J. Am. Ceram. Soc.
, 85 [9] 2161 65 (2002)
journal
Electrophoretic Deposition Behavior of
Aqueous Nanosized Zinc Oxide Suspensions
National Institute for Materials Science, Tsukuba, Ibaraki 305-0047, Japan
The rheological behavior and electrophoretic deposition (EPD)
of ZnO nanopowder (nano-ZnO) in aqueous media have been
described. A cationic polyelectrolyte (polyethylenimine, PEI)
was used to disperse and modify the surfaces of the ZnO
nanoparticles. The rheological properties of the ZnO aqueous
suspension were investigated by measuring the viscosity versus
the pH and amount of dispersant. The EPD processing was
conducted via cathodic electrodeposition, using stable suspen-
sions with low viscosity, and the depositional behavior was
investigated. Bubble-free nano-ZnO deposits with uniform
microstructures were successfully obtained, which was an
indication of good sintering behavior.
I. Introduction
I NTEREST increasingly has been focused on the processing strat-
egies of inorganic ceramic materials.1 Colloidal processing,
which is one of the recently developed approaches, has been
proved to be a very promising method in the processing of
nanostructured powders, because of its advantages in diminishing
agglomeration and defect populations.2,3 Electrophoretic deposi-
tion (EPD) is a straightforward method for the assembly of
charge-carrying particles on an electrode from stable colloidal
suspensions in a dc electric field.4 6 Compared with convention-
ally slip casting, the thickness and morphologies of materials
obtained via EPD can be controlled precisely by varying the
electrochemical parameters; moreover, uniform and complex-
shaped deposits with high green densities on the electrode are
obtained readily at a high deposit speed. This fact is prominent in
the case of nanopowder suspensions. Nonaqueous suspensions
often have been chosen to produce bubble-free materials.57 EPD
in aqueous media should have an advantage over nonaqueous
media, because of the benign environment and low cost.8 11 To
suppress the electrolysis of water, which causes bubble contami-
nation, the applied voltage should be maintained at a value that is
less than the electrolysis voltage of water (1.23 V at 25C);12,13
however, this methodology is not practical, because that restriction
makes the deposition rate too slow. The use of porous molds to
pass the generated gas through the mold is not considered to be
sufficient.14 Our previous studies have shown that bubble-free
deposits can be successfully obtained via deposition onto palla-
dium cathodes, because palladium readily absorbs the hydrogen
that evolves because of the electrolysis of water.15,16
Zinc oxide (ZnO) is well-known as a semiconductor material
with a band gap of 3.2 eV; this material has potential application
in sensors, solar cells, photo-electronic devices, etc.17,18 However,
D. W. Johnson Jr.contributing editor
Manuscript No. 187126. Received February 21, 2002; approved June 10, 2002.
Author FT would like to acknowledge the support from the Science and
Technology Agency of the Japanese Government, through an STA postdoctoral
fellowship.
*Member, American Ceramic Society.
2161
Fengqiu Tang, Tetsuo Uchikoshi, and Yoshio Sakka*
very few studies have focused on the colloidal behavior of ZnO
powders,19,20 and much fewer studies have focused on the EPD
processing of ZnO. We are interested in the rheological and EPD
properties of nanosized ZnO (nano-ZnO) powder, because the
stability of the suspension affects the deposition characteristics.
The rheological properties of the nano-ZnO aqueous suspensions
that have been prepared under various conditions, and the EPD
properties from well-dispersed suspensions, are of great interest to
us. ZnO is well-known to be an amphoteric oxide, and it is known
to be easily dissolved in both acids and bases. ZnO is hydrated
to form the Zn2 cation at acidic pHs of 6.3, and it forms
hydroxide layers in water at basic pHs, where Zn(OH)2 is in
equilibrium with the Zn2, Zn(OH), Zn(OH) 2
3 , and Zn(OH) 4
species. At pH 12, the latter two zincate ions become the
dominant species in solution.19 In this study, the pH range of 711
has been chosen to prepare the suspension; the ZnO particle is
usually negatively charged in this pH range. To obtain bubble-free
deposits using a palladium cathode,15 the ZnO particles should be
positively charged. In this study, a cationic electrolyte has been
used to give the ZnO particles a positive charge potential and good
dispersion in aqueous media. Dense, bubble-free ceramic deposits
with uniform microstructure have been fabricated using nano-ZnO
via aqueous EPD processing.
II. Experimental Procedure
ZnO nanopowder with an average particle size of 40 nm
(Sakai Chemical Industry Co., Ltd., Sakai, Japan) was used.
A cationic polyelectrolytepolyethylenimine (PEI,
O(CH2OCH2ONHO)n)with an average molecular weight of
10000 (Wako Pure Chemical Industry, Ltd., Tokyo, Japan) was
used to modify the ZnO nanopowder, and 5-vol%-ZnO aqueous
suspensions were prepared. Reagent-grade hydrochloric acid
(HCl) and ammonium hydroxide (NH4OH) were used for pH
adjustment. The suspensions were ultrasonicated (Model USP-
600, Shimadzu, Inc., Kyoto, Japan) at a power of 160 kW for 10
min, to break the agglomeration,21 and then were stirred for at least
12 h, to ensure the equilibrium of polymer adsorption. The zeta
potential () versus pH of the suspensions was measured on a
zeta-potential analyzer (Model LEZA-600, Otsuka Electronics
Co., Ltd., Tokyo, Japan) in a 102M sodium chloride (NaCl)
solution with and without PEI addition. The rheological properties
of the ZnO suspensions were studied by measuring the viscosity
versus the pH, as well as the amount of dispersant, on an R-type
rotational viscometer (Model RC-500, Toki Sangyo Co., Ltd.,
Tokyo, Japan). Well-dispersed suspensions with low viscosity
were used for the EPD processing. The deposition was performed
using a dc power source (Model Potentiostat/Galvanostat
HA-3001A, Hokuto Denko, Tokyo, Japan) that was operating at a
constant current of 0.375 mA/cm2. A stainless-steel sheet was used
as an anode, and a palladium sheet with a deposition area of 8 cm2
was used as a cathode. They faced each other at a distance of 2 cm.
The relation between the deposit weight and time was investigated
at different suspension pHs. The obtained deposits were dried in
air at room temperature and weighed together with the cathodic
substrates to determine the deposit weight. After drying, the
deposits were separated from the substrate and sintered at different
2162 Journal of the American Ceramic SocietyTang et al. Vol. 85, No. 9

