the coating.
These double-surfactant-coated nanomagnetic particles were
then used for seed polymerization using 0.3 g N-isopropylacryl-
amide (NIPAM) as the main monomer, 0.03 g N,N-methylene
bis acrylamide (MBA) as the cross-linker, and 0.006 g potas-
sium persulfate (KPS) as the initiator. 2-Mercaptoethanol was
used as the chain transfer reagent. MBA was also used to in-
crease the mechanical strength of the particles. The reaction
was allowed to proceed for 6 h at 70

C under continuous me-

chanical stirring at 150180 rpm. The reaction was conducted
at a higher temperature than the LCST of the polymer in to pro-
mote the precipitation of oligomers formed in aqueous solution.
The particles were washed at least three times before adsorp-
tion experiments were carried out. The reaction procedures of
thermosensitive nanomagnetic particles are shown step by step
in Fig. 1.
2.2.2. Characterization of surface-modied magnetic particles
The size and morphology of the freeze-dried magnetic par-
ticles were determined by TEM. A bright-eld TEM (Model
JEM 2010) was used for the size measurement. TEM samples
were prepared by coating a copper grid (200 mesh and covered
with Formvar/carbon) with a thin layer of dilute particle suspen-
sion. The copper grid was then dried at room temperature for
24 h before the measurement. Freeze-dried magnetic particles
were used for VSM measurements (Model 1606, DMS). The
zeta potentials of bared and thermosensitive nanomagnetic par-
ticles were measured at different pH using a Brookhaven Zeta
Plus 90 analyzer. Samples were prepared by diluting 20 mg of
both particles in 10
3
M NaNO
3
solution at different pHs ad-
justed with diluted HNO
3
and NaOH.
2.2.3. Adsorption/desorption of BSA on
thermosensitive-polymer-coated magnetic particles
Adsorption of BSA onto thermosensitive-(PNIPAM) poly-
mer-coated magnetic particles was carried out by mixing 5 ml
of BSA solution of a certain concentration and 130 mg of wet
sample particles. The mixture was incubated above and be-
low the lower critical solution temperature (LCST) of PNIPAM
(LCST 32

C) at 30 and 40

C in 10 mM phosphate buffer so-

lution for about 2 h. The solid content of the wet magnetic
particles was measured to be 15.6%.
The effect of temperature on protein adsorption on NIPAM-
coated nanomagnetic particles was studied at 40 and 30

C at a
pH of 4.7 and an ionic strength of 0.01 M. BSA concentration
in the supernatant was measured by UV spectrometer analysis
(Shimadzu UV 1601 PC) at 280 nm. The effect of pH was stud-
ied in the pH range of 3.368.18 at 40

C and ionic strength

0.01 M. To investigate the inuence of ionic interactions on
the protein adsorption at a given pH and temperature, selected
concentrations (0.05, 0.1, 0.5, 1.0, 1.5 M) of electrolyte [NaCl]
were added to the buffers.
Desorption was carried out under alkaline conditions using
0.5 M Na
2
HPO
4
(pH 9.35) when adsorption was carried out in
acidic medium. On the other hand, 0.05 M NaH
2
PO
4
(pH 4.25)
was used as a desorption medium when protein was adsorbed
under basic conditions. After equilibrium was achieved for ad-
sorption, the supernatant was separated from the latex particles
by the help of a magnet. Then the particles were washed using
Milli-Q water. Magnetic particles containing BSA were then
mixed with 5 ml of Na
2
HPO
4
solution. After 2 h of incuba-
tion below the lower critical solution temperature, 20

