Colloids and Surfaces A: Physicochem. Eng. Aspects 419 (2013) 125–132
Contents lists available at SciVerse ScienceDirect
Colloids and Surfaces A: Physicochemical and
Engineering Aspects
journal homepage: www.elsevier.com/locate/colsurfa
Atomic-scale interactions of the interface between chitosan and Fe3 O4
Linhui Qiang a , Zhanfeng Li a , Tianqi Zhao a , Shuangling Zhong b , Hongyan Wang a , Xuejun Cui a,∗
a
b
College of Chemistry, Jilin University, Changchun, 130012, PR China
College of Resources and Environment, Jilin Agricultural University, Changchun, 130118, PR China
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
We built the interface model
between chitosan and Fe3 O4 .
The (1 1 1) interface has strongest
interaction than other interfaces.
There was high probability of formed
hydrogen bonds of (1 1 1) interface.
There were interactions between
Fe3 O4 and nitrogen atoms of chitosan.
a r t i c l e
i n f o
Article history:
Received 13 August 2012
Received in revised form 2 November 2012
Accepted 24 November 2012
Available online 4 December 2012
Keywords:
Molecular dynamics simulation
Adsorption
Chitosan
Fe3 O4
Interface
a b s t r a c t
Molecular dynamics (MD) simulation was employed to study chitosan adsorption on different Fe3 O4
crystallographic planes at the atomic level. The interaction energy between chitosan and different Fe3 O4
surfaces indicates that the interaction of chitosan and Fe3 O4 (1 1 1) surfaces is stronger than that of
(1 1 0) and (0 0 1) surfaces. The concentration profiles show that hydrogen and amino groups of chitosan
could form strong interactions with the surfaces of Fe3 O4 . The radial distribution function show that the
probability of forming hydrogen bonds between groups of chitosan and oxygen atoms on Fe3 O4 (1 1 1)
surface is more than that of Fe3 O4 (1 1 0) surface, and the nitrogen atoms and iron atoms on the surface
have weak physical interactions. This study provides useful information in understanding the interfacial
interaction mechanism at the atomistic scale for polymer and mineral.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
The iron oxide nanoparticles (Fe3 O4 NPs) were a kind of important magnetic material. It has been widely used in bio-separation,
hyperthermia and magnetic guided drug targeting [1–4]. For
biomedical applications, Fe3 O4 NPs were often treated by surface
modification which could increase the functionality and improve
the biocompatibility of the Fe3 O4 NPs [5]. Chitosan obtained by
deacetylation of chitin was a unique natural linear cationicpolymer,
which was the structural element in the exoskeleton of crustaceans
∗ Corresponding author at: College of Chemistry, Jilin University, Qianjin Street
2699#, Changchun, 130012, PR China. Tel.: +86 431 85168470;
fax: +86 431 85168470.
E-mail address: cui xj@jlu.edu.cn (X. Cui).
0927-7757/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.colsurfa.2012.11.055
and cell walls of fungi. Chitosan had many beneficial properties
such as a low toxicity, low immunogenicity, excellent biocompatibility, biodegradability and as well as a high positive charge density.
Due to its high positive charge density, it could easily form polyelectrolyte complexes with negatively charged nucleotides or drugs by
electrostatic interaction [6–8].
Recently, chitosan modified magnetic particles (MPs) have
gained significant attention for biomedical applications. Li et al.
has reported the MPs can provide excellent biocompatility,
biodegradability and low toxicity without compromising their
magnetic targeting [9]. Sun et al. have prepared magnetic targeting chitosan nanoparticles as a drug delivery system and
imaging agents for photodynamic therapy to cancer treatment [10].
Nguyen and co-workers have developed a novel chitosan Fe3 O4
nanobiocomposite-based platform for electrochemical detection
of HIV-1 [11]. Although, MPs have applied in many fields, the
126
L. Qiang et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 419 (2013) 125–132
interaction between chitosan and Fe3 O4 are rarely reported that
of is importance both from the academic viewpoint and in applied
fields.
Molecular dynamic (MD) simulation has been proven a valuable
tool for investigating polymer–inorganic interface and the stability. Jia et al. have simulated the interface between self-assembled
monolayer on Au (1 1 1) surface and epoxy resin [12]. Zhang et al.
have researched the chitosan behaviors on different hydroxyapatire crystallographic planes at the atomic level [13]. Yue et al.
have reported the deposition processes of gold nanoparticles on
the magnetite surface through a molecular dynamics method [14].
