Preparation of Nickel Powders by Spray Pyrolysis of Nickel Formate

Article abstract of DOI:10.1111/j.1151-2916.2001.tb00855.x

The preparation of nickel powders by the ultrasonic spray pyrolysis of Ni(HCOO)₂ was studied. Phase-pure nickel powder was obtained at as low as 350°C. HCOOH was a reducing source for nickel formation. Moreover, metallic nickel was obtained at a residence time as short as 0.1 s at 600°C. A broad range of particle morphologies, which included agglomerated nanoparticles, nonagglomerated submicrometer particles, hollow particles, and spherical dense particles, were obtained from Ni(HCOO)₂ pyrolysis and were shown to depend on the precursor solution and the operating condition.

Full text of DOI:10.1111/j.1151-2916.2001.tb00855.x

J. Am. Ceram. Soc., 84 [7] 1425–32 (2001)
journal
Preparation of Nickel Powders by Spray Pyrolysis of Nickel Formate
Bin Xia, I. Wuled Lenggoro, and Kikuo Okuyama†
Department of Chemical Engineering, Hiroshima University, Kagamiyama, Higashi–Hiroshima, 739–8527, Japan
The preparation of nickel powders by the ultrasonic spray
pyrolysis of Ni(HCOO)2 was studied. Phase-pure nickel pow-
der was obtained at as low as 350°C. HCOOH was a reducing
source for nickel formation. Moreover, metallic nickel was
obtained at a residence time as short as 0.1 s at 600°C. A broad
range of particle morphologies, which included agglomerated
nanoparticles, nonagglomerated submicrometer particles, hol-
low particles, and spherical dense particles, were obtained
from Ni(HCOO)2 pyrolysis and were shown to depend on the
precursor solution and the operating condition.
be formed at as low as 500°C. The formation mechanisms have
been studied.13
In this study, we prepared nickel powders by the spray pyrolysis
of Ni(HCOO)2. Different from other nickel precursors that require
external reducing sources such as H2, Ni(HCOO)2 can directly
decompose to nickel and a gas species,14 which is a self-redox
reaction. This characteristic, as well as its low solubility in water
and the reductive nature, distinguishes Ni(HCOO2) from other
nickel precursors studied earlier. We show that HCOOH can be
used to provide a reductive atmosphere during spray pyrolysis and
thus make it unnecessary to use H2 or NH3.
I. Introduction
II. Experimental Procedure
PRAY PYROLYSIS is a powerful tool for preparing powders and
films, because of the easy control of target composition, the
(1) Spray Pyrolysis
S
Nickel formate dihydrate, Ni(HCOO)2⅐2H2O, was used to
prepare nickel by ultrasonic spray pyrolysis, using the apparatus
shown in Fig. 1. The apparatus consisted of an ultrasonic atomizer
(Model NE-U11B, at 1.7 MHz; Omron, Kyoto, Japan), a ceramic
tubular reactor (heating length of 1.0 m and inner diameter of 13
mm), and an electrostatic precipitator. Five furnaces (each 0.20 m
in length) were used to control the temperatures of the ceramic
reactor. The total gas flow rate was 20 L/h, unless otherwise stated.
The residence times in the reactor were 9–7 s, corresponding to the
pyrolysis temperatures of 600°–1000°C, unless otherwise stated.
Powders were collected in an electrostatic precipitator maintained
at 160°C, and the exhaust gas was emitted after washing. All
chemicals used in this research were analytical reagent grade
(Kanto Chemicals, Tokyo, Japan).
In the preparation of nickel powder (also other metal powders),
H2 has been frequently used to reduce metal salts or metal oxides
to the metallic form. Because of the safety problem and the high
cost of H2 gas, our research has been motivated by the need to
prepare nickel powders in the absence of H2. Therefore, N2 gas
was used as carrier gas in this research. The sole experiments using
H2–N2 carrier gas (H2:N2 ϭ 1:10, or a H2 concentration of 9.1
vol%) were conducted to compare the particle morphologies
formed from the solutions with and without HCOOH, which are
shown in Section III(2)(A).
excellent availability of the precursors, and the high quality of the
products.1,2 In the preparation of a powder by spray pyrolysis, a
solution is atomized into a hot reactor, where the aerosol droplets
undergo drying, droplet shrinkage, solute precipitation, thermoly-
sis, and sintering to form final particles.1 Residence times from
several seconds to tens of seconds are most frequently used to
ensure the formation of the desired product. Metals, metal oxides,
and non-oxides can be readily produced by spray pyrolysis.3–5
For electronic products such as multilayer ceramics and batter-
ies, nickel is used for electrodes.6 Because of its good performance
and lower cost, compared with noble metals such as palladium and
silver, nickel is replacing them and becoming one of the most
important electrode materials.
