The roles of hydrochloric acid and polyethylene glycol in magnetic fluids

Article abstract of DOI:10.1016/j.jmmm.2007.10.014

Increasing interest has been drawn to the studies of magnetic fluids due to their multiple applications from industry to medicine. However, further exploration is still required for the techniques of preparing satisfying, convenient and stable magnetic fluids. We explored characteristics of magnetic liquids prepared by employing co-precipitation techniques of hydrochloric acid (HCl) and polyethylene glycol (PEG), and the functions of HCl and PEG in the magnetic liquid. According to the improved technique, after preparing Fe₃O₄ by a co-precipitation method, hydrochloric acid and PEG2000 react with magnetic particles at a certain temperature to generate the anticipated magnetic nanoparticles. The process could be under an air atmosphere rather than a N₂ atmosphere. Compared with traditional techniques, the magnetic nanoparticles prepared by this method have smaller size, better dispersion and stability, with the average hydrodynamic diameter adjustable between 8 and 50 nm. This study revealed that reduction of nanoparticles size is not mainly due to a [Cl^-] coating over the magnetic nanoparticles, but that HCl reacts with Fe₃O₄ particles after being heated. Meanwhile, PEG can stabilize or coat Fe₃O₄ nanoparticles as a dispersing and stabilizing agent.

Full text of DOI:10.1016/j.jmmm.2007.10.014

ARTICLE IN PRESS
Journal of Magnetism and Magnetic Materials 320 (2008) 1328–1334
www.elsevier.com/locate/jmmm
The roles of hydrochloric acid and polyethylene glycol in magnetic fluids
1
Zhang Yangde , Zeng Zhaowu , Zhou Weihua, Liu Xingyan, Li Zhenfa, Li Jun,
1
Ã
Xie Jianfeng, Luo Yulin, Hu Tiehui, Pan Yifeng
National Key Laboratory of Nanobiological Technology, Institute of Biomedical Engineering of Central South University, No. 87 Xiangya Road,
Changsha, Hunan 410008, PR China
Received 27 July 2007; received in revised form 11 October 2007
Available online 25 October 2007
Abstract
Increasing interest has been drawn to the studies of magnetic fluids due to their multiple applications from industry to medicine.
However, further exploration is still required for the techniques of preparing satisfying, convenient and stable magnetic fluids. We
explored characteristics of magnetic liquids prepared by employing co-precipitation techniques of hydrochloric acid (HCl) and
polyethylene glycol (PEG), and the functions of HCl and PEG in the magnetic liquid. According to the improved technique, after
preparing Fe O by a co-precipitation method, hydrochloric acid and PEG2000 react with magnetic particles at a certain temperature to
3
4
2
generate the anticipated magnetic nanoparticles. The process could be under an air atmosphere rather than a N atmosphere. Compared
with traditional techniques, the magnetic nanoparticles prepared by this method have smaller size, better dispersion and stability, with
the average hydrodynamic diameter adjustable between 8 and 50 nm. This study revealed that reduction of nanoparticles size is not
ꢀ
mainly due to a [Cl ] coating over the magnetic nanoparticles, but that HCl reacts with Fe O particles after being heated. Meanwhile,
4
3
3 4
PEG can stabilize or coat Fe O nanoparticles as a dispersing and stabilizing agent.
r 2007 Elsevier B.V. All rights reserved.
Keywords: Magnetic fluid; Hydrochloric acid; PEG
1
. Introduction
already been studied thoroughly [9–13]. The most common
magnetic fluids are formed by ultrafine particles with
superparamagnetism through dispersion with the aid of
surfactants in a liquid carrier. In general, magnetic fluids
have high intensity of saturation magnetization and good
stability under operating temperature. First of all, the
preparation of magnetic nanoparticle is of significance [14].
As for their applications in biomedicine, the particle size,
distribution of particle size, shape, and surface conditions
of magnetic nanoparticles are the key factors. At present,
many approaches have been developed for preparing
magnetic Fe O nanoparticles, such as polyols [15],
microemulsions [16], laser pyrolysis [17], sonochemical
synthesis [18], and chemical co-precipitation [19], etc.