In this paper, MD simulation was employed to study the interaction
mechanism of chitosan and Fe3 O4 in atom scale. It was beneficial
to choose a better condition of producing MPs. Moreover, it could
provide evidence for choosing a proper drug or nucleotide, which
was used for targeting delivery systems.
Table 1
The chitosan models used to calculate the solubility parameter ı.
Repeat units
Molecule
numbers
Atom
numbers
Cell
lengths
Parallel
models
2
5
10
15
20
25
40
60
80
100
34
14
7
5
4
3
2
1
1
1
1564
1498
1554
1660
1768
1656
1764
1322
1762
2202
30.12
30.37
30.36
31.07
31.74
31.06
31.74
28.83
31.73
34.18
5
5
5
5
5
5
5
5
5
5
qi qj
+
i>j
Aij
2. Simulation details
The simulation was conducted with the Materials Studio v4.3
package (Acclrys) using the condensed-phase optimized molecular
potentials for atomistic simulation studies (COMPASS) force field.
Moreover, in MD simulation, the cut-off distance of 10 Å was used
in our study. The COMPASS force field was a general all-atom force
field for molecular dynamic simulation that was developed using
the state-of-the-art ab initio and empirical parameterization techniques [15]. The functional form of COMPASS force field was given
as follows [16]:
=
pot
+
[K2 (b − b0 )2 + K3 (b − b0 )3 + K4 (b − b0 )4 ]
(a)
b
H2 ( − 0 )2 + H3 ( − 0 )3 + H4 ( − 0 )4
(b)
+
i>j
(l)
εri,j
rij9
−
Bij
rij6
(m)
This equation includes the terms that count the effects of bond
stretching (a), bending (b), torsion (c), out-of-plane coordinate (d),
cross effects of (a), (b), (c) and (d)((e)–(k)), Columbic interaction
(l) and van der Waals interaction (m). K, H, F and V are force field
parameters b, and ϕ stand for bond length, bending angle and torsion angle, respectively. COMPASS force field has been successfully
used for a long time on the prediction of structural, conformational,
and adsorption for polymer–inorganic interface system under a
wide range of conditions of temperature and pressure, such as the
adsorption behavior of alkenes on Al2 O3 surface by Li and Choi [17].
2.1. Chitosan model
+
[1 [1 − cos(ϕ
− ϕ10 )] + 2 [1 − cos(2ϕ
In this work, the chain length of chitosan in this simulation was
determined by calculating the solubility parameter ı according to
Eq. (n):
− ϕ20 )]
ϕ
+3 [1 − cos(3ϕ
+
− ϕ30 )]]
K 2
(c)
(d)
+
b
b
b
b
ϕ
+
+
+
+
b
(e)
F ( − 0 )( − 0 )
(f)
Fb (b − b0 )( − 0 )
(g)
(b − b0 )[V1 cosϕ + V2 cos 2ϕ + V2 cos 3ϕ]
(h)
(b − b 0 )[V1 cos ϕ+V 2 cos 2ϕ + V2 cos 3ϕ]
(i)
( − 0 )[V1 cos ϕ + V2 cos 2ϕ + V3 cos 3ϕ]
(j)
ϕ
+
Fbb (b − b0 )(b − b 0 )
ϕ
+
ϕ
Kϕ cos ϕ( − 0 )( − 0 )
(k)
ı=
√
CSD =
Ecoh
V
(n)
where CED represents cohesive energy density, Ecoh represents the
cohesive energy and V is the volume of polymer. To reduce the
statistical error, the parallel experiments were conducted using
for each system investigated. While for every system, five simulations have been conducted. Table 1 presented the details of the
chitosan models used to calculate the solubility parameter ı. First,
the amorphous structures of chitosan with different chain lengths
were built. Then chitosan molecule was optimized by the steepest
descent and conjugate gradient method. During the procedure of
structure optimization, the maximum number for the minimization
was 10,000. At last, all of the chitosan molecules have reached the
convergence before iteration times achieve 10,000. To obtain an
energy-minimized state, the chitosan structure was annealed by
performing dynamics simulations for 100 ps at each temperature,
which increased from 298 K to 500 K and then decreased to 298 K
with a step of 20.2 K. The simulations have performed with the
time step of 1 fs at the isothermal-isobaric ensemble (NPT) which
the pressure is 0.0001 Gpa. Finally, the chitosan chain was performed to calculate the solubility parameter ı of each component.
When a stable value, ı, was obtained, it was confirmed that the
number of repeating units was sufficient for this simulation [18].