Nagashima and co-workers,7,8 prepared nickel powders by
spray pyrolysis of nickel precursors ten years ago. They showed
that nickel particles could be prepared from Ni(NO3)2 and NiCl2 in
a H2–N2 (3%–20% H2) atmosphere with a residence time of
several seconds, but porous and hollow nickel particles were
obtained below its melting point, i.e., 1600°C. Using Ni(NO3)2 as
the precursor in a H2–N2 (15% H2) atmosphere, Che et al.9
obtained nickel at Ն600°C with residence times of 30–150 s, and
dense nickel particles were formed at 1200°C. Stopic and co-
workers10,11 produced nickel particles from NiCl2 and Ni(NO3)2 in
a H2–N2 (16.5% H2) atmosphere. They showed that nickel
particles could be obtained at Ն900°C with residence times of
5–20 s. Typically, nickel has been produced in the presence of H2
in carrier gas (to reduce the Ni(II), e.g. intermediate NiO) at
Ն900°C with residence times of 5 seconds to tens of seconds.
In our earlier study, we reported nickel particle formation by the
spray pyrolysis of nickel ammine complex formed from NiCl2,
aqueous NH3, and ammonium bicarbonate.12 It was shown that
nickel could be obtained both in the presence and in the absence of
H2 carrier gas at residence times of 7–9 s. Phase-pure nickel could
(2) Characterization
Thermal gravimetric analysis (TGA) (Model TGA-50, Shi-
madzu, Kyoto, Japan) was conducted from 25°C to 1000°C at a
heating rate of 10°C/min in a N2 atmosphere using a ϳ18 mg
sample. Crystalline phases of the spray-pyrolysis powders were
examined by X-ray diffraction (XRD) (Model Rint 2200V,
Rigaku, Tokyo, Japan) with CuK␣ radiation operated at 40 kV and
20 mA. The particle morphology was observed using scanning
electron microscopy (SEM) (Model JSM-5600, JEOL, Tokyo,
Japan).
A. Carim—contributing editor
III. Results and Discussion
(1) Thermal Analysis of Nickel Formate
Manuscript No. 188550, Received May 22, 2000; approved February 20, 2001.
Supported by the ministry of Education, Science, Sports and Culture of Japan, with
Grant-in-Aid for scientific research.
Figure 2 shows the TGA plot of Ni(HCOO)2⅐2H2O. The heating
schedule was 25°–1000°C at a rate of 10°C/min in N2, which is
denoted as heating schedule 1. It consists of three stages:
†Author to whom correspondence should be addressed.
1425

1426
Journal of the American Ceramic Society—Xia et al.
Vol. 84, No. 7
atmosphere. In the absence of HCOOH in the solution shown in
Fig. 3(a), very weak nickel peaks were indicated at 300°C, but the
main phase was the undecomposed Ni(HCOO)2. The nickel
coexisted with NiO at 350°C (and also at 400°C), and the
Ni(HCOO)2 was completely decomposed. The nickel was further
oxidized at 400°C, and the oxidation process was almost complete
at 500°C. When 6.0M HCOOH was used with 0.11M Ni(HCOO)2,
phase-pure nickel was obtained at as low as 350°C, as shown in
Fig. 3(b).
The formic acid decomposes by two simultaneous processes
with different kinetics,15,16
HCOOH 3 CO2 ϩ H2
HCOOH 3 CO ϩ H2O
(1)
(2)
In addition to the thermal decomposition, there is abundant
literature that reports the catalytic decomposition of HCOOH on
the surfaces of metals17,18 and metal oxides.19,20 In brief, HCOOH
can be catalytically decomposed to H2, CO, H2O, and CO2 at much
lower temperature (e.g. Ͻ100°C).21 Therefore, H2 and CO are
believed to be generated by the decomposition of the HCOOH
during spray pyrolysis.
Figures 4(a) and (b) show SEM images of the nickel powders
obtained from Ni(HCOO)2 solution with HCOOH in a N2 atmo-
sphere. Figure 4(a) indicates that the nickel powder formed at
350°C consisted of nickel nanocrystals and pores; many particles
do not have regular shapes. The TEM observation showed that the
nickel powder was comprised of agglomerated nickel nanocrystals
15–40 nm in size (not shown here). At 700°C, nickel crystals grew
to larger sizes with similar particle morphology. The pores
decreased remarkably in amount, and the particles became denser;
however, the particles were still far from full sintering. The nickel
particles sintered significantly at 1000°C, and solid particles were
obtained. The particles were spherical and solid with smooth
surfaces, which are notably different from those formed at 350°
and 700°C. The average particle size is 0.43 ␮m.
Fig. 1. Schematic diagram of the experimental apparatus.
