Zirconia Nanoparticles Made in Spray Flames at High Production Rates

Article abstract of DOI:10.1111/j.1551-2916.2004.00197.x

Synthesis of zirconia nanoparticles by flame spray pyrolysis (FSP) at high production rates is investigated. Product powder is collected continuously in a baghouse filter unit that is cleaned periodically by air-pressure shocks. Nitrogen adsorption (BET), X-ray diffractometry (XRD), transmission electron microscopy (TEM), and thermogravimetric analysis (TGA) are used to characterize the product powder. The effect of powder production rate (up to 600 g/h), dispersion gas flow rate, and precursor concentration on product particle size, crystallinity, morphology, and purity is investigated. The primary particle size of zirconia is controlled from 6 to 35 nm, while the crystal structure consists of mostly tetragonal phase (80-95 wt%), with the balance monoclinic phase at all process conditions. The tetragonal crystal size is close to the primary particle size, which indicates weak agglomeration of single crystals.

Full text of DOI:10.1111/j.1551-2916.2004.00197.x

J. Am. Ceram. Soc., 87 [2] 197–202 (2004)
Zirconia Nanoparticles Made in Spray Flames at High Production Rates
Roger Mueller, Rainer Jossen, and Sotiris E. Pratsinis
journal
Particle Technology Laboratory, Department of Mechanical and Process Engineering, ETH Zurich,
CH-8092 Zurich, Switzerland
Mark Watson and M. Kamal Akhtar
Research Center, Millennium Chemicals, Inc., Glen Burnie, Maryland 21060
Hartmann et al.8 used the same process and precursor and
obtained white and fluffy ZrO2 powders with average particle
diameters between 10 and 50 nm. These particles were of the
metastable tetragonal phase with traces of the monoclinic
phase. The properties of the fumed oxides were primarily
controlled by the flame temperature, which depended on the
hydrogen to oxygen ratio. However, the most severe limiting
factor of the gas-phase synthesis is the lack of volatility of the
starting materials.7
Synthesis of zirconia nanoparticles by flame spray pyrolysis
(FSP) at high production rates is investigated. Product powder
is collected continuously in a baghouse filter unit that is
cleaned periodically by air-pressure shocks. Nitrogen adsorp-
tion (BET), X-ray diffractometry (XRD), transmission electron
microscopy (TEM), and thermogravimetric analysis (TGA)
are used to characterize the product powder. The effect of
powder production rate (up to 600 g/h), dispersion gas flow
rate, and precursor concentration on product particle size,
crystallinity, morphology, and purity is investigated. The
primary particle size of zirconia is controlled from 6 to 35 nm,
while the crystal structure consists of mostly tetragonal phase
(80–95 wt%), with the balance monoclinic phase at all process
conditions. The tetragonal crystal size is close to the primary
particle size, which indicates weak agglomeration of single
crystals.
Another promising aerosol technology is flame spray pyrolysis
(FSP), which overcomes most of the precursor limitations of the
gas-phase process. FSP can uniquely fulfill the physical and
chemical requirements for synthesis of a broad spectrum of
functional nanoparticles in a low-cost, single-step process,9 be-
cause each droplet contains the precursor in the same stoichiom-
etry as desired in the product powder.10 Nielsen et al.11 made,
among other things, up to 5 ␮m ZrO2 particles by atomizing a
solution consisting of zirconium sulfate (Zr(SO4)2) and water or
alcohol into a natural gas, carbon monoxide, or hydrogen flame.
Using FSP, Karthikeyan et al.12 made ZrO2 consisting of tetrago-
nal as the dominant phase and monoclinic as the minor phase with
crystal sizes of 12–21 nm at production rates up to 1.2 g/h. They
used zirconium n-butoxide (Zr(C4H9O)4) in butanol as precursor.
Yuan et al.13 used FSP of zirconium n-propoxide (Zr(C3H7O)4) to
make ZrO2 powders in the micrometer and submicrometer range.
These powders consisted of a mixture of tetragonal and monoclinic
phases. Laine et al.14 made, among other things, a range of
CeO2/ZrO2 compositions varying from pure ZrO2 to pure CeO2.
The specific surface area of these powders was ϳ10 m2/g. This
system was further investigated by Sutorik and Baliat,15 who
found that, in pure agglomerated ZrO2 (typical agglomerate size of
ϳ220 nm), the monoclinic phase was dominant, although signif-
icant amounts of tetragonal-phase ZrO2 were also observed. Kilian
and Morse16 used FSP of zirconium n-butoxide and zirconium
n-propoxide to make spherical cubic-phase and monoclinic-phase
ZrO2 at production rates of up to 180 g/h. The particle diameter in
these experiments was between 30 and 120 nm. Very recently,
Limaye and Helble17 used FSP of zirconium n-butoxide to make
tetragonal-phase ZrO2 particles having mean number diameters
ranging from 10 to 90 nm.