Fig. 1(a) displayed the dependence of ı for chitosan on the number
of repeating unites. We can find that when the number of repeating units reaches 20, ı approached a constant value and was close
to the experimental data (10.6 ± 0.8) [19], so we chose 20 as the
L. Qiang et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 419 (2013) 125–132
127
Fig. 1. (a) The solubility parameter depend on the number of repeating unites and
(b) the 3D model of the chitosan chain (20 monomers). Carbon: gray; hydrogen:
white; nitrogen: blue; oxygen: red. (For interpretation of the references to color in
this figure legend, the reader is referred to the web version of the article.)
representative of chitosan chain length in this work. The 3D structure of the optimized chitosan chain was shown in Fig. 1(b).
2.2. Fe3 O4 model
Fe3 O4 has a cubic inverse spinal structure with a lattice constant of 0.8397 nm and eight formula units (Fe24 O32 ). In this study,
the initial coordinates were taken form the crystallography open
database which the CIF number is 9006184. In order to validate
the applicability of the COMPASS force field to Fe3 O4 , we compared the experimental lattice parameters of Fe3 O4 crystal with
those obtained after the MD simulation using COMPASS. MD calculation was carried out on the Fe3 O4 system for 200 ps with the
time step of 0.5 fs the isothermal Cisobaric ensemble (NPT) at 298 K.
The lattice parameters after the MD simulation of the Fe3 O4 are
a = b = c = 0.8394 nm; ˛ = ˇ = = 90◦ , which were very close to those
obtained from the experimental data [20]. This demonstrated that
the COMPASS force field was applicable to the Fe3 O4 simulation.
In this work, the (1 1 1), (1 1 0) and (0 0 1) surfaces of Fe3 O4
were chosen to study the interaction with chitosan, because they
were the dominant surfaces observed in experiments. The surfaces
of Fe3 O4 were built by cleaving the crystal along the crystallographic planes. The dimensions of the Fe3 O4 surfaces were
as follows: Fe3 O4 (1 1 1), 95 Å × 95 Å × 15 Å (a × b × c), ˛ = ˇ = 90◦ ,
= 120◦ ; Fe3 O4 (1 1 0), 101 Å × 71 Å × 15 Å, ˛ = ˇ = = 90◦ ; Fe3 O4
(0 0 1), 95 Å × 95 Å × 15 Å, ˛ = ˇ = = 90◦ (see Fig. 2). The minimization process was completed by the steepest descent method
with the convergence of 1000 kcal/mol and the conjugate gradient
method with the convergence of 10 kcal/mol.
2.3. The interaction between chitosan and Fe3 O4
The optimized chitosan chain was added onto the built Fe3 O4
surfaces, and then the ‘vacuum slab’ with the height of 200 Å was
put upon the chitosan–Fe3 O4 system with 3D periodic boundary
conditions. The initialization model of the Fe3 O4 surface with the
Fig. 2. The model of the Fe3 O4 (1 1 1), (1 1 0) and (0 0 1) surfaces. Iron: blue; oxygen:
red. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of the article.)
chitosan was shown in Fig. 3. The surfaces of Fe3 O4 were fixed
before simulations. The MD simulations of the adsorption of chitosan on Fe3 O4 surface were studied in the canonical ensemble
(NVT) at 298 K. The temperature of the systems was kept through
the Andersen method. The time step was 1.0 fs and the dynamics
balance time was 500.0 ps. The further 20 ps MD simulation was
performed under the same conditions to record the trajectory of all
the atoms in the system. After reaching equilibrium state, the binding energy between the chitosan chain and the Fe3 O4 surfaces were
calculated to evaluate the interfacial interactions. The concentration profiles of the atoms (in the oxygen and amino groups of the
chitosan) and the atoms (in the chitosan backbone) were used to
examine the activity of groups and backbone of the chitosan during the adsorption process. The radial distribution function (RDF)
of different atoms could be used to understand the interaction of
chitosan/Fe3 O4 on the atom scale.
128
L. Qiang et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 419 (2013) 125–132
Fig. 4. The configuration of chitosan chain on Fe3 O4 (1 1 1) surface: (a) and (1 1 0)
surface: (b) after MD calculation. Carbon: gray; hydrogen: white; nitrogen: blue;
oxygen: red; iron:blue. (For interpretation of the references to color in this figure
legend, the reader is referred to the web version of the article.)
and (1 1 0) surfaces for the following interaction study of chitosan/
Fe3 O4 interfaces.