C, the
supernatant was collected and analyzed in a UV spectrometer.
Desorption was studied as a function of temperature and salin-
ity in the desorption medium.
N. Shamim et al. / Journal of Colloid and Interface Science 304 (2006) 18 3
Fig. 1. Preparation of thermosensitive (PNIPAM)-coated nanomagnetic particles.
The amounts of protein adsorbed were expressed in mg of
protein per g of latex particles.
3. Results and discussion
3.1. Characterization of magnetic particles
Typical TEM micrographs for magnetic particles with and
without poly(NIPAM) coating are shown in Fig. 2. The shapes
of both coated and uncoated particles are spherical. The effec-
tive mean diameter of the uncoated particles is 9.3 nm, which
is comparable to the reported value of 8.5 nm [17]. After bind-
ing with the polymer, the particles remained discrete, with a
mean diameter of 12 nm. The mean diameter increased due to
the polymer coating on the surface of the particles. The forma-
tion of small aggregates of the magnetite particles results from
the magnetic dipolar interaction among the magnetite nanopar-
ticles. It is known that particles smaller than 30 nm shows su-
perparamagnetic properties [7]. Therefore, both the magnetic
particles and the polymer-coated magnetic particles show su-
perparamagnetic properties.
The magnetization curves of magnetic particles and of
surfactant-coated and thermosensitive-polymer-coated mag-
netic particles are shown in Fig. 3. The magnetization versus
applied magnetic eld (MH loop) curve shows zero coercivity
and remanence, which proves the superparamagnetic properties
of the particles. The saturation magnetization of these particles
is 76 and 52 emu/g of particles, respectively. Therefore, from
the gure we can observe that the saturation magnetization of
coated magnetic particles is lower than that of the uncoated
magnetic particles. This is because the surfactant changes the
surface magnetic anisotropy, which leads to an increase of the
surface spins disorientation [18], resulting in a decrease of
the magnetic moment.
3.2. Adsorption of BSA
Thermosensitive-polymer-coated nanomagnetic particles
were used for adsorption of BSA at different temperatures,
pHs, and ionic concentrations. It is known that the hydrophobic
interaction has a major role in protein adsorption phenomena
[16,19]. The thermosensitive polymer PNIPAM offers a hy-
4 N. Shamim et al. / Journal of Colloid and Interface Science 304 (2006) 18
(a)
(b)
Fig. 2. Transmission electron micrograph of magnetic nanoparticles of (a) un-
coated and (b) surface coated magnetic particles.
Fig. 3. Magnetization curve at 393 K.
Fig. 4. Adsorption and desorption scheme of protein on thermosensi-
tive-polymer-coated magnetic particles.
drophobic surface above the lower critical solution temperature
(LCST) of 32

C. When the temperature increases above the

lower critical solution temperature the surface becomes hy-
drophobic; therefore, the particles deswell and are susceptible
to adsorbing larger amounts of protein, which are desorbed at
lower temperatures. The schematic diagram of protein adsorp-
tion and desorption is illustrated in Fig. 4.
3.3. Effect of thermosensitive surface properties on protein
adsorption
The effects of BSA adsorption on thermosensitive-polymer
(PNIPAM)-coated nanomagnetic particles at 40, 30, and 25

C
are shown in Fig. 5a at pH 4.66 and ionic strength 0.01 M. It
is observed that a larger amount of protein was adsorbed at a
higher temperature. This adsorption behavior is principally ob-
served because of the presence of PNIPAM at the outer shell
of the particles surface. The afnity for protein adsorption
tends to increase because the particle surface changes from hy-
drophilic to hydrophobic as the temperature increases above
the lower critical solution temperature (32

C) of the polymer,
which results in a larger amount of BSA adsorption [16]. In
contrast to this, adsorption of BSA on bared nanomagnetic par-
ticles in Fig. 5b showed no signicant effect on the amount of
adsorbed BSA at the two different temperatures (40 and 25