When no HCOOH was added to the solution, the solution was
spray-pyrolyzed in a H2–N2 atmosphere to ensure the formation of
nickel. Nickel particles also were formed at Ն350°C. Here, H2 was
not used to reduce Ni(HCOO)2 to nickel. Instead, it formed a
protective atmosphere, similar to the HCOOH. Different from the
particle morphologies observed in Figs. 4(a) and (b), no nanocrys-
tals/pores can be identified from the SEM images, as seen in Figs.
4(c) and (d). At 1000°C, nickel particles have similar morpholo-
gies to Fig. 4(b), which is attributed to the high-temperature
sintering. When the temperature was raised from 350°C to 1000°C,
the particle morphology evolved from an almost-spherical shape
with rough surfaces to a spherical shape with smooth surfaces. The
particle sizes decreased from 0.54 ␮m at 350°C to 0.42 ␮m at
1000°C because of sintering.
Nickel particles were also prepared from Ni(HCOO)2 solution
with 6M formic acid in a H2–N2 atmosphere at 700°C. The
resultant nickel particles had a morphology similar to that formed
in the absence of H2 in carrier gas. Clearly, the presence of
HCOOH in the solution was responsible for the formation of the
nanocrystals/pore structures.
(B) Effect of Residence Time: Regarding the nickel prepa-
ration by spray pyrolysis, gas–solid reactions were shown to be
necessary in earlier studies to form nickel, in which intermediate
NiO (or other Ni(II) compounds) particles react with a reducing
gas such as H2.7–13 For example, the intermediate NiO is initially
formed from the pyrolysis of Ni(NO3)2 precursor, followed by
reduction by H2 to form nickel.9
Figure 5 shows the XRD patterns of the powders formed at
short residence times from a solution of 0.11M Ni(HCOO)2. At
500°C for 0.5 s (Curve “(a)”), metallic nickel was obtained, but
there was a weak peak at ϳ2␪ ϭ 12°, indicating the incomplete
decomposition of the formate. Phase-pure nickel was obtained at
600°C with a residence time of 0.1 s. Therefore, nickel can be
formed at the residence times that are 1–2 orders of magnitude
smaller than that used in earlier spray-pyrolysis studies (100–102
Fig. 2. TGA measurement of nickel formate (Ni(HCOO)2).
(I) Dehydration: The weight loss, started from ϳ150°C to
220°C, is attributed to the dehydration of crystallized water in the
precursor.
(II) Decomposition: From 260°C to ϳ293°C, Ni(HCOO)2
decomposed rapidly. The sample weight at 293°C, i.e., 31.5%,
agrees well with the theoretical nickel content (31.8%) in
Ni(HCOO)2⅐2H2O. Clearly, the precursor directly decomposed to
metallic nickel in a N2 atmosphere.
(III) Oxidation: The sample weight increased to 39.6% at
728°C, and to 40.0% at 1000°C, which corresponds to the
theoretical content of NiO (40.4%) in Ni(HCOO)2⅐2H2O. The
oxidation of nickel was confirmed by the appearance of a green
residue that remained after TGA measurement. The oxidation
could be significantly reduced by first preheating the precursor at
200°C for 1 h. By removing the water of hydration, no oxidation
was present when the Ni(HCOO)2 decomposed to nickel. There-
fore, nickel can be directly formed by the thermal decomposition
of Ni(HCOO)2 in a N2 atmosphere.
(2) Spray Pyrolysis of the Precursor
(A) Effect of Formic Acid: Figure 3 shows XRD patterns of
the powders formed from a 0.11M Ni(HCOO)2 solution in a N2

July 2001
Preparation of Nickel Powders by Spray Pyrolysis of Nickel Formate
1427
Fig. 3. XRD patterns of the powders formed in a N2 atmosphere from [Ni(HCOO)2] ϭ 0.11M solution (a) without and (b) with formic acid, [HCOOH]
ϭ 6M.
HCOOH ͑aq͒ 3 Hϩ ͑aq͒ ϩ COOHϪ ͑aq͒
s). Clearly, the formation kinetics contradict those of the gas–solid
interaction process.
