Journal of Colloid and Interface Science 310 (2007) 456–463

www.elsevier.com/locate/jcis

Evaluation on dispersion behavior of the aqueous copper nano-suspensions

Xinfang Li ∗ , Dongsheng Zhu, Xianju Wang
Key Lab of Enhanced Heat Transfer and Energy Conservation, Ministry of Education, School of Chemical and Energy Engineering,
South China University of Technology, Guangzhou 510641, PR China
Received 15 September 2006; accepted 2 February 2007
Available online 1 March 2007

Abstract
This paper presents a procedure for preparing a nanofluid which is solid–liquid composite material consisting of solid nanoparticles with sizes
typically of 1–100 nm suspended in liquid. By means of the procedure, Cu–H2 O nanofluids with and without dispersant were prepared, whose
sediment photographs and particle size distribution were given to illustrate the stability and evenness of suspension with dispersant. Aiming
at the dispersion of nano-Cu is regarded as the guide of heat transfer enhancement, the dispersion behavior of Cu nanoparticles in water were
studied under different pH values, different dispersant types and concentration by the method of zeta potential, absorbency and sedimentation
photographs. The results show that zeta potential has good corresponding relation with absorbency, and the higher absolute value of zeta potential
and the absorbency are, the better dispersion and stability in system is. The absolute value of zeta potential and the absorbency are higher at
pH 9.5. Hexadecyl trimethyl ammonium bromide (CATB) and sodium dodecylbenzenesulfonate (SDBS) can significantly increase the absolute
value of zeta potential of particle surfaces by electrostatic repulsions, and polyoxyethylene (10) nonyl phenyl ether (TX-10) can form a thick
hydration layer on the particle surfaces by steric interference, which leads to the enhancement of the stability for Cu suspensions. In the 0.1%
copper nano-suspensions, the optimizing concentrations for TX-10, CATB, and SDBS are 0.43, 0.05, and 0.07%, respectively, which have the
best dispersion results.
© 2007 Elsevier Inc. All rights reserved.

Keywords: Copper nano-suspensions; Dispersion and stability; Zeta potential; Absorbency

1. Introduction nanofluids are (1) better stability compared to those fluids

With rapid development of nanomaterial technology and
growing visibility of energy resource problems [1–3], re-
searchers started to apply nanomaterial technologies to the area
of heat transfer enhancement in order to develop a homoge-
neous, stable, new heat exchange medium of high conductivity.
Nanofluids, produced by dispersing nanoparticles into conven-
tional heat transfer fluids such as water, glycol, or oil, was
coined by Choi in 1995 at Argonne National Laboratory of
USA [4]. Nanofluids have been proposed to work as heat ex-
change medium of high efficiency, which have various expected
superior properties relative to those fluids without particles or
suspensions with large-size particles, and they offer exciting
possibilities due to their enhanced heat transfer performance
compared to ordinary fluids [5–14]. The advantages of these
* Corresponding author. Fax: +86 20 87114185.
E-mail address: xtulxf@163.com (X. Li).
0021-9797/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcis.2007.02.067
containing micro- or milli-sized particles and (2) higher ther-
mal conductive capability than the basic fluids. Nanofluids are
proposed for various uses in important fields such as elec-
tronics, transportation, medicine, and heating, ventilating, and
air-conditioning (HVAC) [15]. However, due to the high surface
energy of nanoparticles, it is easy for nanoparticles to coagu-
late and difficult to disperse in water. Therefore, controlling the
coagulation of nanoparticles in the nanofluid has become the
primary issue for the initial research of nanofluid. It is also im-
portant to investigate dispersion and stability of nanofluids in
order to exploit their potential benefits and applications.