I. Introduction
IRCONIA (ZrO2) has a broad range of applications and has
Z
become one of the most industrially important ceramic mate-
rials of the present time. The traditional applications of ZrO2 and
ZrO2-containing materials are foundry sands and flours, refractory
ceramics, and abrasion resistance materials. Other ZrO2 applica-
tions include catalysts, oxygen sensors, fuel cells, resistive heating
elements, and jewelry because of its high oxygen-ion conduction
and high refractive index.1–4 Along with high strength and
toughness, ZrO2 also possesses good hardness, wear resistance,
and thermal shock resistance. These properties have led to the use
of ZrO2-based components in many engineering applications, such
as automobile engine parts, wire-drawing dies, and cutting tools.
The low thermal conductivity and relatively high coefficient of
thermal expansion make ZrO2 a suitable material for thermal
barrier coatings on metal components.2,5
ZrO2 particles have been produced by a variety of techniques,
including vapor deposition, mechanical milling, laser ablation,
flame-based methods, conventional and flame spray pyrolysis,
sol–gel methods, and microwave plasma synthesis. The aerosol
technology is advantageous, because it does not require the
multiple steps, high liquid volumes, and surfactants of wet chem-
ical processes.6 Fumed tetragonal-phase ZrO2 particles have been
made7 by evaporating zirconium tetrachloride (ZrCl4) into a
hydrogen/air (or oxygen) flame (Aerosil process) consisting of
agglomerated primary particles of ϳ8 nm in diameter. Later,
However, systematic investigations into the impact of produc-
tion rate, dispersion gas flow rate, and precursor concentration on
ZrO2 particle characteristics are limited in the above studies. In
contrast, Ma¨dler et al.18 systematically studied the effect of
oxidant and precursor–solvent composition on the size of silica
(SiO2) primary particles (7–39 nm) at production rates of 12 g/h
using FSP. The objective of the present study is to investigate
continuous synthesis of crystalline nanostructured ZrO2 particles
by FSP at high production rates in a pilot plant using baghouse
filters. Especially, the focus is on the size control of ZrO2
nanoparticles by varying production rate, precursor concentration,
and dispersion gas flow rate.
A. Krell—contributing editor
Manuscript No. 186459. Received December 30, 2002; approved September 26,
2003.
Supported in part by the Swiss National Science Foundation under Grant No.
2100-063632.00/1.
197

198
Journal of the American Ceramic Society—Mueller et al.
Vol. 87, No. 2
II. Experimental Procedure
vacuum pump (Model RE 16, Vaccubrand GmbH and Co.,
Wertheim Germany) on a glass-fiber filter (Type GF/A, Whatman,
Dietikon, Switzerland) 150 mm in diameter located in a stainless-
steel holder that samples the product powder by a bypass con-
nected to the inlet pipe (Fig. 1). Powders collected on the small
glass-fiber filter are identical (same specific surface area and
morphology) to the powders collected on the baghouse filters.
(1) Apparatus
Figure 1 shows a schematic of the experimental setup. The
spray apparatus consists of a commercially available external-
mixing stainless-steel gas-assisted nozzle (Model 970/4-S32,
Schlick-Du¨sen, Gustav Schlick GmbH and Co, Untersieman,
Germany) having a capillary tube with an inside diameter of 0.5
mm and an annular gap that can be adjusted to keep a constant
pressure drop (1 bar (1 ϫ 105 Pa)) across the nozzle tip regardless
of the dispersion gas (Ͼ99.95% O2, PanGas, Zurich, Switzerland)
flow rate (25 and 50 L/min). The dispersion gas flow rate is
metered by a mass flow controller (Bronkhorst, Ruurlo, the
Netherlands). The nozzle is surrounded by two stainless-steel
annuluses having inside and outside diameters of 18–19 and
20–25 mm, respectively. These annuluses form a diffusion flame
when 2 L/min of methane (Ͼ99.5% CH4, PanGas) flows through
the inner annulus and 4.5 L/min of O2 flows through the outer
annulus. These gas flow rates are also metered by mass flow
controllers (Bronkhorst). Additional sheath O2 (15 L/min) is
metered by a calibrated rotameter (Vo¨gtlin Instruments AG,
Aesch, Switzerland) and fed through a sintered metal plate ring
with inner and outer diameters of 28 and 50 mm, respectively, that
surrounds the previous outer annulus. Zirconium n-propoxide, 70
wt% in n-propanol (ChemPur, Karlsruhe, Germany), is used as
precursor and is dissolved in ethanol (Ͼ99.8% C2H5OH, EtOH;
Fluka Chemie AG, Buchs SG, Switzerland), which results in
precursor solutions of 0.5M and 1M. The liquid precursor solution
feed rate ranges from 6.8 to 81.1 mL/min, which results in ZrO2
production rates from 50 to 600 g/h. A 1 L precision piston pump
(Model 1000D, Isco, Inc., Lincoln, NE) provides a pulsation-free
supply of the precursor solution through the capillary tube. Before
the particle synthesis is started, an additional syringe pump (Model
Syringe Infusion Pump 22, Harvard Apparatus, Holliston, MA) is
used for initial heating of the nozzle (Fig. 1) by feeding and
burning 5 mL/min of EtOH for 10 min. This helps to assure a
constant pressure drop across the nozzle.