Fig. 1. TEM micrograph of ZnO powder.

temperatures. The green and sintered densities of the ZnO com-

pacts were measured using the Archimedes method. The micro-
structure of original ZnO powders was characterized via transmis-
sion electron microscopy (TEM) (Model 2000FX, JEOL, Tokyo,
Japan). The microstructure of the obtained green and sintered
compacts was observed via scanning electron microscopy (SEM)
(Model JSM-5400, JEOL).

III. Results and Discussion

(1) Zeta Potential of Zinc Oxide Powder
A TEM micrograph of the ZnO used in this experiment is
shown in Fig. 1. The ZnO particles are rodlike; the average length
of a particle is 50 nm, and its average width 25 nm. The
powder seemed well dispersed, and no large agglomerates were
present. ZnO is easily dissolved in acid; therefore, the zeta
potential of the ZnO nanopowder was measured at basic pHs, as
shown in Fig. 2. The isoelectric point (IEP) of the powder is pH
9.6. ZnO has a relatively high positive charge at pH 8, and that
positive charge decreases up to the IEP and then becomes a
negative charge with further increasing pH. To modify the posi-
tively charged ZnO particles, a cationic polyelectrolyte was chosen
as a dispersant. PEI with an average molecular of 10000 was
determined to be the most effective polyelectrolyte, compared to
other amine electrolytes in this study. Figure 2 shows that PEI
addition gives the ZnO surface a high positive charge in the pH
region of 711. Then, the surface charge changes to a negative
charge rapidly at pH 11, and the IEP shifts to pH 11.3 (Fig. 2),
which indicates that the PEI is an effective dispersant to modify
the ZnO nanopowder. PEI has imido (ONH) groups that readily

Fig. 3. Shear stress of 5-vol%-ZnO suspensions with various PEI

additions ((a) 1 wt%, (b) 1.5 wt%, and (c) 2 wt% PEI) versus shear rate at
different pH.

adsorb protons in the solution; PEI becomes charged positively,

according to the following reaction:22
O(CH2OCH2ONHO)n H3O
3 O(CH2OCH2ONH
2 O) n H 2O

