Coating fine nickel particles with Al2O3 utilizing an atomic layer deposition-fluidized bed reactor (ALD-FBR)

Article abstract of DOI:10.1111/j.1551-2916.2004.00762.x

An atomic layer deposition-fluidized bed reactor (ALD-FBR) method has been developed to deposit ultrathin and conformal coatings on fine particles. Experiments of Al₂O₃ deposition on 150-μ-diameter nickel particles were conducted. Th

Full text of DOI:10.1111/j.1551-2916.2004.00762.x

J. Am. Ceram. Soc., 87 [4] 762–65 (2004)
journal
Coating Fine Nickel Particles with Al2O3 Utilizing an
Atomic Layer Deposition-Fluidized Bed Reactor (ALD–FBR)
Jeffrey R. Wank,† Steven M. George,‡ and Alan W. Weimer*,†
Department of Chemical Engineering and Department of Chemistry, University of Colorado,
Boulder, Colorado 80309
An atomic layer deposition–fluidized bed reactor (ALD–FBR)
method has been developed to deposit ultrathin and conformal
coatings on fine particles. Experiments of Al2O3 deposition on
150-␮m-diameter nickel particles were conducted. The fluid-
ized bed was constructed to operate under vacuum, and the
fluidizing gas used was nitrogen. Trimethylaluminum and
water were used as dosing reagents. The reactions were
conducted at 450 K. Successful deposition of alumina films,
with thickness controllable at the nanometer level, was ob-
served based on transmission electron microscopy imaging,
inductively coupled plasma atomic emission spectrometry,
X-ray photoemission spectroscopy, particle-size distributions,
and wavelength-dispersive spectrometry imaging.
e.g., nickel. Because the substrate has a limited number of reactive
sites, each reaction continues until all the sites have been reacted,
and no further. This approach has been successful in coating very
fine boron nitride (BN) particles on a very small (Ͻ0.1 g) scale9,10
as well as flat silicon substrates;8,11,12 however, until now, ALD
on nickel has not been reported.
The theory and practice of fluidized bed reactor technology has
been discussed for many years.13 The practice of using a fluidized
bed reactor to produce coatings via a deposition method was first
described in the 1960s for nuclear fuels,14 but relatively little effort
went into this method until the late 1980s and early 1990s.15–18 All
these efforts describe chemical vapor deposition–fluidized bed
reactor (CVD–FBR) technologies in one fashion or another; for a
review, see Chen and Chen.5
The major goal of the research described here is to demonstrate
and develop an understanding of ALD–FBR. Although ALD and
FBR technologies have been discussed separately, combining the
two has not been previously reported. The nickel particles were
selected because of their relative ease of fluidization, whereas the
alumina coatings were chosen because of the availability of
reactants and prior work with alumina ALD.6,8–10
I. Introduction
HE ability to coat particles with an ultrafine layer of another
T
substance has recently been given more attention.1–5 Many
reasons for this type of coating can be given, but the overriding
theme for each of them is the alteration of the surface properties of
the particles without affecting the bulk properties. Atomic layer
deposition (ALD)6,7 provides a unique method for performing this
task, because the films generated using this method are conformal
and can be controlled down to the atomic level.
II. Experimental Procedures
ALD is based on standard chemical vapor deposition (CVD)
technology, except that it splits the binary reaction into two
half-reactions that occur on the surface of the substrate. In the case
of aluminum using trimethylaluminum (TMA) and water as
reactants, the binary reaction is
As shown in Fig. 1, the apparatus for ALD–FBR is fairly
complex. The reactor itself is composed of stainless steel and is
encased by a clamshell-type furnace. The reactant bubblers are
attached to the system via their vent lines, and are operated by the
2Al͑CH3͒3 ϩ 3H2O 3 Al2O3 ϩ 6CH4
Thus, the two half-reactions are
(1)
AlOH* ϩ Al͑CH͒3 3 Al-O-Al͑CH3͒2* ϩ CH4
(1a)
(1b)
Al͑CH3͒* ϩ H2O 3 AlOH* ϩ CH4
where the starred species are the surface species.6,8 Note that this
indicates growth of the film; were it to represent the initial growth
on a particle surface, the alumina species on the right-side of
reaction (1a) (as well as the first alumina species on the left-side
of reaction (1b)) would be whatever the particle is comprised of,
T. M. Besman—contributing editor
Manuscript No. 187223. Received January 4, 2002; approved January 29, 2003.
Funded by the National Science Foundation Center for Ceramics and Composite
Materials (University of New Mexico/Rutgers University) and Integrated Sciences
(Tulsa, OK).
Fig. 1. Simplified schematic diagram of the ALD–FBR apparatus: (1)
mass flow controller; (2) main vacuum pump; (3) reactor; (4) fluidized bed
of powder; (5) distributor; (6) dosing line purge pumps.
*Member, American Ceramic Society.
†Department of Chemical Engineering.
‡Department of Chemistry.
762