(2) Characterization
The droplet-size distribution is measured by Fraunhofer laser
diffraction spectrometry (Model Helos, Sympatec GmbH,
Claustal-Zellerfeld, Germany) 5 cm above the nozzle (1 bar
pressure drop across the nozzle) in the absence of combustion
when pure EtOH is atomized. The droplet mass median diameter
is in the range of 10–37 ␮m. The spray flame height is determined
visually as the distance from the nozzle tip to the end of the
luminous flame zone.
The powder specific surface area (SSA) is determined from a
five-point N2 adsorption isotherm in the relative pressure range of
0.05–0.25 at 77.3 K (BET analysis; Model Gemini III 2375,
Micromeritics Instruments Corp., Norcross, GA). Before the ad-
sorption, the samples are degassed (Model Flow Prep 060, Micro-
meritics Instruments Corp.) under N2 atmosphere at 150°C for 1 h,
to remove water bound to the particle surface from air moisture.
Assuming monodisperse spherical primary particles, the BET-
equivalent average primary particle diameter, dBET, is calculated
by dBET ϭ 6/(␳ SSA), where ␳ is the density of tetragonal-
p
p
phase ZrO2, 6.1 g/cm3. Error bars shown in the figures are two
times the standard deviation obtained from the results of
multiple experiments.
The powder X-ray diffractometry (XRD) spectra are recorded
using an advanced diffractometer (Model D8, Bruker Instruments,
Billercia, MA) over a 2␪ range from 20° to 70°, steps of 0.02°, and
a scan speed 0.24°/min. Crystalline characteristics are obtained
from the XRD spectra using TOPAS 2.0 software (Model 2000,
Bruker AXS) based on the fundamental parameter approach
(Rietveld method),19,20 in which the effects of the equipment (e.g.,
X-ray source and slits) are incorporated. The crystal size, dXRD, is
calculated from the full-width at half-maximum (FWHM) of the
peak, using Scherrer’s equation,21 as it is typically done using
dXRD ϭ 0.9␭/(␤ cos ␪), where ␭ is a wavelength of the X-ray
(0.154186 nm) and ␤ and ␪ represent the measured FWHM and a
diffraction angle, respectively.
Product powders are collected in a commercial jet filter (Model
FRR 4/1.2, Friedli AG, Burgdorf, Switzerland) consisting of four
polytetrafluoroethylene- (PTFE-, Teflon-) coated baghouse filters
(total surface area of 1.7 m2, Nomex, DuPont, Wilmington, DE),
which are cleaned periodically by air pressure shocks. Small
samples (ϳ1 g) of product particles are collected with the aid of a
Samples of the product powder are analyzed by transmission
electron microscopy (TEM; Model 2000FX II, JEOL, Tokyo,
Japan) using an electron microscope operated at 200 kV, with
magnifications between 50 and 800 000. The holey carbon-coated
copper TEM grids are dipped into the powder, which is collected
onto the filter. For each product powder, typically 700–1000
primary particles are counted manually using OPTIMAS 6.51 (Media
Cybernetics, Webster, NY) software. Statistical analysis of the
data are performed according to Hinds.22 The composition of the
collected powder is determined by thermogravimetric analysis
(TGA) in a thermobalance (Model TGA7SDTA851e, LF/1100°C,
Mettler–Toledo, Greifensee, Switzerland) coupled with a mass
spectrometer (Model Quadstar 422, Balzers, Liechtenstein). The
powder is heated in N2 from 25° to 120°C at 10°C/min, held at this
temperature for 10 min, and then heated at 20°C/min to 800°C and
held at this temperature for 10 min. Then, N2 is replaced by O2,
and the sample is heated at 20°C/min to 1000°C and held there for
5 min.23,24
III. Results and Discussion
For all flame spray conditions, only perfectly white ZrO2
powders are made. The visual indication for carbon-free powders
is verified by TGA coupled with mass spectrometry (MS), accord-
ing to Mueller et al.24 This analysis shows also no indication of
remaining carbonaceous species (no weight loss under oxidizing
conditions and no CO2 signal in the MS) in the as-prepared ZrO2
powders.