C).
It is proposed that either [OH

] [20] or [NH
2
] [21] groups on
the surface of the nanomagnetic particles are responsible for
protein adsorption on bared nanomagnetic particles. Thus ex-
perimental result indicates that temperature-dependent adsorp-
tion of BSA onto thermosensitive magnetic particles is mainly
attributable to the changes in the properties of the particles sur-
face.
3.4. Effects of pH
The electrostatic interaction mechanism may be explained
by the results of zeta potential. The zeta potentials (or elec-
trophoretic mobilities) of the bared and PNIPAM-covered mi-
crospheres (determined for an ionic strength of 0.001 MNaNO
3
N. Shamim et al. / Journal of Colloid and Interface Science 304 (2006) 18 5
(a)
(b)
Fig. 5. BSA adsorption equilibrium isotherms on (a) PNIPAM-coated and
(b) bared magnetic nanoparticles at different temperatures (pH 4.66 and ionic
strength 0.01 M).
Fig. 6. Zeta potentials of magnetic nanoparticles in 10
3
M NaNO
3
at different
pHs.
using a Brookhaven Zeta Plus 90 analyzer) are shown in Fig. 6.
As the thermosensitive PNIPAM covers the surface of the par-
ticles and buries the ionic group inside, the absolute value of
zeta potential for the PNIPAM-coated particles is smaller than
that for the bared magnetic particles. The result showed that
the isoelectric point of polymer-coated particles was about 6.2.
On the other hand, the isoelectric point of bared nanomagnetic
particles was 6.74. This value agrees well with the literature
value of pI = 6.5 for bared magnetic particles. Based on the
Fig. 7. Adsorption equilibrium of BSA at different pHs (40

C and ionic
strength 0.01 M).
Fig. 8. Effect of pH on adsorption of BSA at 40

C and initial concentration

2.03 mg/ml.
literature value and the results obtained from Fig. 6, we can
conclude that magnetic particles coated with thermosensitive
polymer have positive charges below the isoelectric point 6.2
and negative charges above it. This larger value of zeta poten-
tial may greatly affect the adsorption of BSA.
The adsorption of BSA at different pHs is shown in Figs. 7
and 8. The amount of BSA adsorbed decreased dramatically
as the pH increased (pH 4.668.18). The observed adsorption
behavior shows that adsorption of protein onto thermosensitive-
polymer-coated nanomagnetic particles is governed not only by
hydrophobic interaction but also by electrostatic interactions.
Moreover, the results showed that among different pHs the
maximum amount of BSA was adsorbed near the isoelectric
point of BSA (pI =4.7) [22,23].
The pH dependence of BSA adsorption may be explained
solely by the electrostatic interaction force between BSA and
magnetic nanoparticles. At higher pH of 8.18 both BSA and
nanomagnetic particles are negatively charged and there is an
electrostatic repulsive force between them that resists the ad-
sorption of protein. In the same way, Yoon et al. [24] re-
ported electrostatic repulsion between microspheres (negatively
charged) and BSA (positively charged) molecules at pH 7. At
pH 5.23 BSA is negatively charged and nanomagnetic particles
are positively charged; therefore, there should be an electrosta-
6 N. Shamim et al. / Journal of Colloid and Interface Science 304 (2006) 18
tic attraction between BSA molecules and nanomagnetic parti-
cles. This attractive force can promote the adsorption of BSA
onto the surface of the nanomagnetic particles. However, the
results showed that less BSA was adsorbed on the nanopar-
ticles surface. This is probably due to the expansion of the
BSA molecule at acidic pH, which causes the reduction in ad-
sorption. Thus, it seems that conformational alteration of BSA
molecules has some effect on adsorption. From Fig. 8 it can be
seen that the maximum amount of protein was adsorbed at pH
4.66, which is near the isoelectric point of BSA pI = 4.7. This
is because the conformational alteration of the protein mole-
cule is not affected at the isoelectric point and the structure of
BSA remains in its compact state. Moreover, the net charge be-
tween the molecule and nanomagnetic particles is zero at the
isoelectric point of the protein, which promotes adsorption. Ba-
jpai and Shrivastava [25] found that at the isoelectric point,
pI 4.7, BSA undergoes minimum conformational changes and
therefore the adsorption of BSA on glass powder with PVA
reaches the maximum. Chun and Stroeve [23] also suggested
that at the isoelectric point of protein, electrostatic repulsion
between the protein and the membrane surface was minimized
and subsequently resulted in a larger ux than that at other pH
values. Recently, Peng et al. [26] reported results using bared
magnetic particles for BSA separation. They observed that an
increase of pH from 4.64 to 9.07 causes the amount of adsorbed
BSA to decrease signicantly due to surface charge of magnetic
particles and BSA and showed that maximum adsorption oc-
curred at the isoelectric point of BSA. Therefore, the inuence
of pH on the isotherms shows decreasing afnity with increas-
ing electrostatic repulsion between protein and adsorbent and a
maximum plateau value at the isoelectric point of the protein
[2729]. Elaissari and Bourrel [10] also conrmed maximum
adsorption of human serum albumin (HAS) on thermosensi-
tive coreshell magnetic latex with magnetic polystyrene as
core. Thus, these results imply that electrostatic repulsion is
another dominant factor, like temperature, for adsorption on
thermosensitive-polymer-coated nanomagnetic particles.
3.5. Effect of ionic strength
The effect of ionic strength on BSA adsorption was in-
vestigated at two different feed concentrations (0.505 and
1.525 mg/ml at pH 4.66) and different pHs (4.66 and 8.18
at feed concentration 1.525 mg/ml). Both experiments were
carried out at 40