The protonation of PEI has been reported to occur at pH

10.8.23 Here, probably because of the formation of zinc hydrox-
ide, PEI will absorb more protons that remain in the suspension,
which causes the ZnO particles to have a positive charge over a
Fig. 2. Zeta potential of ZnO with and without PEI addition. larger pH range. Figure 2 shows that the surface-modified ZnO
September 2002 Electrophoretic Deposition Behavior of Aqueous Nano-ZnO Suspensions
Fig. 4. Viscosities of ZnO suspensions versus pH, at different PEI
contents, at a shear rate of 200 s1.
particles with PEI could be deposited on the cathode at pH 8 11
when a dc electric current is supplied.
(2) Rheology of Zinc Oxide Suspensions
The rheological behavior of the ZnO suspensions was charac-
terized by measuring their viscosity and shear stress. In the
absence of PEI or a small amount (0.5 wt%) of PEI, the
suspensions were too viscous for the viscosity to be measured with
our equipment at any pH. Figures 3(a), (b), and (c) show the shear
stress versus shear rate of 5-vol%-ZnO suspensions, with the
variation in pH and the amount of PEI (1, 1.5, and 2 wt%,
respectively). For pH 8.5 and pH 10, regardless of the amount
of PEI, the viscosities of the suspensions are shear thinning, and
the shear stresses increase nonlinearly as the shear rate increases;
these traits are characteristic of pseudoplastic fluids with a thixo-
tropic structure, which indicates particle aggregation and suspen-
sion flocculation. Within the pH range of 8.510, the viscosities of
the suspensions remain very low and almost constant under
different shear rates, and the shear stresses are proportional to the
shear rate; these traits are characteristic of Newtonian fluids, which
indicates that the particles are well dispersed and the suspension is
stable.
Figure 4 clearly shows the change in viscosity as a function of
pH at a shear rate of 200 s1. With increasing pH, the viscosity of
the suspensions sharply decreases (up to pH 8), whereas the
viscosity remains at a very low level (pH 8.510), which is
indicative of good dispersion of the suspensions. However, the
Fig. 5. Deposit weight of 5 vol% ZnO versus time, for different PEI contents ((a) 1.5 and (b) 2 wt%).
2163
Fig. 6. Deposit weight of ZnO suspensions, as a function of pH, for a
deposition time of 2 min.
viscosity increases rapidly at pH 10. At higher pH (pH 10), as
more PEI was added, a lower viscosity was observed. This is due
to the protonation of PEI, which occurs at pH 10.8;23 therefore,
the viscosity decreases as the uncharged PEI content increases at
higher pHs.
Figure 4 also indicates that the viscosities of the suspensions
with PEI contents of 1.5 and 2 wt% are lower than that with 1 wt%
PEI added. Stable suspensions that have the lowest viscosity (in
the range of pH 8.510) are considered to be suitable for EPD.
Hence, EPD properties have been investigated for the suspensions
in this pH region with PEI additions of 1.5 and 2 wt%.
(3) Electrophoretic Deposition Behavior of
Zinc Oxide Particles
Figure 5 shows the deposit weight of the 5-vol%-ZnO suspen-
sions versus time during EPD prepared at different pH and PEI
contents. The deposition rate is dependent on deposition time,
suspension pH, and the amount of dispersant added. The deposit
weight increases almost linearly as the deposition time increases in
the measured pH region; the deposition rate also becomes fast
from pH 8.5 to pH 10, and the deposition rate with 1.5 wt% PEI
added is slightly greater than that with 2 wt% PEI added. Figure 6
clearly shows the tendency of deposit weight versus pH after
deposition for 2 min. The deposition rate first decreases from pH
7.8 to pH 8.5, then begins to gradually increase from pH 8.5 to pH
10.5, which is related to the viscosity of the suspension (see Fig.
4). In this experiment, no deposition occurred at pH 7.8 and pH
10.5, probably because it is very difficult for the particles to
2164 Journal of the American Ceramic SocietyTang et al.
Fig. 7. Voltage variation of the 5-vol%-ZnO suspension with 1.5 wt%
PEI addition at pH 9 in a electric current of 0.375 mA/cm2.
move to the oppositely charged electrode in such a viscous
suspension. The deposition rate in the pH ranges of 7.5 8.5 and