Fig. 1. Schematic of the flame spray pyrolysis (FSP) process for the
synthesis of ZrO2 nanoparticles at high production rates using a commer-
cially available nozzle.

February 2004
Zirconia Nanoparticles Made in Spray Flames at High Production Rates
199
(1) Particle Morphology
formation of nonagglomerated ZrO2 particles formed within a size
range of 30–120 nm. However, some qualitative TEM pictures
were missing.
Representative TEM micrographs of ZrO2 powders made at 200
g/h using dispersion/oxidant gas flow rates of 25 and 50 L/min and
0.5M and 1M zirconium n-propoxide in EtOH are shown in Fig. 2.
The ZrO2 particles are agglomerated and consist of uniform,
spherical primary particles. The BET-equivalent particle diameter
of 26 nm (Fig. 2(a)), 20 nm (Fig. 2(b)), 23 nm (Fig. 2(c)), and 14
nm (Fig. 2(d)) are in good agreement with TEM observations.
Furthermore, some primary particles stick together by sinter
bridges, but they are not fully coalesced, which indicates too low
flame temperatures and/or too short residence times of the particles
in the high-temperature region. Lattice planes are discernible,
which indicates that even the smallest particles are reasonably well
crystallized (Fig. 2(a)). Similar ZrO2 powder morphology was
observed by Karthikeyan et al.,12 who showed large agglomerates
that consisted of mostly nanosized particles of ϳ8 nm. Fractal-like
agglomerates and nonagglomerated ZrO2 particles were observed
by Limaye and Helble.17 The formation was dependent on the
flame temperature. At low flame temperatures, they observed the
formation of nonagglomerated, large ZrO2 particles, while highly
agglomerated fractal-like particles were made at high flame
temperatures. Very recently, Kilian and Morse16 reported the
(2) Particle Size
Figure 3 shows the BET-equivalent average diameter of the
ZrO2 primary particles as a function of the ZrO2 production rate at
two constant dispersion/oxidant gas flow rates of 25 L/min
(triangles) and 50 L/min (circles). A dispersion gas flow rate of 50
L/min allows liquid feed rates up to 81.1 mL/min, while, for a
dispersion gas flow rate of 25 L/min, the maximum liquid feed rate
is 54.1 mL/min. Increasing the liquid feed rate at a constant
dispersion gas flow rate decreases the gas to liquid mass ratio
(GLMR), which results in an insufficient dispersion of the liquid at
too low GLMRs.25,26
The zirconium n-propoxide concentration in EtOH is 0.5M
(filled symbols) and 1M (open symbols) (Fig. 3). As the liquid feed
rate of the 0.5M precursor solution is increased from 13.5 to 54.1
or 81.1 mL/min, which corresponds to a ZrO2 production rate from
50 to 200 or 300 g/h, the particle diameter increases from 12 to 29
nm using a dispersion/oxidant gas flow rate of 25 L/min, while the
particle diameter increases from 7 to 26 nm when using an O2 flow
rate of 50 L/min. Likewise, using a precursor concentration of 1M,
Fig. 2. TEM micrographs of FSP-made ZrO2 nanoparticles at a production rate of 200 g/h using dispersion gas flow rates of 25 and 50 L/min and 0.5M
and 1M zirconium n-propoxide in EtOH at (a) 25 L/min and 0.5M, (b) 50 L/min and 0.5M, (c) 25 L/min and 1M, and (d) 50 L/min and 1M.

200
Journal of the American Ceramic Society—Mueller et al.
Vol. 87, No. 2
the enthalpy content of the flame and mass concentration, and, as
a result, the high-temperature particle residence time increases,
which favors higher sintering and coagulation rates that result in
larger primary and aggregated particles. However, Kilian and
Morse,16 observed the formation of larger particles at more diluted
zirconium n-butoxide in n-butanol solutions, which is in disagree-
ment with the observation in this study.
