RF plasma synthesis of nickel nanopowders via hydrogen reduction of nickel hydroxide/carbonate

Article abstract of DOI:10.1016/j.jallcom.2009.03.054

Nickel hydroxide/carbonate was chosen as the precursor to synthesize nickel nanopowders via an RF plasma-assisted hydrogen reduction route. Thermodynamic analysis revealed that the reaction could take place spontaneously. XRD patterns showed that metallic

Full text of DOI:10.1016/j.jallcom.2009.03.054

Journal of Alloys and Compounds 481 (2009) 563–567
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
RF plasma synthesis of nickel nanopowders via hydrogen reduction of
nickel hydroxide/carbonate
Liuyang Baia,b, Junmei Fana, Peng Hua, Fangli Yuana,∗, Jinlin Lia, Qing Tanga
a State Key Laboratory of Multi-phase Complex System, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, PR China
b Graduate University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Nickel hydroxide/carbonate was chosen as the precursor to synthesize nickel nanopowders via an RF
plasma-assisted hydrogen reduction route. Thermodynamic analysis revealed that the reaction could
take place spontaneously. XRD patterns showed that metallic nickel powders could be obtained within
the limited residence time. FESEM images indicated that the products consisted of well-dispersed spheres
with an average diameter of about 60–100 nm. The obtained nickel powders exhibited high tap density.
The use of nickel hydroxide/carbonate as the precursor guaranteed the products of high purity and the
processing of environmental safety. The present plasma-assisted hydrogen reduction of nickel hydrox-
ide/carbonate is an ideal route for large-scale synthesis of well-dispersed metallic nickel nanospheres
used as electrode materials.
Received 17 January 2009
Received in revised form 4 March 2009
Accepted 4 March 2009
Available online 21 March 2009
Keywords:
Electrode materials
Metals
Chemical synthesis
Scanning and transmission electron
microscopy
© 2009 Elsevier B.V. All rights reserved.
X-ray diffraction
1. Introduction
micron spheres while evaporation–condensation leads to spheri-
cal nanoparticles. When the particle size of the input materials is
Fine nickel powders have great potential technological applica-
tions such as conducting paints, rechargeable batteries, chemical
catalysts, microwave absorbing materials, and magnetic recording
media [1–5]. In addition, they are attracting increasing attention
as the substitute for noble metals Pd or Pd/Ag used in the inter-
nal electrodes of multilayer ceramic capacitor (MLCC) due to their
good electrical conductivity, high melting point, and low cost [6].
Nickel powders as the electrode materials need to be prepared as
well-dispersed spheres with a narrow size distribution. Practical
application always promotes the synthetic development. A vari-
ety of techniques such as chemical reduction, spray pyrolysis, and
plasma technique have been employed to produce fine spherical
nickel powders [7–14].
Radio-frequency (RF) thermal plasma is an easy way to prepare
spherical powders [15–18]. The plasma system has a high tempera-
ture flame (up to 10,000 K) with a rapidly cooled tail (105–106 K/s).
The high temperature region can provide enough energy for
the melting/evaporation of the raw materials and the rapidly
cooled tail could help rapid solidification. Melting–spheroidization
and evaporation–condensation often coexist during the forma-
tion of spherical powders. Melting–spheroidization results in
small enough and the feedrate is proper, all the raw materials would
evaporate instantly upon injection into the plasma flame and sub-
sequently condensate into nanoparticles in the quenching region.
The plasma gas does not come in contact with electrodes, which
can eliminate possible sources of contamination. Furthermore, the
operation environment is flexible from oxidizing to reducing atmo-
sphere [18]. So RF thermal plasma is also an effective way to prepare
high purity powders. Using the plasma-assisted physical vapor
deposition (PVD) method, various nanopowders including metal
and ceramic were synthesized [19,20]. RF thermal plasma could
also be used in the chemical synthesis due to its chemically active
species [18,21]. In this regard, Chau [22] attempted to synthesize Ni
and FeNi nanopowders using nickel chloride as precursor in reduc-
tion atmosphere provided by hydrogen. But the generated HCl in
the exhausting gas would be harmful to the equipments as well
as human health and environment. It is also difficult to obtain high
purity powders using chloride as precursor for the strong Cl adsorp-
tion.
