Effects of pH on heat transfer nanofluids containing ZrO2 and TiO2 nanoparticles
ABSTRACT In this paper, p H influences of zeta potential, particle size distribution, rheology, viscosity, and stability on heat transfer nanofluids are studied. Significant enhancement of thermal conductivity (TC) (>20 % ) containing 3 wt % zirconium dioxide ( ZrO 2) and titanium dioxide ( TiO 2) are observed near the isoelectric point (IEP). Meanwhile, at this IEP (p H ) , particle sizes, and viscosities of these nanofluids demonstrate a significant increase to maximum values. Experimental results also indicate that the stabilities of these nanofluids are influenced by p H values. The reasonable explanation for these interesting phenomena is that at this IEP, the repulsive forces among metal oxides are zero and nanoparticles coagulate together at this p H value. According to the Derjaguin–Landau–Verwey–Overbeek theory, when the p H is equal to or close to the IEP, nanoparticles tend to be unstable, form clusters, and precipitate. The resulting big clusters will trap water and the structures of trapped water are varied due to the strong atomic force among nanoparticles. Water is packed well inside and volume fraction of the nanoparticles will be larger. In addition, shapes of clusters containing trapped water will not be spherical but rather has irregular structure (like chains). Such structure favors thermal transport because they provide a long link. Therefore, overall TC of nanofluids is enhanced. Some literature results and conclusions related to p H effects of nanofluids are discussed and analyzed. Understanding p H effects may enable exploration of fundamental nature of nanofluids.
Effects of pH on heat transfer nanofluids containing ZrO2and TiO2
Carine Tchamakam Wamkam,1Michael Kwabena Opoku,1Haiping Hong,1,a?and
1Department of Material and Metallurgical Engineering, South Dakota School of Mines and Technology,
Rapid City, South Dakota 57701, USA
2U.S. Army Research Laboratory, Aberdeen Proving Ground, Maryland 21005, USA
?Received 10 September 2010; accepted 26 November 2010; published online 19 January 2011?
In this paper, pH influences of zeta potential, particle size distribution, rheology, viscosity, and
stability on heat transfer nanofluids are studied. Significant enhancement of thermal conductivity
?TC? ??20%? containing 3 wt % zirconium dioxide ?ZrO2? and titanium dioxide ?TiO2? are
observed near the isoelectric point ?IEP?. Meanwhile, at this IEP ?pH?, particle sizes, and viscosities
of these nanofluids demonstrate a significant increase to maximum values. Experimental results also
indicate that the stabilities of these nanofluids are influenced by pH values. The reasonable
explanation for these interesting phenomena is that at this IEP, the repulsive forces among metal
oxides are zero and nanoparticles coagulate together at this pH value. According to the Derjaguin–
Landau–Verwey–Overbeek theory, when the pH is equal to or close to the IEP, nanoparticles tend
to be unstable, form clusters, and precipitate. The resulting big clusters will trap water and the
structures of trapped water are varied due to the strong atomic force among nanoparticles. Water is
packed well inside and volume fraction of the nanoparticles will be larger. In addition, shapes of
clusters containing trapped water will not be spherical but rather has irregular structure ?like chains?.
Such structure favors thermal transport because they provide a long link. Therefore, overall TC of
nanofluids is enhanced. Some literature results and conclusions related to pH effects of nanofluids
are discussed and analyzed. Understanding pH effects may enable exploration of fundamental
nature of nanofluids. © 2011 American Institute of Physics. ?doi:10.1063/1.3532003?
Fluids containing nanoparticles, e.g., carbon nanotubes,
metal oxides ?called “nanofluids” hereafter? should exhibit
substantially improved thermal conductivity ?TC? values1–9
and could be useful for a variety of heat transfer related
applications including coolants and lubricants. However,
nanofluids utilizing a simple composite structure do not en-
hance the TC effectively. For instance, nanofluids with low
percentage metal oxide loading show no significant improve-
ments in TC, while loading of 4–5 vol % metal oxide only
results in about 10%–20% TC increase.