C. Results are reported in Figs. 9 and

10, respectively. Results from Fig. 9 show that adsorption
decreases with increasing ionic strength. First, the increase
in the salt concentration of the medium decreases the sol-
vency of the poly(NIPAM) chain and consequently decease
the corresponding lower critical solution temperature of the
poly(NIPAM) [30]. Thus, the accessible sites for protein ad-
sorption are reduced and a smaller amount of protein was
adsorbed. Second, the electrostatic attraction between BSA and
the latex particles decreases with increasing ionic strength and
therefore the protein adsorption decreases.
To investigate the effect of ionic strength on electrostatic ad-
sorption, the effect of ionic strength was studied at acidic and
Fig. 9. Effects of electrolyte concentration [NaCl] on BSA adsorption at
pH 4.66 and temperature 40

C for two feed concentrations (0.505 and

1.525 mg/ml).
Fig. 10. Effect of salt [NaCl] concentration at different pHs for feed concen-
tration 1.525 mg/ml and temperature 40

C.
basic pH levels. At both acidic and basic pH the amount of
BSA adsorbed decreased with increasing ionic strength. Fig. 10
shows that at higher ionic strengths the amounts adsorbed for
pH 4.66 and 8.18 are about the same. The reason behind this
may be that at basic pH8.18 and at high temperature (40

C) the
increase in salinity reduces the electrostatic repulsion between
the protein molecules and the latex particles and vice versa.
3.6. Adsorption equilibrium of BSA
The adsorption isotherms of BSA protein on PNIPAM-
coated magnetic nanoparticles are shown in Figs. 5a and 7.
Although the shapes of the isotherms are Langmuir-type, the
isotherms did not show any initial rapid increase, and these
Langmuir isotherms failed to t the experimental data. It is
found that the experimental data was tted by a Langmuir
Freundlich isotherm; the LangmuirFreundlich equation is ex-
pressed as
(1) C
s
=C
m
KC
1/n
b
1 +KC
1/n
b
where C
b
(mg/ml) and C
s
(mg/g solid) are BSA concentra-
tion in the aqueous solution and absorbed BSA on the solid
N. Shamim et al. / Journal of Colloid and Interface Science 304 (2006) 18 7
Table 1
LangmuirFreundlich parameters for BSA adsorption on thermosensitive and bared nanomagnetic particles at different pHs and temperatures
PNIPAM-coated nanomagnetic particles Bared nanomagnetic particles
40