Interestingly, even without HCOOH in the solution, nickel also
͑pKaϭ3.75 at 20°C22͒ (4)
could be formed at 600°C with a residence time of 0.1 s, as seen
in Fig. 5(c). This is ascribed to the relatively slow oxidation
process, which is also related to gas–solid interaction, compared to
where Ka is the acid-dissociation constant of HCOOH and Ksp is
the solubility-product constant of Ni(HCOO)2. Ksp was obtained
from its solubility in water at 25°C. To simplify the calculation, the
Ni(HCOO)2 decomposition. The particles formed at these condi-
concentration and activity of ions are assumed to be equivalent. In
tions have morphologies similar to the nickel nanocrystals/pores
acidic and neutral solution, the formation of Ni(OH)2 precipitate is
shown in Fig. 4(a).
negligible. Based on Eqs. (3) and (4), we calculated the equilib-
rium solubility (S) of Ni(HCOO)2 in aqueous solution as a function
(3) Spray Pyrolysis of Acidic Solution
The solubility of Ni(HCOO)2 in water at 25°C is 2.2 g
Ni(HCOO)2⅐2H2O per 100 mL H2O (or 0.12M), which is much
smaller than those of Ni(NO3)2⅐6H2O (8M) and NiCl2⅐6H2O (10M)
precursors.22 High precursor solubility is preferred for spray
pyrolysis, because this enables the production of nickel particles
with adjustable sizes and higher production rates. To increase the
solubility in water, we adjust the acidity, or pH value, of the
solution. The following equilibria are considered in aqueous
solution,
of pH at 25°C:
S͑M͒ ϭ ͓Ni2ϩ͔ ϭ 0.12 ͑1 ϩ 5.62 ϫ 103ϪpH
͒
(5)
2/3
Accordingly, S increases with decreasing pH, or increasing acidity.
For example, S Ͼ 8.0M at pH 1.0, which is much higher than its
S value in pure water, i.e., 0.12M. Obviously, S can be adjusted
continuously within a wide range to obtain powders with the
desired sizes.
One practical method to raise the value of S is the addition of
acids, such as HNO3, HCl, or other acids, to the Ni(HCOO)2
solution. Another practical method is the direct addition of
HCOOH to nickel salts such as Ni(NO3)2 and NiCl2. The latter
Ni(HCOO)2 ͑s͒ 3 Ni2ϩ ͑aq͒ ϩ 2 COOHϪ ͑aq͒
͑Ksp ϭ 6.74 ϫ 10Ϫ3 at 25°C͒ (3)

1428
Journal of the American Ceramic Society—Xia et al.
Vol. 84, No. 7
Fig. 4. SEM images of nickel powders formed from nickel formate, [Ni(COOH)2] ϭ 0.11M ((a) and (b) formed with formic acid in solution, [HCOOH]
ϭ 6.0 M, in N2 atmosphere; (c) and (d) formed in an H2–N2 atmosphere without HCOOH in solution). Figures 4(a) and (c) were formed at 350°C and 4(b)
and (d) at 1000°C. (Bar ϭ 1 ␮m.)
was used in this study because of the cost considerations and the
operation convenience (to add HCOOH).
The XRD analysis of the powders, formed at 700° and 1000°C
from Ni(NO3)2–HCOOH solution in N2 atmosphere, indicated the
formation of phase-pure nickel powders. Figure 6(a) shows the
nickel powder formed from Ni(NO3)2–HCOOH aqueous solutions
at 700°C. Many hollow particles with rough surfaces were formed.
The proportion of solid particles increased notably at 1000°C, but
a portion of hollow particles still remained. In addition, the particle
size distributions in these two cases are very wide. This situation
is considerably different from that of Ni(HCOO)2 without
Ni(NO3)2 in the solution.
Redox reactions take place between the oxidative NOϪ3 group
and the reductive HCOOϪ group, which was confirmed by the
following experiment. A solution containing ϳ2M Ni(NO3)2 and
ϳ12M HCOOH was heated on an electric plate. When the solution
temperature increased from room temperature to 40°–50°C, a
substantial amount of gas bubbles formed. The solution tempera-
ture increased abruptly to ϳ100°C and the solution was boiled
vigorously, giving off a brown fume and then precipitate. The
XRD pattern of the precipitate showed that it was well-crystallized
Ni(HCOO)2⅐2H2O.‡ Clearly, NOϪ3 decomposed to gas species
Fig. 5. XRD patterns of powders formed at short residence times. Under
the condition of 600°C for 0.1 s, only two heating zones (T1 and T2) were
such as NOx, followed by the precipitation of Ni(HCOO)2 from the
used and set at 700°C. The gas flow rate was 5.3 L/min (room tempera-
solution. The similar vigorous processes can also occur within
ture). The gas temperatures at the exit of the T2 reactor were measured to
be ϽϽ500°C, because of the short residence time. The measured average
temperatures in the T1 reactor were 600°C on the reactor inner wall and
‡Powder Diffraction File, No. 15-917. International Centre for Diffraction Data,
Newtown Square, PA.
575°C in the center. Thus, 600°C was assigned to the reaction temperature,
and the residence time in the T1 reactor was calculated to be 0.10 s.