3.2. The chain behavior of chitosan on Fe3 O4 surfaces
Fig. 3. The configuration of chitosan chain on Fe3 O4 (1 1 1) surface: (a) before MD
calculation and (b) after MD calculation. Carbon: gray; hydrogen: white; nitrogen:
blue; oxygen: red; iron:blue. (For interpretation of the references to color in this
figure legend, the reader is referred to the web version of the article.)
3. Result and discussion
3.1. The interaction energy between chitosan and Fe3 O4 surface
The interaction energy (Ei ) reflected the interfacial compatibility
and interaction. Ei between Fe3 O4 and chitosan could be calculated
by the following equation (o):
Ei =
ECS + EFe − EFeCS
S
(o)
where EFeCS represents for the energy of Fe3 O4 with chitosan after
MD simulation, EFe is the energy of the Fe3 O4 by removing the chitosan and ECS is the energy of the chitosan chain after removing the
Fe3 O4 surface. S denotes the area of the Fe3 O4 surface. The greater
value of Ei implied stronger interaction, and more compatibility of
the two components.
Table 2 presented the interaction energy of chitosan/Fe3 O4 . It
can be found that the value of Ei between chitosan and (1 1 1)
surface was larger than that of the other surface. Therefore, the
interaction between chitosan and (1 1 1) surface was greater than
those of (1 1 0) and (0 0 1) surfaces, the interface interactions
between chitosan/(1 1 0) surface and chitosan/(0 0 1) surface were
very similar. The present results were in agreement with the results
of theoretical calculation of Yang et al. [21]. Moreover, the surface
free energy was nearly equal between (1 1 0) and (0 0 1) surfaces
and they had similar surface structures. Therefore, we choose (1 1 1)
After adsorption, there might be some physical or chemical
interactions between polymer and inorganic surface. These interactions might lead to the configuration changes of polymer. The
changes of molecular models before and after adsorption simulation could be observed by MD simulation. In this study, all the
chitosans of different interfaces had the same initial configuration
(see Fig. 3(a)). Fig. 4 showed the configuration of the chitosan on
(1 1 1) and (1 1 0) surfaces after MD simulation. It could be seen that
the chitosan chain before simulation shows the helical conformation. And amino and hydrogen groups distributed in both sides of
the chitosan. After MD simulation, the whole chitosan chain was
adsorbed on (1 1 1) surface. The helical conformation changed to a
planar conformation (see Fig. 4(a)). However, the configuration of
chitosan/(1 1 0) surface was different from chitosan/(1 1 1) surface
after MD simulation (see Fig. 4(b)). A part of chitosan chain on the
(1 1 0) surface showed a distorted planar. This phenomenon might
be due to the different interaction of two interfaces. We can preliminary investigated the interaction between the polymer and the
surface using the concentration profile.
The concentration profile was a function of the positions along
the axes of the Fe3 O4 surface and a distribution of the specific atom
or group in the polymer. In this study, we calculated the concentration profile of the hydrogen, oxygen, carbon and nitrogen atoms
on the chitosan chain along the axis normal to Fe3 O4 surface. We
investigated the concentration profile of hydrogen atoms from the
hydrogen and amino groups on chitosan chain on the different surface. We could find that the concentration profiles were located
in the range of 8–20 Å before the MD simulation (see Fig. 5(a)).
After the MD simulation, the concentration profile of the hydrogen atoms on (1 1 1) surface (see Fig. 5(a)) was located in the range
of 2–3 Å. This was a very close distance between hydrogen atoms
and Fe3 O4 surfaces. There might be some interaction. The location
Table 2
The interaction energy between different Fe3 O4 surfaces and chitosan.
Fe3 O4 surface
EFeCS
(kcal mol−1 )
EFe
(kcal mol−1 )
ECS
(kcal mol−1 )
Area
(Å2 )
Ei
(kcal mol−1 Å2 )
(1 1 1)
(1 1 0)
(0 0 1)
9476231
5004281
6747614
9477341
5003214
6746660
1384
1295
1278
9025
7171
9025
0.277
0.032
0.036
L. Qiang et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 419 (2013) 125–132
129
in Fig. 5(b), the concentration profile of the oxygen atoms/(1 1 1)
surface was located in range of 2–4 Å and the first peak was located
in 3.2 Å. The concentration of the oxygen atoms/(1 1 0) surfaces was
located in range of 2–5 Å, the first peak was located in 3.2 Å. It means
that the strongest interactions of the oxygen atoms between the
different Fe3 O4 surfaces were similar. From the concentration profile of chitosan/(1 1 0) surface, there was some small peak located
in range of 6–12 Å. Because chitosan chain had a part of distortion,
some of hydrogen group was far away from the (1 1 0) surface (see
Fig. 4(b)). The same phenomenon of concentration profile of other
atoms could be observed (The weaker interaction of (110) interface
might cause of this phenomenon). Then the normalized first peak
area of (1 1 1) interface was 0.54 and that of (1 1 0) interface was
0.43. Therefore, there were more oxygen atoms of hydrogen group
on the (1 1 1) surface in the closest distance.