Theoretical studies have indicated that aggregation of
nanoparticles ?metal oxide, nanotube, etc.? into clusters
could enhance the TC significantly.10–12According to
mean field theory, pH could notably influence nanoparticle
Several efforts have been conducted related to the pH
effect in the heat transfer nanofluids. Xie et al.13reported
that enhanced TC ratio decreases with an increase in pH
value from 2 to 12 in aluminum oxide ?Al2O3? nanofluids. Li
et al.14showed that TC ratio increases as pH value increases
from 3 to 8.5–9.5 in a cupric oxide-water ?CuO–H2O? sys-
tem. Lee et al.15showed that effective TC ?Keff? increases by
a factor of 3 as pH decreases from 8 to 3 in CuO-water
nanofluids. Timofeeva et al.16reported that variation in the
pH between 5.5 and 10.3 results in significant ??34%? vis-
cosity drop in silicon carbide ?SiC?/water nanofluids, while
TC is not affected within the experimental uncertainty.
Unfortunately, the published results are controversial.
All authors discussed these results using the DLVO linear-
ized mean field theory, though, particle size distribution and
stability related to aggregation are obscured. According to
DLVO theory, nanoparticles become unstable and aggregated
when the pH is equal to or close to the isoelectric point
?IEP?, and it is well known that nanoparticles aggregation
could largely enhance the TC of nanofluids. Therefore, more
systematic studies of heat transfer nanofluids related to pH,
viscosity, particle size, TC, etc., become necessary.
ZrO2and TiO2nanoparticles are good components for
a?Electronic mail: email@example.com.
FIG. 1. TEM micrograph of ZrO2nanoparticles prepared by rapid drying of
a diluted sample.
JOURNAL OF APPLIED PHYSICS 109, 024305 ?2011?
0021-8979/2011/109?2?/024305/5/$30.00© 2011 American Institute of Physics
nanofluids.17–21Their TC and viscosity properties in aqueous
solutions are widely studied. Nevertheless, no further results
related to pH, particle size, and zeta potentials have been
In this paper, the pH effects of the TC, zeta potential,
particle size, viscosity of nanofluids containing TiO2and
ZrO2were investigated. Understanding these physical prop-
erties and behavior of such nanofluids at various pH values
would be quite valuable in analysis and synthesis of novel
Nanosize metal oxides ZrO2 ??100 nm? and TiO2
??100 nm? were purchased from Sigma Aldrich. Sonication
was performed using a Branson Digital Sonifier, Model 450.
TC data was obtained by the Hot Disk™ thermal con-
stants analyzer,22using the following parameters: measure-
ment depth of 6 mm, room temperature, power of 0.012 W,
measurement time of 15 s, sensor radius of 3.189 mm, tem-
perature coefficient of resistance of 0.0471/K, disk type Kap-
ton, and temperature drift rec yes.
pH values were measured by pH Mettler Toledo model
SevenEAsy S20. Viscosity and rheology data were recorded
by Brookfield LV DV-II+Prov Viscometer and Bohlin CVO
rheometer, a gap of 150 ?m was experimentally determined.
Zeta potentials and average particle sizes were measured by
Zetasizer Nano ZS and MPT-2 Autotitrator. Transmission
electron microscopy ?TEM? images were acquired with a Hi-
tachi H-7000 FA.
ZrO2and TiO2particles were dispersed in deionized
?DI? water by sonication. No surfactants or dispersants were
added to the fluids. Stability of fluids was observed by the
naked eye and recorded by a digital camera. Adjustment of
pH value was reached through addition of drops of acid ?HCl
or HSCH2CO2H? and base ?NaOH or CH3NH2?.
III. RESULTS AND DISCUSSION
TEM micrographs of ZrO2nanoparticles prepared by
rapid drying of a diluted sample are shown in Fig. 1. It is
apparent that ZrO2particles form large clusters consisting of
small primary particles and these particles are spherical or
near spherical. The average diameter of individual ZrO2par-
ticles is around 20–30 nm. The particle diameter measure-
ments from the TEM results are consistent with the data
provided by vendor. ??100 nm?.
ZrO2nanoparticle zeta potential and average particle
size versus varying pH values are shown in Fig. 2. It is clear
that the IEP ?zero charge? occurred around pH 6. More in-
teresting is that average particle size distribution is almost
symmetrical and reaches the maximum size near IEP. When
the pH shifts away from IEP, either increase or decrease, the
average particle size reduces significantly, from around 1200
to only 100 nm. The reduced particle size indicates that the
particles are less aggregated.