July 2001
Preparation of Nickel Powders by Spray Pyrolysis of Nickel Formate
1429
Fig. 6. Nickel powders formed from (a) Ni(NO3)2–HCOOH, and (b) NiCl2–HCOOH aqueous solutions in N2 at 700°C. [Ni2ϩ] ϭ 1.0M; [HCOOH] ϭ 15M
and 9M for nitrate and chloride, respectively. (Bar ϭ 1 ␮m.)
droplet scale during spray pyrolysis, which is responsible for the
deteriorated particle morphology.
is negligible in alkaline solution. Therefore, the solubility of
Ni(COOH)2 in alkaline solution can be written as
Figure 6(b) shows the nickel powder formed from NiCl2–
HCOOH aqueous solution at 700°C. In this case, no hollow
particles were formed, and the average particle size is 0.83 ␮m.
The particles were further densified at 1000°C, and they became
more spherical with smoother surfaces and an average particle size
of 0.80 ␮m. Obviously, this situation is remarkably different from
that of Ni(NO3)2–HCOOH solution. It is suggested that the
absence of strong redox reactions between NiCl2 and HCOOH is
responsible for the formation of solid particles.
S ϭ 0.12͓1 ϩ ͑1.54 ϫ 10Ϫ23ϩ2pH͒ ϩ ͑1.56 ϫ 10Ϫ46ϩ4pH
ϩ ͑1.18 ϫ 10Ϫ69ϩ6pH͒ ϩ ͑7.81 ϫ 10Ϫ93ϩ8pH
͒
͒
1/3
ϩ ͑4.79 ϫ 10Ϫ116ϩ10pH͒ ϩ ͑2.70 ϫ 10Ϫ139ϩ12pH͔͒
(10)
As an example, the solubility goes up from 0.12M at pH 10 to
Ͼ8.0M at pH 12.
Figure 7(a) shows XRD patterns of the powders formed from
Ni(HCOO)2–NH3⅐H2O solution in a N2 atmosphere. It indicates
that the dehydrated Ni(HCOO)2 did not completely decompose at
200° and 300°C, which is similar to the results shown previously.
Ni(OH)2 or NiO phases were expected to be obtained at these low
temperatures, because we have shown that [Ni(NH3)x]2ϩ complex
ions can lose its NH3 content rapidly during solution heating,
leaving a fine Ni(OH)2 precipitate in solution, followed by
decomposition to NiO.13 It is interesting that neither Ni(OH)2 nor
NiO phase was detected at 200° and 300°C. The competition of
HCOOϪ with OHϪ for Ni2ϩ, because of the high concentration
and small solubility of the HCOOH, and the decrease in OHϪ
concentration, because of the NH3 release during heating and the
acidic product of HCOONH4 hydrolysis, may be responsible for
the absence of Ni(OH)2 and NiO in the products.
As shown in Fig. 7(a), an apparent nickel phase (with very weak
Ni3N) was indicated at as low as 300°C. However, a substantial
amount of Ni(HCOO)2 did not decompose at this temperature (see
inset). At 400°C, the formate decomposed completely, and Ni3N
and NiO (possibly NiO1Ϫx, 0 Ͻ x Ͻ 1) formed, whereas the nickel
phase became very weak. At 500°C, phase-pure nickel was
obtained, which is also a very low formation temperature among
the recent literature.7–12,24 The reductive nature of the formate and
the presence of high NH3 concentration in the gas phase are
considered to be responsible for such results.