April 2004
Communications of the American Ceramic Society
763
Fig. 2. TEM images of Al2O3-coated nickel particles. The first picture shows 25 AB cycles, yielding 2.3 nm. The second picture shows 50 AB cycles,
yielding 4 nm.
driving force of their vapor pressure. The reactor itself is main-
tained at a vacuum using a large pump (Model 2063AC, Alcatel
Vacuum Technology, Paris, France), and the dosing lines can also
be pumped down using smaller separate vacuum pumps (Model
2008A, Alcatel Vacuum Technology, Paris, France). The fluidiz-
ing gas flow is maintained using a mass flow controller (MFC)
(MKS Instruments, Andover, MA). An additional MFC can
control a separate purge flow through the dosing lines, which is
only used to clean the system after operation. The dosing line
entrance into the reactor is just beneath the removable porous
metal distributor plate. Operationally, the bed (while being fluid-
ized) is operated at three times minimum fluidization velocity. The
method used to calculate minimum fluidization velocity for the
conditions encountered in our system is described elsewhere.19,20
Micrograph images of Al2O3-coated nickel particles are per-
formed using transmission electron microscopy (TEM; Model
EM430, Philips, Eindhoven, Netherlands), operating at 300 kV.
The chemical composition of the surface is analyzed using X-ray
photoelectron spectroscopy (XPS; Axis Kratos Analytical, Hof-
heim, Germany). Wavelength dispersive spectroscopy (WDS) and
inductively coupled plasma–auger electron spectroscopy (ICP–
AES) were performed using an electron microprobe (Model JXA
8600, JEOL, Tokyo, Japan) and an (ICP–AES 3410ϩ, Applied
Research Laboratories). Particle size analysis was performed using
a particle size analyzer (Model 3225 Aerosize, TSI Inc., Shore-
view, MN).
coatings based on peak areas (Table I). These results show the
expected trend, i.e., as the number of cycles increases, the
thickness of the Al2O3 layer increases. This can be seen from the
ratio of area for Al 2s and Ni 2p peaks. As the number of cycles
increases, the intensity of Al 2s increases; simultaneously, the
intensity of Ni 3p decreases.
ICP-AES gives parts of aluminum per million parts of nickel.
The sample is initially dissolved in HF. Using the average particle
size of the nickel particles, combined with their density and the
assumed density of the alumina film (also taking into account the
stoichiometry of aluminum in the film), a film thickness could be
calculated. This method was chosen because it is very quick and
easy to perform, and yields acceptable data given the above
analyses (most notably the TEM images). For a 50 AB cycle run,
the average film thickness is 4.6 nm, very close to the observed 4
nm. ICP–AES runs on the 25-, 75-, and 100-cycle films yielded 3,
8.1, and 11.1 nm, respectively. A graphical representation of
ICP–AES results is shown in Fig. 3.
WDS mapping of the surface of the coated particles shows that
all particles have been coated. This is important because it shows
III. Results and Discussion
Several runs of different numbers of AB cycles, giving different
film thickness at ϳ0.1 nm per AB cycle, are performed to verify
the method. Runs were performed with 25, 50, 100, and 200
cycles. TEM images of coated particles are then obtained; the
results are shown in Fig. 2. The 25-cycle run yields a coating of 2.3
nm, while the 50-cycle run yields a coating of 4 nm. The XPS data
accumulated shows that the 75 and 100 cycles have thicker
Table I. ALD Observations and Calculations
Thickness
(observed)
Thickness
AAl 2s/
Sample
(calculated) AAl 2s AAl 2p ANi 2p ANi 2p
Fig. 3. Graphical representation of ICP-AES calculated thickness versus
the number of AB cycles performed. The slope of the line is approximately
one, indicating that ϳ.01 nm of Al2O3 is being deposited per cycle. Error
bars were created by taking the high and low values for observed particle
diameters (see Fig. 5).
25 AB cycles
50 AB cycles
75 AB cycles
100 AB cycles
2.3 nm
4.0 nm
3.0 nm
4.6 nm
8.1 nm
11.1 nm
1.0
1.4
2.5
3.1
0.9
1.4
2.2
2.7
2.8 0.3
1.3 1.1
1.3 1.9
0.7 4.6