DLVO theory23developed by Derjaguin, Verwey, Lan-
dau, and Overbeek in the 1940s could be used to explain the
particle size changes under different pHs. Figure 3 illustrates
the schematic diagram of the variation in free energy accord-
ing to DLVO theory. The theory recognizes that particle sta-
bility depends on its total potential energy ?VT?, i.e., the sum
of the van der Waals attractive ?VA?, and electrical double
layer repulsive ?VR? forces.
VT= VA+ VR.
The attractive forces or van der Waals forces ?VA? are
weak forces which only dominate at short distance and are
VA= − A/?12?D2?,
where A is the Hamaker constant and D is the particle sepa-
The repulsive forces come from the electric surface
charge, which is influenced by the double layer and it is
VR= 2??a?2exp?− ?D?,
where a is the particle radius, ? is the solvent solubility, ? is
the zeta potential, ? is a function of the ionic composition
??−1is the length of the electric double layer?, and D is the
In DLVO theory, zeta potential plays a very important
role in the particle dispersion and stability. At the proper zeta
potential, particles can aggregate when they collide with suf-
ficient energy to overcome the energy barrier resulting from
the repulsive forces. The attractive forces can then prevail
and favor the irreversible aggregation.
FIG. 2. ?Color online? ZrO2zeta potentials and average particle size vs
different pH values.
FIG. 3. ?Color online? Schematic diagram of the variation in free energy
according to DLVO theory.
024305-2 Wamkam et al.J. Appl. Phys. 109, 024305 ?2011?
The average particle size distribution and zeta potential
measured in Fig. 2 are strongly coincident with the DLVO
theory. The zeta potential near the IEP demonstrates that the
repulsive force among ZrO2nanoparticles is close to zero
and nanoparticles should coagulate together under this zeta
potential value. Since zeta potential is correlative with pH
value, it is interesting to observe the dispersion properties of
ZrO2nanoparticels in water at different pH values.
Figure 4 shows the dispersion behavior of 3 wt % ZrO2
nanofluids at different pH ?4, 6, 8, 10? at 0 min and 24 h. The
reason to choose high weight concentration ZrO2instead of
low weight concentration ZrO2is that it is easier to observe
the particle aggregation and precipitation process with high
concentration. Initially, after appropriate sonication, all fluids
containing ZrO2nanoparticles with different pH ?4, 6, 8, 10?
show excellent dispersion and homogeneity. After 24 h, the
fluids with pH 4 and 10 maintain good dispersion and homo-
geneity. At the same time, precipitations are clearly observed
with fluids at pH 6 and 8. The dispersion results further
supported our explanation: pH 6 and 8 are relatively equal to
or close to the ZrO2IEP ?pH 6.2?. According to DLVO
theory, nanoparticle size increases significantly around the
IEP. Therefore, these nanoparticles become unstable, aggre-
gate, form clusters, and finally precipitate.
Table I lists the TC and viscosity of 3 wt % ZrO2at
different pH levels ?4, 6, 8, 10?. The density of ZrO2nano-
particles at 25 °C is measured at 5.89 g/ml. It is clearly seen
that TC of 3 wt % ZrO2nanofluids show significant enhance-
ment at pH 6 and 8. TC increased ratio is around 25%. At the
same time, TC of 3 wt % ZrO2nanofluids only show slight
enhancement ?about 6%–7%? at pH 4 and 10.
The above results can be explained as follows: the at-
tractive forces or Van der Waals forces among particles in-
crease with the pH of a suspension close to the IEP ?pH 6
and 8? and particles tend to aggregate and form clusters.
These factors result in the trapped water and irregular struc-
ture inside the clusters. The large volume fraction of the
nanoparticles and chain like structure due to trapped water
favors TC. The same phenomenon is observed in the struc-
ture of carbon nanotube grease.24Slight TC increase ?around
6%–7%? of 3 wt % ZrO2at pH 4 and 10 is primarily due to
the nanoparticles loading in water based fluids according to
the effective medium theory.