In aqueous medium, the interaction between the colloid
particles is influenced by the local electric field developing
around the particles. The relation between the local electric field
and the stability of the lyophobic colloids is described in the
DLVO theory [16–18]. The theory considers the London–van
der Waals attractive forces (VA ) and the electrostatic repulsion
(VR ) between two charged particles due to the overlapping of
 X. Li et al. / Journal of Colloid and Interface Science 310 (2007) 456–463 457

electric double layers. In any colloidally stable suspensions the

overall particle interaction is repulsive. The thermal motion of
particles results in a certain collision frequency and different
rates of coagulation in unstable systems. In the case of slow
coagulation only few collisions result in dimer formation; how-
ever, if the attraction outweighs the repulsion, each collision
leads to the adhesion of particles in fast coagulation. The criti-
cal coagulation concentration (CCC) separates the fast from the
slow coagulation regime. Coagulation kinetics measurements
provide correct coagulation rates, stability ratio (W ) and CCC
values, and so the experimental characterization of colloidal
stability.
Fig. 1. TEM micrograph of nano-copper.
Studies to date have shown that the amount and the charge of
nanoparticles in the nanofluid, and the interaction between the
particles and the dispersant directly affect the stability of the
suspension. Agitation [19,20] and adding surfactants [21,22]
have been applied to reduce the coagulation of nanoparticles in
the nanofluid. Liu et al. [23] simulated the interaction between
particles based on the theory of stabilization by electrostatic
repulsion, and studied the stabilization mechanism adopting a
small-molecule anionic dispersant; Tang et al. [24] concluded
that nanoparticles exist as individually dispersed particles at
points far away from the isoelectric point. At points near the
isoelectric point, nanoparticles show various degrees of coagu-
lation. Studies of Jia et al. [25] showed that only under certain Fig. 2. Mass fraction vs particle size for nano-copper.
concentrations of dispersant can a suspension with good sta-
bility be obtained. Lisunova et al. [26] studied the stability of
the aqueous suspensions of nanotubes in the presence of non-
ionic surfactant, and found the addition of surfactant exerts a
stabilizing effect at surfactant concentration Cs proportional to
the weight concentration C of multiwalled carbon nanotubes
(a)
(MWNTs), Cs ≈ C mol/dm3 . These studies have given a good
foundation for highly homogeneous, industrial enhanced heat
transfer suspension fluids.
In this study, the dispersion behavior of Cu nanoparticles in
water were studied under different pH values, different disper-
sant types and concentration by the method of zeta potential, (b) (c)
absorbency and sedimentation photographs. It is expected to Fig. 3. Chemical structures of surfactants: (a) TX-10; (b) CATB; (c) SDBS.
provide guidance to design nanofluids with excellent perfor-
mance.
factant, hexadecyl trimethyl ammonium bromide (referred to as
CATB) in analytical grade, and an anionic surfactant, sodium
2. Experimental
dodecylbenzenesulfonate (referred to as SDBS) in chemical
grade, all from Guangzhou Chemical Reagent Factory (China),
2.1. Chemical
were used. The surfactant structures are shown in Fig. 3. The
water was purified by a Milli-Q Academic Millipore system.
Cu powder (Shenzhen Junye Nano Material Ltd, China) with
The pH was controlled using hydrochloric acid (HCl) and
copper content >99.9% was used in the study. The transmission
sodium hydroxide (NaOH) in analytical grade. All chemicals
electron micrograph (TEM) of Cu powder is shown in Fig. 1.
were used as received without any further purification.
The histogram of mass fraction vs particle diameter is shown
in Fig. 2. In Fig. 1, there are a few larger particles, which are
likely aggregates of the smaller ones, but the whole distribution 2.2. Evaluation of dispersion by sedimentation method
of the particles is relatively well-dispersed. The particles are
basically spherical or near spherical. Particle size is relatively Evaluation of dispersion characteristics of the copper sus-
consistent with a unimodal distribution and an average diameter pension was carried out using the conventional sedimentation
of 25 nm. method in the test tubes. The experiments were conducted us-
A non-ionic surfactant, polyoxyethylene (10) nonyl phenyl ing 0.1% copper nano-suspensions. Different concentrations
ether (referred to as TX-10) in chemical grade, a cationic sur- of the dispersant were added to the suspensions, which were
458 X. Li et al. / Journal of Colloid and Interface Science 310 (2007) 456–463
stirred thoroughly and ultrasonicated (KQ2200DE Ultrasonic
Cleanser, 100 W, Kunshan of Jiangsu Equipment Company,
China) for at least 1 h. 10 ml of each suspension was then
poured into test tubes, and then the samples were allowed to de-
posit for several days. The above solution for certain height of
the suspensions was taken and then absorbency of the suspen-
sions was measured on a U-3010 spectrophotometer (Hitachi
Instrument Inc., Japan) after depositing for 24 h. Meanwhile,
the states of sedimentation were observed every 24 h, and the
sedimentation photographs were then taken directly from the
test tubes. The pH value of system was adjusted with HCl and
NaOH solution by precise pH Meter (PHS-25, China).