Most importantly, Fig. 3 shows the operation window for the
synthesis of tailor-made ZrO2 nanoparticles as dispersion gas
flow rates lower than 25 L/min result in an insufficient
dispersion of the precursor solution, while, at 50 L/min, the exit
gap is at its widest opening to maintain a pressure drop across
the nozzle tip of 1 bar. Precisely controlled product primary
particles can be made by selecting the dispersion gas flow rate
(between 25 and 50 L/min) and the precursor concentration at a
constant production rate (Fig. 3).
Figure 4 shows the primary particle-size distribution (PPSD)
obtained by TEM for ZrO2 production rates of 100 (triangles) and
300 g/h (circles) using a dispersion gas flow rate of 50 L/min and
0.5M zirconium n-propoxide in EtOH. The PPSD is shifted to
larger average particle diameters as the production rate increases
from 100 to 300 g/h (Fig. 4), while the corresponding geometric
standard deviation, ␴g, is almost identical (1.64 and 1.63 for the
100 and 300 g/h spray, respectively) but is above 1.45, the
self-preserving limit for coagulation.31 The TEM-counted Sauter
mean primary particle diameters, ds, are 15.3 and 26.7 nm for the
100 and 300 g/h rates, respectively, which are consistent with the
corresponding BET-equivalent particle diameters (Fig. 3) of 12.0
nm (100 g/h) and 25.8 nm (300 g/h).
Fig. 3. BET-equivalent diameter of ZrO2 nanoparticles made by FSP
from zirconium n-propoxide in EtOH as a function of powder production
rate at precursor concentrations of (filled symbols) 0.5M and (open
symbols) 1M and at O2 dispersion/oxidant gas flow rates of (triangles) 25
L/min and (circles) 50 L/min.
the particle diameter increases from 10 to 35 nm when increasing
the liquid feed rate from 6.8 to 54.1 or 81.1 mL/min (production
rate from 50 to 400 or 600 g/h) at a dispersion gas flow rate of 25
L/min, while the particle diameter increases from 6 to 31 nm at a
dispersion gas flow rate of 50 L/min.
(3) Phase Composition
A representative XRD pattern of ZrO2 powder made by FSP is
shown in Fig. 5. The ZrO2 is made at a production rate of 200 g/h
using a dispersion gas flow rate of 25 L/min and 1M zirconium
n-propoxide in EtOH. The ZrO2 crystal structure consists of
mainly tetragonal (T) phase and limited monoclinic (M) phase.
The calculation of dXRD (Scherrer’s equation) from the strongest
tetragonal-phase peak results in a crystal size of 22 nm, which is
in agreement with the BET-equivalent average particle diameter of
23 nm, while the dXRD of the monoclinic structure results in a
crystal size of 26 nm. Using the fundamental parameter approach,
the phase composition is 95 wt% tetragonal and 5 wt% monoclinic
(Fig. 5).
A mixture of mostly tetragonal (85–95 wt%) and monoclinic
phases of ZrO2 is detected by XRD of all FSP-made powders using
zirconium n-propoxide as precursor. The highest content of tet-
ragonal phase is detected for the smallest ZrO2 particles, while it
decreases slightly with increasing particle diameter. Normally,
An increase in feed (or ZrO2 production) rate increases the
spray flame height. For example, the flame height increases from
4 to 40 cm when the feed rate of the 1M precursor solution
increases from 6.8 to 81.1 mL/min, and the dispersion gas flow
rate is 50 L/min. The increase in precursor solution feed rate
results in higher enthalpy content (from 10.9 to 130 MJ/h at these
conditions) of the flame. Hence, the particles have longer resi-
dence times at high temperatures. The increase in particle sintering
rate at high residence time and high temperature contributes to the
formation of larger primary particles. Furthermore, as the zirco-
nium n-propoxide concentration increases, the ZrO2 particle con-
centration increases. This leads to more particle collisions and,
therefore, enhanced particle growth, which increases the particle
size, especially when complete coalescence takes place.27 These
results are in agreement with FSP studies of other pure oxides at
lower production rates, where an increase of the liquid feed (or
28
production) rate increases the primary particle diameter of Bi2O3
29
and CeO2 as well as with SiO2 formation studies at high
production rates using FSP26 and vapor-fed turbulent hydrogen–air
diffusion flames.23
Increasing the dispersion/oxidant gas flow rate from 25 to 50
L/min intensifies mixing and accelerates combustion,30 thus de-
creasing the spray flame height. The flame height decreases, for
example, from 35 to 26 cm when the dispersion gas flow rate
increases from 25 to 50 L/min at a 1M liquid feed rate of 54.1
mL/min. Additionally, increasing the dispersion gas flow rate
decreases the droplet concentration of the spray flame; thus, the
particle concentration and the particle residence time at high
temperature decrease as the spray flame height decreases. This
leads to faster quenching of particle growth and, therefore, smaller
particles (Fig. 3). These results are in agreement with other FSP
studies.26,28
Increasing the precursor concentration from 0.5M to 1M in-
creases the primary particle diameter. For example, at a liquid feed
rate of 40.6 mL/min, the primary particle diameter increases from
23 to 30 nm (corresponding to powder production rates of 150 and
300 g/h, Fig. 3) when increasing the precursor concentration from
0.5M to 1M. Again, increasing precursor concentration increases
Fig. 4. Primary particle-size distributions of ZrO2 made at 0.5M zirco-
nium n-propoxide in EtOH at powder production rates of (‚) 100 and (E)
300 g/h using a dispersion gas flow rate of 50 L/min.