764
Communications of the American Ceramic Society
Vol. 87, No. 4
Fig. 4. (a) WDS pattern for aluminum; (b) simple SEM image of the same frame of nickel particles.
that ALD is a site-specific surface reaction; it takes place on all
available surfaces (as long as the surface is reactive). By pulsing
each reactant twice, complete surface coverage is achieved before
switching reagents. Spectrum for 50 AB cycles is shown in Fig. 4.
Particle-size distributions are observed for the nickel particles
used with no coating, and 25, 50, 75, and 100 AB cycles. The
distributions are all virtually identical; the mean particle size for
every run, including the “no-coating” run, are all within experi-
mental error of each other, ranging from 151–152 Ϯ 1 ␮m. The
curves for no coating and 100 AB cycles are shown in Fig. 5, with
the others left out for clarity. Both curves are normalized to one.
This result indicates that the particles are not coated as agglomer-
ates during the coating process, rather as individual particles.
Transmission electron microscopy images show the conformal
nature of the films, as well as indicating the level of control for the
thickness of the film. X-ray photoelectron spectroscopy and
inductively coupled plasma-auger electron spectroscopy data indi-
cate that varying the number of cycles performed can control the
thickness of the films. Wavelength dispersive spectroscopy spec-
tral mapping shows that each particle is coated in approximately
the same way, i.e., the particles are all coated equally.
Acknowledgments
The authors thank Dr. Clint Dutcher for his time, resources, and valuable
discussion. The authors would also like to thank Dr. John Drexler in the Department
of Geology at the University of Colorado for his analytical expertise. Additionally, the
authors thank the National Science Foundation’s GAANN fellowship program.
IV. Conclusions
An atomic layer deposition–fluidized bed reactor (ALD–FBR)
method has been developed to coat fine nickel particles with an
ultrafine, conformal coating of Al2O3. This coating method pro-
vides a method to encapsulate fine particles with an atomic-level
thickness controlled film.
References
1B. Golman, S. Ishino, and K. Shinohara, “CVD Coating of Si3N4 Ultra-Fine
Particles with AlN in Co-Current Moving Bed Reactor,” J. Chem. Eng. Jpn., 30 [6]
1138–40 (1997).
2T. Hanabusa, S. Uemiya, and T. Kojima, “Production of Si3N4/Si3N4 and
Si3N4/Al2O3 Composites by CVD Coating of Fine Particles with Ultrafine Powder,”
Chem. Eng. Sci., 54 [15–16] 3335–40 (1999).
3B. Hua and C. Z. Li, “Production and Characterization of Nanocrystalline SnO2
films on Al2O3 Agglomerates by CVD in a Fluidized Bed,” Mater. Chem. Phys., 59
[2] 130–35 (1999).
4Y. W. Yen and S. W. Chen, “Nickel and Copper Deposition on Fine Alumina
Particles by Using the Chemical Vapor Deposition-Circulation Fluidized Bed Reactor
Technique,” J. Mater. Sci., 35 [6] 1439–44 (2000).
5C. C. Chen and S. W. Chen, “Nickel and Copper Deposition on Al2O3 and SiC
Particulates by Using the Chemical Vapor Deposition Fluidized Bed Reactor
Technique,” J. Mater. Sci., 32 [16] 4429–35, (1997).
6S. M. George, A. W. Ott, and J. W. Klaus, “Surface Chemistry for Atomic Layer
Growth,” J. Phys. Chem., 100 [31] 13121–31 (1996).
7S. M. George, A. W. Ott, and J. W. Klaus, “Atomic Layer Growth Using Binary
Reaction Sequence Chemistry”; p. 41-PHYS in Abstracts of Papers of the American
Chemical Society, 213. American Chemical Society, Columbus, OH, 1997.
8S. M. George, O. Sneh, A. C. Dillon, M. L. Wise, A. W. Ott, L. A. Okada, and
J. D. Way, “Atomic Layer Controlled Deposition of SiO2 and Al2O3 Using
Abab–Binary Reaction Sequence Chemistry,” Appl. Surf. Sci., 82 [3] 460–67 (1994).
9J. D. Ferguson, A. W. Weimer, and S. M. George, “Atomic Layer Deposition of
Ultrathin and Conformal Al2O3 films on BN Particles,” Thin Solid Films, 371, 95–104
(2000).
10J. D. Ferguson, A. W. Weimer, and S. M. George, “Atomic Layer Deposition of
Al2O3 and SiO2 on BN Particles Using Sequential Surface Reactions,” Appl. Surf.
Sci., 162–163, 280–92 (2000).
11G. S. Higashi and C. G. Fleming, “Sequential Surface Chemical Reaction Limited
Growth of High Quality Al2O3 Dielectrics,” Appl. Phys. Lett., 55 [19] 1963–65
(1989).
12A. C. Dillon, A. W. Ott, J. D. Way, and S. M. George, “Surface Chemistry of
Al2O3 Deposition Using Al(CH3) and H2O in a Binary Reaction Sequence,” Surf.
Sci., 322, 230–42 (1995).
13D. Kunii and O. Levenspiel, Fluidization Engineering; pp. 534. Wiley, New
York, 1969.
Fig. 5. Normalized particle size distributions for 150 ␮m fine nickel
particles, uncoated and coated with ϳ100 nm of aluminum.