C 30

C 25

C 40

C 25

C
pH 4.66 pH 8.18 pH 4.66 pH 8.18 pH 4.66 pH 4.66 pH 4.66
C
m
(mg/g solid) 279.97 145.47 223.04 89.81 56.77 137.86 127.24
K (ml/mg) 1.62 3.06 1.12 3.09 0.13 37.38 37.01
R
2
0.93 0.97 0.91 0.95 0.98 0.99 0.99
Fig. 11. Effect of temperature on adsorption and desorption (pH 9.35 and feed
concentration 1.525 mg/ml).
at equilibrium, respectively. C
m
in the maximum adsorption
amount, K is the adsorption constant, and n is the exponen-
tial factor. The values of n were relatively constant between
0.4 and 0.5, and hence n could be xed to its average value of
0.45. Experimental data are tted to the LangmuirFreundlich
equation using nonlinear regression. The tted parameters of
the model for thermosensitive-polymer-coated nanomagnetic
particles and bared nanomagnetic particles are summarized in
Table 1. High R
2
values indicate that the model predicts the
adsorption behavior well.
3.7. Effect of temperature, pH, and ionic strength on BSA
desorption
Desorption of BSA was investigated at a function of temper-
ature and ionic strength. Desorption was carried out at a tem-
perature of 20

C and under alkaline conditions with an initial

feed concentration of 1.525 mg/ml. Fig. 11 shows the effects of
adsorption and desorption with temperature. It is observed that
more than 80% of protein can be desorbed from the latex parti-
cles when the adsorption was done near the lower critical solu-
tion temperature (30

C) of PNIPAM. On the other hand, about

60% of BSA was desorbed from the latex particles that were
adsorbed above the lower critical solution temperature (40

C).
Although the amount of protein adsorbed increases with in-
creasing temperature, lesser amounts of protein were desorbed.
This is because at higher temperature the protein molecules
are easily deformed either by the interactions between polymer
Fig. 12. Desorbed amount of BSA as a function of ionic strength (pH 8.18 and
feed concentration 0.503 mg/ml).
chains and the molecules or by the electrostatic force at higher
temperature.
The effects of salt concentration on desorption of BSA from
the PNIPAM-coated latex particles are shown in Fig. 12. The
amount of desorbed BSA increases with the increase of salt
concentration in the desorption medium at 20

C. The result
observed may be attributed to decrease in electrostatic attrac-
tive force with the increase in ionic strength, which leads to an
enhancement of protein desorption.
4. Conclusions
Nanosized thermosensitive-polymer (PNIPAM)-coated mag-
netic particles were prepared by seed polymerization using
surface-modied magnetic particles as the seed. The attachment
of polymer chains to the nanomagnetic particles was conrmed
by FTIR. The adsorption of BSA onto these nanomagnetic par-
ticles coated with thermosensitive polymer was investigated
as a function of temperature, pH, and ionic strength. It was
observed that more protein was adsorbed at higher tempera-
tures and less below the lower critical solution temperature;
this behavior is attributed to the hydrophobic and hydrophilic
characteristics of the nanomagnetic particles above and below
the LCST of PNIPAM, respectively. Therefore, we can say
that adsorption is principally governed by the surface proper-
ties of the polymer-coated nanomagnetic particles. However,
the contribution of electrostatic interaction was evident from
the effect of pH and ionic concentration. As the pH and ionic
strength increased the electrostatic attractive force decreased
between the polymer-coated nanomagnetic particles and the
8 N. Shamim et al. / Journal of Colloid and Interface Science 304 (2006) 18
protein molecules; consequently the amount of protein adsorp-
tion was reduced. The adsorption equilibrium results are tted
by the LangmuirFreundlich model. Desorption of BSAwas in-
vestigated as a function of temperature and ionic strength. The
results showed that more than 80% of protein was desorbed
from the preadsorbed protein at 30

