1090
Chemistry Letters 2002
Synthesis of Thermally Stable Cs-Doped Alumina Nanoparticles by Microemulsion Method
Zhixiong You, Ioan Balint, and Ken-ichi Aikaꢀ
Department of Environmental Chemistry and Engineering, Interdisciplinary Graduate School of Science and Technology,
Tokyo Institute of Technology, 4259Nagatsuta-cho, Midori-ku, Yokohama 226-8502
(Received June 24, 2002; CL-020529)
Cs-doped alumina nanoparticles were synthesized by micro-
emulsion method and then characterized. Cs-doping greatly
improved aluminas thermal stability. For an optimum molar ratio
of Cs/Al = 1/11 the structure of ꢀ-alumina can be preserved even
at temperatures as high as 1330 ꢁC.
doped alumina nanoparticles were then recovered by removing
the solvents by rotoevaporation and then by decomposing the
surfactant at 500 ꢁC (‘‘as-prepared’’ samples). After character-
ization, these ‘‘as-prepared’’ samples were calcined at 1100 ꢁC
for 24 h.
Nitrogen adsorption measurements were performed at
À196 ꢁC (BELSORP 28SA). The results are listed in Table 1.
The specific surface area of the as-prepared samples decreased
monotonically with increasing the Cs amount. The surface area of
the high temperature calcined samples (1100 ꢁC for 24 h) was also
related to the Cs content but the highest surface area (102 m2/g)
was observed for the material with Cs to Al molar ratio of 1/11.
Lower or higher Cs/Al molar ratios led to a decrease in the final
surface area. The structure of Cs free sample collapsed after high
temperature calcination (Sa ꢂ 14 m2/g). The experimental re-
sults are clearly proving that Cs-doping significantly improved
the thermal stability of alumina.
Active alumina, especially ꢀ-alumina, has been widely used
as adsorbents, supports, and catalysts.1;2 However, it transforms
to ꢁ-alumina when it is used in high temperature processes such
as partial oxidation, three-way catalysts, methane combustion, or
steam reforming.3 This transformation leads to drastic reduction
in specific surface area and the collapse of pore structure which
are essential elements in catalytic reactions. Two strategies can be
adopted to overcome this sintering problem. One is to incorporate
large cations, e.g. Ba2þ,4 La3þ,5 etc., into the matrix of active
alumina to prevent the phase transition to ꢁ-alumina. The other
way is to convert alumina into barium hexaaluminate (BHA) at
the lowest temperature possible. BHA particles have been
reported to grow anisotropically and thus they show better
thermal stability than alumina.4 If BHA is formed at low
temperature, the growth of BHA particles can be suppressed.
Reverse microemulsion technique is now extensively used
for the synthesis of nanoparticles. The uniformly nanosized water
droplets of the microemulsions can be used as ‘‘constrained
reactors’’ for tailoring the size of nanoparticles.6{8
The large ionic radius of Ba2þ or La3þ, Csþ is expected to
effectively retard the phase transition of ꢀ to ꢁ-alumina. In this
letter we report the synthesis of the nanostructured alumina doped
with different amounts of Cs, i.e. Cs/Al molar ratio = 0, 1/24,
1/11, 1/9, and 1/6, by reverse microemulsion method. The Cs-
doping effect on the ꢀ to ꢁ-alumina phase transition was studied
by nitrogen adsorption, powder X-ray diffraction (XRD), and
thermogravimetry (TG)/differential thermal analysis (DTA)
techniques.
Table 1. Morphology of Cs-doped aluminas
Sab
a
c
d
Cs/Al
Tc
Dp
Vp
ꢂl/gÀ1
De
molar ratio
ꢁC
m2/g
nm
nm
500
1100
500
1100
500
1100
500
1100
500
263
14
208
89
164
102
122
80
109
55
5.5
mic.f
9.6
15
9.6
12
9.6
9.6
9.6
12
861
13
5.0
22
0
929
706
807
710
699
573
573
475
5.1
6.8
5.0
7.0
5.4
7.8
4.9
7.3
1/24
1/11
1/9
1/6
1100
aTc: Calcination temperature; bSa: BET specific surface area;
cDP: Average pore size; dVp: Pore volume; eD: XRD crystallite
size; fmic.: micropores.
We have already presented elsewhere the typical procedure
for synthesis of barium-doped alumina nanocomposites.9 There-
fore, the synthesis procedure for Cs-doped alumina nanoparticles
will be only briefly described. Appropriate amount of CsNO3
(Wako) was dissolved in 50 ml distilled water, and then this
CsNO3 aqueous solution was emulsified in 140 ml isooctane
(2,2,4-trimethylpentane, Aldrich) with the help of surfactant
(20 ml polyethylene glycol 200, Wako) and cosurfactant (ca.
260 ml n-propanol, Wako) to obtain a stable W/O microemulsion.
1.836 g aluminum isopropoxide (Aldrich) was dissolved in 20 ml
isopropanol (Wako) and 20 ml isooctane by vigorous stirring.
This mixture was heated to 80 ꢁC in order to increase the rate of
alkoxide dissolution. After that, the solution of aluminum
precursor was added to the microemulsion and hydrolyzed for
about 20 h. Subsequently, the mixture was transferred to an
autoclave and hydrothermally treated at 150 ꢁC for 24 h. The Cs-
The pore size distribution (not shown here) of the analyzed
samples, determined by D-H method on desorption branch, was
approximately of Gaussian type. The average values of the pore
size were located in the mesopore region (2–50 nm). After high
temperature calcination (1100 ꢁC for 24 h) the Cs-free alumina
contained only micropores indicating that an extensive sintering
took place. In contrast, the Cs-doped aluminas preserved their
mesoporous structure (average pore size ꢂ12 nm) even after high
temperature calcination. This thermally stable mesoporosity may
have in future interesting applications in catalytic as well as in
other fields.
The crystalline structure of the materials was investigated
with a Rigaku Multiflex diffractometer using Cu K radiation.
ꢁ
Figure 1 shows the XRD patterns of the samples calcined at 500,
1100, and 1330 ꢁC. The crystalline phase of the as-prepared
Copyright Ó 2002 The Chemical Society of Japan
You, Zhixiong