A small amount of Ni4N was detected in the preparation of
NiFe2O4 by spray pyrolysis at 750°C.25 However, Ni3N has not
been identified in that report and also in another study using
Ni(NO3)2–NH3⅐H2O solution,24 which is similar to our earlier
study using NiCl2–NH3⅐H2O precursor solution.12,13 Figure 7(a)
illustrates that nickel was initially formed by the decomposition of
Ni(HCOO)2. Then, it was nitrided to Ni3N and oxidized to NiO at
400°C. A portion of NH3 gas might have been catalytically
decomposed to H2 by generated nickel.13 It was found that when
the concentrations of aqueous NH3 and Ni2ϩ decreased to 3.6M
and 0.3M, respectively, nickel could be formed as a major phase at
the same temperature, because of the decomposition of Ni3N at
(4) Spray Pyrolysis of Alkaline Solution
As indicated by Eq. (5), the dissolution of Ni(HCOO)2 is sup-
pressed in alkaline solution. However, when Ni(HCOO)2 (1.0M) was
mixed with aqueous NH3 (8.0M), it dissolved rapidly, even faster than
Ni(NO3)2 and NiCl2 in water. This is due to the formation of
nickel-ammine complex ions ([Ni(NH3)x]2ϩ, x ϭ 1–6).23
Here, we consider the Ni(HCOO)2 dissolving in aqueous NH3
solution. Ni2ϩ and NH3 can form nickel-ammine complex ions in
successive steps, yielding Ni(NH3)2ϩ, Ni(NH3)22ϩ, Ni(NH3)23ϩ
,
Ni(NH3)24ϩ, Ni(NH3)52ϩ, and Ni(NH3)26ϩ ions, with overall stabil-
ity constants of ␤1, ␤2, . . . ␤6 for each step.23 The S value of
Ni(HCOO)2 thus is increased with NH3 concentration in the
solution:
1/3
Ksp
6
S ϭ
͑1 ϩ ␤ ͓NH ͔ ϩ ␤ ͓NH ͔2 ϩ · · ·ϩ ␤ ͓NH ͔ ͒
ͩ ͪ
ͫ
ͬ
1
3
2
3
6
3
4
(6)
Again, the concentration and activity of ions are assumed to be
equivalent to simplify the calculation. Because of the equilibrium
NH3͑aq͒ ϩ H2O͑l͒ ϭ NHϩ4 ͑aq͒ ϩ OHϪ͑aq͒
͑pKb ϭ 4.751 at 25°C22͒
(7)
where Kb is the protonation constant of NH3, the equilibrium
concentration of NH3 (aq) can be expressed as a function of
solution pH at equilibrium:
͑Ϫ28ϩ2pH͒
10
͓NH3͔ ϭ
(8)
Kb
The hydrolysis of HCOOϪ(aq) ions,
HCOOϪ͑aq͒ ϩ H2O͑aq͒ ϭ HCOOH͑aq͒ ϩ OHϪ͑aq͒
͑K ϭ 10Ϫ14/Ka ϭ 5.66 ϫ 10Ϫ11 at 25°C͒ (9)

1430
Journal of the American Ceramic Society—Xia et al.
Vol. 84, No. 7
Fig. 7. XRD patterns of the powders formed from [Ni(NH3)6]2ϩ complex solution in a N2 atmosphere ((a) normal residence times and (b) short residence
times). [Ni(II)] ϭ 1.0M, and [NH3⅐H2O] ϭ 8M.
lower NH3 concentration. Ni3N completely decomposed to nickel
and the intermediate NiO was reduced to nickel at 500°C; thus,
phase-pure nickel was obtained at this low temperature.
Figure 7(b) shows XRD patterns of the powders obtained at
short residence times. Nickel and Ni3N mixture phases are
identified. Interestingly, no NiO can be identified from the
powders formed at 0.1 s (500°C) and 0.5 s (500°C). Phase-pure
nickel was obtained at a longer residence time, i.e., 5.5 s.
Undecomposed formate was detected in the former two cases but
not in the latter case. These results imply that some of the particles
did not reach the nominal 500°C in the short-residence-time
process; thus, Ni3N phase formed and increased with residence
time in the range of 0.1–0.5 s. Clearly, heat-transfer resistance
must be considered to explain the incomplete decomposition of
Ni(HCOO)2 at 500°C for 0.5 s. The powder formed at 600°C for
0.1 s shows that major nickel phase coexists with very weak NiO,
without the undecomposed formate.
The short residence time experiments indicate that after nickel
formation by formate decomposition, Ni3N formed through nitrid-
ing and then NiO (possibly NiO1Ϫx, 0 Ͻ x Ͻ 1) formed through
oxidation. Ni3N decomposed at higher temperatures and NiO was
reduced by NH3 (or by H2, through NH3 catalytic decomposition
by Ni13). At a residence time of 5.5 s at 500°C, phase-pure nickel
was obtained.
Based on the above results, it is proposed that nickel may be
formed from nickel-ammine complex solution through the follow-
ing steps: (i) precipitation of Ni(HCOO)2 from solution; (ii)
dehydration and decomposition of the formate, yielding nickel and
gaseous products with the resultants being nickel with undecom-
posed formate; (iii) nitriding of nickel, yielding Ni3N with the
resultants being nickel and Ni3N; (iv) oxidation of nickel, yielding
NiO with the resultants being Ni3N, NiO, and nickel; (v) decom-
position of Ni3N, yielding nickel with the resultants being nickel
and NiO; (vi) reduction of NiO, yielding nickel with resultant
nickel; and (vii) formation of phase-pure nickel followed by
sintering. These steps take place successively or/and simulta-
neously.