Among these reported methods, the chemical co-precipita-
tion might be the most promising one because of its
simplicity and productivity [20]. However, it is very difficult
to control particle size and its distribution by this method,
which leads to its disadvantage in further applications in
biomedicine. Moreover, the ferromagnetic nanoparticles
Magnetic fluid (MF) is a kind of stable colloid system
composed by magnetic nanoparticles dispersed in a suitable
liquid carrier. The magnetic fluid generally used today was
first synthesized in early 1960s [1]. Magnetic fluids consist
of colloid suspensions of nanoparticles of ferromagnetic or
ferrimagnetic materials dispersed and stabilized in a carrier
liquid through some suitable organic surfactant. The
coated surfactant forms an entropic repulsion between
particles, so that thermal agitation alone is sufficient to
maintain the stable dispersion among them [2]. Because of
the special properties of magnetic fluids, they are widely
applied in the area of sealing, bearing, grinding, speaker,
damper and so on [3–8]. Their properties, such as
magnetism, heat transfer and optical properties, have
3
4
Ã
Corresponding author. Tel.: +86 731 4327991; fax: +86 731 4327987.
E-mail address: artgreenking@126.com (P. Yifeng).
These authors contributed equally to this paper.
1
0304-8853/$ - see front matter r 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmmm.2007.10.014

ARTICLE IN PRESS
Z. Yangde et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 1328–1334
1329
are difficult to disperse in aqueous solution. As a result,
large amount of dispersing agents are required, as is
reported in some researches. In this study, we have
improved the co-precipitation procedure: NaOH is used
biofilms and pits on the coupon surface. The radius of
curvature of the AFM tip was ꢃ50 nm. The scan rate
varied between 1.0 and 2.0 Hz depending on the image size.
The Digital Nanoscope software was used to analyze the
topographic images of the coupon surface. The magnetic-
particle-size distribution was investigated with Zetasizer
Nano (Malvern Instrument) based on dynamic laser
scattering, in which the particle size was measured by
detecting the Brownian motion of the particles through
probing the Doppler frequency shift of a scattered light
with respect to the incident light. It was noted that the
diameter detected via dynamic laser scattering was hydro-
dynamic diameter. The magnetization properties of the
magnetic fluid were examined by a vibration sample
magnetometer (VSM) with model HH-20 (Nanjing Uni-
versity Instrument Plant).
to reduce the irritation caused by NH OH; HCl and
4
PEG2000 are adopted to control the average particle size
and the distribution of Fe O magnetic nanoparticles.
3
4
Because of their non-toxicity, high hydrophilicity and good
controllability, PEG2000 and HCl are selected to prepare
the water-based magnetic fluid. In the experiment, we tried
to discuss the functions of HCl and PEG2000 in magnetic
fluid, the adjustability of average size and its distribution of
particle obtained by adjusting preparation parameters, and
the magnetism of PEGylated magnetic nanoparticles.
2
. Experiment
The magnetic Fe O particles were synthesized by an
3
3. Results and discussion
4
improved chemical co-precipitation procedure. There were
three stages. (1) Primary stage: Ferrous chloride tetra-
3.1. The role of HCl in the preparation process of magnetic
fluids
hydrate (FeCl ꢁ 4H O) and ferric chloride hexa-hydrate
2
(
1
FeCl ꢁ 6H O) with molar ratio of 1:1.5 were dissolved in
3
2
00 ml deionized water under vigorous stirring to prepare
The crystalline of magnetic particles was analyzed by
powder XRD. Fig. 1 shows the XRD pattern of the
synthesized and dried Fe O particles in the initial phase. It
ferrite solution of total concentration of 0.60 M. Sodium
hydroxide was then dissolved in deionized water to obtain
3
4
1
M sodium hydroxide (NaOH) as a base source. Then,
was clear that only the phase of Fe O was detectable.