C. The effect of salinity

revealed the contribution of electrostatic interaction to protein
desorption.
Acknowledgment
This work was nancially supported by the National Univer-
sity of Singapore Research Fund (R-279-000-085-112).
References
[1] X.B. Ding, Z.H. Sun, G.X. Wan, Y.Y. Jiang, React. Funct. Polym. 38
(1998) 1115.
[2] R.F. Dekker, Appl. Biochem. Biotechnol. 22 (1989) 289310.
[3] A. Kondo, H. Fukuda, J. Ferment. Bioeng. 84 (1997) 337341.
[4] R.S. Molday, L.L. Molday, FEBS Lett. 170 (1984) 232238.
[5] Y. Haik, V. Pai, C.-J. Chen, J. Magn. Magn. Mater. 194 (1999) 254261.
[6] T. Abudiab, R.R. Beitle Jr., J. Chromatogr. A 795 (1998) 211217.
[7] S.M. OBrien, O.R.T. Thomas, P. Dunnill, J. Biotechnol. 50 (1996) 13
25.
[8] A.N. Rusetski, E.K. Ruuge, J. Magn. Magn. Mater. 85 (1990) 299302.
[9] X. Ding, Z. Sun, W. Zhang, Y. Peng, A.S.C. Chan, P. Li, Colloid Polym.
Sci. 278 (2000) 459463.
[10] A. Elaissari, V. Bourrel, J. Magn. Magn. Mater. 225 (2001) 151155.
[11] I. Safarik, M. Safarikova, J. Biochem. Biophys. Methods 27 (1993) 327.
[12] X.D. Tong, Y. Sun, Biotechnol. Prog. 17 (2001) 738.
[13] X.D. Tong, B. Xue, Y. Sun, Biotechnol. Prog. 17 (2001) 134.
[14] Z.G. Peng, K. Hidajat, M.S. Uddin, Colloids Surf. B Biointerfaces 33
(2004) 1521.
[15] M.H. Liao, D.H. Chen, Biotechnol. Lett. 24 (2002) 1913.
[16] X.B. Ding, Z.H. Sun, W.C. Zhang, Y.X. Peng, G.X. Wan, Y.Y. Jiang,
J. Appl. Polym. Sci. 77 (2000) 29152920.
[17] Y.S. Kang, S. Risbud, J.F. Rabolt, P. Stroeve, Chem. Mater. 8 (1996)
22092211.
[18] S. Morup, Europhys. Lett. 28 (1994) 671676.
[19] D. Duracher, R. Veyret, A. Elaissari, C. Pichot, Polym. Int. 53 (2004)
618626.
[20] R.V. Mehta, R.V. Upadhyay, S.W. Charles, C.N. Ramchand, Biotechnol.
Techn. 11 (1997) 493496.
[21] D.H. Chen, M.H. Liao, J. Mol. Catal. B Enzymat. 16 (2002) 283291.
[22] W.-K. Lee, J.-S. Ko, H.-M. Kim, J. Colloid Interface Sci. 246 (2002)
7077.
[23] K.Y. Chun, P. Stroeve, Langmuir 18 (2002) 46534658.
[24] J.-Y. Yoon, J.H. Lee, J.-H. Kim, W.-S. Kim, Colloids Surf. B Biointer-
faces 10 (1998) 365377.
[25] A.K. Bajpai, M. Shrivastava, J. Macromol. Sci. Pure Appl. Chem. 38
(2001) 11231139.
[26] Z.G. Peng, K. Hidajat, M.S. Uddin, J. Colloid Interface Sci. 271 (2004)
277283.
[27] A.V. Elgersma, R.L.J. Zsom, W. Norde, J. Lyklema, J. Colloid Interface
Sci. 138 (1990) 145156.
[28] P.G. Koutsoukos, C.A. Mumme-Young, W. Norde, J. Lyklema, Colloids
Surf. 5 (1982) 93104.
[29] W. Norde, J. Lyklema, J. Colloid Interface Sci. 66 (1978) 266276.
[30] H. Inomata, N. Wada, Y. Yagi, S. Goto, S. Saito, Polymer 36 (1995)
875877.