SEM images of the powders formed from nickel-ammine
complex solution are shown in Fig. 8. Spherical particles with
smooth surfaces were formed at 300°C. Numerous bowl-like
(almost half-bowl) particles were formed at 600°C (also at 800°C),
and the bowls were comprised of smaller grains. When the
precursor solution is ultrasonically atomized into the reactor,
aqueous NH3 evaporates rapidly. When the formate precipitates
out at the droplet surfaces, the particle interiors may still contain
aqueous NH3, because of the high concentration used in this
research. Another reason may be associated with HCOONH4
decomposition. It has a large solubility in water, i.e., 531 g per

July 2001
Preparation of Nickel Powders by Spray Pyrolysis of Nickel Formate
1431
formed with smooth surfaces. The average particle diameter is
0.65 ␮m, which is much smaller than that formed from NiCl2–
HCOOH solution at the same concentration at 1000°C. In addition
to the improved sintering at 1200°C, the explosion of some
particles may form smaller particles, which also contributes to the
decrease in the average particle size.
The nickel particle morphology and its formation mechanism
from nickel-ammine complex solution are remarkably different
from those shown previously and those in our earlier studies.12,13
IV. Summary
In this article, inorganic Ni(COOH)2 was used to prepare nickel
powder by spray pyrolysis. Dramatically different particle mor-
phologies and various crystalline phases were formed during spray
pyrolysis, which can be primarily attributed to the chemical
reactions occurring in the spray-pyrolysis process.
Phase-pure nickel can be prepared at temperatures as low as
350°C by the decomposition of Ni(HCOO)2; this is the lowest
nickel formation temperature to date using spray pyrolysis.
HCOOH is used as a reducing agent for nickel formation (probably
also for other metals). It is thought that during spray pyrolysis,
HCOOH can decompose to reducing gases such as H2 and CO,
which can protect the nickel from being oxidized. Nickel can also
be obtained at a residence time as short as 0.1 s at 600°C, which
is 1–2 orders of magnitude smaller than earlier reports (100–102 s).
The solubility of Ni(HCOO)2 increases remarkably both in
acidic and alkaline solutions. This enables the practical production
of nickel with adjustable particle sizes in a wide range and high
production rates. Metallic nickel can also be formed from other
nickel salts such as Ni(NO3)2 and NiCl2 in a N2 atmosphere,
simply by adding HCOOH to these nickel salt solutions. The
external reducing sources, such as H2 and NH3, are nonessential.
Ni3N was identified in the product when aqueous NH3 was
added to Ni(HCOO)2 solution. Mixture phases of nickel, Ni3N,
and NiO were present in the product at Ͻ500°C, depending on the
operation conditions. Phase-pure metallic nickel was formed at
500°C, which is also a very low temperature for nickel formation,
and the formation mechanism was proposed.
Acknowledgement
Author BX thanks the Japan Society for the Promotion of Sciences (JSPS) for a
postdoctoral fellowship. The authors are also grateful to Prof. T. Tsuru (Dept. of
Chemical Engineering, Hiroshima University) for the thermal analysis and useful
discussion.
References
1G. L. Messing, S. C. Zhang, and G. V. Jayanthi, “Ceramic Powders Synthesis by
Spray Pyrolysis,” J. Am. Ceram. Soc., 76, 2707 (1993).
2P. S. Patil, “Versatility of Chemical Spray Pyrolysis Technique,” Mater. Chem.
Phys., 59, 185 (1999).
3D. Majumdar, T. T. Kodas, and H. D. Glicksman, “Gold Particle Generation by
Spray Pyrolysis” Adv. Mater., 8, 1020 (1996).
4K. Okuyama, I. W. Lenggoro, N. Tagami, S. Tamaki, and N. Tohge, “Preparation
of ZnS and CdS Fine Particles with Different Particle Sizes by a Spray-Pyrolysis
Method,” J. Mater. Sci., 32, 1229 (1997).
5I. W. Lenggoro, T. Hata, F. Iskandar, M. M. Lunden, and K. Okuyama, “An
Experimental and Modeling Investigation of Particle Production by Spray Pyrolysis
Using a Laminar Flow Aerosol Reactor,” J. Mater. Res., 15, 733 (2000).
6W. Chen, L. Li, J. Qi, Y. Wang, and Z. Gui, “Influence of Electroless Nickel
Plating on Multilayer Ceramic Capacitors and the Implications for Reliability in
Multilayer Ceramic Capacitors,” J. Am. Ceram. Soc., 81, 2751 (1998).
7K. Nagashima, M. Wada, and A. Kato, “Preparation of Fine Particles by the
Spray-Pyrolysis Technique and Their Film Forming Properties in the Thick Film
Method,” J. Mater. Res., 5, 2828 (1990).
Fig. 8. SEM images of the powders formed from [Ni(NH3)6]2ϩ complex
solution in a N2 atmosphere ((a) 300°C, (b) 600°C, and (c) 1200°C).
[Ni(HCOO)2] ϭ 1.0M, [NH3⅐H2O] ϭ 8M. (Bar ϭ 1 ␮m.)