In the present paper, we reported the RF plasma synthesis of
spherical nickel nanoparticles via hydrogen reduction of nickel
hydroxide/carbonate. Both thermodynamic calculation and exper-
imental results showed that metallic nickel powders could be
obtained. Different influencing parameters were examined. The for-
mation process of the nickel particles was discussed based on the
experimental results.
∗
Corresponding author. Tel.: +86 10 82627058; fax: +86 10 62561822.
E-mail address: flyuan@home.ipe.ac.cn (F. Yuan).
0925-8388/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2009.03.054

564
L. Bai et al. / Journal of Alloys and Compounds 481 (2009) 563–567
Fig. 2. ꢁG–T curves of the possible reduction reactions.
gaseous H2O and form gaseous Ni(OH)2 [24]. Accordingly, the pos-
sible reaction during the plasma processing can be expressed as
below:
NiCO3(s) → NiO(s) + CO2(g)
Ni(OH)2(s) → NiO(s) + H2O(g)
NiO(s) → NiO(g)
(1)
(2)
(3)
(4)
(5)
(6)
(7)
NiO(s) + H2O(g) → Ni(OH)2(g)
NiO(s) + H2(g) → Ni(g) + H2O(g)
NiO(g) + H2(g) → Ni(g) + H2O(g)
Ni(OH)2(g) + H2(g) → Ni(g) + 2H2O(g)
Fig. 1. The schematic illustration of the experimental setup.
2. Experimental
Nickel hydroxide and basic nickel carbonate were used as the precursor. Nickel
hydroxide was synthesized using ammonia-precipitation method. Na has low sur-
face energy and is difficult to be removed from the resulting powders. The present
ammonia-precipitation method guaranteed the final products of high purity. Basic
nickel carbonate was provided by JNMC (Jinchuan Group Ltd., China). Commercial
nickel hydroxide microspheres were obtained from WESTAIM.
The experiments were carried out in an RF thermal plasma system under atmo-
spheric pressure. The experimental setup consists of an RF generator (30 kW, 4 MHz),
a plasma generator, a reactor, and a powder-collecting filter. The schematic illustra-
tion of the setup is shown in Fig. 1 [23]. The screw feeder was homemade and the
feedrate could be controlled by setting the screw speed.
In a typical experimental procedure, stable plasma was first generated using
argon and nitrogen as the plasma-forming gas and sheath gas, respectively. Nickel
hydroxide or basic nickel carbonate was supplied into the plasma flame in a contin-
uous way with the use of hydrogen as carrier gas. Hydrogen also played the role of
reducing agent. The precursor underwent vaporization and reduction, and then the
resulting metallic nickel species condensed and formed nanoparticles as a result of
the high quenching rate. Most of the products were collected from the filters and
some were collected from the chamber wall. Detailed processing parameters are
given in Table 1.
The crystalline phase of the as-prepared samples was characterized by X-ray
diffractometry (XRD, X’pert PRO, Panalytical, Cu K␣ radiation) in a 2ꢀ range from
10◦ to 110◦. The purity of the product was demonstrated by energy dispersive X-ray
spectrometry (EDX). Their size and morphology were inspected with field emission
scanning electron microscopy (FESEM, JEOL JSM-6700F). The thermal stability of the
products was recorded using thermal analyzer (Netzsch STA 449 TG–DTA/DSC). The
tap density of the nickel powders was examined with the help of a Hall flowmeter.
The change of the Gibbs free energy ꢁG is usually used to judge
the spontaneous direction of a reaction. The corresponding ꢁG–T
curves of the reduction are shown in Fig. 2. (The curves were drawn
based on the data given in Ref. [24].) The ꢁG values for both reac-
tions between NiO and H2 are below zero and declined as the
temperature increased, which means that both of them can take
place spontaneously. The ꢁG value for Eq. (7) is below zero at the
temperature higher than 1550 K, indicating that the reactions can
also take place spontaneously at high temperature.