February 2004
Zirconia Nanoparticles Made in Spray Flames at High Production Rates
201
Fig. 6. Crystal size of (filled symbols) monoclinic-phase and (open
symbols) tetragonal-phase ZrO2 made by FSP as a function of the ZrO2
powder production rate using O2 dispersion gas flow rate of (triangles) 25
and (circles) 50 L/min at 0.5M zirconium n-propoxide in EtOH.
Fig. 5. XRD pattern of ZrO2 powder made at a production rate of 200 g/h
using a dispersion gas flow rate of 25 L/min and 1M zirconium
n-propoxide in EtOH. ZrO2 crystal structure consists of mainly tetragonal
(T, 95 wt%) and monoclinic (M, 5 wt%) phases.
ZrO2 has a monoclinic structure at room temperature and atmos-
pheric pressure, which transforms martensitically to a tetragonal
structure at 1170°C and then to a fluorite-type cubic structure at
2370°C.2,32 The presence of tetragonal-phase ZrO2 at room tem-
perature can be explained by the Gibbs–Thomson effect. In small
particles, the tetragonal structure is energetically favored, while in
larger particles, the monoclinic structure is more stable. During the
rapid cooling of the nanoparticles, as they flow out of the
high-temperature spray flame, the tetragonal structure is quenched.
Structural transformation occurs at a critical particle size of 8–18
nm, below which the tetragonal structure is thermodynamically
stable.2,33,34 Formation of tetragonal-phase ZrO2, even in particles
Ͼ18 nm (Fig. 5) may result from fast quenching of the particles
during FSP. This is in agreement with other FSP studies at much
lower production rates12,13 but is in disagreement with Kilian and
Morse,16 who reported the formation of cubic and monoclinic
structures of ZrO2 particles by FSP without quantifying that phase
composition.
quenched more slowly, which results in a decrease of the tetragonal
structure and the formation of larger monoclinic crystals (Fig. 6).
The tetragonal crystal sizes (Fig. 6) are astonishingly close to
the BET-equivalent particle diameters (Fig. 3), which indicates
that the primary particles are weakly agglomerated single crystals.
For ZrO2 powders made at production rates up to 150 g/h, the
BET-equivalent particle diameter is slightly smaller than the
average tetragonal crystal size. A possible reason for this differ-
ence is polydispersity, because the BET-equivalent particle diam-
eter dBET is a surface-weighted particle property, whereas dXRD is
a mass-weighted particle property, which leads to dBET/dXRD Ͻ 1.
At higher ZrO2 production rates (Ͼ150 g/h), the dBET is slightly
larger than the tetragonal-phase dXRD. Here, the tetragonal-phase
content of the powders is slightly lower (down to ϳ85 wt%),
which leads to a larger average crystal size, because monoclinic
crystals are, on average, larger than the tetragonal crystals (Fig. 6).