100 g H2O at 80°C; therefore, it can remain in the particle interiors
during drying, followed by the decomposition and the release of
gases above its decomposition temperature (180°C). The smooth
surfaces of the formed particles do not appear permeable enough,
as shown in Fig. 8(a). The subsequent gas evolution from aqueous
NH3 and HCOONH4 can result in particle explosion, which may
be responsible for the formation of bowl-like particles. The
particles can be sintered at 1200°C, and dense particles were
8K. Nagashima, T. Himeda, and A. Kato, “Properties of Conductive Films Made
from Fine Spherical Silver–Palladium Alloy Particles,” J. Mater. Sci., 26, 2477
(1991).
9S. L. Che, K. Takada, K. Takashima, O. Sakurai, K. Shinozaki, and N. Mizutani,
“Preparation of Dense Spherical Nickel Particles and Hollow NiO Particles by Spray
Pyrolysis,” J. Mater. Sci., 34, 1313 (1999).
10S. Stopic, I. Ilic, and D. P. Uskokovic, “Preparation of Nickel Submicron Powder
by Ultrasonic Spray Pyrolysis,” Int. J. Powder Metall., 32, 59 (1996).

1432
Journal of the American Ceramic Society—Xia et al.
Vol. 84, No. 7
11S. Stopic, J. Nedeljkovic, Z. Rakocevic, and D. Uskokovic, “Influence of
19M. A. Borowiak, M. H. Jamro´z, and R. Larsson, “Catalytic Decomposition of
Formic Acid on Oxide Catalysts—An Impulse–Oscillation Model Approach to the
Unimolecular Mechanism,” J. Mol. Catal., 139, 97 (1999).
Additives on the Properties of Spherical Nickel Particles Prepared by Ultrasonic
Spray Pyrolysis,” J. Mater. Res., 14, 3059 (1999).
12B. Xia, I. W. Lenggoro, and K. Okuyama, “Preparation of Ni Particles by
Ultrasonic Spray Pyrolysis of NiCl2⅐6H2O Precursor Containing Ammonia,” J. Mater.
Sci., 36 [7] 1701–705 (2001).
20M. Lintuluoto, H. Nakatsuji, M. Hada, and H. Kanai, “Theoretical Study of the
Decomposition of HCOOH on an MgO(100) Surface,” Surf. Sci., 429, 133 (1999).
21J. Kubota, A. Bandara, A. Wada, K. Domen, and C. Hirose, “IRAS Study of
Formic Acid Decomposition on NiO(111)/Ni(111) Surface: Comparison of Vacuum
and Catalytic Conditions,” Surf. Sci., 368, 361 (1996).
13B. Xia, I. W. Lenggoro, and K. Okuyama “The Roles of Ammonia and
Ammonium Bicarbonate in Preparation of Nickel Particles from Nickel Chloride,” J.
Mater. Res., 15, 2157 (2000).
22R. C. Weast, M. J. Astle, and W. H. Beyer, CRC Handbook of Chemistry and
Physics, 69th Ed.; pp. B-110, D-162 and D-163. CRC Press, Inc., Boca Raton, FL,
1988–89.
14S. Budavari, M. J. O’Neil, A. Smith, and P. E. Heckelman, The Merck Index; p.
1028. Merck and Co., Inc., Rahway, 1989.
15P. G. Blake and S. C. Hinshelwood, “The Homogeneous Decomposition
Reactions of Gaseous Formic Acid,” Proc. R. Soc. London, A, 255, 444 (1960).
16C. H. Bamford and C. F. H. Tipper, Comprehensive Chemical Kinetics. Vol. 4,
Elsevier, Amsterdam, The Netherlands, 1972.
23S. Kotrly and L. Sucha, Handbook of Chemical Equilibrium in Analytical
Chemistry; pp. 197 and 216. Ellis Horwood Limited, England, 1985.
24S. Che, K. Takada, and N. Mizutani, “Formation of Spherical Dense Nickel
Particles by Pyrolyzing the Aerosol of an Ammine Complex Solution in Nitrogen
Atmosphere,” J. Mater. Sci. Lett., 17, 1227 (1998).
17S. Haq, J. G. Love, H. E. Sanders, and D. A. King, “Adsorption and Decompo-
sition of Formic Acid on Ni{110},” Surf. Sci., 325, 230 (1995).
18M. Bowker, E. Rowbotham, F. M. Leibsle, and S. Haq, “The Adsorption and
Decomposition of Formic Acid on Cu{110},” Surf. Sci., 349, 97 (1996).
25H.-F. Yu and A. M. Gadalla, “Preparation of NiFe2O4 Powder by Spray Pyrolysis
of Nitrate Aerosols in NH3,” J. Mater. Res., 11, 663 (1996).
Ⅺ