3 4
through rigorous stirring, drops of 1 M sodium hydroxide
were added at 50 1C until pH value of 11 was reached. It
could be observed that the solution turned black due to the
formation of Fe O particles. The black mixture was then
There were not other phases such as Fe(OH) or Fe O ,
3 2 3
which are usual co-products in the chemical co-precipita-
tion procedure. The results shown in Fig. 1 reveal the high
purity of the synthesized Fe O magnetic particles. With
3
4
3
4
heated at 80 1C in an aqueous bath for 30 min to ripen. The
sediment obtained was washed for five times with distilled
water until pH value reached 7. (2) Pre-centrifugation stage:
The volume of upper-layer of solution is kept at 100 ml.
Fifty mililiters 0.5 M HCl and 1 g PEG2000 were then
added at 50 1C with intensive stirring for 1 h; then the
mixture was laid still for 24 h and then the upper layer of
solution was dumped out and kept for later use with
precipitation removed. (3) Post-centrifugation stage: High-
speed centrifugation at 15,000 ꢂ g for 15 min was carried
out for the above-mentioned upper layer solution, and
precipitates were collected. It was repeated one more time;
after that, distilled water was added to suspend the
precipitates with ultrasonic agitation, and then the suspen-
sion was laid still for 24 h. After that, it was filtrated with
the XRD pattern, the average core diameter of the particles
can be calculated from Scherrer equation [21]:
0
:94l
L ¼
,
(1)
Bð2yÞ cos y
where L is equivalent to the average core diameter of the
particles; l is the wavelength of the incident X-ray; B(2y)
denotes the full width in radian subtended by the half
maximum intensity width of the powder peak. l is
0.15406 nm; L is obtained as about 10 nm via Eq. (1) and
fit-size-only is 10.8 nm.
Fig. 2 shows the XRD pattern of the Fe O particles
3
4
synthesized with HCl but without PEG2000 and dried from
magnetic fluid before centrifugation. It can be noticed that
several phases including Fe O , FeCl and FeO(OH) were
3
4
2
2
00 nm filtrator. Finally, the purified water-based magnetic
fluid containing PEGylated Fe O nanoparticles collected in
detectable. The results shown in Fig. 2 reveal that the
synthesized Fe O magnetic particles reacted with HCl and
3
4
3
4
the void volume had a concentration of about 10mg Fe/ml
through adjustment. The whole process above was not
necessary to be in N atmosphere.
The crystallinity of magnetic particles was analyzed with
powder X-ray diffraction (XRD) of a Rigaku D/max2500
X-ray diffractometer by using Cu Ka radiation. The
morphology of the particles and their diameters were
observed by atomic force microscopy (AFM). A Nano-
scope AFM (MDF-3D, AR Company, USA) operating in
a contact mode in air was used to capture the images of
were oxidized. Fig. 3 shows the XRD pattern of the Fe O
3 4
particles synthesized with both HCl and PEG2000 and
dried from magnetic fluid before centrifugation. It can be
noticed that several phases including Fe O , FeO(OH) and
2
3
4
FeOCl were detectable. Fig. 4 shows the X-ray powder
diffraction pattern of the dried particles from magnetic
fluid by using HCl and PEG2000 after centrifugation. It
can be noticed that only the phase of Fe O was detectable.
3
4
There were no other phases such as FeO(OH), FeOCl and
FeCl . If HCl was added to aqueous solution containing
2

ARTICLE IN PRESS
Z. Yangde et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 1328–1334
1330
Fig. 1. X-ray powder diffraction pattern of the synthesized magnetic particles in the initial phase.
Fig. 2. X-ray powder diffraction pattern of the Fe
centrifugation.
3 4
O particles synthesized with HCl but without PEG2000 and dried from magnetic fluid before
magnetic particles under heating, it would become trans-
parent after reflux condensation, but almost lose its
magnetism. Obviously, Fe O particles reacted with HCl
further demonstrated that their particle size could not be
reduced by adding sodium chloride into the suspension
containing Fe O particles. If HCl reacted excessively with
3
4
3
4
and then generated FeCl and FeO(OH) after being heated;
2
Fe O particles after being heated, they would become
3 4
however, when HCl and PEG2000 were added simulta-
neously, it mainly generated Fe O magnetic nanoparticles.