Experimental results showed that metallic nickel powders could
be obtained whether nickel hydroxide or basic nickel carbonate
was used as precursor. The accomplishment of complete reduc-
tion within the limited residence time in plasma flame might be
attributed to the unique reaction field provided by the thermal
nonequilibrium effects in plasma [25,26]. Fig. 3 shows the typi-
cal XRD patterns of the product obtained with proper feeding rate,
3. Results and discussion
NiCO3 (s) decomposes at the temperature of 381 K and Ni(OH)2
(s) decomposes at the temperature of 600 K. NiO can react with
Table 1
Detailed parameters for plasma processing.
Parameters
Values
Central gas, argon
1.0 m3/h
Sheath gas, nitrogen
Carrier gas, hydrogen
Quenching gas, nitrogen
Powder feedrate
5.0 m3/h
0.7 m3/h
0–5.0 m3/h
2–10 g/min
Fig. 3. XRD patterns of the product obtained using nickel hydroxide (a) and basic
nickel carbonate (b), respectively.

L. Bai et al. / Journal of Alloys and Compounds 481 (2009) 563–567
565
shape, the uniform size and thickness of which is about 500 nm
and 50 nm. The basic nickel carbonate particles shown in Fig. 5b
are irregular in shape and loosely aggregated. Powders with big-
ger particle size always have better fluidity, which is helpful for the
feedrate control. However, bigger particles are difficult to evaporate
and would influence the following reduction. Both the homemade
nickel hydroxide and commercial basic nickel carbonate could be
used as the raw material, while the commercial nickel hydroxide
microspheres are too big and dense for complete evaporation and
reduction. Fig. 5c shows the FESEM images of the nickel powders
obtained using homemade nickel hydroxide as the precursor with
a feeding rate of 2 g/min. The nickel particles are well-dispersed
nanospheres and most particles have a diameter in the range of
50–100 nm and only a few coarse particles exist. The products
obtained using commercial basic nickel carbonate as the precursor
shown in Fig. 5d exhibit similar results.
Fig. 4. EDX spectra of the product.
As mentioned previously, the main products were collected from
the filters, and some were collected from the chamber wall. The
products collected from the chamber account for about 20%. Fig. 6
shows the FESEM images of the nickel powders collected from the
chamber wall using homemade nickel hydroxide as precursor. The
powders collected from the high temperature region (shown in
Fig. 6a) are more aggregated and have bigger particle size com-
pared to that collected from the low temperature region (shown
in Fig. 6b). The reason is that the particles deposited on the wall
in high temperature region keep active and can absorb more nickel
species and grow. Meanwhile, the active nickel particles have a ten-
dency to aggregate. On the contrary, the particles staying in the low
temperature region lose the activity partly.
using nickel hydroxide and basic nickel carbonate, respectively. The
five diffraction peaks of (1 1 1), (2 0 0), (2 2 0), (3 1 1), and (2 2 2) of
face-centered cubic (fcc) Ni can be easily observed. No peaks of
nickel oxide are detected, indicating high purity of the products.
Fig. 4 shows the EDX spectra of the product. Besides that of Ni
confirmed by XRD patterns, the peaks associated to O is present. The
other peaks originate from the Cu grid and the TEM equipment. The
mole content of O was 1.2%. Taking the air adsorbed on the surface
of the Ni particles into consideration, the mole content of O in the
product was about 1%.
Fig. 5 shows the FESEM images of the homemade nickel hydrox-
ide and commercial basic nickel carbonate, and the products
obtained using them as precursors, respectively. The nickel hydrox-
ide particles shown in Fig. 5a are well-dispersed and flake-like in
The cooling rate has great effect on both shape and size of the
resulting particles. Rapid cooling is necessary for the formation
Fig. 5. FESEM images of the homemade nickel hydroxide (a) and commercial basic nickel carbonate (b), and the product obtained using them as procedure (c and d).

566
L. Bai et al. / Journal of Alloys and Compounds 481 (2009) 563–567
Fig. 6. FESEM images of the nickel powders collected from the chamber wall: (a) high temperature region and (b) low temperature region.
Fig. 7. FESEM images of the nickel powders obtained with (a) and without (b) quenching gas.
of spherical particles. The plasma system itself has a high cooling
rate, so the products exhibited spherical shape whether quench-
ing gas was introduced or not. Fig. 7 shows the FESEM images
of the nickel powders obtained in case with (a) and without (b)
quenching gas. Both powders consist of spherical nanoparticles.