IV. Conclusions
Figure 6 shows the average XRD tetragonal (open symbols) and
monoclinic (filled symbols) crystal size as a function of the ZrO2
powder production rate using a dispersion gas flow rate of 25
(triangles) and 50 L/min (circles) at 0.5M zirconium n-propoxide
in EtOH. Both crystal sizes increase with increasing production
rate, which is in agreement with Fig. 3, where the BET-equivalent
primary particle diameter increases also with increasing powder
production rate. Here, the tetragonal crystal size increases from 8.7
to 24.3 nm, while the monoclinic crystal size increases from 9 to
34 nm when increasing the ZrO2 production rate from 50 to 300
g/h at an O2 gas flow rate of 50 L/min. Likewise, at 25 L/min O2
flow rate, the tetragonal crystal size increases from 13.9 to 27.3 nm
and the monoclinic crystal size increases from 18.2 to 35.1 nm
(powder production rate increases from 50 to 200 g/h). The
monoclinic crystal sizes are always larger than the tetragonal
crystal sizes produced under the same process conditions. This
difference increases with increasing crystal size (higher ZrO2
production rate) and decreasing dispersion gas flow rate (from 50
to 25 L/min). The fast quenching of the spray flame is more
efficient at high dispersion gas flow rates and at low powder
production rates. Here, the residence time of the particles in the
spray flame is very short, which leads to small primary particles
(Fig. 3), where the high-temperature tetragonal phase is quenched
in becoming the dominant structure. Longer particle-residence
times in the spray flame lead to larger particles, which are
A systematic investigation of flame spray synthesis of ZrO2
nanoparticles was conducted at high production rates up to 600 g/h
using a commercially available external-mixing stainless-steel
gas-assisted nozzle. The influence of ZrO2 powder production rate,
precursor concentration, and oxidant dispersion gas flow rate was
investigated on the product morphology, average primary particle
diameter and size distribution, crystallinity, and purity using 0.5M
and 1M zirconium n-propoxide in EtOH. The average particle
diameter of pure ZrO2 was controlled from 6 to 35 nm by varying
the production rate, precursor composition, and dispersion gas
flow rates. The crystal structure consisted of mostly tetragonal
phase (80–95 wt%) and the balance monoclinic phase at all used
process conditions. XRD-determined tetragonal crystal sizes were
close to the BET-determined primary particle sizes, which indi-
cated that the primary particles were weakly agglomerated single
crystals.
References
1M. J. Mayo, J. R. Seidensticker, D. C. Hague, and A. H. Carim, “Surface
Chemistry Effects on the Processing and Superplastic Properties of Nanocrystalline
Oxide Ceramics,” Nanostruct. Mater., 11 [2] 271–82 (1999).
2T. Chraska, A. H. King, and C. C. Berndt, “On the Size-Dependent Phase
Transformation in Nanoparticulate Zirconia,” Mater. Sci. Eng. A-Struct. Mater. Prop.
Microstruct. Process., 286 [1] 169–78 (2000).

202
Journal of the American Ceramic Society—Mueller et al.
Vol. 87, No. 2
3M. Gell, “Application Opportunities for Nanostructured Materials and Coatings,”
17A. U. Limaye and J. J. Helble, “Morphological Control of Zirconia Nanoparticles
through Combustion Aerosol Synthesis,” J. Am. Ceram. Soc., 85 [5] 1127–32 (2002).
18L. Ma¨dler, H. K. Kammler, R. Mueller, and S. E. Pratsinis, “Controlled Synthesis
of Nanostructured Particles by Flame Spray Pyrolysis,” J. Aerosol. Sci., 33 [2]
369–89 (2002).
Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 204 [1–2] 246–51
(1995).
4H. Gleiter, “Materials with Ultrafine Microstructures: Retrospectives and Perspec-
tives,” Nanostruct. Mater., 1 [1] 1–19 (1992).
5R. Nielsen and T. W. Chang, “Zirconium and Zirconium Compounds”; pp.
543–67 in Ullmann’s Encyclopedia of Industrial Chemistry, Vol. A28. Wiley–VCH
Verlag, Weinheim, Germany, 1996.
19R. W. Cheary and A. Coelho, “A Fundamental Parameters Approach to X-ray
Line-Profile Fitting,” J. Appl. Crystallogr., 25, 109–21 (1992).
20R. W. Cheary and A. A. Coelho, “Axial Divergence in a Conventional X-ray
Powder Diffractometer. I. Theoretical Foundations,” J. Appl. Crystallogr., 31,
851–61 (1998).
6S. E. Pratsinis and S. V. R. Mastrangelo, “Material Synthesis in Aerosol
Reactors,” Chem. Eng. Prog., 85 [5] 62–66 (1989).
7A. T. Liu and P. Kleinschmit, “Production of Fumed Oxides by Flame Hydroly-
sis”; pp. 1–10 in Novel Ceramic Fabrication Processes and Applications, Vol. 38.
Edited by R. W. Davidge. Institute of Ceramics, Staffs, U.K., 1986.
8W. Hartmann, A. T. Liu, D. Peuckert, and P. Kleinschmit, “Fumed Oxides as Base
Materials for Ceramic Applications,” Mater. Sci. Eng. A-Struct. Mater. Prop.
Microstruct. Process., 109, 243–46 (1989).
21R. C. Rau, “Routine Crystallite-Size Determination by X-ray Diffraction Line
Broadening,” Adv. X-ray Anal., 5, 104–15 (1962).
22W. C. Hinds, Aerosol Technology; pp. 75–110. Wiley, New York, 1999.