2.3. Measurement of zeta potential
In investigation of both absorbency and sedimentation pho-
tographs, 0.1% copper nano-suspensions was used. However,
for measurement of zeta potential, a higher content of cop-
per nanoparticles was not suitable. Instead, dilute 0.05% (mass
fraction) copper nano-suspension was selected. After the sus-
pensions were stirred thoroughly and ultrasonicated for at least
15 min, 2–4 ml of suspensions was transferred into a measuring
cell. Then zeta potential was measured by a Malvern ZS Nano
S analyzer (Malvern Instrument Inc., London, UK). The mea-
surement was run at V = 10 V, T = 25 ◦ C with switch time at
t = 50 s. Each experiment was repeated at least ten times to cal-
culate the mean value of the experimental data. The pH value
of system was adjusted with HCl and NaOH solution by precise
pH Meter (PHS-25, China).
3. Results and discussion
3.1. Preparation of nanofluids
Two-step method was selected to prepare the nanofluids.
Preparation of nanofluids is the first key step in applying
nanophase particles to changing the heat transfer performance
of conventional fluids. The nanofluid does not simply refer to a
liquid–solid mixture. Some special requirements are necessary,
such as even suspension, stable suspension, durable suspension,
low agglomeration of particles, no chemical change of the fluid.
In general, these are effective methods used for preparation of
suspensions: (1) to change the pH value of suspensions; (2) to
use surface activators and/or dispersants; (3) to use ultrasonic
vibration. All these techniques aim at changing the surface
properties of suspended particles and suppressing formation of
particles cluster in order to obtain stable suspensions. It depends
upon the application case how these techniques are used.
Ultrasonication was used for preparation of mixed aqueous
nano-suspensions, which is an accepted technique for dispers-
ing the highly entangled or aggregated nanoparticle samples
[27,28], but longer time of high-energy sonication can introduce
defects. In the study, copper nanoparticle (0.1 g) and a water
solution (99.8 g) with CATB surfactant (0.1 g) were directly
mixed in a 150-ml beaker. The suspension was transferred into
an ultrasonic vibrator and sonicated for 1 h at a frequency of
40 kHz and an output power of 100 W at 25–30 ◦ C. For the
Fig. 4. Photographs of Cu–H2 O suspensions in the absence (a) and in the pres-
ence (b) of CATB dispersant depositing for a week. Concentration of Cu and
CATB are 0.1% (mass fraction).
comparison, the suspension without CATB dispersant was soni-
cated for 1 h in the same way. Their sedimentation photographs
are shown in Fig. 4, which shows that the stabilization of the
suspension with dispersant can last about 1 week in the sta-
tionary state and no sediment is found, while the suspension
without dispersant exhibits weaker dispersion and quickly oc-
curs aggregatation.
Fig. 5 illustrates the particle size distributions of Cu–H2 O
nano-suspensions in the absence (a) and in the presence (b) of
CATB dispersant, which shows that there are obvious variations
in the particle size characteristics between two samples. The
average particle sizes obtained are (a) in the absence of CATB
dispersant: 5560 nm, (b) in the presence of CATB dispersant:
130 nm. Therefore, the stabilization of Cu–H2 O suspension
with CATB dispersant is better, which exhibits considerable ac-
cordance as seen in Fig. 4.
3.2. Influence of pH on stability of copper nano-suspensions
The stability of copper powder suspension in aqueous so-
lution is closely related to its electrokinetic properties. Well-
dispersed suspension can be obtained with high surface charge
density to generate strong repulsive forces. Therefore, the study
of the electrophoretic behavior through measurement of the
zeta potential becomes important for understanding the disper-
sion behavior of copper particles in a liquid medium [29–31].