10 10.5 is higher than that in the pH range of 8.510, depending
on the amount of PEI addition. This phenomenon is because some
flocculation of the suspension can enhance the deposition rate to
some extent.11 In the case of pH 8.510, although the deposition
rate is slightly lower, steady and thick deposits were prepared.
Therefore, the deposits obtained in this pH region were sintered to
observe the microstructure.
Figure 7 shows the voltage variation of the suspension with 1.5
wt% PEI added at pH 9 and a current of 0.375 mA/cm2. The
voltage jumped to 2.7 V in the first 10 s, followed by a gradually
slow increase with the deposition time; the voltage is maintained at
3 V. The voltage is much higher than the electrolysis voltage of
water; however, uniform and bubble-free deposits can be obtained,
because of the excellent hydrogen adsorption property of the
Fig. 8. Microstructure of green deposits with 1.5 wt% PEI addition (a) at pH 9 and sintered for 2 h at temperatures of (b) 1200, (c) 1250, and (d) 1350C).
Vol. 85, No. 9
palladium electrode. If the current is 1 mA/cm2 in this experi-
ment, the voltage will be 50 V and increase rapidly as the
deposition time increases; bubbles that are produced from elec-
trolysis of the water will appear on the deposits, because of the
high generation rate of hydrogen, the limitation of the hydrogen
adsorption rate, and the capacity of the palladium electrode.
(4) Density and Microstructure of Green and Sintered Bodies
The green densities of the deposits prepared in the pH range of
8.510 with different PEI contents have been measured; all relative
densities of the green bodies are 43% 45%, which is sufficient for
nanosized powders. All the surfaces of the green bodies appear to
be uniform, and they contain no bubbles that have been caused by
water electrolysis. Figure 8 shows typical SEM micrographs of
green deposits that have been prepared at pH 9 with 1.5 wt% PEI.
The microstructure of the deposits is homogeneous and dense; no
macroscopic or microscopic pores that are caused by water
electrolysis could be observed on the deposit surface. The SEM
images show that bubble-free deposits with uniform surfaces are
readily prepared using the EPD processing.
The deposits with 1.5 wt% PEI added were sintered at different
temperatures. The density of the samples prepared at pH 8.5 and 9
attained 99% of the theoretical density (TD) when sintered at
1250C; however, that of the sample prepared at pH 10 attained
only 98.5% TD when sintered at 1350C. This discrepancy may be
due to the poor particle packing and the formation of more
hydroxide and zincate ions at high pH; the latter would be
decomposed during sintering and small pores would remain at the
grain boundaries. As a result, a higher sintering temperature was
required to eliminate those pores. Figures 8(b), (c), and (d) show
the microstructure of the deposits with 1.5 wt% PEI added at pH
9, sintered at different temperatures. In the microstructure of the
compact that was sintered at 1200C, uniformly distributed small
pores, which did not originate from hydrogen bubbles, are mostly
observed at the grain boundaries. This condition may be related to
September 2002 Electrophoretic Deposition Behavior of Aqueous Nano-ZnO Suspensions
the particle packing; sintering at high temperature can diminish
these pores. Figures 8(c) and (d) show a dense and uniform
microstructure. The grain sizes of the compacts sintered at 1250
and 1350C are 15 and 25 m, respectively. Further careful
control of the sintering procedure can produce sintered bodies with
a denser and more homogeneous microstructure.
IV. Conclusions
Polyethylenimine (PEI) is an effective surfactant to disperse
zinc oxide nanopowder (nano-ZnO) suspensions and modify the
particles to become positively charged. Well-dispersed ZnO sus-
pensions have been obtained in the pH range of 8.510 with the
addition of 12 wt% PEI; these suspensions are suitable for EPD
processing. Uniform and bubble-free nano-ZnO deposits with
good sinterability have been obtained in this experiment. Aqueous
EPD processing has been demonstrated to be an easy approach to
produce dense, uniform, bubble-free nanosized ceramic deposits