Fig. 5(c) showed the concentration profile of nitrogen atoms
both chitosan/(1 1 1) surface and chitosan/(1 1 0) surface. The first
peak of chitosan/(1 1 1) surface was located in 3 Å, most of nitrogen
atoms distribute in range of 2–5.5 Å. The closest distance between
the nitrogen atoms and 1 1 0 surface was 3.5 Å, and many nitrogen atoms was located in range of 2–8 Å. The first peak areas
were normalized. The peak area of chitosan/(1 1 1) was about two
times than that of chitosan/(1 1 0) surface, the normalized peak
area of chitosan/(1 1 1) was 0.20 while the normalized peak area
of chitosan/(1 1 0) was 0.11. Therefore the interactions of the nitrogen/surface have bigger difference between chitosan/(1 1 1) surface
and chitosan/(1 1 0) surface.
The concentration profiles of carbon, oxygen and hydrogen
atoms from the chitosan backbone on different surface were shown
in Fig. 6. Compared with the atoms of hydrogen and amino groups
on chitosan chain, hydrogen, oxygen and nitrogen atoms which
belong to backbone had a longer distance from Fe3 O4 surface.
It means that the interaction between carbon, oxygen, hydrogen
atoms of chitosan backbone and surface was weaker than that of
hydrogen and amino groups. Therefore, the interactions of between
chitosan and Fe3 O4 may mainly come from the groups of chitosan.
Moreover, the atoms of hydrogen and amino groups were closer
than the atoms of chitosan backbone.
3.3. Atom-scale interaction analysis
Fig. 5. Concentration profiles of different atoms of hydrogen and amino groups of
chitosan on Fe3 O4 (1 1 1) surface and (1 1 0) surface: (a) hydrogen, (b) oxygen and
(c) nitrogen atoms before and after the MD calculation.
The interaction of different atoms could be investigated by the
distance of them. Therefore, the interaction between chitosan and
Fe3 O4 surface could be further investigated by calculating the distance between the atoms of chitosan and the atoms of Fe3 O4 surface
on atom scale. The radial distribution function (RDF) could calculate
this distance between different atoms. RDF represents the probability density of A and B at a distance of r with respect to the bulk
phase in a completely random distribution. RDF is defined as (p):
V
of first peak indicated the closet distance between the atoms and
surfaces. Moreover, the strongest interaction has occurred in the
closest distance. And the normalized peak area was used to compare the concentration distribution of the atom which have the
closest distance on the surface. From Fig. 5(a), the first peak of
chitosan/(1 1 1) surface was located in 2 Å; and the first peak of
chitosan/(1 1 0) surface was located in 2.5 Å. Then, The normalized
first peak area of (1 1 1) surface was 0.71 which is higher than that
of (1 1 0) surface (0.42). That was to say there were more hydrogen
atoms on the (1 1 1) surface in the closest distance than on the (1 1 0)
surface. Therefore, there might be stronger interaction of hydrogen
atoms/(1 1 1) surface than hydrogen atoms/(1 1 0) surface.
The concentrations of oxygen atoms that belong to the hydrogen
group of chitosan on different surface were investigated. As shown
gAB (r) =
ı(r − |rAi − rBj |)
i=
/ j
(NA NB − NAB )4r 2 dr
(p)
where i and j refer to the ith and jth atoms of group A of NA atoms
and group B of NB atoms, NAB is the number of atoms common to
both groups A and B, angle brackets imply averaging over different
configurations.