The viscosity values show the same trend as TC values
could be explained in the same way. The viscosities of 3
wt % ZrO2in water at pH 6 and 8 are much higher than that
at pH 4 and 8. At pH 8, the ratio of viscosity enhancement is
around 65%, while at pH 4, the enhancement is only around
is 20%, which is consistent with the viscosity of normal
nanofluids under 3 wt % ZrO2loading.
Due to the easy precipitation nature of nanofluids with
high weight percentage ZrO2loading, it is very difficult to
measure the rheology curve of such fluids. Therefore, the
low ZrO2concentration is used to avoid the precipitation
issue. Figure 5 shows the rheology curve of 0.1 wt % ZrO2at
different pH ?4, 6, 8, 10? in water. The rheology results in-
dicate that all of the nanofluids with different pH values are
non-Newtonian fluids. This indicates that viscosity decreases
with the increase in shear stress. As explained, viscosities at
pH 6 and 8 are higher than that of pH 4 and 10.
It is important to determine whether or not this behavior
occurs in other metal oxides besides ZrO2, or if these results
are independent of the type of nanoparticle. Therefore, we
choose TiO2nanoparticles to replace ZrO2and measure all
physical properties, since TiO2shows a different IEP and
density. Also, its average particle size distribution is asym-
Figure 6 shows TEM micrograph of TiO2nanoparticles
prepared by rapid drying of a diluted sample. A large cluster
is observed similar to the one found in ZrO2. The particles
are also spherical or near spherical. The average diameter of
TiO2particles is around 10–15 nm. The particle diameter
obtained from TEM results is consistent with the data pro-
vided by vendor ??100 nm?.
Figure 7 shows TiO2zeta potential and average particle
size versus different pH values. It is clearly shown that IEP
?zero charge? occurred around pH 4.5, and the average par-
ticle size reaches the maximum value near IEP. When the pH
FIG. 4. ?Color online? Dispersion behavior of 3 wt % ZrO2nanofluids at
different pH ?4, 6, 8, 10? at 0 min and 24 h.
TABLE I. TC and viscosities of 3 wt % ZrO2aqueous solutions at different
pH. ?Note: Density of ZrO2: 5.89 g/ml at 25 °C. TC of water: 0.60 W/mK.
Viscosity of water: 1 cp.?
Viscosity ?cP? shear rate 50 ?1/s?
FIG. 5. ?Color online? Rheology curve of 0.1 wt % ZrO2at different pH ?4,
6, 8, 10? in water.
024305-3 Wamkam et al. J. Appl. Phys. 109, 024305 ?2011?
shifts away from IEP, the average particle size is reduced
significantly, from around 1600 to about 400 nm, which in-
dicates that particles are less aggregated. It is interesting to
note that the particle size distribution of TiO2is not sym-
metrical. Furthermore, the particle size of TiO2is larger than
that of ZrO2at the least aggregation.
Dispersion behavior of 3 wt % TiO2nanofluids at differ-
ent pH ?4.5, 5.5, and 8.5? at 0 min and 24 h is also shown in
Fig. 8. Just as in ZrO2, 3 wt % high loading of TiO2fluids
are used to exaggerate the precipitation process. After appro-
priate sonication, all TiO2nanoparticles disperse well in wa-
ter and all fluids confirm homogeneity. After 24 h, the fluids
with pH 8.5 maintain good dispersion and homogeneity. At
the same time, precipitations are clearly observed with fluids
at pH 4.5 and 5.5. The explanation is the same as ZrO2.
Since IEP of TiO2is 4.5, pH 4.5 and 5.5 is quite equivalent
to or close to the IEP. Hence, TiO2nanoparticles are unstable
according to the DLVO theory. They form clusters, aggre-
gates, and precipitate. The dispersion results further support
our assumption. Since the particle size distribution is asym-
metrical, it will be quite interesting to study the dispersion
properties of the other side ?pH below 4.5?.
The significance of the above ZrO2and TiO2results is
that we have found the pH influence of zeta potential, par-
ticle size distribution, rheology, viscosity, and stability. These
factors finally affect the TC value of heat transfer nanofluids
and make significant TC enhancement possible ??20%?. The
experimental results demonstrate that TC enhancement is in-
dependent of the type of nanoparticle used. ZrO2and TiO2
show the same trend and similar enhancement.