on a palladium cathode; this procedure has significant potential for
the economic mass production of a wide variety of components.
References
1 18
M. Trau, D. A. Saville, and I. A. Aksay, Field-Induced Layering of Colloidal
Crystals, Science, 272, 706709 (1996).
2 19
F. F. Lange, Powder Processing Science and Technology for Increased Reliabil-
ity, J. Am. Ceram. Soc., 72 [1] 315 (1989).
3 20
F. Q. Tang, L. Yu, X. Huang, and J. K. Guo, Characterization of Adsorption and
Distribution of Polyelectrolyte on Stability of Nano-Zirconia Suspensions by Auger
Electron Spectroscopy, Nanostruct. Mater., 11 [4] 44150 (1999).
4 21
P. Sarkar and P. S. Nicholson, Electrophoretic Deposition (EPD): Mechanics,
Kinetics, and Application to Ceramics, J. Am. Ceram. Soc., 79 [8] 19872002
(1996).
5 22
T. Ishihara, K. Shimose, T. Kudo, H. Nishiguchi, T. Akbay, and Y. Takita,
Preparation of Yttria-Stabilized Zirconia Thin Films on Strontium-Doped LaMnO3
Cathode Substrates via Electrophoretic Deposition for Solid Oxide Fuel Cells, J. Am.
Ceram. Soc., 83 [8] 192127 (2000).
6
B. D. Hatton and Y. Sakka, Influence of Washing on Zirconia Powder for
Electrophoretic Deposition, J. Am. Ceram. Soc., 84 [3] 666 68 (2001).
2165
7
A. M. Affoune, B. L. V. Prasad, H. Sato, and T. Enoki, Electrophoretic
Deposition of Nanosized Diamond Particles, Langmuir, 17 [2] 54751 (2001).
8
R. Moreno and B. Ferrari, Effect of the Slurry Properties on the Homogeneity of
Alumina Deposits Obtained by Aqueous Electrophoretic Deposition, Mater. Res.
Bull., 35, 88797 (2000).
9
R. C. Bailey, K. J. Stevenson, and J. T. Hupp, Assembly of Micropatterned
Colloidal Cold Thin Films via Microtransfer Molding and Electrophoretic Deposi-
tion, Adv. Mater., 12 [24] 1930 34 (2000).
10
M. Holgado, F. Garcia-Santamaria, A. Blanco, M. Ibisate, A. Cintas, H. Miguez,
C. J. Serna, C. Molpeceres, J. Requena, A. Mifsud, F. Meseguer, and C. Lopez,
Electrophoretic Deposition to Control Artificial Opal Growth, Langmuir, 15 [14]
4701704 (1999).
11
F. Q. Tang, T. Uchikoshi, K. Ozawa, and Y. Sakka, Electrophoretic Deposition
of Aqueous Nano--Al2O3 Suspensions, Mater. Res. Bull., 37 [4] 653 60 (2002).
12
A. T. Kuhn (ed.), Industrial Electrochemical Processes; p. 128. Elsevier,
Amsterdam, The Netherlands, 1971.
13
A. Formeto, L. Montanaro, and M. V. Swain, Micromechanical Characterization
of Electrophoretic-Deposited Green Films, J. Am. Ceram. Soc., 82 [12] 3521 49
(1999).
14
W. Ryan, E. Massoud, and C. Perera, Pottery Section-Fabrication by Electro-
phoresis, Trans. Br. Ceram. Soc., 80 [2] 46 47 (1981).
15
T. Uchikoshi, K. Ozawa, B. D. Hatton, and Y. Sakka, Dense, Bubble-Free
Ceramic Deposits from Aqueous Suspensions by Electrophoretic Deposition, J.
Mater. Res., 16 [2] 32123 (2001).
16
Y. Sakka, T. Uchikoshi, and B. Hatton, Processing of Zirconia and Alumina
Fine Particles through Electrophoretic Deposition, Stud. Surf. Sci. Catal., 132,
233236 (2001).
17
H. Nanto, T. Minami, and S. J. Takata, Zinc-Oxide Thin-Film Ammonia Gas
Sensors with High Sensitivity and Excellent Selectivity, Appl. Phys, 60 [2] 482 84
(1986).
J. S. C. Prentice, Optical Generation Rate of ElectronHole Pairs in Multilayer
Thin Film Photovoltaic Cells, J. Phys. D: Appl. Phys., 32 [17] 2146 50 (1999).
A. Degen and M. Kosec, Effect of pH and Impurities on the Surface Charge of
Zinc Oxide in Aqueous Solution, J. Eur. Ceram. Soc, 20, 66773 (2000).
P. Benezeth, D. A. Palmer, and D. J. Wesolowski, The Solubility of Zinc Oxide
in 0.03m NaTr as a Function of Temperature with in Situ pH Measurement,
Geochim. Cosmochim. Acta, 63 [10] 1571 86 (1999).
T. S. Suzuki, Y. Sakka, K. Nakano, and K. Hiraga, Effect of Ultrasonication on
the Microstructure and Tensile Elongation of Zirconia Dispersed Alumina Ceramics
Prepared by Colloidal Processing, J. Am. Ceram. Soc., 84 [9] 213234 (2001).
F. Q. Tang, X. X. Huang, Y. F. Zhang, and J. K. Guo, Effect of Dispersants on
Surface Chemical Properties of Nano-Zirconia Suspensions, Ceram. Int., 26 [1]
9399 (2000).
23
G. M. Lindquist and R. A. Stratton, The Role of Polyelectrolyte Charge Density
and Molecular Weight on the Adsorption of Colloidal Silica with Polyethylenimine,
J. Colloid Interface Sci., 55 [1] 4559 (1976).