The zeta potential values of Cu–H2 O suspensions with differ-
ent dispersants at different pH values are presented in Fig. 6.
According to the zeta potential values of copper powders, pH
9.5 can be selected as an operating pH. Because, in this pH, the
absolute values of zeta potential for Cu–H2 O suspensions with
three kinds of dispersants are higher, ξCu (CATB) = 28.1 mV,
ξCu (SDBS) = −43.8 mV, and ξCu (TX-10) = −8.3 mV, re-
spectively. In the case of copper suspension with CATB dis-
persant, the copper powder is positively charged in the whole
pH ranges (from 2 to 12). At pH below 2, the zeta potential
of the particle surface (see Fig. 6) is at the minimum, there-
fore the force of electrostatic repulsion between particles is not
sufficient to overcome the attraction force between particles;
the absorbency (see Fig. 7) is smaller, and dispersion stabil-
 X. Li et al. / Journal of Colloid and Interface Science 310 (2007) 456–463
(a)
(b)
Fig. 5. Particle size distributions of Cu–H2 O suspensions in the absence (a) and in the presence (b) of CATB dispersant. Concentration of Cu and CATB dispersant
are 0.05% (mass fraction).
Fig. 6. Effect of pH on zeta potential of Cu–H2 O suspensions with different
dispersants. Concentration of Cu and dispersants are 0.05% (mass fraction).
Fig. 7. Effect of pH on absorbency of Cu–H2 O suspension with CATB disper-
sants. Concentration of Cu and CATB dispersant are 0.1% (mass fraction).
ity is poor. As pH increases, the zeta potential of the particle
surface increases, so the electrostatic repulsion force between
particles becomes sufficient to prevent attraction and collision
between particles caused by Brownian motion. Greater electro-
static force can also lead to more free particles by increasing
459
particle–particle distance so that the distance exceeds the hy-
drogen bonding range between particles and further reduces the
probability of particle coagulation and settling. The absorbency
also becomes higher with pH increase, improving the dispersion
stability of Cu nanoparticles. At pH around 9.5, the zeta poten-
tial and the absorbency become even higher; the electrostatic
repulsion force between particles is stronger, and the coagu-
lated particles can be redispersed through mechanical force,
so the dispersion stability of Cu nanoparticles is the best. As
pH value continues to further increase, the concentration of the
pH adjustment reagent (NaOH) in the system increases, which
causes the compression of electrical double-layer, so lowering
the zeta potential of the particle surface and electrostatic re-
pulsion force, and the suspension exhibits a poorer dispersion.
This analysis shows that the dispersion behavior of Cu nanopar-
ticles is affected by electrostatic repulsion force and is basically
agreement with the classical DLVO theory.
The sediment photographs of Cu–H2 O suspensions with
CATB dispersant as a function of pH is shown in Fig. 8. As
shown in this figure, copper particles tend to form agglomerated
structures at pH below 7, which result in rapid sedimentation of
particles and instability of the suspension. The copper suspen-
sions at pH 9–10 shows a good dispersion of copper particles
and the dispersion is maintained for a long period, which is at-
tributed to the higher charge on the surface of copper particles
(see Fig. 6). Therefore, the suspension at pH 9–10 is more sta-
ble compared to others which are far from this pH value. This
result is in good agreement with zeta potential and absorbency
tests.
According to the DLVO theory, the total interaction poten-
tial (VT ) was determined by a repulsive interaction energy due
to the overlap between the diffuse double layers of the particles
(VR ) and the attractive interaction energy due to London–van
460 X. Li et al. / Journal of Colloid and Interface Science 310 (2007) 456–463

Fig. 8. Sediment photographs vs pH depositing for a week. Concentration of

Cu and CATB are 0.1% (mass fraction).

der Waals dispersion (VA ). In principle, the presence of surfac- Fig. 9. The calculated DLVO interparticle interaction potentials between copper
particles at pH 3.
tant covering the external surface of copper particles should be
taken into account by means of an electrosteric potential. How-
ever, it has been shown that the adsorption of ionic surfactants
onto charged copper particles alters mainly the electrical inter-
action between the particles [32]; as a consequence, and for the
sake of clarity, the steric contribution could be neglected in or-
der to calculate the total interaction potential.