However, the average particle size was reduced from 100 nm to
60 nm when nitrogen was introduced as quenching gas at a rate
of 5 m3/h.
tion of the nickel particles. The proper feeding rate was about
10 g/min.
We believe that the reduction and formation of nickel nanopar-
ticles were achieved via vapor instead of liquid phase. In order
to verify this process, commercial nickel hydroxide microspheres
with the size range of 2–10 ␮m were used as precursor to synthe-
size nickel powders. Two separated categories of particles existed
in the resulting products. Fine particles similar with that shown
in Fig. 5c and d were collected from filters, which were indicated
by XRD as pure metallic Ni. Big particles collected from the bot-
tom of the reactor remained the shape and size of the nickel
hydroxide precursor as shown in Fig. 8. XRD pattern showed that
the main phase of the big particles was hydroxide. It is known
that the surface of the particles becomes smooth and the size
The effect of feedrate on the resulting powders was also
examined. Particle size increased
a bit and the aggregation
became serious with the feeding rate increased. The con-
centration of the reactive species would increase when the
feedrate was high and the resulting Ni species would increase
accordingly, which would lead to the growth and aggrega-
Fig. 8. FESEM images of the commercial nickel hydroxide (a) and the product collected from the bottom of the reactor (b).

L. Bai et al. / Journal of Alloys and Compounds 481 (2009) 563–567
567
4. Conclusions
Spherical nickel nanoparticles with uniform size were syn-
thesized using RF plasma-assisted hydrogen reduction route.
Nickel hydroxide/carbonate was chosen as the precursor, which
guaranteed the products of high purity and the processing of envi-
ronmental safety. The product consists of well-dispersed particles
in the size range of 60–100 nm and has a tap density as high
as 3.7 g/cm3. The present plasma-assisted hydrogen reduction is
an ideal route for large-scale synthesis of well-dispersed metallic
nickel nanospheres used as electrode materials.
Acknowledgements
This work was supported financially by the National Natural Sci-
ence Foundation of China (No. 50574083) and the National High
Technology Research and Development Program of China (863) (No.
2008AA03Z308).
Fig. 9. TG curve of the obtained Ni powders.
References
[1] H. Shoji, Y. Nakano, H. Matsushita, A. Onoe, H. Kanai, Y. Yamashita, J. Mater.
Synth. Process. 6 (1998) 415.
[2] J.Z. Gao, F. Guan, Y.C. Zhao, W. Yang, Y.J. Ma, X.Q. Lu, J.G. Hou, J.W. Kang, Mater.
Chem. Phys. 71 (2001) 215.
[3] H.J. Zhang, H.T. Zhang, X.W. Wu, Z.L. Wang, Q.L. Jia, X.L. Jia, J. Alloys Compd. 419
(2006) 220.
[4] H. Niu, Q. Chen, M. Ning, Y. Jia, X. Wang, J. Phys. Chem. B 108 (2004) 3996.
[5] F. Davar, Z. Fereshteh, M. Salavati-Niasari, J. Alloys Compd. 476 (2009) 797.
[6] K.H. Kim, H.C. Park, S.D. Lee, W.J. Hwa, S.S. Hong, G.D. Lee, S.S. Park, Mater. Chem.
Phys. 92 (2005) 234.
[7] L.Y. Bai, F.L. Yuan, Q. Tang, J.L. Li, H. Ryu, J. Mater. Sci. 43 (2008) 1769.
[8] J. Park, E. Chae, S. Kim, J. Lee, J. Kim, S. Yoon, J. Choi, Mater. Chem. Phys. 97 (2006)
371.
[9] Y.Q. He, X.G. Li, M.T. Swihart, Chem. Mater. 17 (2005) 1017.
[10] B. Xia, I.W. Lenggoro, K. Okuyama, J. Mater. Sci. 36 (2001) 1701.
[11] Z.Q. Wei, T.D. Xia, L.F. Bai, J. Wang, Z. Wu, P.X. Yan, Mater. Lett. 60 (2006) 766.