It probably was because that HCl eroded some Fe O
ferric ion or ferrous ion and almost lose their magnetism.
It is presumed generally that Fe O particles generated
3
4
3
4
under alkalinity condition are colloids with positive charge
that played an important role in the stability of magnetic
fluids. When Zeta electric potential is high, the particles
retain distance from each other, weakening and counter-
acting van der Waals force to improve the stability of
particles in fluids; on the contrary, when Zeta electric
potential is low, the repulsion of particles decreases and
3
4
magnetic nanoparticles and reduced their particle size, and
then PEGylated surface layer was generated to prevent
Fe O nanoparticles from further eroding. So, it showed
3
4
that reduction of nanoparticles size was not mainly due to
ꢀ
a [Cl ] coating over magnetic nanoparticles but that HCl
reacted with Fe O particles after being heated. It could be
3
4

ARTICLE IN PRESS
Z. Yangde et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 1328–1334
1331
Fig. 3. X-ray powder diffraction pattern of the Fe
centrifugation.
3 4
O particles synthesized with both HCl and PEG2000 and dried from magnetic fluid before
Fig. 4. X-ray powder diffraction pattern of the dried particles from magnetic fluid by using HCl and PEG2000 after centrifugation.
particles are approaching gradually, therefore they will
agglomerate through attracting each other in the range of
van der Waals force. pH value of the solution decreases
and its electropositivity increases after HCl is added. The
larger the amount of HCl is, the higher the positive charge,
the higher Zeta electric potential and the better stability
will be [22]. When dealing Fe O nanoparticles with HCl,
particles agglomeration as a big particle is corroded slightly
by HCl to single particles, and superficial sticky layer is
generated which can prevent the decomposition and
further agglomeration of nanoparticles, furthermore, the
layer can remove coordinated water molecules on the
surface of nanoparticles to prevent the generation of
hydrogen bond among them. They suspend in aqueous
solution because of thermal motion. Moreover, PEG can
coat the nanoparticles or be absorbed on their surface to
carry the same electric charge when HCl and PEG act
together. At the same time, PEG can overcome the surface
free energy of nanoparticles to prevent their agglomeration
and further eroding by HCl, so the size of nanoparticles
varies over a range and they suspend in the base solution
without precipitation. Thus, the effect of HCl and PEG
together is better than that of HCl or PEG alone. But it
brings some by-products by dissolving Fe O particles after
3
4
3
4
HCl is added, so the amount of HCl has to be restricted,
neither excessive nor too thick. It may increase the
concentration of chloridions in the solution after adding
HCl and thus affects Zeta electric potential; however, this
ꢀ
is smaller than the [OH ]’s elevation effect on electric
potential. Therefore, they have to be removed after
centrifugation to reduce the amount of HCl and weaken
ꢀ
the effect of [Cl ].

ARTICLE IN PRESS
Z. Yangde et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 1328–1334
1332
3
.2. The role of PEG in magnetic fluids
ism, the mutual magnetic effects among them, and the
tendency of mutual aggregation increase, then the particles
deposit finally [23]. It was found that the amount of PEG
used together with HCl reduced tens of times than that
used alone, thus it would benefit further application. Fe O
magnetic fluids generated by adding HCl and PEG are still
stable after standing for several months.
It is common to add surfactants in magnetic fluids when
a mass of agglomeration is generated during precipitating,
and standing due to that nanoparticles absorb large
amount of adsorbed water, coordinated water, bridge
3
4
ꢀ
ꢀ
[
stability and dispersibility, we select PEG as a surfactant.
OH ] and non-bridge [OH ] and so on. For safety,
ꢀ
After PEG is added, it can displace H O, [OH ] on the
2
3.3. The size and size distribution of magnetic fluids
surface of particles, generating a sticky layer by adhering to
them through coordination bond or hydrogen bond,
reducing the surface free energy, preventing agglomeration
from small particles to big particles, and dispersing
particles stably in base solution. So PEG acts as a
dispersing and stabilizing agent. The larger the amount of
PEG is, the better the effect of dispersing and stabilizing of
nanoparticles, the smaller their sizes, and the better the
stability of magnetic fluids will be. However, when the
amount of PEG is large, the effect of mutual penetration
and winding should not be ignored, because it will aid in
the agglomeration from small particles to big particles.