As mentioned Section 3.2, the atoms of groups were closer on
the surface than the atoms of chitosan backbone. The distance of
the hydrogen atoms of groups on Fe3 O4 surfaces were very close
to the bond length of the hydrogen bond. Hydrogen bond is the
interaction of a hydrogen atom with an electronegative atom, such
as nitrogen, oxygen or fluorine that comes from another molecule
or chemical group. Hydrogen bond could occur between molecules,
or within different parts of a single molecule. There were many
130
L. Qiang et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 419 (2013) 125–132
Fig. 6. Concentration profiles of different atoms of chitosan backbone on Fe3 O4
(1 1 1) surface and (1 1 0) surface: (a) hydrogen, (b) oxygen and (c) nitrogen atoms
before and after the MD calculation.
intramolecular hydrogen bonds in the chitosan chain, which was
responsible for the helical conformation. The distance of different
atoms have been calculated by RDF (see Figs. 7 and 8). The first
peak position is the closest distance between the two atoms, and
the peak height indicates the probability that the two atoms appear
in this distance.
Fig. 7(a) shows the RDF between hydrogen atoms, from hydrogen and amino groups of chitosan, and oxygen atoms of surfaces.
The first peak is located at 1.9 Å. And the peak height which belongs
to chitosan/(1 1 1) surface was higher than the chitosan/(1 1 0) surface. It is known that the mean bond length of ‘O H· · ·O’ formed
by two water molecules is 1.9 Å. The distance between two oxygen atoms is close to the value of 2.8 Å, and the distance between
oxygen atoms and nitrogen atoms is about 2.8 Å [22,23]. Thus, we
Fig. 7. The radial distribution function between different atoms of the hydrogen and
amino groups and the oxygen atoms of the (1 1 1) and (1 1 0) surface: (a) hydrogen,
(b) oxygen and (c) nitrogen.
consider that the hydrogen atoms of groups on chitosan chain have
formed hydrogen bonds with the oxygen atoms of Fe3 O4 surfaces.
Moreover, the probability of formed hydrogen bonds between the
hydrogen atoms and the oxygen atoms of (1 1 1) surface is higher.
The reason may be that the oxygen atoms of the (1 1 1) surface
had the high activity, they could interact with other atoms easily.
This was consistent with the experimental results [24]. Another
distance was between the acceptor and donor of forming hydrogen
bonds need to discuss. Then we could find which kind of hydrogen atoms easily formed hydrogen bonds with the oxygen atoms
on Fe3 O4 surface. The possible acceptor of X H bond was nitrogen
and oxygen atoms of chitosan and the possible donor Y is oxygen
atoms of the Fe3 O4 surface. Fig. 7(b) presented the RDF between
the oxygen atoms of hydrogen group in chitosan chain and oxygen
L. Qiang et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 419 (2013) 125–132
131
between nitrogen atoms of amino group on chitosan chain and iron
atoms of (1 1 1) surface is higher than that of (1 1 0) surface. It can
be seen from Fig. 8(b), the closest distance between oxygen atoms
and iron atoms of 1 1 1 surfaces is 3.5 Å and that of (1 1 0) surface is
located at 3.8 Å. The radius of the oxygen atoms are 0.48 Å. The sum
of the radius values of the two atoms is smaller remarkably than
the distance between oxygen atoms and iron atoms. Therefore the
iron atoms of (1 1 1) and (1 1 0) surfaces have weak interaction with
the oxygen atoms of hydrogen groups on chitosan chain.
4. Conclusions
Fig. 8. The radial distribution function between different atoms of the hydrogen
and amino groups and the iron atoms of the (1 1 1) and (1 1 0) surface: (a) oxygen
and (b) nitrogen.
atoms of the Fe3 O4 surfaces. The first peaks were located at 2.8 Å,
the peak height of (1 1 1) surface was higher than that of (1 1 0) surface. Thus, the hydrogen groups have more probability of forming
hydrogen bonds with the oxygen atoms of (1 1 1) surface than that
of (1 1 0) surface. It means that the difference of interactional energy
from the different surface (from Section 3.1) may be caused by the
hydrogen bonds which formed by hydrogen groups of chitosan and
the oxygen of Fe3 O4 surface. Then we further investigated the RDF
between nitrogen atoms of amino groups and oxygen atoms of the
Fe3 O4 surface, as show in Fig. 7(c). Their first peaks are located at
2.8 Å, and the peak height of (1 1 1) surface is higher than (1 1 0)
surface. It means that the hydrogen atoms of amino group are hard
to forming hydrogen bonds with the oxygen atoms of (1 1 0) surface. The oxygen atoms of (1 1 1) surface have higher activity than
other oxygen atoms which from (1 1 0) surface.