As discussed in the introduction, Timofeeva et al.16re-
ported that viscosity of SiC/water nanofluids decreased with
a pH increase between 5.5 and 10.3 ??34% viscosity drop?,
while TC was not affected within the experimental uncer-
tainty. Several other authors13–15who studied the metal oxide
systems ?Al2O3and CuO? found significant TC changes with
the pH shifts, regardless of changes in TC values. The pos-
sible explanation for these inharmonic results may be due to
dispersion and aggregation. Timofeeva used the commercial
SiC nanoparticles which showed excellent dispersion and
stability in aqueous solutions. The fluids demonstrated the
Newtonian behavior and well dispersed particles were not
ideal for the aggregation. On the other hand, Al2O3and CuO
were poorly dispersed in water according to the author’s ex-
perimental descriptions. Hence, these nanoparticles were
more likely to form cluster and aggregation or agglomera-
tion. It is understandable that varied TC could occur in these
clusters and agglomerations.
Also, previous publications13,14explained the cause of
TC using the following terms: “The hydration forces among
particles increase near the IEP, which results in the enhanced
mobility of nanoparticles in the suspension. The microscopic
motions of the particles cause microconvection that enhances
the heat transfer process, or TC.” These explanations using
the micromotion do not apply much. When the particles are
larger, the micromotion will be much slower ?the Brownian
motion velocity is proportional to the inverse of the square
root of the particle mass?.
The exciting pH effect results discussed in this paper
may open a new way to better understand the fundamental
nature of engineering nanofluids with enhanced heat transfer
properties. With optimum concentration and pH value, nano-
particles could be suspended ?dispersed? well in fluids with-
out precipitation, trap as much as possible water, and vigor-
ously enhance thermal properties of nanofluids. Of course,
viscosity issue should be taken into consideration seriously.
It is well known that chemical surfactants could improve
the dispersion and stability of nanofluids. Therefore, more
studies to examine the pH influence of various physical pa-
rameters such as TC, zeta potential, particle size, viscosity,
etc., on nanofluids with the existence of chemical surfactants
are necessary. The work could possibly lead to better under-
standing of the fundamental nature of these novel engineer-
ing fluids ?nanofluids?.
Since Ti and Zr are transition metals, it will be quite
interesting to investigate more metals such as transition met-
als: Cu, Fe, Zn; alkali earth metal: Mg; other metal: Al. The
FIG. 8. ?Color online? Dispersion behavior of 3 wt % TiO2nanofluids at
different pH ?4.5, 5.5, 8.5? at 0 min and 24 h.
FIG. 6. TEM micrograph of TiO2nanoparticles prepared by rapid drying of
a diluted sample.
FIG. 7. ?Color online? TiO2zeta potentials and average particle size vs
different pH values.
024305-4Wamkam et al.J. Appl. Phys. 109, 024305 ?2011?
test results of those metal oxides should reveal more secrets
of physical properties of nanofluids and confirm our assump-
tion that at the IEP, repulsive forces among metal oxides is
zero and nanoparticles coagulate together under this pH
In summary, pH influence of zeta potential, particle size
distribution, rheology, viscosity, and stability on heat transfer
nanofluids were studied and significant enhancement of ther-
mal conductivities ??20%? containing 3 wt % metal oxides
ZrO2and TiO2were observed near the IEP. Meanwhile, par-
ticle sizes and viscosities of these nanofluids were increased
significantly until they reached a maximum.
Experimental results also indicated that the stabilities of
these nanofluids were influenced by pH values ?IEP?. The
reasonable explanation for these interesting phenomena is
that at the IEP, the repulsive forces among metal oxides are
zero and nanoparticles coagulated together under this pH
value. According to the DLVO theory, when the pH was
equal to or close to the IEP, nanoparticles were unstable,
tended to aggregate and form clusters, and finally precipitate.
The resulting big clusters will trap water and the structures of
trapped water are varied due to the strong atomic force
among nanoparticles. Therefore, water is packed well inside
and volume fraction of the nanoparticles will be larger. In
addition, shapes of clusters containing trapped water will not
be spherical but rather has irregular structure ?like chains?.