When their sizes become larger and larger, their magnet-
In order to investigate the shape and size of magnetic
particles, AFM analysis system was used. The typical shape
and size of the magnetic particles are shown in Fig. 5; the
PEGylated Fe O nanoparticles are close to spherical, with
3
4
average particle size of 42.53 nm. In order to investigate the
size distribution of magnetic particles, a dynamic laser
scattering (DLS) analysis system was used in this study. It
was notable that the diameter detected by DLS system
referred to a hydrodynamic diameter of particles. Accord-
ing to DLS results, the poly index of the size distribution of
the Fe O particles ranged from 0.159 to 0.359. The typical
3
4
size distribution of the magnetic fluid is shown in Fig. 6.
2
1
1
0
0
.0
.5
.0
.5
.0
2
1
0
-
-
1
2
0.0
0.5
1.0
µm
1.5
2.0
dxy = 42.53 nm
6
nm
4
2
0
0.0
0.5
1.0
1.5
2.0
µm
Fig. 5. AFM images of the dried particles from magnetic fluid by using both HCl and PEG2000 after centrifugation. The average size of magnetic particles
is 42.53 nm.

ARTICLE IN PRESS
Z. Yangde et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 1328–1334
1333
Size Distribution by Intensity
6
4
2
0
0
0
0
1
1
2
0
8
6
4
2
0
-
-
-
20
40
60
0
.1
1
10
100
1000
10000
Diameter (nm)
Fig. 6. Hydrodynamic diameter distribution of magnetic PEGylated
Fe particles in magnetic fluid. The average size is 44.76 nm and the
poly index is 0.359.
3 4
O
-
6000
-4000
-2000
0
2000
4000
6000
Applied field (Oe)
Size Distribution by Intensity
Fig. 8. The magnetic hysteretic curve of the dried PEGylated Fe
particles at room temperature.
3 4
O
1
1
5
0
5
0
8
6
4
2
0
0.1
1
10
100
1000
10000
Diameter (nm)
Fig. 7. Hydrodynamic diameter distribution of magnetic PEGylated
Fe particles in magnetic fluid. The average size is 11.77 nm and the
poly index is 0.159.
-
2
3 4
O
-
4
-
6
8
The average particle size is 44.76 nm for this size
distribution.
In order to the reduce size distribution, Fe O particles
-
3
4
-6000
-4000
-2000
0
2000
4000
6000
of various average sizes can be obtained through control-
ling the amount of HCl and PEG2000. Fig. 7 plots the size
distribution of Fe O particles, via greater amount of HCl
and PEG2000 in the synthesis process or extending the
duration at 50 1C. It shows a smaller average size
Applied field (Oe)
Fig. 9. The magnetic hysteretic curve of magnetic fluid at room
temperature.
3
4
(
¼ 11.77 nm). Through the preparation procedure devel-
superparamagnetism of magnetic fluids is also observed
when dispersing the PEGylated Fe O particles into
oped in this study, the average size of the synthesized
Fe O particles could be adjusted from 8 to 50 nm,
depending on the decomposition of HCl and PEG2000 in
the ferrite suspension. And it was found that after adding
HCl and PEG2000, the size of the Fe O particles as end-
3
4
aqueous solution, as shown in Fig. 9. In our experiment,
the PEGylated Fe O particle can be stably dispersed in
water without self-agglomeration.
3
4
3
4
3
4
product was weakly related to that of the initial Fe O
3
4. Conclusion
4
particles, even if it agglomerated to micrometer particles in
the initial phase, it could generate nanometer particles
eventually. Thus, it facilitated the preparation of magnetic
fluids.