The total interaction energy as a function of the shortest
distance between the particle surfaces (H ) is assumed to be
the sum of all the attractive and repulsive potentials mentioned
above,
VT (H ) = VR (H ) + VA (H ). (1)
For small particle’s surface potentials (ψ0 ) and ka > 10, the
electrostatic repulsive energy (VR ) is given by the expression:
 
VR (H ) = 2πεaψ02 ln 1 + exp(−kH ) , (2)
where ε is the permittivity of the medium, and k −1 is the thick- Fig. 10. The calculated DLVO interparticle interaction potentials between cop-
ness of the electrical double layer surrounding the particle, or per particles at pH 9.5.
Debye’s length.
The van der Waals attractive energy (VA ) between spherical forces are dominant interparticle forces. This leads to forma-
particles is given by tion of flocculated structures, which result in rapid settling of
    particles in the copper suspensions. When the pH value reaches
A 2a 2 2a 2
VA (H ) = − + to 9.5 (see Fig. 10), calculation indicates a net repulsive in-
6 H 2 + 4aH H 2 + 4aH + 4a 2
  teraction between copper nanoparticles. The repulsion barrier
H 2 + 4aH between nanoparticles hinders agglomeration of the particles,
+ ln , (3)
H 2 + 4aH + 4a 2 which results in stable copper suspension.
where H is the separation distance between the surfaces of the
spheres, and A is the Hamaker constant, whose value depends 3.3. Influence of the dispersant on stability of copper
on the properties of the particles and the dispersing medium. At nano-suspensions
close separations (H  a) the above equation may be simpli-
fied considerably: 3.3.1. Nonionic dispersant
The measurement of the absorbency has become an impor-
aA tant evaluation for understanding the dispersion behavior of
VA (H ) = − . (4)
12H copper particles in a liquid medium. The greater the absorbency
According to Eqs. (1), (2), and (4), the calculated DLVO in- is, the higher the particle concentration in the suspension is,
terparticle interaction potentials [33] between copper nanopar- hence the more stable the suspension is. Fig. 11 shows the ef-
ticles at pH 3 and 9.5 are shown in Fig. 9 and Fig. 10, respec- fect of nonionic dispersant TX-10 concentration on the stability
tively. As shown in Fig. 9, at pH 3, the attractive van der Waals of Cu nanoparticles at pH 9.5 for the 0.1% nano-suspensions.
 X. Li et al. / Journal of Colloid and Interface Science 310 (2007) 456–463
Fig. 11. Effect of TX-10 concentration on the zeta potential and absorbency
(pH 9.5). Concentration of Cu and TX-10 for the measurement of zeta potential
are 0.05% (mass fraction), and concentration of Cu and TX-10 for the measure-
ment of absorbency are 0.1% (mass fraction).
Fig. 12. Sediment photographs vs TX-10 concentration depositing for a week:
(a) 0.03, (b) 0.10, (c) 0.20, (d) 0.43, and (e) 0.62%. Concentration of Cu is 0.1%
(mass fraction).
The results indicate: as the concentration of TX-10 increases,
Cu particles maintain a negative surface potential with no obvi-
ous change in zeta potential, i.e. dispersant TX-10 has no effect
in modifying the surface characteristics of Cu particles. As the
TX-10 concentration increases, the absorbency is observed to
increase, resulting in a maximal value, thereby suggesting good
dispersion system. The optimum concentration of the TX-10 is
found to be 0.43% (mass fraction). Variation of sediment pho-
tographs at different TX-10 concentration is due to the different
extents of agglomeration/dispersion of the particles. The higher
sediment height occurred at 0.43% TX-10 (see Fig. 12), which
is in good agreement with absorbency tests (see Fig. 11).