Such structure favors thermal transport because they provide
a long link. Therefore, overall TC of nanofluids is enhanced.
These conclusions could also help to explain some of the
earlier literature results. This research could possibly open
new doors to exploration of fundamental nature of nano-
The financial support of Army Research Laboratory ?Co-
operative Agreement No. W911NF-08-2-0022? and NASA
EPSCoR ?Award No. NNX09AU83A? are acknowledged.
The authors also thank Mark Horton of Material and Metal-
lurgical Engineering Department, South Dakota School of
Mines for providing help in the TC measurements of nano-
fluids, and Dr. William Cross for his valued comments.
1S. Choi, Z. Zhang, W. Yu, F. E. Lockwood, and E. A. Grulke, Appl. Phys.
Lett. 79, 2252 ?2001?.
2C. H. Li and G. P. Peterson, J. Appl. Phys. 99, 084314 ?2006?.
3P. Keblinski, J. A. Eastman, and D. G. Cahill, Mater. Today 8, 36 ?2005?.
4X. Wang, X. Xu, and S. Choi, J. Thermophys. Heat Transfer 13, 474
5D. S. Wen and Y. L. Ding, J. Nanopart. Res. 7, 265 ?2005?.
6J. Buongiomo, D. C. Venerus, N. Prabhat, T. McKrell, J. Townsend et al.,
J. Appl. Phys. 106, 094312 ?2009?.
7H. Hong, B. Wright, J. Wensel, S. Jin, X. Ye, and W. Roy, Synth. Met.
157, 437 ?2007?.
8B. Wright, D. Thomas, H. Hong, L. Groven, J. Puszynski, D. Edward, X.
Ye, and S. Jin, Appl. Phys. Lett. 91, 173116 ?2007?.
9M. Horton, H. Hong, C. Li, B. Shi, G. P. Peterson, and S. Jin, J. Appl.
Phys. 107, 104320 ?2010?.
10R. Prasher, W. Evans, P. Meakin, J. Fish, P. Phelan, and P. Keblinski,
Appl. Phys. Lett. 89, 143119 ?2006?.
11P. E. Gharagozloo, J. K. Eaton, and K. E. Goodson, Appl. Phys. Lett. 93,
12J. Wensel, B. Wright, D. Thomas, W. Douglas, B. Mannhalter, W. Cross,
H. Hong, J. Kellar, P. Smith, and W. Roy, Appl. Phys. Lett. 92, 023110
13H. Xie, J. Wang, T. Xi, Y. Liu, F. Ai, and Q. Wu, J. Appl. Phys. 91, 4568
14X. F. Li, D. S. Zhu, X. J. Wang, N. Wang, J. W. Gao, and H. Li, Thermo-
chim. Acta 469, 98 ?2008?.
15D. Lee, J. W. Kim, and B. G. Kim, J. Phys. Chem. B 110, 4323 ?2006?.
16E. V. Timofeeva, D. S. Smith, W. Yu, D. M. France, D. Singh, and J. L.
Routbort, Nanotechnology 21, 215703 ?2010?.
17X. Zhang, H. Gu, and M. Fujii, Int. J. Thermophys. 27, 569 ?2006?.
18W. Williams, J. Buongiorno, and L. W. Hu, J. Heat Transfer 130, 042412
19A. Turgut, I. Tavman, M. Chirtoc, H. P. Schuchmann, C. Sauter, and S.
Tavman, Int. J. Thermophys. 30, 1213 ?2009?.
20S. M. S. Murshed, K. C. Leong, and C. Yang, Int. J. Therm. Sci. 44, 367
21Y. He, Y. Jin, H. Chen, Y. Ding, D. Cang, and H. Lu, Int. J. Heat Mass
Transfer 50, 2272 ?2007?.
22Detail information see www.hotdisk.se
23R. J. Hunter, Foundations of Colloid Science, 1st ed. ?Clarendon, Oxford,
24H. Hong, D. Thomas, A. Waynick, W. Yu, P. Smith, and W. Roy, J.
Nanopart. Res. 12, 529 ?2010?.
024305-5Wamkam et al.J. Appl. Phys. 109, 024305 ?2011?