April 2004
Communications of the American Ceramic Society
765
14J. M. Bolcher, “Nuclear Fuel Particles Coated with Mixture of Pyrolytic Carbon
and Silicon Carbide,” 1966.
18A. Kobata, T. Okubo, K. Kusakabe, and S. Morooka, “Preparation of Compos-
iteCeramic Particles by Gas-Phase Reactions in Fluidized-Bed,” Kagaku Kogaku
Ronbunshu, 16 [3] 543–50 (1990).
15A. Sanjurjo, K. Lau, and B. Wood, “Chemical Vapor-Deposition in Fluidized-
Bed Reactors,” Surf. Coat. Technol., 54 [1–3] 219–23 (1992).
19J. R. Wank, A. W. Weimer, and S. M. George, “Vibro-Fluidization of Fine Boron
Nitride Powder at Low Pressure,” Powder Technol., 121 [2–3] 195–204 (2001).
20M. F Llop, F. Madrid, J. Arnaldos, and J. Casal, “Fluidization at Vacuum
Conditions: A Generalized Equation for the Prediction of Minimum Fluidization
16S. Morooka, T. Okubo, and K. Kusakabe, “Recent Work on Fluidized-Bed Process-
ing of Fine Particles as Advanced Materials,” Powder Technol., 63 [2] 105–12 (1990).
17K. Kusakabe, T. Kuriyama, and S. Morooka, “Fluidization of Fine Particles at
Reduced Pressure,” Powder Technol., 58 [2] 125–30 (1989).
Velocity,” Chem. Eng. Sci., 51 [23] 5149–57 (1996).
Ⅺ