[12] G. Shanmugavelayutham, V. Selvarajan, Bull. Mater. Sci. 27 (2004) 453.
[13] X.H. Luo, Y.Z. Chen, G.H. Yue, D.L. Peng, X.T. Luo, J. Alloys Compd. 476 (2009)
864.
[14] L.Y. Bai, F.L. Yuan, Q. Tang, Mater. Lett. 62 (2008) 2267.
[15] N. Kobayashi, Y. Kawakami, K. Kamada, J.G. Li, R. Ye, T. Watanabe, T. Ishigaki,
Thin Solid Films 516 (2008) 4402.
[16] S. Kumar, V. Selvarajan, P.V. Padmanabhan, K.P. Sreekemar, J. Mater. Process.
Technol. 176 (2006) 87.
[17] X.L. Jiang, M. Boulos, Trans. Nonferrous Met. Soc. China 16 (2006) 13.
[18] M. Boulos, IIEEE Trans. Plasma Sci. 19 (1991) 1078.
[19] B.M. Goortani, P. Proulx, S. Xue, N.Y. Mendoza-Gonzalez, Powder Technol. 175
(2007) 22.
[20] P. Hu, S.K. Yan, F.L. Yuan, L.Y. Bai, J.L. Li, Y.F. Chen, Plasma Sci. Technol. 9 (2007)
611.
[21] S. Son, M. Taheri, E. Carpenter, V.G. Harris, M.E. McHenry, J. Appl. Phys. 91 (2002)
7589.
[22] J.L.H. Chau, Mater. Lett. 61 (2007) 2753.
[23] P. Hu, F.L. Yuan, L.Y. Bai, J.L. Li, Y.F. Chen, J. Phys. Chem. C 111 (2007) 194.
[24] O. Knacke, O. Kubaschewski, K. Hesselmann, Thermochemical Properties of
Inorganic Substances, Springer-Verlag, Berlin/New York; Verlag Stahleisen,
Dsseldorf, 1991, pp. 309–1458.
[25] M.I. Boulos, P. Fauchais, E. Pfender, Thermal Plasma: Fundamentals and Appli-
cations, vol. 1, Plenum Press, New York, 1994.
[26] J.G. Li, X.H. Wang, K. Watanabe, T. Ishigaki, J. Phys. Chem. B 110 (2006) 1121.
[27] R. Liao, D.Z. Zhou, Y. Zhang, G.F. Pan, Chin. J. Electron. Compon. Mater. 22 (2003)
28.
keeps almost constant after plasma melting–spheroidization. If
the reduction had been achieved via liquid phase, the obtained
nickel particles should be in micron size when commercial nickel
hydroxide microspheres were used as precursor. As a matter of
fact, nickel nanoparticles were obtained instead, and the newly
formed Ni were not on the surface but separated from the pre-
cursor. It can be induced that the reduction and formation of
nickel nanoparticles were achieved via vapor instead of liquid
phase.
Thermal stability and tap density are two important param-
eters for the electrode materials. The sample obtained with the
feedrate of 10 g/min and cooling gas rate of 5 m3/h was exam-
ined. The thermal stability of the products was recorded using
thermal analyzer. As shown in Fig. 9, obvious weight gain starts
at about 250 ◦C, indicating thermal oxidation of the obtained
nickel powders began at about 250 ◦C. Compared with that
reported (300 nm, 340 ◦C), the obtained nickel nanoparticles are
easily oxidized [7]. The low oxidation temperature is related to
the small particle size. Most of the electrodes need sintered
at relatively high temperature (above 800 ◦C), so inert/reducing
atmosphere is necessary for the materials with poor oxidation resis-
tance.
In order to measure the tap density with the help of a
Hall flowmeter, the container was vibrated regularly when the
powders were flowing down. The accordingly calculated result
was 3.7 g/cm3, which was much higher than that reported in
the literature (1000 nm, 3.0 g/cm3) [27]. We also measured the
commercial carbonyl nickel powders and the spherical nickel
powders synthesized via wet chemical reduction [7], using the
same flowmeter-assisted method. Both samples are no more than
2.5 g/cm3 in tap density. So the present plasma-assisted hydrogen
reduction is an effective route to synthesize nickel nanopowders
with a high tap density.