The mechanism of TX-10 on the copper suspension can be
steric hindrance. It exists as charge neutral molecules in wa-
ter: there is no ionization. Due to its affinity for water, there
is a thick hydration layer around each molecule. The effect of
steric stabilization is apparent. At low concentration, there are
not enough surfactant molecules to completely cover the sur-
face of the particles. The adsorption saturation is not achieved
and the side chain of a surfactant molecule on one particle can
461
Fig. 13. Effect of CATB concentration on the zeta potential and absorbency
(pH 9.5). Concentration of Cu and CATB for the measurement of zeta potential
are 0.05% (mass fraction), and concentration of Cu and CATB for the measure-
ment of absorbency are 0.1% (mass fraction).
attach to a different particle and cause coagulation and settle-
ment, lowering the dispersion stability. With the increase of
TX-10 concentration, the surfaces of particles are covered more
and more by dispersant molecules, increasing the thickness of
electrical double-layer. The dispersion stability of Cu nanopar-
ticles is the best at 0.43% TX-10 in the suspension. Here, the
steric stabilization mechanism is achieved. When more TX-10
is added into the suspension, the amount of surfactant adsorbed
on the surface of particles reaches the stage of super satura-
tion. Parts of the TX-10’s hydrophobic groups may extend into
water, therefore the excessive TX-10 molecules can bridge be-
tween particles and cause coagulation, which is of no advantage
for dispersion.
3.3.2. Cationic dispersant
Fig. 13 shows the effect of cationic dispersant CATB con-
centration on the stability of 0.1% copper nano-suspensions
at pH 9.5. The results show: with increasing CATB concen-
tration, the surface charges on Cu particles remain positive,
and the zeta potential and the absorbency both increase ini-
tially and then decrease, correspondingly, dispersion stability
also increases initially and then decreases. The optimum con-
centration of the CATB is found to be 0.05% (mass fraction).
At this concentration, the zeta potential and the absorbency are
maximal, and the system is more stable, which is also corrobo-
rated from the sediment photographs (see Fig. 14). This is due
to CATB, a quaternary ammonium cationic dispersant, can par-
tially ionize in water and give cationic species. The mechanism
of CATB on the copper suspension can be electrostatic stabi-
lization. When CATB concentration is lower, the amount of
the dispersants adsorbed on the particle surface is lower, hence
the zeta potential on the particle surface is lower, resulting in
weak dispersion system. With increasing CATB concentration,
the amount of the dispersants adsorbed on the particle surface
increases, and the zeta potential also increases, improving the
dispersion stability of Cu nanoparticles. But when excessive
amount of CATB is added, due to increasing ionic strength, the
electrical double-layer becomes compressed and leads to de-
creasing the dispersion stability of Cu nanoparticles.
462 X. Li et al. / Journal of Colloid and Interface Science 310 (2007) 456–463
Fig. 14. Sediment photographs vs CATB concentration depositing for a week:
(a) 0.02, (b) 0.04, (c) 0.05, (d) 0.08, and (e) 0.12%. Concentration of Cu is 0.1%
(mass fraction).
Fig. 15. Effect of SDBS concentration on the zeta potential and absorbency
(pH 9.5). Concentration of Cu and SDBS for the measurement of zeta potential
are 0.05% (mass fraction), and concentration of Cu and SDBS for the measure-
ment of absorbency are 0.1% (mass fraction).
3.3.3. Anionic dispersant
The influence of anionic dispersant SDBS concentration on
the zeta potential and the absorbency of 0.1% copper nano-
suspensions at pH 9.5 is shown in Fig. 15. It can be seen that
the copper powder is negatively charged by adding SDBS dis-
persant. It may be due to dissociation of SDBS in the solution
producing phenyl sulfonic group adsorbed on the copper parti-
cle, which in turn increases the net negative charge of powder
surface and also increases the repulsive forces. As the disper-
sant concentration increases, the zeta potential is observed to
increase, resulting in a maximal value, thereby suggesting good
dispersion system. The optimum concentration of the SDBS is
found to be 0.07% (mass fraction), which is also corroborated
from absorbency test. At this concentration, the zeta potential
and the absorbency is maximal, and the system is more stable,
which suggests SDBS dispersant yields well dispersed system.