23H. K. Kammler, R. Mueller, O. Senn, and S. E. Pratsinis, “Synthesis of
Silica–Carbon Particles in a Turbulent H2–Air Flame Aerosol Reactor,” AIChE J., 47
[7] 1533–43 (2001).
9H. K. Kammler, L. Ma¨dler, and S. E. Pratsinis, “Flame Synthesis of Nanopar-
ticles,” Chem. Eng. Technol., 24 [6] 583–96 (2001).
24R. Mueller, H. K. Kammler, K. Wegner, and S. E. Pratsinis, “OH Surface Density
of SiO2 and TiO2 by Thermogravimetric Analysis,” Langmuir, 19 [1] 160–65 (2003).
25A. H. Lefebvre, Atomization and Sprays; pp. 201–72. Taylor and Francis,
London, U.K., 1989.
10G. L. Messing, S. C. Zhang, and G. V. Jayanthi, “Ceramic Powder Synthesis by
Spray Pyrolysis,” J. Am. Ceram. Soc., 76 [11] 2707–26 (1993).
11M. L. Nielsen, P. M. Hamilton, and R. J. Walsh, “Ultrafine Metal Oxides by
Decomposition of Salts in a Flame”; pp. 181–95 in Ultrafine Particles. Wiley, New
York, 1963.
26R. Mueller, L. Ma¨dler, and S. E. Pratsinis, “Nanoparticle Synthesis at High
Production Rates by Flame Spray Pyrolysis,” Chem. Eng. Sci., 58 [10] 1969–76
(2003).
12J. Karthikeyan, C. C. Berndt, J. Tikkanen, J. Y. Wang, A. H. King, and H.
Herman, “Nanomaterial Powders and Deposits Prepared by Processing of Liquid
Precursors,” Nanostruct. Mater., 8 [1] 61–74 (1997).
27S. E. Pratsinis, “Flame Aerosol Synthesis of Ceramic Powders,” Prog. Energy
Combust. Sci., 24 [3] 197–219 (1998).
28L. Ma¨dler and S. E. Pratsinis, “Bismuth Oxide Nanoparticles by Flame Spray
Pyrolysis,” J. Am. Ceram. Soc., 85 [7] 1713–18 (2002).
13F. L. Yuan, C. H. Chen, E. M. Kelder, and J. Schoonman, “Preparation of
Zirconia and Yttria-Stabilized Zirconia (YSZ) Fine Powders by Flame-Assisted
Ultrasonic Spray Pyrolysis (FAUSP),” Solid State Ionics, 109 [1–2] 119–23 (1998).
14R. M. Laine, T. Hinklin, G. Williams, and S. C. Rand, “Low-Cost Nanopowders
for Phosphor and Laser Applications by Flame Spray Pyrolysis”; pp. 500–510 in
Metastable, Mechanically Alloyed, and Nanocrystalline Materials, Parts 1 and 2,
Vols. 343–346. TransTech Publications, Zurich-Uetikon, Switzerland, 2000.
15A. C. Sutorik and M. S. Baliat, “Solid-Solution Behavior of CexZr1–xO2
Nanopowders Prepared by Flame Spray Pyrolysis of Solvent-Borne Precursors”; pp.
371–76 in Metastable, Mechanically Alloyed and Nanocrystalline Materials, Vols.
386–388. TransTech Publications, Zurich-Uetikon, Switzerland, 2002.
16A. Kilian and T. F. Morse, “A Novel Aerosol Combustion Process for the High
Rate Formation of Nanoscale Oxide Particles,” Aerosol Sci. Technol., 34 [2] 227–35
(2001).
29L. Ma¨dler, W. J. Stark, and S. E. Pratsinis, “Flame-Made Ceria Nanoparticles,”
J. Mater. Res., 17 [6] 1356–62 (2002).
30I. Glassman, Combustion; pp. 1–34. Academic Press, San Diego, CA, 1996.
31S. Vemury and S. E. Pratsinis, “Self-Preserving Size Distributions of Agglom-
erates,” J. Aerosol. Sci., 26 [2] 175–85 (1995).
32R. C. Garvie, “The Occurrence of Metastable Tetragonal Zirconia as a Crystallite
Size Effect,” J. Phys. Chem., 69 [4] 1238–43 (1965).
33R. Ramamoorthy, D. Sundararaman, and S. Ramasamy, “X-ray Diffraction Study
of Phase Transformation in Hydrolyzed Zirconia Nanoparticles,” J. Eur. Ceram. Soc.,
19 [10] 1827–33 (1999).
34G. Skandan, “Processing of Nanostructured Zirconia Ceramics,” Nanostruct.
Mater., 5 [2] 111–26 (1995).
Ⅺ