In this study, we have developed a route for the
preparation of magnetic fluid with an improved chemical
co-precipitation procedure. Compared with the traditional
techniques, the PEGlyated-Fe O particles prepared by this
3
4
3
.4. The magnetism of magnetic fluids
method have smaller size, better dispersion and stability,
and need less amount of PEG. It was found that, with the
addition of HCl and PEG2000 into a ferrite suspension, the
average size of magnetic Fe O nano-particles in magnetic
The magnetic properties of magnetic fluid were examined
by using the VSM. Fig. 8 shows the magnetic hysteresis
curve of dried particles of PEGylated Fe O at room
temperature. The magnetic hysteretic curve exhibits
super-paramagnetic behavior and strong magnetism. The
3
4
fluid could be significantly reduced. In addition, by
adjusting the amount of HCl and PEG2000, the average
hydrodynamic diameter of magnetic Fe O particles could
3
4
3
4

ARTICLE IN PRESS
Z. Yangde et al. / Journal of Magnetism and Magnetic Materials 320 (2008) 1328–1334
1334
be controlled between 8 and 50 nm. This study also
revealed that reduction of nanoparticles size is not mainly
[7] D. Li, H. Xu, X. He, et al., J. Magn. Magn. Mater. 289 (2005)
19.
[8] V. Polevikov, L. Tobiska, J. Magn. Magn. Mater. 289 (2005) 379.
9] G.S. Park, S.H. Park, IEEE Trans. Magn. 38 (2002) 957.
10] F.R. Cunha, Y.D. Sobral, Physica A 343 (2004) 36.
4
ꢀ
due to a [Cl ] coating over the magnetic nanoparticles, but
that HCl reacts with Fe O particles after being heated.
[
3
4
[
Meanwhile, PEG can stabilize or coat the Fe O nano-
3
particles as a dispersing and stabilizing agent.
4
[11] K. Shimada, H. Oka, J. Magn. Magn. Mater. (2005) 290.
[12] S. Shuchi, T. Mori, H. Yamaguchi, IEEE Trans. Magn. 38 (2002)
3
13] S. Li, Y. Song, Mechatronics 17 (8) (2007) 442.
234.
[
[
[
[
[
References
14] L.B. Yu, et al., Mater. Chem. Phys. 66 (2000) 6.
15] G. Viau, F. Fievet-Vicent, F. Fievet, J. Mater. Chem. 6 (1996) 1047.
16] P. Tartaj, C. Serna, Chem. Mater. 14 (2002) 4396.
17] S. Veintemillas-Verdaguer, M. Morales, C. Serna, Mater. Lett. 35
[
[
[
[
[
[
1] J. Chen, X. Chen, S. Pu, Z. Di, Y. Xia, Opt. Commun. 276 (2) (2007)
68.
2] P.C. Fannin, C. MacOireachtaigh, J. Magn. Magn. Mater. 316 (2)
2007) e802.
3] X. Chen, Q.D. Zhang, H.C. Chong, et al., IEEE Trans. Magn. 37
2001) 805.
4] E.J. Rhee, M.K. Park, R. Yamane, et al., J. Magn. Magn. Mater. 252
2002) 350.
5] E. Uhlmann, G. Spur, N. Bayat, et al., J. Magn. Magn. Mater. 252
2002) 336.
6] A. Nethe, T. Scholz, H.D. Stahlmann, et al., IEEE Trans. Magn. 38
2002) 1177.
2
(
1998) 227.
(
[
[
18] K. Suslick, M. Fang, T. Hyeon, J. Am. Chem. Soc. 118 (1996) 11960.
19] D. Kim, Y. Zhang, W. Voit, K. Rao, M. Muhammed, J. Magn.
Magn. Mater. 225 (2001) 30.
(
[
20] W. Jiang, H.C. Yang, S.Y. Yang, et al., J. Magn. Magn. Mater. 283
(
(
2004) 210.
[
[
[
21] B.E. Warren, X-ray Diffraction, Dover, Canada, 1990.
22] C. Li, F. Pan, Y. Yang, J. Wuhan Eng. Inst. 17 (2005) 18.
23] G. Qu, Z. Sun, Chem. Reagents 15 (1993) 234.
(
(