Fig. 16 shows the sediment photographs of the copper sus-
pension at different SDBS concentration after depositing for
a week. As seen from Fig. 16, the following results can be
reached: the stability of the suspension becomes better and bet-
ter when SDBS concentration changes from 0 to 0.07%, and the
best stability is reached with 0.07% SDBS. Furthermore, the
Fig. 16. Sediment photographs vs SDBS concentration depositing for a week:
(a) 0.03, (b) 0.05, (c) 0.07, (d) 0.09, and (e) 0.12%. Concentration of Cu is 0.1%
(mass fraction).
stability shows decreasing trend when SDBS concentration in-
creases further. A plausible dispersion mechanism of SDBS on
the copper suspension is as follows: SDBS can partially ionize
in water and give anionic species, while Cu nanoparticles carry
positive charge in a neutral aqueous medium, having a strong
affinity for anionic groups. When SDBS concentration is lower,
the negatively charged phenyl sulfonic group dissociated from
SDBS is adsorbed on the positively charged copper surface and
consequently the surface is negatively charged because of the
ionization of SDBS, thus effect of electrostatic stabilization is
obtained. With the increase of SDBS, on the hand, the anion
groups pushed into the adsorbed layer makes the absolute value
of zeta potential increase, resulting in increasing the repulsive
forces between the particles, thereby increasing the thickness of
electrical double-layer; on the other hand, large numbers of an-
ion groups in the aqueous suspension hinders the particles from
colliding and then reduces the collision probability between the
particles. Here, the adsorbed phenyl sulfonic group and a sig-
nificant electrical double-layer were simultaneously presented.
When more SDBS is added into the suspension, the concentra-
tion of Na+ increases with SDBS concentration increase, and
the Na+ group entering into the absorbed layer reduces the net
charge of powder surface and makes the absolute value of zeta
potential decrease, resulting in weak dispersion system.
4. Summary
Based on the work done in this research, the following gen-
eral conclusions can be presented:
• Zeta potential and absorbency are important basis for se-
lecting conditions for dispersing particles. There is a good
correlation between absorbency and zeta potential. The
higher the absolute value of zeta potential is, the greater
the absorbency is, and the better the stability of the system
is.
• Cu–H2 O nanofluids by two-step method were prepared.
The observed sediment photographs and particle size dis-
 X. Li et al. / Journal of Colloid and Interface Science 310 (2007) 456–463 463

tribution show better dispersion behavior in the suspension [4] S.U.S. Choi, ASME FED 231 (1995) 99.
with the addition of dispersant. [5] J.A. Eastman, Appl. Phys. Lett. 78 (2001) 718.
• The effect of pH on the stability of the copper suspension [6] S.U.S. Choi, Appl. Phys. Lett. 79 (2001) 2252.
[7] H.E. Patel, Appl. Phys. Lett. 83 (2003) 2931.
was critical. At pH 9.5, a good dispersion of copper par- [8] H.Q. Xie, J.C. Wang, T.G. Xi, Y. Liu, F. Ai, J. Appl. Phys. 91 (2002) 4568.
ticles was obtained which is attributed to charge build up [9] H.T. Zhu, Y.S. Lin, Y.S. Yin, J. Colloid Interface Sci. 277 (2004) 100.
on the surface of copper particles due to adding the disper- [10] P. Vassallo, Int. J. Heat Mass Trans. 47 (2004) 407.
sants. The calculated DLVO interparticle interaction poten- [11] P. Keblinski, J.A. Eastman, D.G. Cahill, Mater. Today 8 (2005) 36.
tials verified the experimental results of the pH effect on [12] Y. Yang, Z.G. Zhang, E.A. Grulke, W.B. Anderson, G. Wu, Int. J. Heat
Mass Trans. 48 (2005) 1107.
the stability behavior of the copper suspensions. Therefore, [13] Y.L. Ding, H. Alias, D.S. Wen, R.A. Williams, Int. J. Heat Mass Trans. 49
the high charged formation around copper particles and a (2006) 240.
good dispersion of suspension was verified at pH 9.5. [14] S.Z. Heris, S.G. Etemad, M.N. Esfahany, Int. Commun. Heat Mass Trans-
• The stability behavior of the 0.1% copper suspensions with fer 33 (2006) 529.