There might be some others interactions between chitosan and
surface of Fe3 O4 except hydrogen bonds. The interactions might be
formed between the iron atoms of Fe3 O4 surface and the nitrogen or oxygen atoms of groups in chitosan. Fig. 8(a) shows the
RDF between iron atoms and nitrogen atoms. We found that the
first peak which belongs to (1 1 1) surface is located at 2.9 Å. The
radius of the nitrogen atoms and iron atom are 0.75 Å and 1.65 Å
respectively. Their distance is similar to the sum of the radius values of the two atoms. Thus, some interactions between nitrogen
and iron atoms might be occured. The same experimental results
are obtained by FT-IR [24]. As shown in Fig. 8(a), the first peak of
(1 1 0) surface is located at 3 Å, which is similar with (1 1 1) surface, but the height of peak (1 1 1) surface is about two times than
that of (1 1 0) surface. Thus, the probability of occurred interaction
The MD simulation was used to study the behaviors of chitosan
chain on the (1 1 1), (1 1 1) and (1 1 0) surfaces of Fe3 O4 based on
atomic scale. The interaction energy of different interfaces, RDF
and the concentration profiles between different atoms on chitosan
chain with the atoms on the Fe3 O4 surfaces was studied. The result
of the interaction energy indicated that the chitosan chain has
more intensely interacts with the (1 1 1) surface than the (1 1 0) and
(0 0 1) surfaces of Fe3 O4 . The concentration profiles show that there
are more atoms of groups absorbed on the surfaces than the atoms
of chitosan backbone. The groups on chitosan chain have higher
concentration on the (1 1 1) surface than (1 1 0) surface at the same
distance. It implied that the interaction mainly occurred between
atom of the groups on chitosan chain and the (1 1 1) surface of
Fe3 O4 . RDF further revealed the interaction among the atoms. The
hydrogen atoms of hydrogen and amino groups could form hydrogen bonds with the oxygen atoms on (1 1 1) and (1 1 0) surfaces.
However, the hydrogen atoms of groups have more probability of
forming hydrogen bonds with (1 1 1) than that of (1 1 0) surfaces.
Moreover, nitrogen atoms has stronger interaction with the iron
atoms of (1 1 1) than that of (1 1 0) surface. In summary according
to these results, the chitosan have more probability of adsorbing
on the (1 1 1) surface than (1 1 0) and (0 0 1) surfaces of Fe3 O4 .
Acknowledgments
This work was supported by the National Natural Science Foundation of China (nos. 21104023 and 21106052), and the Specialized
Research Fund for the Doctoral Program of Higher Education of PR
China (no. 20090061120078).
References
[1] R.Y. Hong, T.T. Pan, H.Z. Li, Microwave synthesis of magnetic
Fe3 O4 nanoparticles used as a precursor of nanocomposites and ferrofluids, J.
Magn. Magn. Mater. 303 (2006) 60–68.
[2] H. Zhou, J. Lee, T.J. Park, S.J. Lee, J.Y. Park, J. Lee, Ultrasensitive DNA monitoring
by Au–Fe3 O4 nanocomplex, Sens. Actuat. B Chem. 163 (2012) 224–232.
[3] J. Wang, Y. Chen, B.A. Chen, J.H. Ding, G.H. Xia, C. Gao, J.A. Cheng, N. Jin, Y.
Zhou, X.M. Li, M. Tang, X.M. Wang, Pharmacokinetic parameters and tissue
distribution of magnetic Fe3 O4 nanoparticles in mice, Int. J. Nanomed. 5 (2010)
861–866.
[4] F. Yan, X.W. Zheng, Z.M. Sun, A.H. Zhao, Effect of surface modification of
Fe3 O4 nanoparticles on the preparation of Fe3 O4 /polystyrene composite particles via miniemulsion polymerization, Polym. Bull. 68 (2012) 1305–1314.
[5] D. Stamopoulos, V. Gogola, E. Manios, E. Gourni, D. Benaki, D. Niarchos, M.
Pissas, Biocompatibility and solubility of Fe3 O4 –BSA conjugates with human,
Curr. Nanosci. Blood 5 (2009) 177–181.
[6] R. Riva, H. Ragelle, A. des Rieux, N. Duhem, C. Jerome, V. Preat, Chitosan and
chitosan derivatives in drug delivery and tissue enginnering, Adv. Polym. Sci.
244 (2011) 19–44.
[7] Y.J. Park, Y.M. Lee, S.N. Park, S.Y. Sheen, C.P. Chung, S.J. Lee, Platelet derived
growth factor releasing chitosan sponge for periodontal bone regeneration,
Biomaterials 21 (2000) 153–159.
[8] T. Ishii, Y. Okahata, T. Sato, Mechanism of cell transfection with plasmid/chitosan complexes, Bba-Biomembranes 1514 (2001) 51–64.
[9] L.L. Li, D. Chen, Y.Q. Zhang, Z.T. Deng, X.L. Ren, X.W. Meng, F.Q. Tang, J. Ren, L.
Zhang, Magnetic and fluorescent multifunctional chitosan nanoparticles as a
smart drug delivery system, Nanotechnology 18 (2007) 405102.
132
L. Qiang et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 419 (2013) 125–132
[10] Y. Sun, Z.L. Chen, X.X. Yang, P. Huang, X.P. Zhou, X.X. Du, Magnetic chitosan
nanoparticles as a drug delivery system for targeting photodynamic therapy,
Nanotechnology 20 (2009) 135102.
[11] D.T. Lam, H.N. Binh, V.H. Nguyen, V.T. Hoang, L.N. Huy, X.N. Phuc, Electrochemical detection of short HIV sequences on chitosan/Fe3 O4 nanoparticle based
screen printed electrodes, Mater. Sci. Eng. C Mater. 31 (2011) 477–485.
[12] J. Jia, Y.D. Huang, J. Long, J.M. He, H.X. Zhang, Molecular dynamics simulation
of the interface between self-assembled monolayers on Au(1 1 1) surface and
epoxy resin, Appl. Surf. Sci. 255 (2009) 6451–6459.
[13] H.P. Zhang, X. Lu, L.M. Fang, S.X. Qu, B. Feng, J. Weng, Atomic-scale interactions
at the interface of biopolymer/hydroxyapatite, Biomed Mater 3 (2008) 044110.
[14] J. Yue, X.C. Jiang, A.B. Yu, Molecular dynamics study on
Au/Fe3 O4 nanocomposites and their surface function toward amino acids, J.
Phys. Chem. B 115 (2011) 11693–11699.
[15] H. Sun, COMPASS: an ab initio force-field optimized for condensed-phase
applications-overview with details on alkane and benzene compounds, J. Phys.
Chem. B 102 (1998) 7338–7364.
[16] K.L. Tung, K.T. Lu, R.C. Ruaan, J.Y. Lai, Molecular dynamics study of the effect of
solvent types on the dynamic properties of polymer chains in solution, Desalination 192 (2006) 380–390.
[17] C.L. Li, P. Choi, Molecular dynamics study of the adsorption behavior of normal
alkanes on a relaxed alpha-Al2 O3 (0 0 0 1) surface, J. Phys. Chem. C 111 (2007)
1747–1753.
[18] S.S. Jawalkar, K.V.S.N. Raju, S.B. Halligudi, M. Sairam, T.M. Aminabhavi, Molecular modeling simulations to predict compatibility of poly(vinyl alcohol) and
chitosan blends: a comparison with experiments, J. Phys. Chem. B 111 (2007)
2431–2439.
[19] R. Ravindra, K.R. Krovvidi, Solubility parameter of chitin and chitosan, Carbohydr. Polym. 36 (1998) 121–127.
[20] H.S.C. O¡N̄eill, W.A. Dollase, Crystal structures and cation distributions in
simple spinels from powder XRD structural refinements: MgCr2 O4 , ZnCr2 O4 ,
Fe3 O4 and the temperature dependence of the cation distribution in ZnAl2 O4 ,
Phys. Chem. Miner. 20 (1994) 541–555.
[21] T. Yang, X. Wen, J. Ren, Y. Li, J. Wang, Surface structures of Fe3 O4 (1 1 1), (1 1 0)
and (0 0 1): a density functional theory study, J. Fuel Technol. Chem. 38 (1)
(2010) 121–128.
[22] S. Dong, X. Cui, S. Zhong, Y. Gao, Effects of temperature and concentration
on the structure of ethylene oxide–propylene oxide–ethylene oxide triblock
copolymer (Pluronic P65) in aqueous solution: a molecular dynamics simulation study, Mol. Simulat. 37 (2011) 1014–1022.
[23] M. Mirzaei, K. Ahmadi, Investigation of CH· · ·OC and NH· · ·OC hydrogenbonding interations in crystalline thmine by DFT calculations of O-17, N-14,
and H-2 NQR parameters, Biophys. Chem. 125 (2007) 411–415.
[24] N.D. Cuong, T.T. Hoa, D.Q. Khieu, T.D. Lam, N.D. Hoa, Synthesis, characterization, and comparative gas-sensing properties of Fe2 O3 prepared from Fe3 O4 and
Fe3 O4 –chitosan, J. Alloys Compd. 523 (2012) 120–126.