different dispersant type and concentration was investi- [15] S. Zussman, Public communication, Argonne National Laboratory, IL,
USA, 2002.
gated at pH 9.5 by the techniques of zeta potential, ab- [16] R.J. Hunter, Foundations of Colloid Science, vol. 1, Clarendon Press, Ox-
sorbency and sedimentation photographs. It was indicated ford, 1987.
that CATB and SDBS dispersants can significantly increase [17] D.J. Shaw, Introduction to Colloid and Surface Chemistry, Butterworths,
the absolute value of zeta potential of the particle surfaces London, 1966.
by electrostatic repulsions, and TX-10 can form a good [18] M. Elimelech, J. Gregory, X. Jia, R.A. Williams, Particle Deposition and
Aggregation: Measurement, Modeling and Simulation, Butterworths, Ox-
hydration layer around the particle surfaces by steric in- ford, 1995.
terference, which lead to the enhancement of the stability [19] D.G. Wei, R. Dave, R. Pfeffer, J. Nanopart. Res. 4 (2002) 21.
of the copper suspensions at pH 9.5. In the 0.1% cop- [20] S.M.S. Murshed, K.C. Leong, C. Yang, Int. J. Thermal Sci. 44 (2005) 367.
per nano-suspensions, the optimizing concentrations for [21] J.P. Li, K.H. Guo, D.Q. Liang, R.Z. Wang, Int. J. Refrig. 27 (2004) 932.
[22] Q. Kang, B.Y. Gao, J.T. Hu, D.Z. Shen, Adsorption 11 (2005) 519.
TX-10, CATB, and SDBS are 0.43, 0.05, and 0.07%, re-
[23] Y.Q. Liu, L. Gao, Chem. Res. Chin. Univ. 23 (2002) 2155.
spectively, which have the best dispersion results. [24] F.Q. Tang, X.X. Huang, Y.F. Zhang, J.K. Guo, Mater. Sci. Eng. 17 (1999)
8.
Acknowledgments [25] C.Z. Jia, Y.Q. Fan, H. Qi, N.P. Xu, Chem. Eng. 33 (2005) 47.
[26] M.O. Lisunova, N.I. Lebovka, O.V. Melezhyk, Y.P. Boiko, J. Colloid In-
terface Sci. 299 (2006) 740.
The authors would like to acknowledge the financial sup-
[27] M. Lindheimer, E. Keh, S. Zaini, S. Partyka, J. Colloid Interface Sci. 138
ports from the National Natural Science Foundation of China (1990) 83.
(Grant No. 20346001), Program for New Century Excellent [28] P. Cheol, O. Zoubeida, K.A. Watson, R.E. Crooks, J. Smith, S.E. Lowther,
Talents in University (Grant No. NCET-04-0826), Specialized J.W. Connell, E.J. Siochi, J.S. Harrison, T.L. Clair, Chem. Phys. Lett. 364
Research Fund for the Doctoral Program of Higher Education (2002) 303.
[29] H. Karimian, A.A. Babaluo, J. Eur. Ceram. Soc. 27 (2007) 19.
(Grant No. 20050561017) for the research work.
[30] H. Bala, W.Y. Fu, J.Z. Zhao, X.F. Ding, Y.Q. Jiang, K.F. Yu, Z.C. Wang,
Colloids Surf. A Physicochem. Eng. Aspects 252 (2005) 129.
References [31] C.R. Mayer, S. Neveu, F. Secheresse, V. Cabuil, J. Colloid Interface
Sci. 273 (2004) 350.
[1] L. He, D.S. Zhu, J. Heat Transfer 11 (2004) 371. [32] R. Porcel, A.B. Jódar, M.A. Cabrerizo, R. Hidalgo-Álvarez, A. Martín-
[2] S.W. Wang, D.S. Zhu, Adsorption 10 (2004) 47. Rodríguez, J. Colloid Interface Sci. 239 (2001) 568.
[3] D.S. Zhu, S.W. Wang, Solar Energy 73 (2002) 177. [33] D. Kong, H. Yang, Y. Yang, S. Wei, B. Cheng, Mater. Lett. 58 (2004) 3503.